PDA

View Full Version : Dual Core PowerPC 970s


Pages : [1] 2

MacRumors
Jul 23, 2004, 04:26 PM
ThinkSecret reports (http://www.thinksecret.com/news/antares.html) that a dual-core processor code-named "Antares" is in the works at IBM.

The new processor, officially dubbed the PowerPC 970MP, will provide two interconnected micrpprocessors each with its own 1MB L2 Cache. In comparison, the current 970FX has only 512KB of L2 cache.

Apple is said to have already committed to their use in future Mac products. Timeframe of new products is "sometime in 2005".

fredwick
Jul 23, 2004, 04:28 PM
sounds good, real good :eek:

365
Jul 23, 2004, 04:29 PM
ThinkSecret reports (http://www.thinksecret.com/news/antares.html) that a dual-core processor code-named "Antares" is in the works at IBM.

The new processor, officially dubbed the PowerPC 970MP, will provide two interconnected micrpprocessors each with its own 1MB L2 Cache. In comparison, the current 970FX has only 512KB of L2 cache.

Apple is said to have already committed to their use in future Mac products. Timeframe of new products is "sometime in 2005".

I'll have to wait again... NOT!

MattG
Jul 23, 2004, 04:33 PM
Wow...imagine having a G5 with dual-duals :-)

:cool:

ProfSBrown
Jul 23, 2004, 04:34 PM
It says both cores will share a 1ghz bus...right now each CPU in a dual 2.5 gets 1.25ghz of it's own bandwith, but with that each 3ghz core would have to share only 1ghz, isn't that a step backwards? Or are we getting dual dual-cores? :confused: :(

AmigoMac
Jul 23, 2004, 04:35 PM
How may we start "PB G5 next tuesday" again? :p

Go Apple, Go IBM ... Let's kick some A$$%&@S!

cslewis
Jul 23, 2004, 04:37 PM
Awesome! G6 Powermacs, G5 iMacs and Powerbooks. :p
Hey, now THAT would be a lineup.

hcuar
Jul 23, 2004, 04:39 PM
May I say.... Wow! I'll take 2. I've been a PC user for 17 years. I'm ready to switch. Maybe first with a PB, but I'd have to say a Dual Core - Dual G5 would be very interesting. PC what?

:D

Photorun
Jul 23, 2004, 04:39 PM
Very cool, a two processor G6 with, well, two cores per processor... kinda like having a four processor two processor mean machine. now if only OS X would take advantage of all that 64 bit goodness I'd be giddy as a school girl.

GFLPraxis
Jul 23, 2004, 04:41 PM
I'm imagining dual dual core processor powermacs...I'm also imagining shocked PC fans when a G6 smashes their Alienwares into the ground :)

Grimace
Jul 23, 2004, 04:42 PM
Would this processor be considered a G5 or a G6??

trbeat
Jul 23, 2004, 04:43 PM
I just ordered a new Mac with one of those new fangled chips, my order says on or before July 30th ... :cool:

cslewis
Jul 23, 2004, 04:44 PM
Apple should break with the G-series processors and try something new.
Imagine a PowerMac 970. :p

Maxx Power
Jul 23, 2004, 04:44 PM
I'm steaming in my own shell I am, it's that hot it is. - Zoidberg

So we can have a all in one iMac with a built in toaster ? Or how about a G5 that also provides winter heating ? Then comes the summer....

My next computer, PC and Mac are going to be very efficient on power and MUST have power state transitions so when it is not being used, it doesn't eat up power, and this has to be flexible too, not just TWO states, manually configured is nice like my old P3 laptop, that was sweet. I'm sick of the general population wasting away power like its duty to.

Make an ultra low voltage dual core G5 with p-state transitions and a Mac with a bios that allows shifting of core/bus/voltage with a key combination (with or without a OS), and I'm in for a good deal.

Maxx Power
Jul 23, 2004, 04:46 PM
I'm imagining dual dual core processor powermacs...I'm also imagining shocked PC fans when a G6 smashes their Alienwares into the ground :)

Try playing any latest game and tell PC gamers that... nuff said

kenaustus
Jul 23, 2004, 04:47 PM
When the PM G5 was announced I couldn't help but feel that the Apple-IBM relationship would be one that could really take on the WinTel. It's taken a while to overcome of the challenges in the 90nm fabrication process and engineering (G5 PB) but things are starting to move forward and look very exciting. Apple has definitely become the computer company to watch.

cslewis
Jul 23, 2004, 04:47 PM
I just ordered a new Mac with one of those new fangled chips, my order says on or before July 30th ... :cool:

Yeah, Sure. :rolleyes:

manu chao
Jul 23, 2004, 04:48 PM
Would this processor be considered a G5 or a G6??

I would guess they would only call a Power 5 derived PPC a G6. If these news are correct (or, 'this news is correct'? might a native speaker please tell me what is the correct version), than this rumored Power 5 derived PPC might not see the light before the end of 2005 or even 2006.

Grimace
Jul 23, 2004, 04:51 PM
I would guess they would only call a Power 5 derived PPC a G6. If these news are correct (or, 'this news is correct'? might a native speaker please tell me what is the correct version), than this rumored Power 5 derived PPC might not see the light before the end of 2005 or even 2006.
news is singular, so it is "this" news. thanks for asking!! There are a slew of native speakers that don't put any thought into grammar on these forums. then again, I don't use capital letters correctly either! :p

ddtlm
Jul 23, 2004, 04:52 PM
Excellent, this is what they should be doing. The 1MB caches would be wonderful, 512k is yesterday's high-end, not today's.

Fender2112
Jul 23, 2004, 04:53 PM
I would guess they would only call a Power 5 derived PPC a G6. If these news are correct (or, 'this news is correct'? might a native speaker please tell me what is the correct version), than this rumored Power 5 derived PPC might not see the light before the end of 2005 or even 2006.

a) This is good news - singular
b) These are good news - plural

I think a) is more correct and quite exciting. ;)

MacinDoc
Jul 23, 2004, 04:53 PM
I would guess they would only call a Power 5 derived PPC a G6. If these news are correct (or, 'this news is correct'? might a native speaker please tell me what is the correct version), than this rumored Power 5 derived PPC might not see the light before the end of 2005 or even 2006.
"If this news is correct" would be the proper way to say this in English.
Edit: Sorry, didn't realize CarltonMusic and Fender beat me to it.

musicpyrite
Jul 23, 2004, 04:53 PM
1 MB L2 cache per processor sounds good to me.


Does anybody know if it will have 2 MB L3 cache per probessor?

strider42
Jul 23, 2004, 04:55 PM
Wow...imagine having a G5 with dual-duals :-)

:cool:

Apple may not have dual duals if this is indeed true. They may simply replace the dual's with one of the dual core processors. The advatage for apple is that they can keep the powermac performance significantly above the iMac line still, without the added cost a second separate processor incurs (all the connections for it plus the cost of the actual processor, presumably a single dual core would be cheaper than two single cores). I rather suspect that is what apple will do. They may call it a G6 or whatever just to further differentiate.

hulugu
Jul 23, 2004, 04:56 PM
I would guess they would only call a Power 5 derived PPC a G6. If these news are correct (or, 'this news is correct'? might a native speaker please tell me what is the correct version), than this rumored Power 5 derived PPC might not see the light before the end of 2005 or even 2006.

News is always singular, so it would be "if this news is correct then..."

The reasoning being that news is always a thing that can contain multiple elements, hence the news (current events, sports, weather) on TV, or a newspaper a paper full of news.

blueBomber
Jul 23, 2004, 04:58 PM
this is pretty cool. Someone just asked about this kind of thing in the Doom3 thread, it's sweet that Apple is clawing their way back to the top of the performance heap (like back when the G4 was first released, no pc could touch them)

cslewis
Jul 23, 2004, 04:59 PM
If apple pumps the speed up to 3 or more GHz, PC gamers would have something other than wintel to use. But if apple doesn't ever design macs for gaming they can only realistically gain 30 to 40 percent market share, as 70% :eek: of PCs are sold for GAMING

fluidinclusion
Jul 23, 2004, 05:00 PM
Apple should break with the G-series processors and try something new.
Imagine a PowerMac 970. :p


I agree. the PowerMac 970 would be great. Remember the commercial a year or two ago that had a guy carring a box with a G5 or G6 on it? They were ripping on what Apples current and near future naming scheme is before it even happened. I'd support a new naming system. It's not like you can even figure out what other companies' names for their models even are (e.g. P720GX-421YK Inspirduroncrapon)

A PowerMac 970 seems so much more powerful because of the "970" >> "5" or "6". Marketing 101.

cslewis
Jul 23, 2004, 05:02 PM
I agree. the PowerMac 970 would be great. Remember the commercial a year or two ago that had a guy carring a box with a G5 or G6 on it? They were ripping on what Apples current and near future naming scheme is before it even happened. I'd support a new naming system. It's not like you can even figure out what other companies' names for their models even are (e.g. P720GX-421YK Inspirduroncrapon)

A PowerMac 970 seems so much more powerful because of the "970" >> "5" or "6". Marketing 101.

I'd better torment my PC friend about his Inspirduroncrapon. :D

iris_failsafe
Jul 23, 2004, 05:02 PM
I would like to draw attention to the 4 way smp that Apple has committed to. A 4 processor multi-core machine.

Earth Simulator here we come :cool:

Sun Baked
Jul 23, 2004, 05:02 PM
So what happened to the Power5-UL?

There were some hints that Tiger was updated for SMT (a Power5 feature), with Tiger keeping track of Physical/Logical processors.

mpopkin
Jul 23, 2004, 05:03 PM
your getting the fx, not the mp


I just ordered a new Mac with one of those new fangled chips, my order says on or before July 30th ... :cool:

AoWolf
Jul 23, 2004, 05:06 PM
Cool imagine a duel duel g5 :D

Mord
Jul 23, 2004, 05:07 PM
some guy mentioned L3 cache but that will never happen because with a massive system bus the ram acts as a level 3 cache because ram is now fast enough to outpace L3 cache (when comined with a bus that allows for that speed)

Elan0204
Jul 23, 2004, 05:11 PM
1 MB L2 cache per processor sounds good to me.


Does anybody know if it will have 2 MB L3 cache per probessor?

If you read the full article at ThinkSecret (http://www.thinksecret.com/news/antares.html) it says that the new chip will not support L3 cache.

From the article:
Known officially as the PowerPC 970MP, the chip will feature two interconnected microprocessors on a single 13.225mm x 11.629mm die -- a first for the 970 processor family. Each core will have its own 1MB L2 cache, sources said; the 970FX has only 512KB. L3 cache will not be supported.

Elan0204
Jul 23, 2004, 05:13 PM
I just ordered a new Mac with one of those new fangled chips, my order says on or before July 30th...

2005, maybe.

rcs128
Jul 23, 2004, 05:18 PM
Pontiac is coming out with a new car next year – the replacement for the Grand Am – and it's called the G6. I can't imagine there hasn't already been some copyright issues addressed on this –*Pontiac must be in the clear to use the name –*and I'd be willing to bet Apple has no intention to let the name of their high-end processor be confused with a Pontiac.

cslewis
Jul 23, 2004, 05:20 PM
Pontiac is coming out with a new car next year – the replacement for the Grand Am – and it's called the G6. I can't imagine there hasn't already been some copyright issues addressed on this –*Pontiac must be in the clear to use the name –*and I'd be willing to bet Apple has no intention to let the name of their high-end processor be confused with a Pontiac.

Who would confuse a car with a computer processor? :confused:

People aren't confused about G4TechTV and the apple processor- or at least i'm not, rcs128 :rolleyes:

Sun Baked
Jul 23, 2004, 05:23 PM
Who would confuse a car with a computer processor? :confused:

People aren't confused about G4TechTV and the apple processor. :rolleyes:But you're talking about Pontiac owners here. :p

nuckinfutz
Jul 23, 2004, 05:25 PM
Pontiac is coming out with a new car next year – the replacement for the Grand Am – and it's called the G6. I can't imagine there hasn't already been some copyright issues addressed on this –*Pontiac must be in the clear to use the name –*and I'd be willing to bet Apple has no intention to let the name of their high-end processor be confused with a Pontiac.

Incorrect. You think "Apple" is the only company name out there. Why do you think Apple Records sued Apple Computer? It's because trademarks can be shared as long as the markets are totally different. Apple just had to give up the TM "Rendezvous" not because 30 other people used it but because TIBCO is also involved in computers. The G6 is coming unless another computer company has TM'd "G6"

cslewis
Jul 23, 2004, 05:31 PM
Dell or some other computer company probably TMed it so gullible users would associate "Faster than apple's G5" with "Dell"

applekid
Jul 23, 2004, 05:31 PM
If apple pumps the speed up to 3 or more GHz, PC gamers would have something other than wintel to use. But if apple doesn't ever design macs for gaming they can only realistically gain 30 to 40 percent market share, as 70% :eek: of PCs are sold for GAMING

Think again. A bestselling PC game that tops the sales chart would get about 500,000 to 1 million units sold. That will slowly increase, but let's go with the initial sales. Now, toss that into existing 22 million Mac users, that would give us a 3 to 6% increase. Doubles our population, but it's not like Mac sales are reliant on such games.

70% of PCs sold for gaming? Doubtful. Business buy PCs in bulk, not gamers.

Abstract
Jul 23, 2004, 05:32 PM
"Does Pontiac use an Apple to process all those calculation thingimajiggies!?!?" :eek:

Okay, Apple would probably replace the two chips they are using with this one, dual core chip, but I hope they just use 2 dual core chips. Will software even be able to recognize that many microprocessors? Will you need to get new software to take advantage of two dual core chips?

AndrewMT
Jul 23, 2004, 05:34 PM
By that time, pci-express 16x and and ddr2 memory will actually show some performance improvements over agp 8x and ddr memory, so these technologies better be included in these 2005 G5s.

stcanard
Jul 23, 2004, 05:36 PM
I just ordered a new Mac with one of those new fangled chips, my order says on or before July 30th ...

your getting the fx, not the mp

Hmm, maybe I misread, but I took the original post as a very dry comment on the accuracy of Apple Store ship dates ("I can order non-existent products and it still says ship on or before July 30 ;)")

cslewis
Jul 23, 2004, 05:38 PM
Think again. A bestselling PC game that tops the sales chart would get about 500,000 to 1 million units sold. That will slowly increase, but let's go with the initial sales. Now, toss that into existing 22 million Mac users, that would give us a 3 to 6% increase. Doubles our population, but it's not like Mac sales are reliant on such games.

70% of PCs sold for gaming? Doubtful. Business buy PCs in bulk, not gamers.

Let me rephrase that. 70% of HOME PCs are used to play games more than 3 hrs. per day, according to Janus. So not EXCLUSIVELY for games.

Sun Baked
Jul 23, 2004, 05:39 PM
"Does Pontiac use an Apple to process all those calculation thingimajiggies!?!?" :eek:

Okay, Apple would probably replace the two chips they are using with this one, dual core chip, but I hope they just use 2 dual core chips. Will software even be able to recognize that many microprocessors? Will you need to get new software to take advantage of two dual core chips?Tiger looks like it was set-up for dual Power5-UL processors (ie, keeping track of physical/logical CPUs), so having 4 physical CPUs shouldn't be much different to the OS than the 4 CPUs seen on a dual SMT processor machine -- even though it's really 2 physical and 2 logical making up the 4 CPUs.

ddtlm
Jul 23, 2004, 05:44 PM
Hector:

some guy mentioned L3 cache but that will never happen because with a massive system bus the ram acts as a level 3 cache because ram is now fast enough to outpace L3 cache (when comined with a bus that allows for that speed)
Totally incorrect. You can't look at mhz and conclude that its fast enough, any more than you can look at processor clock speeds. It's latency is not low enough. Sure it can stream a lot of data (bandwidth) but that main RAM has a huge round-trip time (latency). The G4's L3 had lower bandwidth but much lower latency as well.

ddtlm
Jul 23, 2004, 05:49 PM
Sun Baked:

Tiger looks like it was set-up for dual Power5-UL processors (ie, keeping track of physical/logical CPUs), so having 4 physical CPUs shouldn't be much different to the OS than the 4 CPUs seen on a dual SMT processor machine -- even though it's really 2 physical and 2 logical making up the 4 CPUs.
I think this all goes to show that the rumor mill cranks out random crap all the time. Who knows if the next chip is a Power5-lite, or a 970-dual, or something else. Clearly not the rumormongers, they just spew constantly and everyone seems to forget that the story was different last week.

I actually find the dual-core G5 very easy to believe because it would be easy for IBM to design, easy to Apple to use, and sort of easy to produce (though it would be getting a tad large at 90nm). Lots of companies are talking about dual core versions of chips they already have (AMD, Intel, Sun, HP...).

rcs128
Jul 23, 2004, 05:53 PM
Incorrect. You think "Apple" is the only company name out there. Why do you think Apple Records sued Apple Computer? It's because trademarks can be shared as long as the markets are totally different. Apple just had to give up the TM "Rendezvous" not because 30 other people used it but because TIBCO is also involved in computers. The G6 is coming unless another computer company has TM'd "G6"

I don't think *most* people would confuse a chip with a car, but the Pontiac G6 is going to be the carmaker's bread-and-butter vehicle –*and I'm sure they'll throw millions in dollars of advertising to promote it. So shudder to think, when the masses hear G6, they're going to think "Pontiac". :)

cslewis
Jul 23, 2004, 05:54 PM
If PowerMacs become G6, it should follow that iMacs, Powerbooks and current PowerMac G4's will be G5. For eMacs and iBooks pump up the G4's but don't loose too much sleep. They are low end, after all.

Sun Baked
Jul 23, 2004, 05:56 PM
Sun Baked:

I think this all goes to show that the rumor mill cranks out random crap all the time. Who knows if the next chip is a Power5-lite, or a 970-dual, or something else. Clearly not the rumormongers, they just spew constantly and everyone seems to forget that the story was different last week.

I actually find the dual-core G5 very easy to believe because it would be easy for IBM to design, easy to Apple to use, and sort of easy to produce (though it would be getting a tad large at 90nm). Lots of companies are talking about dual core versions of chips they already have (AMD, Intel, Sun, HP...).Problem is it's easy to confirm or kill this rumor by doing sysctl -a in 10.4 and seeing if it turns up two values, specifically the number of physical and logical processors.

This is along the lines of seeing the PowerMac 7,3 and PowerMac 8,1 in the last OS update.

ddtlm
Jul 23, 2004, 05:59 PM
Sun Baked:

Problem is it's easy to confirm or kill this rumor by doing sysctl -a in 10.4 and seeing if it turns up two values, specifically the number of physical and logical processors.
Oh that is interesting. Then perhaps this dual-core 970 thing is nonsense, cause Apple's not gona get both. Neither IBM nor Apple are into throwing away their money. ;)

nuckinfutz
Jul 23, 2004, 06:02 PM
I happen to think both rumors are fine. I do think Apple will be utilizing SMT in a future Powermac that ships during Tiger's reign. I also believe dual core is the way as well. The two aren't mutually exclusive.

Tiger is enhanced to support SMP systems better. Whether that comes from SMT enabled CPU or dual core is irrelevant.

applekid
Jul 23, 2004, 06:10 PM
Calling it a G5 or G6 is all a marketing ploy. Jumping up to a new generation could backfire on Apple or work to its advantage. I don't know which one, but I rather see it stay G5. Jumping up a generation after only a year or so makes it sound like the G5 was a mistake. Make it a G6 and it's only a gimmick.

Does dual core really mean it's exactly like two processors on one chip? I haven't see any of our tech experts correct us, yet. Just checking.

supergod
Jul 23, 2004, 06:11 PM
You're all just wildly guessing on this one guys! Clearly, the only possible solution is to skip G6 and go straight to a four processor dual core G7. That would show those wintellers! :eek:

Bear
Jul 23, 2004, 06:13 PM
Cool imagine a duel duel g5 :DAhhh a real fighting machine. I myself would prefer a dual dual G5.

Bear
Jul 23, 2004, 06:14 PM
I would like to draw attention to the 4 way smp that Apple has committed to. A 4 processor multi-core machine.

Earth Simulator here we come :cool:Actually 2 dual core processors would require the 4 way smp code.

Remember each core looks like a processor to the OS.

Sun Baked
Jul 23, 2004, 06:15 PM
Remember the XBox2 was supposed to be using a PPC970MP variant... ;)

For those that remember the 3 core PPC970 block diagram they tossed out for the XBox2 leak.

Dr. Dastardly
Jul 23, 2004, 06:15 PM
I don't know about everyone else but I am just stoked to get some fresh rumors through here. Seems like its been a real dry spell lately.

And no colored ipods are not considered a good rumor. :p

macnews
Jul 23, 2004, 06:18 PM
A PowerMac 970 seems so much more powerful because of the "970" >> "5" or "6". Marketing 101.

That is Dilbert type of marketing - in other words, the bad kind going with 970 because "it seems more powerful."

I think the public is feed up with the numbers game. In my area local tv stations got in to it just before the year 2000 when they all upgraded their weather system computers. Channel 2 decided to name their's "weather center 2000" - very short sighted given 2000 was only a year a way. Channel 7 thought tried to one up, falling short several times until they really out did channel two by creating "storm tracker 7000." Come on, where does this hype/c*@p end?

I like the G4, G5, G6 naming. Very elegant, simple, easy to understand (a G5 is better than a G4, etc) and catchy to say/think. Think about it, what do you do when 990 doesn't "sound powerful enough?" Go with the powermac 1000? Then how does that compare to the Dell Dimension 2400? What about the PowerMac 9600, is the 9600 a better machine? No thank you, stick with the G series.

I do hope Apple doesn't remove the dual processor. I would prefer the dual-dual option. Talk about super computer options!

~Shard~
Jul 23, 2004, 06:20 PM
Even if Apple implements this chip "sometime in 2005" I doubt they will brand it as a G6. The G5 hasn't been out for that long relatively, even when next year rolls around, and I don't think a processor change like this would really warrant a name change like that. Save the "G6" moniker for something truly big - maybe "The G6 in '06"... :cool:

w00tmaster
Jul 23, 2004, 06:21 PM
Hector:


Totally incorrect. You can't look at mhz and conclude that its fast enough, any more than you can look at processor clock speeds. It's latency is not low enough. Sure it can stream a lot of data (bandwidth) but that main RAM has a huge round-trip time (latency). The G4's L3 had lower bandwidth but much lower latency as well.

And that is exactly why IBM needs to steal an idea from AMD and put the memory controller on chip. That lowers latency by a fair amount from what I have read.

ddtlm
Jul 23, 2004, 06:22 PM
nuckinfutz:

Tiger is enhanced to support SMP systems better. Whether that comes from SMT enabled CPU or dual core is irrelevant.
Lemme give you a secret to estimating the feasibility of a rumor relating to a large R&D investment: you need to establish that the people spending on R&D are going to recoup that cost. For years now I've used that simple rule to correctly predict the non-arrival of crasy things that everyone else seemed to swallow hook, line and sinker. Apply that rule to the idea of there being both a 970-dual and a Power5-lite. When I apply it, I see that the two chips aim at the same market, harming profit. Note, neither chip is aimed at the market that will be increasingly held by the 970fx, so I see no conflict there.

It is possible that a G5-dual could occupy the market for a year or more before being replaced by a Power5-lite. There are some questions there, I'd wonder if going from true dual core to SMT would be a step backwards, and also I wonder why Apple would be designing for SMT already.

Dr. Dastardly
Jul 23, 2004, 06:24 PM
If this new technology is implemented in new powermacs... in say... WWDC '05 then the G5 would have been around for two years. I'm not saying that it will be a G6 by next June, but the G5 won't still be the new kid on the block like it is now.

ddtlm
Jul 23, 2004, 06:28 PM
w00tmaster:

And that is exactly why IBM needs to steal an idea from AMD and put the memory controller on chip.
Mmm, on-die memory controllers. Drool. But anyway Sun, at least, had it before AMD did... for example on the crappy US2i chips. ;)

FlatMac
Jul 23, 2004, 06:29 PM
I think this dual core technology is the new trend. Intel is using it already (what they call HyperThreading) and I just read an article in Dr Dobbs (August) explaining that Intel is struggling with the old Pentium arch to win Mhz and the only think that saves them is the dual core tech they have.

This confirms what S. Jobs explained when he spoke about not reaching 3Ghz for the G5. The CPU frequency is harder and harder to improve, if the processor gurus don't find a solution to this problem, I'm sure multi CPU (logical/physical whatever) will be the future of personal computers (not only for server / pro-desktops)

That's why I bet on the dual core dual CPU G5 (with dual dual link and two 30in displays please, some kind of a dual dual dual dual ...) :)

As for the Gx name, I kind of like it, it's simple and says what it needs to: xth Generation PowerMac

Are we really going to drop this megahertz myth now ?

ddtlm
Jul 23, 2004, 06:32 PM
FlatMac:

I think this dual core technology is the new trend. Intel is using it already (what they call HyperThreading) and I just read an article in Dr Dobbs (August) explaining that Intel is struggling with the old Pentium arch to win Mhz and the only think that saves them is the dual core tech they have.
Actually Hyperthreading ("SMT") is not the same thing as dual cores. SMT is one processor pretending to be more than one, dual cores is two genuine processors jammed into one package.

Intel has spoken about dual-core Prescotts... dual SMT cores, kinda like a Power5 in that regard.

nuckinfutz
Jul 23, 2004, 06:42 PM
It is possible that a G5-dual could occupy the market for a year or more before being replaced by a Power5-lite. There are some questions there, I'd wonder if going from true dual core to SMT would be a step backwards, and also I wonder why Apple would be designing for SMT already.

I agree. If anything the 970MP pretty much shelves the idea of a P5 Lite for at least a year. I'm thinking Apple ships Powermacs in 2005 with single socket 970MP systems. This cuts down on motherboard complexity and obviates the need for dual FSB so the memory controller gets simplified. All these "efficiencies" go right to the bottom line and Apple keeps this up for next 14 months with one refresh in between.

That gives them the ability to really hype the P5 Lite up the right way with ondie memory controller and SMT along with dual cores. I figure 14 months from now IBM and Intel will be talking about moving to 65nm which makes sense for the P5 Lite.

FlatMac
Jul 23, 2004, 06:45 PM
Actually Hyperthreading ("SMT") is not the same thing as dual cores. SMT is one processor pretending to be more than one, dual cores is two genuine processors jammed into one package.

Intel has spoken about dual-core Prescotts... dual SMT cores, kinda like a Power5 in that regard.

Yeah thanks, my mistake.

In fact Intel is also scheduling dual core for 2005 : http://www.xbitlabs.com/news/cpu/display/20040610151158.html

Another question: what is the pentium M exaclty? because they have lower frequency than P4, I think they are closer to the G4 than the old P4

In the DDJ article I mentioned earlier, they said current apps run faster on 64 bits processor, I don't see why but that's good news for you lucky G5 owners

:cool:

Fitzcaraldo
Jul 23, 2004, 06:47 PM
1 MB L2 cache per processor sounds good to me.


Does anybody know if it will have 2 MB L3 cache per probessor?

Hmmm. I was reading as this broke Last Sunday, and No L3 Cache was suported.

ddtlm
Jul 23, 2004, 06:56 PM
FlatMac:

Another question: what is the pentium M exaclty? because they have lower frequency than P4, I think they are closer to the G4 than the old P4
The P-M is something of an enhanced P-3 that is designed for laptops.

In the DDJ article I mentioned earlier, they said current apps run faster on 64 bits processor, I don't see why but that's good news for you lucky G5 owners
When AMD went to 64-bit they took the time to fix some of x86's problems and so they gained performance, however since PPC didn't have those issues, there really isn't an automatic gain. Some PPC programs can benefit if redesigned for 64-bit but generally its not a real big deal.

JohnnyFirpo
Jul 23, 2004, 07:08 PM
1 MB L2 cache per processor sounds good to me.


According to the original article, it says each core has its own 1MB L2 cache. It means 2MB per die, 4 MB per machine, if they will be "dual duals", if I got it right. What a dream...

rdowns
Jul 23, 2004, 07:13 PM
Even if Apple implements this chip "sometime in 2005" I doubt they will brand it as a G6. The G5 hasn't been out for that long relatively, even when next year rolls around, and I don't think a processor change like this would really warrant a name change like that. Save the "G6" moniker for something truly big - maybe "The G6 in '06"... :cool:

With IBM seeming to be plowing forward quite rapidly (recent problems excepted), why not change the name? This isn't Motorola we're talking about here.

jettredmont
Jul 23, 2004, 07:24 PM
But if apple doesn't ever design macs for gaming they can only realistically gain 30 to 40 percent market share, as 70% :eek: of PCs are sold for GAMING

And this BS statistic came from where?

In my experience, while indeed it is likely that 70% of all home PCs are used for (among other things) games, it is incorrect to say that 70% of all PCs are sold for gaming. You miss:

1) Corporate PCs. I can guarantee that these aren't bought for games, and indeed the fact that they can be used to play games is often a negative factor (corps tend to opt for smaller disks and less-powerful video cards for just this reason, even if said options actually cost *more* than the stock machine would).

2) While 70% of home PCs are used to play games on occasion, a good fraction of those are "puzzle" games (the dominant genre of computer games amongst women), "strategy" games, or "popular" games. Puzzle games exist in droves for the Mac; most strategy games are ported over fairly quickly, and "popular" games are tending more towards near-simultaneous release. The ones that never get the Mac lovin' are the second-string FPS games, and the knockoffs of previous hits. Granted, there are notable exceptions, but the trend is fairly clear.

3) While 70% of all PCs might play games, it is a purchase-determining factor in only a small percentage of those. Face it: a $200 Playstation/GameCube/XBox investment will tend to provide a much better gaming experience for most people out there than the extra $500 you spend on a PC to make it the biggest baddest gaming albatross on the street.

Gamers are a vocal and young bunch, which tends to keep this "games are the lifeblood of any computer" meme alive and kicking despite complete lack of reason or evidence in support. That doesn't make it true.

hitchhiker
Jul 23, 2004, 07:25 PM
The G-Series naming sytem is definitely getting too old. Sure, it makes some sense. But I think after the G6, apple should look for some other way to name their subsequent chip and powermac designs. Perhaps changing the first letter would help- "PowerMac X7" sounds much more advanced and interesting than staying with the G system for years to come.

Dave00
Jul 23, 2004, 07:26 PM
Maybe they'll just keep the name Antares. Had to refresh my memory, I knew I remembered the name from sometime years ago when I used to know my astronomy. From Dictionary.com:

"A giant red binary star, the brightest in the constellation Scorpio, about 424 light-years from Earth."

--Dave

cslewis
Jul 23, 2004, 07:28 PM
And this BS statistic came from where?

In my experience, while indeed it is likely that 70% of all home PCs are used for (among other things) games, it is incorrect to say that 70% of all PCs are sold for gaming. You miss:

1) Corporate PCs. I can guarantee that these aren't bought for games, and indeed the fact that they can be used to play games is often a negative factor (corps tend to opt for smaller disks and less-powerful video cards for just this reason, even if said options actually cost *more* than the stock machine would).

2) While 70% of home PCs are used to play games on occasion, a good fraction of those are "puzzle" games (the dominant genre of computer games amongst women), "strategy" games, or "popular" games. Puzzle games exist in droves for the Mac; most strategy games are ported over fairly quickly, and "popular" games are tending more towards near-simultaneous release. The ones that never get the Mac lovin' are the second-string FPS games, and the knockoffs of previous hits. Granted, there are notable exceptions, but the trend is fairly clear.

3) While 70% of all PCs might play games, it is a purchase-determining factor in only a small percentage of those. Face it: a $200 Playstation/GameCube/XBox investment will tend to provide a much better gaming experience for most people out there than the extra $500 you spend on a PC to make it the biggest baddest gaming albatross on the street.

Gamers are a vocal and young bunch, which tends to keep this "games are the lifeblood of any computer" meme alive and kicking despite complete lack of reason or evidence in support. That doesn't make it true.

Didn't you read the response to the post on the third page? 70% OF HOME COMPUTERS ARE USED TO PLAY GAMES FOR MORE THAN THREE HOURS PER DAY.

nuckinfutz
Jul 23, 2004, 07:29 PM
Maybe they'll just keep the name Antares. Had to refresh my memory, I knew I remembered the name from sometime years ago when I used to know my astronomy. From Dictionary.com:

"A giant red binary star, the brightest in the constellation Scorpio, about 424 light-years from Earth."

--Dave


They can't www.antarestech.com audio company but close enough to computers to cause problems with that name.

cslewis
Jul 23, 2004, 07:32 PM
They can't www.antares.com audio company but close enough to computers to cause problems with that name.

When I followed your link, I was informed that the page did not exist... perhaps apple can use the name after all :D

afields
Jul 23, 2004, 07:49 PM
Will there be a g5 update between now and when the dual core is released? Hmmm. I wonder if this an off shoot of that crazy "quad" rumor from macosrumors.com

cslewis
Jul 23, 2004, 07:49 PM
Maybe they'll just keep the name Antares. Had to refresh my memory, I knew I remembered the name from sometime years ago when I used to know my astronomy. From Dictionary.com:

"A giant red binary star, the brightest in the constellation Scorpio, about 424 light-years from Earth."

--Dave

Sounds cool :p

Kelvin
Jul 23, 2004, 07:51 PM
The most obvious use for a dual-core (single FSB) 970 would be the powerbook. Less power consumption than 2 FSBs, smaller package to cram into the thin enclosure. Easier cooling since you only have to pump heat off of one package. I know powerbook g5 rumors are really old by now, but seriously, it makes sense logically. I'm willing to bet that a 970MP has lower power consumption also simply because IBM has a few chips behind their belt at this point.

Sun Baked
Jul 23, 2004, 07:59 PM
Will there be a g5 update between now and when the dual core is released? Hmmm. I wonder if this an off shoot of that crazy "quad" rumor from macosrumors.comProbably... or a new source/angle on it.

Can't count out the XBox either.

Though IBM finally did publish some Maximum Power Dissipation specs for the original PPC970 -- done on a rev A DP 2.0GHz PowerMac G5. ;)

Of course these numbers back up IBMs claim that it dissipates less Watts than Apple says it does, and some have pointed out that IBM did there test using an Integer spec suite -- neglecting the Altivec and FP units, which could boost the power dissipation into the range Apple was stating (for the FX).

P-Worm
Jul 23, 2004, 08:07 PM
Didn't you read the response to the post on the third page? 70% OF HOME COMPUTERS ARE USED TO PLAY GAMES FOR MORE THAN THREE HOURS PER DAY.

Just because 70% of computers play games, does not mean that they were bought for the sole purpose of playing games. I know mine wasn't, but I still play games on it.

P-Worm

jeffbax
Jul 23, 2004, 08:09 PM
Ugh...

Game consoles do not provide a better gaming experience than a PC dammit!

Its the worst Mac cop-out ever!

Games are what keep me locked to PC, thats for sure.

One day my precious... (http://www.apple.com/powerbook/)

Fender2112
Jul 23, 2004, 08:34 PM
I happen to think both rumors are fine. I do think Apple will be utilizing SMT in a future Powermac that ships during Tiger's reign. I also believe dual core is the way as well. The two aren't mutually exclusive.

Tiger is enhanced to support SMP systems better. Whether that comes from SMT enabled CPU or dual core is irrelevant.

Let's not forget about the Xserve. Perhaps Apple is laying the ground work multicore servers. Apples seems to really want to get a foothold in the server market.

Sun Baked
Jul 23, 2004, 08:37 PM
Here was the original thread on this issue...

http://forums.applenova.com/showthread.php?t=999

Though I never saw the proof that was shown, possible somebody grabbed a copy. :confused:

strider42
Jul 23, 2004, 08:42 PM
The most obvious use for a dual-core (single FSB) 970 would be the powerbook. Less power consumption than 2 FSBs, smaller package to cram into the thin enclosure. Easier cooling since you only have to pump heat off of one package. I know powerbook g5 rumors are really old by now, but seriously, it makes sense logically. I'm willing to bet that a 970MP has lower power consumption also simply because IBM has a few chips behind their belt at this point.

Why would a dual core processor make more sense for a powerbook than a single core processor like we have now? It seems much more likely to me that a single core processor will end up in the powerbook first, even if this dual core rumor pans out to be true and available by the time a G5 powerbook shows up.

jettredmont
Jul 23, 2004, 08:45 PM
Didn't you read the response to the post on the third page? 70% OF HOME COMPUTERS ARE USED TO PLAY GAMES FOR MORE THAN THREE HOURS PER DAY.

Correct, I missed your later clarification on home vs business PCs. My apologies. Now please stop shouting.

Please provide a link to that Janus survey. I can't find it anywhere. In the meantime, a few more notes on surveys:

1) Just because some organization comes up with a statistic doesn't mean it is valid.

2) Anecdotal evidence isn't much better, but my experience completely differs from "70% of home computers playing games for more than three hours per day". I know of no one who is not currently in college on their parents' dime who has this kind of time or who would spend it playing computer games. Frankly, as I said before, the folks I know who play video games of any sort for more than a couple of hours per week invested in dedicated gaming hardware. The only cases I can see where a single computer gets that kind of gaming use is when it is used by several children in the family for games (in which case the assertion that games drive computer sales for 70% of the market is obviously incorrect).

3) Battling surveys. Here's a Pew survey which found, among other things, "Seventy percent (70%) of college students reported playing video, computer or online games at least once in a while." If only 70% report of college students report playing games once in a while, even if you were to stretch "once in a while" out to 3+ hours per day because said college students wanted to appear more studious on their anonymous surveys, that leaves the rest of the computer-owning world at a much lower game-play rate; it leaves it mathematically unlikely for the overall average to come out with 70% of the computers getting 3 hours of Doom3-loving per night. http://www.pewinternet.org/PPF/r/93/report_display.asp

4) Battling surveys 2. This one's much older, but from the Interactive Digital Software Association, which would tend to skew the results towards more than actual usage. http://www.idsa.com/consumersurvey2001.html Again, this is from 2001, but at that time the IDSA found that amongst households with computers and consoles, "the typical family uses its console or computer to play games for an average of 10 to 11 hours per week, up about one hour over last year’s figures." This shows a much lower rate than 3-hours-per-day, and also notes that on average the gaming is done by 4.5 individuals (two console players and 2.5 computer players). Assuming 1.5 computers/consoles per household (a conservative estimate), and half of the reported time being spent on computers instead of consoles, that leaves about four hours per week per computer; remember that that's not calculated in the study, so it's not necessarily accurate, but it gives an idea of how those number compare to your purported numbers. Even leaving that entire 11 hours per week on computers and assuming that this is one computer in the house where the gaming is going on, three years later you're claiming 70% of computers spend more than 21 hours per week on games, which would put the "average" at a significantly higher number. Where did the typical family pull an extra 10 hours per week of game playing from?

Again, please provide a link if you have one. That statistic just doesn't look anywhere near right. Either their methodology was screwed up, or you're reading their results wrong.

joeboy_45101
Jul 23, 2004, 09:02 PM
If this this IS true, then HOLY MOSES ON A CRUTCH! But then again I don't trust ThinkSecret. Look at the July 14th headline, it gets the new iPods totally wrong. And I think they were also saying that the new iPods wouldn't be coming until August or September.

A rumor is a rumor, until it hits the stores. :D

jouster
Jul 23, 2004, 09:02 PM
news is singular, so it is "this" news....There are a slew of native speakers

slew is singular, too......so that would be "There is a slew...."

If you're gonna correct people........

:D

Grimace
Jul 23, 2004, 09:23 PM
slew is singular, too......so that would be "There is a slew...."

If you're gonna correct people........

:D

Well done!! That is one that has almost become vernacular in both forms - but you are very right! Thanks for keeping me honest! :p

another common one is "there are a group of people outside"

wdlove
Jul 23, 2004, 09:28 PM
This new rumor seems to be real exciting. I would like to see the G6 name used. It would also seem that is the same recommendation occurs, then would have to wait for Rev. B. That would be really tough to wait for another year.

hassanpr
Jul 23, 2004, 09:31 PM
they cant even ship my dual 2.5. It had a ship date of july 30th and just a couple of days ago it was pushed august 9th. What the hell is going on.

joeboy_45101
Jul 23, 2004, 09:35 PM
they cant even ship my dual 2.5. It had a ship date of july 30th and just a couple of days ago it was pushed august 9th. What the hell is going on.

Apple gives academic institutions more precedence then regular customers. This is nothing new. That's probably why they moved the the ship date back, they've been slammed with orders from schools and universities.

sockeatingdryer
Jul 23, 2004, 10:14 PM
Even if Apple implements this chip "sometime in 2005" I doubt they will brand it as a G6. The G5 hasn't been out for that long relatively, even when next year rolls around, and I don't think a processor change like this would really warrant a name change like that. Save the "G6" moniker for something truly big - maybe "The G6 in '06"... :cool:

Pretty soon, around 2005, it will have been around about as long as the G4.. and remember, that G5 has had production problems. So if we hadn;t had any problems, it would have gone through many different revisions. The G6 would be a great thing to roll out next year... paving the way for all of the Mac lineup to become 64-bit, and not having any more G4. They should be scrapping the G3 completely by now, and they can't do that until they drop the G4 totally as well, due to bit limitations.

paxtonandrew
Jul 23, 2004, 10:22 PM
It seem we are only talking about processor power consumption, and inclusions like the cache. Did anyone stop to think about the speeds of the processors? With the altivec engine, and Single core, this WILL be an amazingly fast computer. When this comes out, I may have to sell my PowerBook, and look at a Dual Core, Dual Processor G6(If Pontiac doesn't copyright it) we may well see it come to fruition.

Tilan
Jul 23, 2004, 10:28 PM
Maybe they'll call it the G5.2.

It would make sense since it is still a 970, dual core, and the third version of the chip (the original and then the 90nm version).

hassanpr
Jul 23, 2004, 10:42 PM
Apple gives academic institutions more precedence then regular customers. This is nothing new. That's probably why they moved the the ship date back, they've been slammed with orders from schools and universities.

I did order it through my job its the Board Of Education In my city so whats there excuse.

Sun Baked
Jul 23, 2004, 10:47 PM
I did order it through my job its the Board Of Education In my city so whats there excuse.The same as always, constrained supply of 90nm product out of the FishKill plant. :(

Processors for the 2.5GHz PowerMac, XServe G5, and the coming iMac G5 in short supply -- no telling if the forthcoming U3-Lite (iMac G5 Memory Controller) is 90nm or 130nm.

phillymjs
Jul 23, 2004, 10:55 PM
For about a month or two, could you maybe shift focus from developing future products and see if you can actually ship some of your current products?

Just asking.

~Philly

dizastor
Jul 23, 2004, 11:11 PM
two words:
Saaaa Weeeeet!

maxvamp
Jul 23, 2004, 11:59 PM
IBM Roadmap (http://www.appleinsider.com/article.php?id=516))


We know that IBM is working on an SMT enabled dual core Power Proc.

We know that the 970 is a derived Power Core.

We know that Apple has an interest in expanding it's business unit with machines like the XServe.

We know that the trend is towards dual core.

We know that the OSX.4 kernel is smp/smt aware making any threaded app ( multiple apps ) perform better on this ( rumored ) machine.

We know that AMD is positioning a consumer Athlon64 that is single core ( 754 ), and a professional / industrial line that can be dual core ( 939 / 940 )


I suspect that there will be a 97x machine with these spec PLUS smt. I suspect that while the single core 970 will be relegated to the iMac / PowerBook (?), and the iBook may stay with the G4 for another year or two, we very well may see the PowerMac and XServe as a single or dual processor dual core.

This is just speculation based on IBM's ability and roadmap, and the AMD / Intel's plans for 2005 / 2006.

As someone said before, it should get interesting to see how these factors affect the importance of GHz ( speed ).

Anyone ready for a 4/8 processor machine?

Max.

BTW: For a wish list, I would love to see IBM integrate the memory controller, but I see no evidence that is going to happen.

Max.

alexrd
Jul 24, 2004, 12:01 AM
I can't believe no one has pointed out that each of the two cores (presumably*) has SMT, so a single chip is 4 logical processors. The proposed 4-way SMP system would appear to the OS as 16 processors (4 chips X 2 cores/chip X 2 Logical procs/core).

Also, there seems to be a sentiment here that Tiger will only be able to handle four processors (logical or otherwise). Where does this perception come from? Is it mentioned somewhere in Apple's Tiger pages that I missed? If not, seeing as Apple has gone to great pains to make OS X multithreaded and multi processor aware from the ground up, I can't think of a reason why Tiger shouldn't be able to scale up to supporting 16 logical processors.

Damn. That's one hell of an Powermac. :D

Does anyone know if OS X has any ugly SMP kludges left that would prevent that kind of scaling? I seem to recall that at some point I/O was locked to one processor, which would decrease the efficiency of scaling above two processors, but I may be wrong or it may be fixed.

*Since each of the two cores is based on the POWER5, which supports SMT. One might argue that it's not worth having the transistors for all that sophisticated branch prediction when you're cramming two cores on one die, but the purpose of SMT is to keep the pipeline[s] full at all times, an idea which should work as well for multi- as single-cores. In any case, that's how multicore POWER5s work (see here (http://www.eetimes.com/story/OEG20030228S0017)) so it's reasonable to assume that the "G6" will work this way as well.

[edit: AARG!! The dreaded simultaneous post! :rolleyes: ]

Sun Baked
Jul 24, 2004, 12:10 AM
I can't believe no one has pointed out that each of the two cores (presumably*) has SMT, so a single chip is 4 logical processors. The proposed 4-way SMP system would appear to the OS as 16 processors (4 chips X 2 cores/chip X 2 Logical procs/core).It's a dual GP-UL (aka, Power4-UL) -- not the GR-UL (aka, Power5-UL).

If you look at the link I provided... http://forums.applenova.com/showthread.php?t=999

hmurchison, GR is a different story...

Antares is Power4 based. There was a time when a dual-core GPUL was the only plan (as mentioned on the NMR linked above), although that changed for various reasons...

As for planned dates, Protos to Apple Aug/2004 and Oct/2004, Prod Jan/2005... as for Apple's plan, that I do not know...

Is it possible to post attachments here?

maxvamp
Jul 24, 2004, 12:13 AM
Isn't OSX based on the MACH kernel?

If I remember history, the MACH kernel could scale to 256 processors ( based on OS/2 PPC ).

If this is true, the OSX kernel should scale very well.

One other thing to note from the slide in my previous post is that the POWER5/6 processor could partition into, not the two processors of the P4, but 64 virtual processors.

For the consumer, not a big deal, but for servers, such as web and Database, this is a very big deal. I doubt that most users / prosumers would benefit from more than 4 -> 6 processors real or virtual.

Just my opinion.

Max.

DMann
Jul 24, 2004, 12:22 AM
Would this processor be considered a G5 or a G6??

Although the chip is still of the 970 generation,
one might imagine that it would retain it's G5
status, but if the performance is this dramatically
improved, they'll likely call it a G6.

DMann
Jul 24, 2004, 12:25 AM
The same as always, constrained supply of 90nm product out of the FishKill plant. :(

Processors for the 2.5GHz PowerMac, XServe G5, and the coming iMac G5 in short supply -- no telling if the forthcoming U3-Lite (iMac G5 Memory Controller) is 90nm or 130nm.

Also consider supply and DEMAND for those little demons :cool:

ddtlm
Jul 24, 2004, 12:26 AM
alexrd:

I can't believe no one has pointed out that each of the two cores (presumably*) has SMT, so a single chip is 4 logical processors. The proposed 4-way SMP system would appear to the OS as 16 processors (4 chips X 2 cores/chip X 2 Logical procs/core).
Hold on there, Apple is not going to suddenly jump to that many threads. Such a system would be very expensive and noone would have anything to use it (other than as a server), so it wouldn't sell. I could see 4 logical processors, and no more.

ddtlm
Jul 24, 2004, 12:28 AM
maxvamp:

If this is true, the OSX kernel should scale very well.
There is a difference between being able to run 256 CPUs, and being able to run 256 CPUs well. That many processors represents a non-trivial challenge, to say the least.

alexrd
Jul 24, 2004, 12:35 AM
It's a dual GP-UL (aka, Power4-UL) -- not the GR-UL (aka, Power5-UL).

If you look at the link I provided... http://forums.applenova.com/showthread.php?t=999


Ah. So much for that, then.

And upon further consideration, it certainly makes sense that the 970MP would be based on the Power4-UL.

At any rate, I would put my money on 2-chip, 4-way systems from apple (rather than saying "well, each chip is two procs, so we'll just use one"). The whole point of mutlicore design is to take advantage of the extra transistors that you can cram in due to process improvements (SOI, 90nm, etc...) The chip should be price competitive with contemporary, state of the art single-core chips; it's not like a two-core chip is twice as big, you're just utilizing your transistor count in a different way.

-alex

ibjoshua
Jul 24, 2004, 12:36 AM
A binary star?
Nice choice of name.

i_b_joshua

alexrd
Jul 24, 2004, 12:43 AM
Hold on there, Apple is not going to suddenly jump to that many threads. Such a system would be very expensive and noone would have anything to use it (other than as a server), so it wouldn't sell. I could see 4 logical processors, and no more.

Why? What's inherently expensive about jumping to so many threads? If the cost per chip is comparable, and the multithreading is already in the OS, there's no reason at all for Apple not to be able scale to 4, 8, 16 or more logical processors.

It's not like there's a Thread Tax.

Besides, the system on your desk already runs a lot more threads than it has processors, each application you run is spawning one or more threads. Chances are, there are more than 16 threads running on your system right now. Once the computation tasks are already broken down into small units (threads) by the OS, parcelling them out to many processors is easy.

-alex

~Shard~
Jul 24, 2004, 12:50 AM
Pretty soon, around 2005, it will have been around about as long as the G4.. and remember, that G5 has had production problems. So if we hadn;t had any problems, it would have gone through many different revisions. The G6 would be a great thing to roll out next year... paving the way for all of the Mac lineup to become 64-bit, and not having any more G4. They should be scrapping the G3 completely by now, and they can't do that until they drop the G4 totally as well, due to bit limitations.

That's fair enough - I guess in my mind it will come down to more of a Marketing decision on branding than anything...

Fukui
Jul 24, 2004, 01:03 AM
Why? What's inherently expensive about jumping to so many threads? If the cost per chip is comparable, and the multithreading is already in the OS, there's no reason at all for Apple not to be able scale to 4, 8, 16 or more logical processors.

It's not like there's a Thread Tax.
-alex
Well, to a certain extent, there can.

ddtlm
Jul 24, 2004, 01:25 AM
alexrd:

Why? What's inherently expensive about jumping to so many threads?
The 4 dual-cored SMT chips.

If the cost per chip is comparable, and the multithreading is already in the OS, there's no reason at all for Apple not to be able scale to 4, 8, 16 or more logical processors.
Yeah and while we're ignoring costs, lets have Apple put Power5's in there too. It won't actually help anyone get work done faster, but the number of threads alone is bound to cause the surrender of Intel. Heh. ;)

afields
Jul 24, 2004, 01:29 AM
For about a month or two, could you maybe shift focus from developing future products and see if you can actually ship some of your current products?

Just asking.

~Philly

*sustained applause*

NSiebenmor
Jul 24, 2004, 01:31 AM
I'm about to order a 2.5 ghz G5. Should I change my mind and wait? I'm tired of waiting and my student discount ends this summer!

Trowaman
Jul 24, 2004, 01:41 AM
I'm about to order a 2.5 ghz G5. Should I change my mind and wait? I'm tired of waiting and my student discount ends this summer!

Personnally, i'd go with yes, buy now. I wouldn;t expect these dual cores until next WWDC (with 3.0 G5 at MWSF or before)

Any Powerbook theories? I'm placing my money on first 1/2 of next year but beyond that . . .

alexrd
Jul 24, 2004, 01:48 AM
The 4 dual-cored SMT chips.


Yeah and while we're ignoring costs, lets have Apple put Power5's in there too. It won't actually help anyone get work done faster, but the number of threads alone is bound to cause the surrender of Intel. Heh. ;)

:) :) :)

Hey, it's all rumors anyway, I can dream, can't I?

Besides, even the Power5 quakes in sight of the multi threaded insanity that is Sun's (http://www.sun.com/processors/throughput/) Niagara (http://www.geek.com/procspec/sun/niagara.htm). ;)

On a somewhat related topic, in the environment I support (graphics and operations for a cable news channel) we would certainly have use for four-way, multi-core, SMT, Power5 based XServes. A few of those would make heck of a render farm, and with the distributed computing API goodness that is XGrid, many other processor intensive tasks should be amenable to running on some sort of back-end cluster, relieving the pressure for insane power at each workstation. So saying that nobody has any use for that sort of power isn't really correct (not that you said that, ddtlm).

-alex.

Mord
Jul 24, 2004, 02:13 AM
Hector:


Totally incorrect. You can't look at mhz and conclude that its fast enough, any more than you can look at processor clock speeds. It's latency is not low enough. Sure it can stream a lot of data (bandwidth) but that main RAM has a huge round-trip time (latency). The G4's L3 had lower bandwidth but much lower latency as well.

tests have shown that it shows very little benefit, i was pointing out the general lack of need for such a cache because the ram is fast enough and so is the system bus.

iNetwork
Jul 24, 2004, 02:31 AM
Wow...imagine having a G5 with dual-duals :-)

:cool:
It's called a Pentium 4 Xeon... :)

iNetwork
Jul 24, 2004, 02:42 AM
Would this processor be considered a G5 or a G6??
I think apple will still call it a G5, I don't think we're ready for G6 yet. I would be willing to bet they'll add an extension to it... probabally call it dual core G5. I wouldn't be surprised if the dual core chip will make it into consumer products and the true dual processor machines will be the professional machines. Intel has done this with their HT technology. A HT Pentium 4 is a dual core chip.

It is much cheaper to make a system with one dual-core processor than it is to have a mainboard with two sockets and seperate processors. I think our race for clock speed ramping is coming to a slow down a little bit. I think what processor manufacturers are going to move to is a more efficient processor and multi-core and multi-processor machines in order to gain computing power until the next advance in micro-electronics comes in order to ramp clock speeds.

Even if the bandwidth is shared on this processor, it makes it perfect for consumer level products. This will also give Apple an edge to further distinguish between consumer market and professional market. Right now there isn't a whole lot of difference. For example, this generations ibooks are just as powerful as the previous generation power book. I call a generation when apple releases a new version of the product; this is usually only a few months.

Windowlicker
Jul 24, 2004, 03:18 AM
yes, please. :)

so would there be any point putting two of these processors in one computer ... making a Dual G5 a Dualdual G5? that would crush competition.

Windowlicker
Jul 24, 2004, 04:08 AM
I agree. the PowerMac 970 would be great. Remember the commercial a year or two ago that had a guy carring a box with a G5 or G6 on it? They were ripping on what Apples current and near future naming scheme is before it even happened. I'd support a new naming system. It's not like you can even figure out what other companies' names for their models even are (e.g. P720GX-421YK Inspirduroncrapon)

A PowerMac 970 seems so much more powerful because of the "970" >> "5" or "6". Marketing 101.

the G-naming is much easier. also thing of a consumer who knows nothing about computers. he/she walks to an apple store/apple reseller and starts looking at the names. There's "PowerMac 980" and "PowerMac G6". Just think which one looks more cryptical -> the consumer will get the G6.

...and this is totally just my guess ;D anyway I think the G-naming is ok. I don't really see any reason for changing it.

ddtlm
Jul 24, 2004, 04:41 AM
Hector:

I spent too much time looking this stuff up. ;)

tests have shown that it shows very little benefit
Two points: first I don't believe that, and second, that was the G4 they tested, not the G5.

Point one: Powerlogix did some SDR vs DDR L3 testing for their upgrades and that PDF is still available. It shows one test (Cinbench for what its worth) that goes 7% faster going from 1MB SDR L3 to 2MB DDR L3, and in those circumstances Photoshop was boosted 4%. If there were no L3 at all, the performance impact would be larger.

http://www.powerlogix.com/downloads/SDRDDR.pdf

Point two: Evidence that L3 matters can be found by examining P4 vs P4EE tests, clearly that 2MB L3 is making a difference in some benchmarks, sometimes a lot of difference, and sometimes none at all. In Civ3 the L3 apparently provides a 16% kick in the pants, on a 3D studio test it can add 12%.

http://www.aceshardware.com/read.jsp?id=60000253
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=1965

There are countless other tests of the P4EE out there which you can find if you'd like.

i was pointing out the general lack of need for such a cache because the ram is fast enough and so is the system bus.
What do you mean "fast enough"? Perhaps you didn't pay any attention to what I said about latency, the silent enemy. I think you should mosey over to Apple's official PR page and ponder the performance scaling going from 2.0ghz to 2.5ghz:

http://www.apple.com/powermac/performance/

Take the "Bibble 3.1a" test for example: the 2.5 is 150% the speed of a P4, the 2.0 is 119% the speed, so (100+150)/(100+119) = 14.2% speedup. That's the best test Apple presented as far as I know, the Photoshop test of all things showed a (100+98)/(100+82) = 8.8% speedup, "Audio Plug-ins" showed a (100+180)/(100+159) = 8.1% speedup. Remember, those are all on a 25% clock speed boost. A hypothetical 3ghz 970fx would be less than 18% faster at Photoshop than a 2ghz G5, on Apple's own test.

(Aside: note that Apple's G5 prices are in line with observable performance differences rather than clockspeed.)

Intel provides us a point of comparison: their 2.0ghz P4 vs the 2.53ghz P4, both Northwood cores with 512k L2. The 2.0ghz chip has a 400mhz FSB and the 2.53ghz chip has a 533mhz FSB, similar FSB scaling to what Apple has done (33% vs 25%, so Intel has a bit of an edge). Like the G5, the RAM on the P4's does not change ("note that we ran all of our CPU tests with PC800 RDRAM", on pg 6).

http://www.anandtech.com/showdoc.aspx?i=1615&p=1

18.3% on "Content Creation" which includes Photoshop.
27.3% on MP3 encoding with Lame (the accuracy of measurement is suspect).
19.7% on MPEG-4 encoding with Xmpeg.
20.8% on Adobe After Effects 5.5
22.0% on 3D Studio Max
19.6% on Maya

Feel free to do more math if you'd like.

So based on the combination of Apple's own evidence and tests performed by Anandtech, I think most people will agree that the 1.25ghz FSB and dual DDR 400 isn't doing the trick. With only 512k of L2 the G5 gets a lot of cache misses (depending heavily on the application of course) and each of those misses will require the G5 to wait on the main RAM. That'll put an end to performance scaling real fast if the RAM is slow to respond, which is what we are seeing.

I claim that despite the massive bandwith, the G5 is bottlenecked on RAM worse than the G4 ever was. Quite a claim, eh?

So repeat after me, pleaseIBMgiveusondiememorycontrollers!. :) Larger caches, be they L2 or L3, would help as well.

french macuser
Jul 24, 2004, 04:42 AM
It's called a Pentium 4 Xeon... :)
Xeon is not dual-core.

maxvamp
Jul 24, 2004, 05:18 AM
pleaseIBMgiveusondiememorycontrollers


Me and my AMD64 second that!!


Max.

Anyone waiting for the next great system/drive/machine will wait forever, since after every release of any product, speculation starts on what will be included on the next version.

Max.

JohnnyFirpo
Jul 24, 2004, 06:12 AM
A HT Pentium 4 is a dual core chip.

No, it's not. It's a single core model, with some extended logic to act as it were dual core. An "advanced" OS see it as two different processor instead just one. I mean XP/W2k is more advanced than W95/98/Me.

As far as I know the IBM implementation of SMT will be more efficient than Intel's HT. IBM's solution will have more "spare" registers to dispose.

rdowns
Jul 24, 2004, 06:37 AM
I did order it through my job its the Board Of Education In my city so whats there excuse.

I can only hope you are not a teacher.

phonic pol
Jul 24, 2004, 07:05 AM
I want one of these in my G5 rev b powerbook :D

flappo
Jul 24, 2004, 07:15 AM
VERY exciting news !

relimw
Jul 24, 2004, 10:02 AM
How may we start "PB G5 next tuesday" again? :p

Hmm, wonder if they'll get to 3GHz with the curent G5 or wait for the dual core...hmm, which to buy? :D

morkintosh
Jul 24, 2004, 10:27 AM
Wow...imagine having a G5 with dual-duals :-)

:cool:

yes, it'd be like having a quad ... amazing!

AidenShaw
Jul 24, 2004, 10:36 AM
We know that the OSX.4 kernel is smp/smt aware making any threaded app ( multiple apps ) perform better on this ( rumored ) machine.

Are you sure that it's SMT aware? Since there aren't any SMT CPUs around, I doubt that the current version is SMT aware.

(Or is OSX.4 your shorthand for Tiger? By the time 10.4 ships, it could be extended to handle SMT of course.)

The important thing about being SMT aware is that the system needs to schedule intelligently for better performance.

For example, my dual Xeons with HT show up as four processor systems - CPU0, CPU1, CPU2 and CPU3.

In reality, CPU0 and CPU2 are the same - they are two logical processors in a single physical processor. CPU1 and CPU3 are the other chip.

Being SMT-aware means:

o if a thread is running on CPU0, it is better to schedule a second thread on CPU1 or CPU3. If it's put on CPU2, then the two threads share one physical processor (the ~25% boost) and the second physical CPU is idle. If this happens, the SMT system can be much slower than the non-SMT system.

o If you have enough threads to keep all logical CPUs busy, then it is better to schedule threads from the same process on the same physical CPU. Most of the time, this will give better cache behaviour, although depending on the code it might be better to be on different physical CPUs. (E.g. two pure floating point number crunching code threads might find that being able to use the floating point units in both CPUs offsets the less efficient cache usage.)

In addition, Windows XP and later gives an API so that a programmer can distinguish logical and physical CPUs, and either give the OS hints about how to schedule - or even to force a particular scheduling.

Windows 2000 didn't know about logical CPUs, so it sees them all as physical and isn't SMT-aware. It uses all the logical CPUs, but not in the most efficient manner.

XP and W2k3 are SMT-aware, and makes smarter choices on its own, and gives the programmer tools to be even better. (W2k3 has more SMT features than XP, as well as being NUMA-aware.)

~Shard~
Jul 24, 2004, 10:36 AM
Hmm, wonder if they'll get to 3GHz with the curent G5 or wait for the dual core...hmm, which to buy? :D

I'm just going to wait for a RevB G6 PowerMac to be sure. :cool:

relimw
Jul 24, 2004, 10:42 AM
Okay, Apple would probably replace the two chips they are using with this one, dual core chip, but I hope they just use 2 dual core chips. Will software even be able to recognize that many microprocessors? Will you need to get new software to take advantage of two dual core chips?

I'd guess that Apple would initially go back to the single processor for low and mid-range, and dual for high end due to low initial quantities of the new processor. As far as needing new software to recognize a dual core, dual processor system, I doubt it'd be much more than an OS revision bump, ie 10.4.4 or such.

Now if we could only get the new machines to move the AGP slot further away from the PCI slots, that would be great (or even to space all the slots out a bit more, since certain companies are bad about double width cards) :(

iMeowbot
Jul 24, 2004, 10:46 AM
tests have shown that it shows very little benefit, i was pointing out the general lack of need for such a cache because the ram is fast enough and so is the system bus.
Note also that this is using the wave pipelining on the bus, so some of the traditional benefits of an L3 cache don't apply anyway.

guifa
Jul 24, 2004, 10:47 AM
"If this news is correct" would be the proper way to say this in English.
Edit: Sorry, didn't realize CarltonMusic and Fender beat me to it.
Actually, all of you are a bit off. It should be, "If this news be correct." You're using the present subjunctive in English, which is the same form as the infinitive.

os10geek
Jul 24, 2004, 10:51 AM
Wow! We'll have dual dual DVI, and dual dual processors!


Woohooooooo!

AidenShaw
Jul 24, 2004, 10:57 AM
Now if we could only get the new machines to move the AGP slot further away from the PCI slots...


The next machines will have PCI Express (PCIe) graphics, no doubt.

MikeTheC
Jul 24, 2004, 10:59 AM
I would guess they would only call a Power 5 derived PPC a G6. If these news are correct (or, 'this news is correct'? might a native speaker please tell me what is the correct version), than this rumored Power 5 derived PPC might not see the light before the end of 2005 or even 2006.

If you want to refer to news in the plural, the correct way to do so would be to say "If these news items are correct..." This would suggest you are referring specficially to multiple news items independant of each other (that is, drawn from seperate sources). However technically correct this may be, the average native speaker probably wouldn't talk like that, rather simply saying "If the news is correct..."

Just my 2 cents. (Not "My .02 cents" as that would be 2 tenths of a cent)

MikeTheC
Jul 24, 2004, 11:01 AM
Actually, all of you are a bit off. It should be, "If this news be correct." You're using the present subjunctive in English, which is the same form as the infinitive.

'Scuse me? What version of English do you use?

MikeTheC
Jul 24, 2004, 11:07 AM
The G-Series naming sytem is definitely getting too old. Sure, it makes some sense. But I think after the G6, apple should look for some other way to name their subsequent chip and powermac designs. Perhaps changing the first letter would help- "PowerMac X7" sounds much more advanced and interesting than staying with the G system for years to come.

Well, if you look at military classifications, X usually means "experimental". I realize that most people probably would overlook that, but I think the G is cool. I don't know what it actually means, but I kind of think of it as "generation", since one could make the argument that the PPC 601/603/603E was a kind of G1, and the 604/604e was a sort-of G2.

applekid
Jul 24, 2004, 11:08 AM
Question: With a dual core, what will be the clock speed?

For example, let's say one of these new 970MPs have a clock speed at 3 GHz. Is each core 3 GHz or is each core 1.5 GHz and adds up to 3 GHz.

Also, would two cores also be able to act as one core? Let's say it is a 3 GHz clock speed for each core, could it also act as one 6 GHz core?

iMeowbot
Jul 24, 2004, 11:17 AM
Actually, all of you are a bit off. It should be, "If this news be correct." You're using the present subjunctive in English, which is the same form as the infinitive.'Scuse me? What version of English do you use?
If it was good enough for Shakespeare, it's good enough for you :p

iMeowbot
Jul 24, 2004, 11:26 AM
Question: With a dual core, what will be the clock speed?

For example, let's say one of these new 970MPs have a clock speed at 3 GHz. Is each core 3 GHz or is each core 1.5 GHz and adds up to 3 GHz.
Each would be at 3.
Also, would two cores also be able to act as one core? Let's say it is a 3 GHz clock speed for each core, could it also act as one 6 GHz core?
No. You really have two separate CPUs on one chip. The nice thing about have two on one chip is that the logic required to couple them is also on that chip, so it's faster, less comlicated to build, potentially cheaper.

mε₯ε
Jul 24, 2004, 11:30 AM
First its a dual core G5

with an elastice bus meaning its defaulted to 1GHz bus to 3GHz speed per core.

remember its elastice so the default is 1GHz bus meaning Apple can crank up the bus to 1.5GHz per core at 3GHz speeds. This is neither good or bad thing if Apple choses to crank up the speed great for us if not they will stick it in low portables and iMac.

Its also has PowerTune Enhanced so its going to control the processor speed equally amoung both cores. At top speed you are looking at 3GHz when you are surfing the net or typing a document it will drop the speed down, so all this means is that even though you are buying a 3GHz G5 makes good uses of it otherwise you are wasting your money. Or donate it to me
;) :D

dontmatter
Jul 24, 2004, 11:50 AM
more power! Great! Wait! so few really have that serious a need for that kind of power, software just isn't keeping up. More than that, c'mon! Apple's alreaedy got the graphics crowd, etc.

The only thing that will seriously get me oohing and aaahhing is something that majorly reduces heat and power consumption. Those are the biggest limiting factors now. Do those, and we've got the ability to do G5's in less than powermac, and not be pushing it, spending extra money on cooling, etc. If you do them really well, apple could actually makes some seriously awesome laptops, that don't make your palms sweat when browsing the web and that have awesome batterylife. B/c now, again, an ibook can handle most everybody's needs, but who wants a computer that, like my PB, I sometimes can hardly keep my hands on long enough to put in it's case, and runs out of power in a little more time than watching a DVD on it? What I want to hear is a G5 that can directly compete with a pentium M, and beat it.

G4-power
Jul 24, 2004, 11:50 AM
Just my 2 cents. (Not "My .02 cents" as that would be 2 tenths of a cent)

Excuse me, don't you mean "2 hundredths of a cent", since 0.2 cents would be 2 tenths of a cent. Correct me if I'm wrong. :o

Little Endian
Jul 24, 2004, 11:53 AM
I really don't like this thinksecret article. Sure having a dual Core PowerPC chip would be great but what it all boils down to is unsubstantiated rumors. Anyone could say that IBM is working on Dual Core chips and be right as they probably are, as every other company is doing the same. What gets me about the article is the timing. Thinksecret claims that IBM will begin volume production of this Dual core 970 in Jan of 2005!! That does not sound good as that could mean we may not see another PowerMac revision shipping until March 2005 or even later like August 2005. If this is true it would mean that it will be another year between updates. I am really hoping for an interim update in the meantime.... I would really like to see a 3Ghz machine shipping or at least announced in Jan 05 regardless if it is a single or Dual core chip.

MikeTheC
Jul 24, 2004, 12:05 PM
Excuse me, don't you mean "2 hundredths of a cent", since 0.2 cents would be 2 tenths of a cent. Correct me if I'm wrong. :o

Darn, must have been using that first-generation Pentium 1 again. :eek:

You're right, thanks for catching my mistake!

wdlove
Jul 24, 2004, 12:07 PM
pleaseIBMgiveusondiememorycontrollers

Anyone waiting for the next great system/drive/machine will wait forever, since after every release of any product, speculation starts on what will be included on the next version.

Max.

All of the speculation is fun as shown by how fast this thread is growing.

Always waiting for the next great thing would mean not ever purchasing.

joeconvert
Jul 24, 2004, 12:08 PM
If apple pumps the speed up to 3 or more GHz, PC gamers would have something other than wintel to use. But if apple doesn't ever design macs for gaming they can only realistically gain 30 to 40 percent market share, as 70% :eek: of PCs are sold for GAMING


Might want to qualify that 70% with consumer market.

nuckinfutz
Jul 24, 2004, 12:37 PM
What I want to hear is a G5 that can directly compete with a pentium M, and beat it.

Then you should be keen on the new Powertune features of this rumored 970MP. If anything is going to bring the G5 to a powerbook it's Powertune.

I really don't like this thinksecret article. Sure having a dual Core PowerPC chip would be great but what it all boils down to is unsubstantiated rumors. Anyone could say that IBM is working on Dual Core chips and be right as they probably are

Thinksecrets record on Apple products is better than "any" other sites. They had the mockups of the LCD displays a month or more before they shipped and they were spot on. While everyone knows Dual Core is coming the article gives us good information. 1MB L2 cache per core. 10 layer process, Powertune and ABIST. That's far more meaty than saying "Dual Cores are coming". An interim refresh would be ok "if" Dual Core Powermacs weren't ready by say early March. It would make no sense to do a refresh that would but the DC Powermacs off until late summer.

I don't think it's really sunk in to people how nifty DC is. Since both CPU cores share the same die they can snoop each others cache without the system controller being involded. Much faster and more efficient. Powertune makes the though of having a Quad CPU system actually obtainable. If the rumors of pipeline changes are also correct it means we're more likely to hit 3Ghz. If you're a professional you can wait or grab a 2.5Ghz Powermac. If Apple is ready to ship DC in Q1 then only AMD would have the chance to match them and that's great marketing opportunity.

~Shard~
Jul 24, 2004, 12:45 PM
Actually, all of you are a bit off. It should be, "If this news be correct." You're using the present subjunctive in English, which is the same form as the infinitive.

Don't get your panties in a knot Shakespeare - you're gonna be one busy guy if all you do on these forums is go around correcting people's English. Just go back to reading your dictionary and calm down... ;) :cool:

gop007
Jul 24, 2004, 01:42 PM
This I find rather interesting. It appears the issues IBM had with the 3ghz chips has created further advantages in other areas. What is worth pointing out is that it appears we will not see a 3 ghz system until mid 2005. IBM is now the new Motorola for Apple. They are a year behind from what they promised Apple. I have seen this before 4 years ago with Motorola. The G5 is here to stay for much longer than the original IBM road map indicated last year.

No big deal, we got used to the G3 for more years than we would have liked along with the G4 as well. This will mean that when the G6 comes out in two or three years we should see the same kick ass movement when we saw the first PowerPC G3 in 1997.

jouster
Jul 24, 2004, 01:50 PM
I really don't like this thinksecret article. Sure having a dual Core PowerPC chip would be great but what it all boils down to is unsubstantiated rumors. Anyone could say that IBM is working on Dual Core chips and be right as they probably are....


And doubtless, anyone does. But thinksecret says a lot more than that.

Sun Baked
Jul 24, 2004, 02:08 PM
[quote=Little Endian]I really don't like this thinksecret article. Sure having a dual Core PowerPC chip would be great but what it all boils down to is unsubstantiated rumors. Anyone could say that IBM is working on Dual Core chips and be right as they probably are....]And doubtless, anyone does. But thinksecret says a lot more than that.Yes, IBM is most likely working on multiple core variants. One semi-public possibility for their use has been Microsoft for the XBox2.

There is a darn good possibility, Apple has been playing with these same chips -- especially if they work with Apple's existing 3 G5 memory controllers (U3, U3H - ECC capable/faster HT, & U3-Lite - iMac3).

However if they do need a new memory controller for this CPU, game play is a bit more interesting -- especially since Apple will be supporting 4 memory controllers (can't forget the G4 chipset) going into 2005 if they don't prune the tree soon.

nuckinfutz
Jul 24, 2004, 02:14 PM
This I find rather interesting. It appears the issues IBM had with the 3ghz chips has created further advantages in other areas. What is worth pointing out is that it appears we will not see a 3 ghz system until mid 2005. IBM is now the new Motorola for Apple. They are a year behind from what they promised Apple. I have seen this before 4 years ago with Motorola. The G5 is here to stay for much longer than the original IBM road map indicated last year.

No big deal, we got used to the G3 for more years than we would have liked along with the G4 as well. This will mean that when the G6 comes out in two or three years we should see the same kick ass movement when we saw the first PowerPC G3 in 1997.

Whoa. I don't see anything that appears to portend 3Ghz not shipping until Mid 2005. In fact information from the Apple Nova forums seems to point to some pipeline changes that might make 3Ghz much easier. I think you're building a case for IBM= Moto based off of naivete. Make your case but try to inject some technical reasoning to back up your diatribe.

Nick dePlume
Jul 24, 2004, 02:23 PM
"Thinksecret claims that IBM will begin volume production of this Dual core 970 in Jan of 2005!! That does not sound good as that could mean we may not see another PowerMac revision shipping until March 2005 or even later like August 2005. If this is true it would mean that it will be another year between updates."

The story never states that the next Power Mac G5 revision will ship with the PowerPC 970MP. While it could certainly be until March or later that we see dual-core G5s shipping, there's nothing that rules out an interim update shipping with a 970FX.

-Nick dePlume
Publisher and Editor in Chief, Think Secret

shyataroo
Jul 24, 2004, 02:23 PM
now all apple needs is to use hyper/multi threading and DDR2 RDRAM imagine the speed at which it will go...it will be a (basically) Quad Processor G5 that processes 30 or so times faster than its Pentium 4 (or prescott) equivlent Apple will be the first company to make a computer that preforms 1 Sextillion (1,000,000,000,000,000,000,000) FLOPS a second mmmmm.....unrealistc vision.,,

nuckinfutz
Jul 24, 2004, 02:38 PM
"Thinksecret claims that IBM will begin volume production of this Dual core 970 in Jan of 2005!! That does not sound good as that could mean we may not see another PowerMac revision shipping until March 2005 or even later like August 2005. If this is true it would mean that it will be another year between updates."

The story never states that the next Power Mac G5 revision will ship with the PowerPC 970MP. While it could certainly be until March or later that we see dual-core G5s shipping, there's nothing that rules out an interim update shipping with a 970FX.

-Nick dePlume
Publisher and Editor in Chief, Think Secret

Nice reply Nick. Seems like there's a whole lot of "hand wringing" over speed. Even if no refresh happens until March that's only 8 months. It would be one thing if we were behind in speed but if you have $3k you can buy a dual 2.5 and that's not a shabby computer by any stretch of the imagination.

Being as how the next revision is likely to significantly change key aspects of the architecture( PCI Express, DDR2 etc) I'm all for Apple extensively testing the platform before shipment. My guess is Tiger ships around end of Q1 if new Powermacs are shipping close to this timeframe they can be synced better IMO.

rt_brained
Jul 24, 2004, 02:56 PM
Pontiac is coming out with a new car next year – the replacement for the Grand Am – and it's called the G6. I can't imagine there hasn't already been some copyright issues addressed on this –*Pontiac must be in the clear to use the name –*and I'd be willing to bet Apple has no intention to let the name of their high-end processor be confused with a Pontiac.
Perhaps the new Pontiac is going to be powered by an Apple G6. The Grand Am hasn't seen that kind of power for years.

edgar_is_good
Jul 24, 2004, 03:31 PM
Don't get your panties in a knot Shakespeare - you're gonna be one busy guy if all you do on these forums is go around correcting people's English. Just go back to reading your dictionary and calm down... ;) :cool:

Sheesh, and everyone's missing the obvious possibility that he was originally talking about an English soccer/football team named "news", in which case it would be "if these news are correct, they are the best team in Europe"

~Shard~
Jul 24, 2004, 03:37 PM
Sheesh, and everyone's missing the obvious possibility that he was originally talking about an English soccer/football team named "news", in which case it would be "if these news are correct, they are the best team in Europe"

Ah, I stand corrected - thanks for the clarification on that one... ;)

recursivejon
Jul 24, 2004, 03:41 PM
The G-Series naming sytem is definitely getting too old. Sure, it makes some sense. But I think after the G6, apple should look for some other way to name their subsequent chip and powermac designs. Perhaps changing the first letter would help- "PowerMac X7" sounds much more advanced and interesting than staying with the G system for years to come.

Oh god NO MORE X's.

everyone is doing that to their products; putting in an "x" or "pro"

</shuddder>

recursivejon
Jul 24, 2004, 03:45 PM
I just ordered a new Mac with one of those new fangled chips, my order says on or before July 30th ... :cool:

If you do mean a 2.5, then you and me both.

~Shard~
Jul 24, 2004, 03:48 PM
Oh god NO MORE X's.

everyone is doing that to their products; putting in an "x" or "pro"

</shuddder>

You mean like "Windows XP Pro"? ;) Yah, I agree, 'X' is used way too much nowadays - OS X is quite enough for me, thank you very much... :cool:

spacemoose
Jul 24, 2004, 03:51 PM
It seems that IBMs does not really have a mobile processor development program.

While this is well and good for servers/workstations, it could hurt Apple in the growing mobile market with Intel working hard to improve their "M" chips.

If IBM is not creating any mobile specific technology, and Apple is forced to retrofit their desktop chips into their mobile products, they will never compete with the already superior Intel mobile technology.

Perhaps this is why Apple has maintained its relationship with Motorola, as it intends to create G4 derivatives that will become its "M" line.

----

I think Apple will surely stick with the "G" identity for its products; it is in complete sync with the clean, simple and easily distinguishing design of their systems.

BrianKonarsMac
Jul 24, 2004, 04:24 PM
1 MB L2 cache per processor sounds good to me.


Does anybody know if it will have 2 MB L3 cache per probessor?

as the article states, L3 cache is not supported...so i'd assume that means NO L3, at all.

so does the dual core share bandwith along the FSB? or does it have seperate channels? making four total for a dual two core system.

BrianKonarsMac
Jul 24, 2004, 04:29 PM
It seems that IBMs does not really have a mobile processor development program.

While this is well and good for servers/workstations, it could hurt Apple in the growing mobile market with Intel working hard to improve their "M" chips.

If IBM is not creating any mobile specific technology, and Apple is forced to retrofit their desktop chips into their mobile products, they will never compete with the already superior Intel mobile technology.

Perhaps this is why Apple has maintained its relationship with Motorola, as it intends to create G4 derivatives that will become its "M" line.

----

I think Apple will surely stick with the "G" identity for its products; it is in complete sync with the clean, simple and easily distinguishing design of their systems.
apparently you are not familiar with IBM's PowerTune technology. They don't need to design a mobile processor if they can get it working correctly, as it will effectively be better than a seperate mobile design, as it will be scalable from 1/64 of clock frequency to full frequency in 3 cycles. this will cut heat and power consumption drastically, because the processor will only be running hard when you're doing heavy computing, and will enter a deep nap mode in between., i.e. when you're typing etc.

iMeowbot
Jul 24, 2004, 04:50 PM
I really don't like this thinksecret article. Sure having a dual Core PowerPC chip would be great but what it all boils down to is unsubstantiated rumors.
There has been some substantiation. The information came from leaked IBM preproduction documents, portions of which have been circulated in public; and it corresponds well with intentionally public information from IBM. The only part that might seriously be in question is the timetable.

Roller
Jul 24, 2004, 04:51 PM
Being as how the next revision is likely to significantly change key aspects of the architecture( PCI Express, DDR2 etc) I'm all for Apple extensively testing the platform before shipment. My guess is Tiger ships around end of Q1 if new Powermacs are shipping close to this timeframe they can be synced better IMO.

To me, the architecture changes are at least as compelling as the CPU clock rate increases. The only question is will we see these changes in the next rev with dual 3 GHz CPUs or in the rev after that?

Multimedia
Jul 24, 2004, 05:30 PM
I Knew It Was Coming But Not As Soon As Next Summer. Excellent. I'll take two x 3 GHz per core for a total of 12 GHz inside. Now that's what I call POWER. :D

mr.steevo
Jul 24, 2004, 05:43 PM
Actually, all of you are a bit off. It should be, "If this news be correct." You're using the present subjunctive in English, which is the same form as the infinitive.
Well well. I suppose this is American English, which is, of course, improper. "If this news is correct..." is far more correct than "be correct". Where DID you go to school?
Oh, and where is the New Cube?

s.

itsa
Jul 24, 2004, 06:01 PM
Apple may not have dual duals if this is indeed true. They may simply replace the dual's with one of the dual core processors. The advatage for apple is that they can keep the powermac performance significantly above the iMac line still, without the added cost a second separate processor incurs (all the connections for it plus the cost of the actual processor, presumably a single dual core would be cheaper than two single cores). I rather suspect that is what apple will do. They may call it a G6 or whatever just to further differentiate.

And it's much more easy to keep one cool.

nuckinfutz
Jul 24, 2004, 06:09 PM
To me, the architecture changes are at least as compelling as the CPU clock rate increases. The only question is will we see these changes in the next rev with dual 3 GHz CPUs or in the rev after that?

Let us hope for the next revision. I'm pefectly content to see the effects of 6 months of PCIe X86 motherboards has. So far there doesn't seem to be a lot of issues but they are just now shipping.

I will say I'm a bit intrigued by Nvidia's SLI technology. I can't wait to see the real world benchmarks coming this fall. Two fast PCI Express GPU strapped together working hard seems to be quite exciting.

~Shard~
Jul 24, 2004, 06:25 PM
I Knew It Was Coming But Not As Soon As Next Summer. Excellent. I'll take two x 3 GHz per core for a total of 12 GHz inside. Now that's what I call POWER. :D

I hope you're saving your pennies for this machine you describe that's coming out next summer!

melgross
Jul 24, 2004, 06:28 PM
OS X supports two logical processors. It's possible, from what I've read, that 10.4 will support four. The idea that it's easy to continue support up to sixteen, or higher, is wrong.

When NT first came out it supported two. It wasn't able to make a real push into the server space until Microsoft was able to support four processors. That took a lot of work, and didn't work well. In fact, support for four only worked well when they got the support to eight-way. Now it supports thirty two way, and works well up to sixteen.

Memory, cache coherency, as well as issues with threading and program allocation, among others, are not easily gotten around.

Most programs have one thread and only benefit from two processors if the OS can move system ops to one processor while keeping program usage on the other. Most programs won't benefit much from hyper-threading, and indeed, might suffer. Intel has found this to be a problem as well.

A program must be compiled for multiple threads, as well as multiple processors. The more threads and processors, the more difficult the problem becomes.

Very few programs will see a benefit from four way machines. If you are running more than one program, however, you might see an advantage if the OS is written properly.

There has been criticism of Apple's server offerings because there are no four way systems. Business's like to have options to move to higher capacity systems, and so far, Apple doesn't offer any. I would assume that Apple would like to correct that.

IBM was talking about dual cores as far back as 2001. They were expecting to go there when switching to 65nm, but maybe now, as Intel and AMD seem to be doing, they will not wait. They also were going to an on board memory controller, but, from what I remember, the work to get Altivec on die stopped that work for the time being. I had understood that the on die controller was a requisite for the chip to be used in a portable. Perhaps that has changed.

Hyper-threading does not make a chip equivalent to a dual chip on die. There are certain operations that can be hyper-threaded, and those that cannot. Threads can be moved between chips or on chip, if more than one is used, but the chip (Xenon) has one cache and one line out to main memory. There have been dual on die chips with one cache, but they have not worked well. When hyper-threading is turned on, some programs see two logical PROCESSES, but not two logical cpu's. This is complex to explain. Going to Intel's or IBM's sites will give a good explanation. Much better that I could do here in such limited space.

HiRez
Jul 24, 2004, 07:29 PM
apparently you are not familiar with IBM's PowerTune technology. They don't need to design a mobile processor if they can get it working correctly, as it will effectively be better than a seperate mobile design, as it will be scalable from 1/64 of clock frequency to full frequency in 3 cycles. this will cut heat and power consumption drastically, because the processor will only be running hard when you're doing heavy computing, and will enter a deep nap mode in between., i.e. when you're typing etc.The old G3 chips had this, and it worked great. But Apple needs to get the software to work with it too. On my 400 MHz G3 Pismo under OS 9, I'd regularly get 4.5 hours per battery, with full screen brightness. Pop in two batteries and I was good to go for 9 hours :eek:!! However, as soon as I loaded OS X on the same machine, battery life dropped to a comparatively-pathetic 2.5 hours and to this day Apple's battery life on their portables is mediocre at best. If I try playing an OpenGL game on my TiBook, I get 45 minutes per battery :mad:

macsrus
Jul 24, 2004, 07:45 PM
When NT first came out it supported two. It wasn't able to make a real push into the server space until Microsoft was able to support four processors. That took a lot of work, and didn't work well. In fact, support for four only worked well when they got the support to eight-way. Now it supports thirty two way, and works well up to sixteen.

Actually NT supported 4 processors way back in Windows NT 3.1 Advanced Server.....
And Believe it or not Unisys was building 32 processor NT servers using Intel pentiums back in 1995/96 (during the NT 3.51 days)

The real problem with NT's support of multiple processors wasnt/isn't NT's problem....
It is the terrible legacy architectural design of IBM PC's and compatables...
These systems were never designed to be anything more than a Personal Computer.....
Actually it is almost amazing that they have worked as well as they have considering there initial design flaws....
Many engineers have spent countless hours trying to mitigate a PC's poor design.



There has been criticism of Apple's server offerings because there are no four way systems. Business's like to have options to move to higher capacity systems, and so far, Apple doesn't offer any. I would assume that Apple would like to correct that.

I agree.... I for one would love to see a 4 way PPC server...




IBM was talking about dual cores as far back as 2001. They were expecting to go there when switching to 65nm, but maybe now, as Intel and AMD seem to be doing, they will not wait. They also were going to an on board memory controller, but, from what I remember, the work to get Altivec on die stopped that work for the time being. I had understood that the on die controller was a requisite for the chip to be used in a portable. Perhaps that has changed.

Actually IBM has been producing dual core processors for 3 years now...
All Power 4 processors are dual core... and the Power 5 is dual core too.

I expect if IBM is making a low cost dual core CPU for Apple, then it will be based on the Power 4+ and maybe the Power 5....
Either way they will more than likely be stripped of cache.

AidenShaw
Jul 24, 2004, 07:46 PM
When hyper-threading is turned on, some programs see two logical PROCESSES, but not two logical cpu's.

No, it's PROCESSORS.

With Windows (since OS X doesn't run on any HT chips, anything there is speculation) when HT is enabled you see twice as many PROCESSORS.

On my dual Xeons, the environment variable NUMBER_OF_PROCESSORS is 4. The Task Manager shows 4 CPU load windows in the performance pane.

With Windows 2000 Pro - you cannot use the additional logical processors because 2k Pro only supports duals. With XP Pro, the "dual" support allows dual HT processors - or 4 logical processors.

As far as "operations that can be hyper-threaded" - the real issue is whether threads are waiting on the same resources. The OS doesn't see hyper-threads as anything other than a normal thread - they're scheduled as they request CPU time. If they're on real duals, they'll get 2 physical CPUs. If HT on a single, they'll get 2 logical CPUs.

The gain from HT is when threads use different chip resources. If one thread is stalled on memory, the other can use the integer unit. While one is using the float unit, the other can access memory. If both threads hit the same units, there's little benefit to HT.

macsrus
Jul 24, 2004, 08:11 PM
It seems that IBMs does not really have a mobile processor development program.

While this is well and good for servers/workstations, it could hurt Apple in the growing mobile market with Intel working hard to improve their "M" chips.

If IBM is not creating any mobile specific technology, and Apple is forced to retrofit their desktop chips into their mobile products, they will never compete with the already superior Intel mobile technology.

Perhaps this is why Apple has maintained its relationship with Motorola, as it intends to create G4 derivatives that will become its "M" line.

----

I think Apple will surely stick with the "G" identity for its products; it is in complete sync with the clean, simple and easily distinguishing design of their systems.

I keep hearing on this board about how the powerbooks cant compete with a Pentium M....
NOW lets get this staight.... Im a long time PC laptop user and I wouldnt take a slow a-- Pentium M system if you gave it to me AND paid me to use it.

Apple will not be hurt in the laptop market by anything intel makes...
because a Powerbook isnt and never really was a competitor to a PC laptop
Apple sells mainly to a niche market...
And it competes in the (I just love this cool well engineered work of art laptop and typical MAC buyer market...)

bombcar
Jul 24, 2004, 09:58 PM
Well well. I suppose this is American English, which is, of course, improper. "If this news is correct..." is far more correct than "be correct". Where DID you go to school?
Oh, and where is the New Cube?


"be" is used correctly in that case when speaking in the subjunctive. It is rarely used correctly anymore, but still exists. An example is in the Three Dog Night song, Joy to the World:

If I were the king of the world,
I'd tell you what I'd do:
I'd throw away the cars and the bars and the wars,
And make sweet love to you.


See the Wikipedia artice (http://en.wikipedia.org/wiki/Subjunctive) for more information about the dying subjunctive. :)

-The former English teacher

maxvamp
Jul 25, 2004, 12:21 AM
OS X supports two logical processors. It's possible, from what I've read, that 10.4 will support four. The idea that it's easy to continue support up to sixteen, or higher, is wrong.



As of OSX.3, the OS supports at least two physical processors. I am not aware that it supports any logical ( hyper threaded ) processors.


When NT first came out it supported two. It wasn't able to make a real push into the server space until Microsoft was able to support four processors. That took a lot of work, and didn't work well. In fact, support for four only worked well when they got the support to eight-way. Now it supports thirty two way, and works well up to sixteen.

Memory, cache coherency, as well as issues with threading and program allocation, among others, are not easily gotten around.




These issues were due to MS's lack of technology. At the same time MS was having a hard time getting past 4 processors, IBM had OS/2 V2 up to 64 processor capable. Yes, the architecture of day was limited since multi proc 386s often shared a common cache, which got cleared often by mistake, but the technology did exist. UNIX and the Mach kernel was even more advanced than the OS/2 kernel.

It is not wise to try to use MS as an example of the latest in technology.

Most programs have one thread and only benefit from two processors if the OS can move system ops to one processor while keeping program usage on the other. Most programs won't benefit much from hyper-threading, and indeed, might suffer. Intel has found this to be a problem as well.

A program must be compiled for multiple threads, as well as multiple processors. The more threads and processors, the more difficult the problem becomes.

Very few programs will see a benefit from four way machines. If you are running more than one program, however, you might see an advantage if the OS is written properly.



Bull!!! Any programmer can simply call for a new thread for any task in their code. Furthermore, the classes that a programmer might call to do tasks such as GUI controls, or file system access will very often spawn threads, due to the nature of the built in classes in the OS ( Win32, .NET, Next Framework, etc ).

Unless you are sitting in Terminal, and launching a simple program such as tar, you will be hard pressed to find a single threaded program for Windows, OSX, or most any modern day OS. Case in point, I have written a small program in JAVA that reads and modifies file attributes. I have intentionally not threaded the app, but when it does launch, it uses 7 threads, because I use the system calls found in the File object. You cannot have a GUI interface without threading.

The compiler does not perform the threading, the programmer does.

About the only thing you are somewhat correct is that just because a system has many physical / logical processors, does not guarantee speed. multi processors are like torque in an engine. More torque doesn't make you go faster, it just takes a lot more of a load to slow you down.

If you can't keep the processors busy enough ( generate enough of a load ), the extra power is wasted.


Max.

Sun Baked
Jul 25, 2004, 12:53 AM
As of OSX.3, the OS supports at least two physical processors. I am not aware that it supports any logical ( hyper threaded ) processors.

Max.Better SMP support is "supposed" to be on the way in OS X v10.4 -- this is reading between the lines on Apple Tiger Preview page (http://www.apple.com/macosx/tiger/unix.html) about FreeBSD 5.xMore Power to Power-Users
The upgraded kernel, based on FreeBSD 5.x, provides optimized resource locking for better scalability across multiple processors, support for 64-bit memory pointers through the System library and standards-based access control lists. The system enhances network services via a next-generation launch daemon and centralized application logging. Tiger also features command-line access to Spotlight for searching application metadata and enables many common UNIX utilities to handle HFS+ resource forks.

Optimized Kernel Resource Locking
Optimized locking provides better SMP performance by allowing two CPUs to simultaneously access different portions of the kernel. This will improve performance of almost every task on multiprocessor machines.Of course no telling how much better things will be just jumping from FreeBSD 4.4 to 5.x, along with extremely recent versions of GCC.

http://www.freebsd.org/releases/

AidenShaw
Jul 25, 2004, 12:56 AM
These issues were due to MS's lack of technology. At the same time MS was having a hard time getting past 4 processors, IBM had OS/2 V2 up to 64 processor capable.

This document OS/2 for SMP V2.11 Reference (http://www.warpspeed.com.au/cgi-bin/inf2html.cmd?..\html\book\ddk\smp.inf+5) says 16 CPUs max.

Theoretical support for many processors is a lot different than good performance with that many. Windows NT always had architectural support for 32 processors, but system locking issues caused a performance bottleneck on many workloads with a quad CPU machine. From release to release, these bottlenecks have been addressed.

And as far as "lack of technology", did you note that Tiger claims:

Optimized Kernel Resource Locking
Optimized locking provides better SMP performance by allowing two CPUs to simultaneously access different portions of the kernel. This will improve performance of almost every task on multiprocessor machines.

So, it seems that Panther must have some "lack of technology" that makes it suffer even on dual processors.


"A program must be compiled for multiple threads, as well as multiple processors. The more threads and processors, the more difficult the problem becomes."

Bull!!! Any programmer can simply call for a new thread for any task in their code.

I think that I must agree with the first post.

Of course creating threads is easy, and of course many OS activities create hidden threads. Overlapped I/O is a common example - the O/S is doing multiple I/Os in the background to speed things up.

What can be very difficult is splitting the program algorithm up so that it can be done in parallel by multiple threads and actually result in a program speedup. For some programs, this is next to impossible - by their nature things have to be done in order, and each step must wait for the previous to complete. This becomes even harder when you try to go beyond a dual processor system.

Some programs are easy to multi-thread - by their nature they consist of lots of small tasks that can be done in parallel. Video encoding is a great example - each keyframe makes an independent starting point, you have a keyframe and then a group of intermediate frames that are based on the keyframe.

You can split the input into these keyframe chunks, farm each chunk out to a separate thread, then reassemble them as they complete. In practice one would use a small number of worker threads, and feed chunks to those threads as they complete the previous chunks.

And of course, a task that can be farmed (or clustered) is probably a task that can be multi-threaded.

I think that your "Bull!!" comment is unjustified.

maxvamp
Jul 25, 2004, 01:36 AM
:)

Ok, I stand corrected on the OS/2 max procs. It has been awhile since I was an OS/2 freak.

The reason I called BULL was over the comment that most programs are single threaded. Aiden, do you support this comment?

I will not dispute that some scheduling can be tricky, but effective threading happens all the time in code. Even in the example you provided with the video encoding, there are at least two threads here. One thread will be reading the source video and hopefully caching some of it, while another will be doing the processing. If you take into account the UI progress, and the File I/O, I doubt that you can say that this is a single threaded app.


A comment I have made several time is that more processors doesn't always mean more power. The claim that people still need to compile all of their apps for dual proc is a fallacy. People need to optimize their apps for the most effective threading.

Finally, the technology comment was made as a direct comment of Microsoft at that time period.

Microsoft did not come into it's own as an innovator until almost 2000 with the re-write of the OS/2-NT kernel into Windows 2000. Before then, in almost every place that they were competing Microsoft was technologically behind all other players. To connect this fact with improved technology that Apple is trying to put into the next version of the OS is failed logic taken out of context. IBM was always better at kernel design than Microsoft. I suspect that it had something to do with the fact that IBM is really a hardware company that happens to be the biggest software company out there.

Feel free to dispute this stuff, but I plan on sticking to my position.

Max. :D

shawnce
Jul 25, 2004, 01:36 AM
So, it seems that Panther must have some "lack of technology" that makes it suffer even on dual processors. Mac OS X from its initial release and to this day (Panther) has used a locking construct called a funnel in the kernel. This funnel is used to protect parts of the kernel from simultaneous modification by multiple threads. Initially only one funnel existed and that funnel protected the majority of the BSD based aspects of the kernel since it was traditionally designed to not support multiple threads running concurrently. Later a second funnel was added (I think in 10.1), one to protect the networking aspects and the other the rest of the BSD derived kernel. This allowed better utilization of the system (at least while in kernel) when running on multiprocessor systems.

Note that IOKit and a few other areas of the kernel are pervasively multi-threaded, using finer grain locking that allow better utilization. Also user mode code can be heavily threaded to leverage available compute resources, etc.

As for Tiger I can not delve into specifics because of NDA but the granularity and locality of locking in the kernel will be changed for the better.

melgross
Jul 25, 2004, 02:10 AM
Sorry, the Quote feature doesn't seem to be working for me today. I'm doing it manually.

Quote:
Originally Posted by melgross
When NT first came out it supported two. It wasn't able to make a real push into the server space until Microsoft was able to support four processors. That took a lot of work, and didn't work well. In fact, support for four only worked well when they got the support to eight-way. Now it supports thirty two way, and works well up to sixteen.


macsrus:
Actually NT supported 4 processors way back in Windows NT 3.1 Advanced Server.....
And Believe it or not Unisys was building 32 processor NT servers using Intel pentiums back in 1995/96 (during the NT 3.51 days)

me:
Yes at 3 they WERE supporting four. But when it first came out, you know, ver 1.0, they were supporting two.
macsrus:
Actually IBM has been producing dual core processors for 3 years now... All Power 4 processors are dual core... and the Power 5 is dual core too.

me:
True, but I was talking about the 970, not the Power series.

AidenShaw:
Quote:

Originally Posted by melgross

When hyper-threading is turned on, some programs see two logical PROCESSES, but not two logical cpu's.


No, it's PROCESSORS.
With Windows (since OS X doesn't run on any HT chips, anything there is speculation) when HT is enabled you see twice as many PROCESSORS.

On my dual Xeons, the environment variable NUMBER_OF_PROCESSORS is 4. The Task Manager shows 4 CPU load windows in the performance pane.

me:
I agree. But the program being run doesn't see it that way.

maxvamp:
Quote:
OS X supports two logical processors. It's possible, from what I've read, that 10.4 will support four. The idea that it's easy to continue support up to sixteen, or higher, is wrong.


As of OSX.3, the OS supports at least two physical processors. I am not aware that it supports any logical ( hyper threaded ) processors.

me:
Sorry, I meant to say physical, not logical.


Quote:
When NT first came out it supported two. It wasn't able to make a real push into the server space until Microsoft was able to support four processors. That took a lot of work, and didn't work well. In fact, support for four only worked well when they got the support to eight-way. Now it supports thirty two way, and works well up to sixteen.

Memory, cache coherency, as well as issues with threading and program allocation, among others, are not easily gotten around.


These issues were due to MS's lack of technology. At the same time MS was having a hard time getting past 4 processors, IBM had OS/2 V2 up to 64 processor capable. Yes, the architecture of day was limited since multi proc 386s often shared a common cache, which got cleared often by mistake, but the technology did exist. UNIX and the Mach kernel was even more advanced than the OS/2 kernel.

It is not wise to try to use MS as an example of the latest in technology.

me:
Actually, all companies had problems getting past two processors, in the beginning. It wasn't just Microsoft. IBM had more experience than Microsoft did (they had none, really.).

Quote:

Most programs have one thread and only benefit from two processors if the OS can move system ops to one processor while keeping program usage on the other. Most programs won't benefit much from hyper-threading, and indeed, might suffer. Intel has found this to be a problem as well.

A program must be compiled for multiple threads, as well as multiple processors. The more threads and processors, the more difficult the problem becomes.

Very few programs will see a benefit from four way machines. If you are running more than one program, however, you might see an advantage if the OS is written properly.



Bull!!! Any programmer can simply call for a new thread for any task in their code. Furthermore, the classes that a programmer might call to do tasks such as GUI controls, or file system access will very often spawn threads, due to the nature of the built in classes in the OS ( Win32, .NET, Next Framework, etc ).

Unless you are sitting in Terminal, and launching a simple program such as tar, you will be hard pressed to find a single threaded program for Windows, OSX, or most any modern day OS. Case in point, I have written a small program in JAVA that reads and modifies file attributes. I have intentionally not threaded the app, but when it does launch, it uses 7 threads, because I use the system calls found in the File object. You cannot have a GUI interface without threading.

The compiler does not perform the threading, the programmer does.

About the only thing you are somewhat correct is that just because a system has many physical / logical processors, does not guarantee speed. multi processors are like torque in an engine. More torque doesn't make you go faster, it just takes a lot more of a load to slow you down.

me:
Bull! My, such a strong word. Please think with your head, and not with your heart. I do understand how threads are spawned. You are not paying attention to the point of the argument. I am talking about parallel execution. Maybe I should have been clearer. Many, if not most, programs can't effectively use multi-threading. Most super computers work on problems that can be broken down into single units that can be worked upon at the same time. That's why supercomputers shuch as the older Cray's had an avarage speed that was close to the peak rating. Not so with massively parralel machines. The peak is MUCH higher than the average, sometimes ten times higher.

Mr. Shaw explains this here in post #189.

maxvamp
Jul 25, 2004, 02:53 AM
Please note that Bull is an adjective to a friendly dispute of a statement.

BullS**t implies aggression and is far less friendly unless prefixed with 'I Call ' .


I believe Shakespeare would agree.


Early days of multiprocessors... Common cache... One processor fills it... Second processor looks in cache for reusable code and finds none for the process it is working on... Clears cache... First processor goes back to cache to fetch work... doesn't find and as a result suffers performance penalty...

Problem got much worse with more procs. Life / Technology is much better now.

Max.

BTW: I still take a dual stance on the Threading thing. Yes, there are many, many threaded apps out there. No people don't need 4 way or more machines, unless they are running MS-Exchange.

Max.

melgross
Jul 25, 2004, 03:26 AM
I'd like to follow through on this, but it's almost 4:30 am here, so I have to finally say

Nitey, nite.

Scottgfx
Jul 25, 2004, 04:14 AM
then again, I don't use capital letters correctly either! :p

And for that, god will strike you down... Oh wait! Crap!

Colonel Panik
Jul 25, 2004, 06:21 AM
I did order it through my job its the Board Of Education In my city so whats there excuse.

You gotta be joking with grammar like that.

AidenShaw
Jul 25, 2004, 07:25 AM
The reason I called BULL was over the comment that most programs are single threaded. Aiden, do you support this comment?

I support it in the common context that most programs do their main flow of execution in a single program thread.

I support it in the sense that hidden threads created by the operating system for various support tasks do not make the main flow of the program multi-threaded.

I support it in the belief that a program is single-threaded unless the programmer designs his algorithm for parallel execution and calls services to spawn additional threads.

I support it in contrast to your first statement about "Bull!!!" which had no mention of the main problem of exploiting parellel threads in the main program to actually speed up the program.

___________________________________
Microsoft did not come into it's own as an innovator until almost 2000 with the re-write of the OS/2-NT kernel into Windows 2000.

The word "re-write" is open to interpretation, but Windows 2000 was an evolved Windows NT - not a clean start.

http://www.winntmag.com/Articles/Index.cfm?IssueID=97&ArticleID=4494

___________________________________
To connect this fact with improved technology that Apple is trying to put into the next version of the OS is failed logic taken out of context.

I see it more as a comment that a design paradigm of every operating system group has been:

First, get it working. Then get it working fast.


Apple is following the same path that Microsoft has been following - but Apple's OS design "problems" aren't as visible as were those of NT. There aren't any 8-way and 32-way PPC970 systems to bring attention to Apple's "problems".

I'd take the comment about improvements in Tiger's design to be a hint that Apple is getting ready for a 4-way (even if it's a HyperThreaded dual).

MacSA
Jul 25, 2004, 07:29 AM
For anyone technologicaly challenged such as my self, could someone please explain what all this means in simpple terms ? :eek: :eek: :eek:

broken_keyboard
Jul 25, 2004, 07:45 AM
I have the original dual 2.0 G5. If a model comes out next year with dual core 3.0, PCI express, time for an upgrade!

IBM rulez.

AidenShaw
Jul 25, 2004, 07:50 AM
For anyone technologicaly challenged such as my self, could someone please explain what all this means in simpple terms ? :eek: :eek: :eek:


The hardware rumour ("Antares") is that IBM may be developing a single chip that has two complete CPUs on it. This is called "dual-core" - referring to 2 CPU cores on one piece of silicon. It means that dual CPU Macs should get cheaper, and a quad processor might be easier to build.

It could also mean that duals could show up in lower-end systems - dual CPU iMacs or even laptops would be more feasible with the new dual-core chip.

It also means that there's a rumour that IBM will do what Intel has announced that it will do - push dual-core chips across the lineup:

http://www.xbitlabs.com/news/cpu/display/20040610151158.html

Within a year or so mainstream PCs will be dual CPU using dual-core chips. Most likely PC laptops, at least the more powerful ones, will be dual CPU.

___________________________________
The comments about SMT (Simultaneous Multi Threading, which is what Intel calls HyperThreading) describe a way where a single CPU can act like two CPUs. It's not as good as two real CPUs, but it's usually better than a single CPU and it's extremely cheap to add to the chip.

The single dual-core chip with SMT would look like a 4 CPU system, and perform like a 2.5 to 3-way system.

The SMT support is very questionable, though, since the rumour says that it's a PPC970, and the 970 does not have SMT.

____________________________________
The other discussions about threading are a digression by geeks arguing about how easy it is to use multiple CPUs to actually make your work run faster.

No need to really understand it - but we're talking about the root reasons that a dual CPU system is not twice as fast as a single, and why a quad wouldn't even to be close to 4 times as fast.

zac4mac
Jul 25, 2004, 08:37 AM
Aiden, Sun Baked and Morpheus(by proxy) et al, for a very interesting read.
Will we ever see this thing in a Mac - who knows??

I got quite a performance leap going from a DP550 G4 to a new G5 - 2x2 last year and I don't see a need, personally, for more power soon. But "soon" is a relative term.
The difference, on current equipment, between single and dual procs is significant. If indeed, more is better, then let's have it.

Z

macsrus
Jul 25, 2004, 08:46 AM
macsrus:
Actually NT supported 4 processors way back in Windows NT 3.1 Advanced Server.....
And Believe it or not Unisys was building 32 processor NT servers using Intel pentiums back in 1995/96 (during the NT 3.51 days)

me:
Yes at 3 they WERE supporting four. But when it first came out, you know, ver 1.0, they were supporting two.


NT 3.1 was the first Windows NT (There wasnt a verson 1.0 thru 3.0)


Another thing Microsoft NOT IBM wrote OS2.... IBM was tricked by Bill and company again on this one..... They wrote OS2 while writting NT at the same time as a replacement.

IBM got mad and took all of the OS2 source code and cancelled the contract with MS for OS2 development....

relimw
Jul 25, 2004, 09:07 AM
The next machines will have PCI Express (PCIe) graphics, no doubt.
Ahh, now see I'm not too much on my PCI standards these days, how much faster is PCI Express over AGP8x and the various other PCI incarnations?

relimw
Jul 25, 2004, 09:10 AM
Well, if you look at military classifications, X usually means "experimental". I realize that most people probably would overlook that, but I think the G is cool. I don't know what it actually means, but I kind of think of it as "generation", since one could make the argument that the PPC 601/603/603E was a kind of G1, and the 604/604e was a sort-of G2.
I know, they should just start naming it: IS1, IRS2, IRRS3, etc (Intel S*cks, Intel Really S*cks, etc) ;p

relimw
Jul 25, 2004, 09:28 AM
more power! Great! Wait! so few really have that serious a need for that kind of power, software just isn't keeping up. More than that, c'mon! Apple's alreaedy got the graphics crowd, etc.
But yet some people *really* need that kind of power. Like those that are building supercomputer clusters, or those that are clustering for mainframe-like, multiple user situations. By now you should know that more than just yourself can make use of your Mac computer if you want/need too. I'd love to have such a beast to work, it'd make things a lot faster, and simpler for myself, and my multiple users to work :)

Yet those that 'just' check email, or do simple word processing, can go back to using the old DOS-based machines or equilivant, since really, you don't need that much power if the software you are using is coded efficiently. I mean, really, who the heck needs all the bells and whistles that openoffice or M$ office has? Down with the stupid talking paperclip! Long live DOS! ;)

Trekkie
Jul 25, 2004, 09:34 AM
Excellent, this is what they should be doing. The 1MB caches would be wonderful, 512k is yesterday's high-end, not today's.

Not necessarily.

Cache is used to speed up the processor by keeping it in cache and not in main memory because main memory was slower than cache by orders of magnitude.

However if you have kick ass speed to main memory (like, oh, 1.25GHz the PM G5 2.5GHz does) cache gets in the way in certain workloads, so keeping the cache smaller is not a bad thing....

relimw
Jul 25, 2004, 09:34 AM
I really don't like this thinksecret article. Sure having a dual Core PowerPC chip would be great but what it all boils down to is unsubstantiated rumors. Anyone could say that IBM is working on Dual Core chips and be right as they probably are, as every other company is doing the same.
Well, first off, IBM is working with, and producing dual core chips, just never before in the lowly PowerPC line. We'll just have to wait and see when they do come out. I'd like a dual core, dual proccessor 3GHz system announced in March '05, shipping immediately :) Will I get it from Apple? Prolly not :)

MarcelV
Jul 25, 2004, 09:44 AM
...now if only OS X would take advantage of all that 64 bit goodness I'd be giddy as a school girl.

They are able to start production in January. Consider the timing of Tiger, which will have full 64-bit support. I think they will build in support for Dual Core. I think 2004 and 2005 will be very interesting years for Apple's hardware development.

AidenShaw
Jul 25, 2004, 10:00 AM
Ahh, now see I'm not too much on my PCI standards these days, how much faster is PCI Express over AGP8x and the various other PCI incarnations?

AGP8X peaks at 2.1 GB/sec.

PCI Express (PCIe) is like HyperTransport - different width busses can be used for differnt needs.

The current PCIe chipsets use "16x" width for PCIe graphics, giving a peak of about 8 GB/sec. (4GB/sec up, 4GB/sec down) 32x is also in the spec, for 16 GB/sec.

So it's double to 4 times AGP8X, depending on whether your load is bidirectional.

Other PCI speeds:

o 32-bit/33 MHz - 133 MB/sec
o 32-bit/66 MHz - 266 MB/sec
o 64-bit/33 MHz - 266 MB/sec
o 64-bit/66 MHz - 533 MB/sec
o 64-bit/100 MHz - 800 MB/sec (PCI-X)
o 64-bit/133 MHz - 1066 MB/sec (PCI-X)
o 64-bit/266 MHz - 2133 MB/sec (PCI-X 2)
o 64-bit/533 MHz - 4266 MB/sec (PCI-X 2)

PCI Express:

o x1 - 250 MB/sec bidirectional - 500 MB/sec total
o x2 - 500 MB/sec bidirectional - 1000 MB/sec total
o x4 - 1000 MB/sec bidirectional - 2000 MB/sec total
o x8 - 2000 MB/sec bidirectional - 4000 MB/sec total
o x16 - 4000 MB/sec bidirectional - 8000 MB/sec total
o x32 - 8000 MB/sec bidirectional - 16000 MB/sec total


Unlike HyperTransport, though, PCIe is also a card interconnect - you'll be able to get PCIe network, I/O and other cards. Initially high end cards will have this (you can find InfiniBand, 10 Gbps Ethernet, FibreChannel PCIe cards today) and a PCIe <-> PCI bridge will be used to put both PCI and PCIe slots on the motherboards.

relimw
Jul 25, 2004, 10:57 AM
AGP8X peaks at 2.1 GB/sec.

PCI Express (PCIe) is like HyperTransport - different width busses can be used for differnt needs.

The current PCIe chipsets use "16x" width for PCIe graphics, giving a peak of about 8 GB/sec. (4GB/sec up, 4GB/sec down) 32x is also in the spec, for 16 GB/sec.

So it's double to 4 times AGP8X, depending on whether your load is bidirectional.

Other PCI speeds:

o 64-bit/100 MHz - 800 MB/sec (PCI-X)
o 64-bit/133 MHz - 1066 MB/sec (PCI-X)
o 64-bit/266 MHz - 2133 MB/sec (PCI-X 2)
o 64-bit/533 MHz - 4266 MB/sec (PCI-X 2)


Yikes! That is a heck of a lot faster than what we're use to on the Mac. So I'd guess that skipping PCI-X 2 would be in Apple's (and our's) best interest. I wonder tho, how many changes would Apple be willing to commit too at one time? Maybe at the next refresh, up the machines to PCI Express and a small speed bump? Then the follow-on refresh would be dual core processor(s) and some other wizzbang snazzy feature?

G4-power
Jul 25, 2004, 11:22 AM
apparently you are not familiar with IBM's PowerTune technology. They don't need to design a mobile processor if they can get it working correctly, as it will effectively be better than a seperate mobile design, as it will be scalable from 1/64 of clock frequency to full frequency in 3 cycles. this will cut heat and power consumption drastically, because the processor will only be running hard when you're doing heavy computing, and will enter a deep nap mode in between., i.e. when you're typing etc.

Heh, wonder how PowerTune scales when doing massive computing like Folding@Home... Will it use 100% processor speed (and burn your thighs) with a process that uses only free CPU cycles? Or 50%? The maximum heat should still be quite low, that you won't burn your thighs with F@H.

Reality
Jul 25, 2004, 11:27 AM
It says both cores will share a 1ghz bus...right now each CPU in a dual 2.5 gets 1.25ghz of it's own bandwith, but with that each 3ghz core would have to share only 1ghz, isn't that a step backwards? Or are we getting dual dual-cores? :confused: :(


That point discredits this story. They would not take a step back like that, I hope not anyways.

No dual cores anytime soon guys, sorry.


Visit my site, http://reality.spymac.net
Red Pipe Backgrounds, spice up your desktop today.
Free 30 inch desktops.

gekko513
Jul 25, 2004, 11:34 AM
AGP8X peaks at 2.1 GB/sec.
... PCIe graphics, giving a peak of about 8 GB/sec...
However, PCIe has higher latency than AGP. That is the reason why current graphics card perform equal or worse on PCIe compared to AGP.

Masao[RY]
Jul 25, 2004, 11:45 AM
Maybe they'll call it the G5.2.

It would make sense since it is still a 970, dual core, and the third version of the chip (the original and then the 90nm version).
G5se, G5fx, G5 Pro, and so on. Apple has a lot of room to play with the marketing a bit. I do hope they don't call it a G6 since it is way to soon for such. I'd prefer G5fx or something of that nature. Anyway, if this bares fruit I will just have to work like a mofo one summer just to get one :D

gekko513
Jul 25, 2004, 11:45 AM
That point discredits this story. They would not take a step back like that, I hope not anyways.

No dual cores anytime soon guys, sorry.


Visit my site, http://reality.spymac.net
Red Pipe Backgrounds, spice up your desktop today.
Free 30 inch desktops.
It's not really a step backward. Dual channel DDR 400 isn't able to fill the 1GHz bus. I think the only data-traffic that's able to fill the current bus capacity is cache lookup on the second processor on a dual setup. The current config has 512KB cache on the processor + 512KB on the other processor.

If they set up that Dual dual-core 3GHz system with 1GHz bus, then each processor will have 1MB of cache by itself + 1MB from the second core on the chip.

So the current config has 512KB of high-speed cache + 512KB@1.25GHz, but the described future system will have 2MB of high-speed cache + 2MB@1GHz. So it would be a huge step forward.

I'm assuming now that one core is allowed to peak at the other core's cache, at least for reads, I think that's the way it works in todays dual configs.

G4-power
Jul 25, 2004, 11:48 AM
']G5se, G5fx, G5 Pro, and so on. Apple has a lot of room to play with the marketing a bit. I do hope they don't call it a G6 since it is way to soon for such. I'd prefer G5fx or something of that nature. Anyway, if this bares fruit I will just have to work like a mofo one summer just to get one :D

Yeah, we'll see. But I don't think it looks cool, "PowerMac G5fx". The "fx" is just not clear enough. And since there is already a fx in use, the PPC 970fx, that would confuse things a bit.

maxvamp
Jul 25, 2004, 12:01 PM
Melgross and Aiden,

You both are trying hard to take the original comments and skew them out of context to make an untrue point.

Threaded tasking happens all the time in many programs. Mail.app collects mail from multiple mailboxes at once while indexing and doing other tasks. Safari does not download graphics sequentially. Framworks used to build a program, such as a UI **is** part of that program.

DOS is dead. People expect more from programs.

We must agree to disagree at this point, as I doubt that we will change each other's minds.

MacsRus :

One day, we'lll have to start an OS/2 thread, as I am sure that the messy history of this doomed, yet excellent OS would take 200 posts by itself.

Thanx all for the lively debate. I look forward to doing it again.

Max.

itsa
Jul 25, 2004, 12:31 PM
Now if we could just get them to offer G5's with a decent graphics card.
The least they can do is sell them without cards so you can use that money for one worth having. It sucks that they force you to buy one that you don't even want.

AidenShaw
Jul 25, 2004, 12:45 PM
Threaded tasking happens all the time in many programs. Mail.app collects mail from multiple mailboxes at once while indexing and doing other tasks. Safari does not download graphics sequentially. Framworks used to build a program, such as a UI **is** part of that program.

I don't think that we disagree that much. I specifically mentioned multi-threaded (overlapped) I/O (mail/safari) in my posts. I also cited video as an example of an application that lends itself to threading across multiple CPUs.

In none of your examples, however, is there any hint at exploiting multiple CPUs.

Does mail.app collect mail faster on a dual? Not appreciably, the task is network and I/O limited!

Does Safari download graphics faster on a dual? Again not significantly - it's a network limited problem.

The original post was talking about the often hard problem of taking a compute intensive task and breaking it into multiple threads to run simultaneously on multiple processors, thereby completing the overall job much faster.

Your statements about overlapped I/O and threaded UI frameworks are true, but you're also skewing the original point with unrelated examples.

ClimbingTheLog
Jul 25, 2004, 01:12 PM
(Or is OSX.4 your shorthand for Tiger?

That seems pretty easy to dissect:

Mac OS X 10.4
(we know it's mac here)-->
OS X 10.4
(factor out redundant 10)-->
OS X.4
(take out space)-->
OSX.4

A nice shorthand - of course Tiger is the same number of characters...

ClimbingTheLog
Jul 25, 2004, 01:15 PM
You mean like "Windows XP Pro"? ;) Yah, I agree, 'X' is used way too much nowadays - OS X is quite enough for me, thank you very much... :cool:

No, XP is just egotistical arrogance from Microsoft.

XP is the traditional Christian abbreviation for "Jesus Christ", from Latin.

Microsoft constantly thinks of itself as the underdog; see Cringely's writings on the subject.

So, Windows XP, the first decent core OS aimed at consumers was going to be Microsoft's Saviour, or "the second coming" of Windows.

So, Microsoft Windows XP.

'P' clearly doesn't stand for 'Professional', since we have 'Microsoft Windows XP Professional' as a marketed product.

ClimbingTheLog
Jul 25, 2004, 01:19 PM
OS X supports two logical processors. It's possible, from what I've read, that 10.4 will support four. The idea that it's easy to continue support up to sixteen, or higher, is wrong.

It's not easy but Mach is purpose-built for it. Mach is the layer under Unix on Mac OS X that does the interfacing with hardware, manages processes, memory, etc. It's been used for Massively Parallel computing, for instance it ran the Cenju-4 MPP machine from NEC which had 1024 processors.

That's not to say you can just load 1024 processors into a Mac - there are bsd kernels, compile-time options, gcc work to be done, etc, but the foundation is there to be built upon.

If Tiger does ship with the reentrant kernel things will get much more interesting.

ClimbingTheLog
Jul 25, 2004, 01:21 PM
As of OSX.3, the OS supports at least two physical processors. I am not aware that it supports any logical ( hyper threaded ) processors.

If SMT works on the Mac like it does on Intel the OS won't know there's an SMT processor - the BIOS presents it as two processors.

ClimbingTheLog
Jul 25, 2004, 01:24 PM
Is mail.app collect mail faster on a dual? Not appreciably, the task is network and I/O limited!

On my 1.2 GHz G4 AppleMail is CPU bound. It downloads much faster than it indexes. If I had 4 or 8 CPU's, real or virtual it would busy one with fetching, leave UI on another, and spend the rest indexing, ideally.

It looks like Mail in Tiger will use OS-level indexing which should also help with speed, assuming it's based on V-Twin.

AidenShaw
Jul 25, 2004, 01:31 PM
If SMT works on the Mac like it does on Intel the OS won't know there's an SMT processor - the BIOS presents it as two processors.

The BIOS also creates a SRAT (Static Resource Allocation Table) as part of the ACPI enumeration. The SRAT contains information that describes physical and logical processors.

If the system doesn't look at the SRAT, you are right that it appears to be two physical processors. (That's why Windows 2000 sees and runs with the two processors - it doesn't know about the SRAT.)

The application programmer can check the GetLogicalProcessorInformation (http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dllproc/base/getlogicalprocessorinformation.asp) API to learn which processors are real, which are logicals sharing the same real, where memory is connected (in a ccNUMA system),....

AidenShaw
Jul 25, 2004, 01:35 PM
On my 1.2 GHz G4 AppleMail is CPU bound. It downloads much faster than it indexes.

That's too bad ;) .

On the other hand, this could be a good example of the problems and successes of multithreading.

If indexing a mailbox is CPU-intensive, and if each mailbox is indexed separately - then threading will be easy and effective. Each mailbox can be assigned a different thread, and on an SMP system the mailboxes can be indexed in parallel. Fairly easy to program, but speeds the real system up noticeably.

On the other hand, if there's only one mailbox - threading doesn't do much for you.

On the third hand - if there are multiple mailboxes and a single index, it can be done somewhat, but it's more work. Each mailbox can be separately indexed (in parallel), and then when all the individual indices are ready they can be merged into the single master index. The merge, of course, is serial and can only be done in one thread (unless you wanted to use a recursive, threaded parallel merge - which would still end up at the end as a single thread).

Note that I've never claimed threading is bad or is not done. I objected to the "Bull!!!" statement and its implication that creating additional threads is trivial. It is *not* trivial to get real speedups with multi-threading in many applications.


If I had 4 or 8 CPU's, real or virtual it would busy one with fetching, leave UI on another, and spend the rest indexing, ideally.

Typically it would just spawn as many threads as it can, and let the OS schedule the threads as they need CPU.

Dedicating processors (logical or physical) to particular threads is usually a net loss compared to letting the OS do dynamic scheduling.

Surreal
Jul 25, 2004, 03:11 PM
It says both cores will share a 1ghz bus...right now each CPU in a dual 2.5 gets 1.25ghz of it's own bandwith, but with that each 3ghz core would have to share only 1ghz, isn't that a step backwards? Or are we getting dual dual-cores? :confused: :(

both cores HAVE a 1MB L2 cache. each. individually. :)

skunk
Jul 25, 2004, 03:12 PM
http://www.linuxinsider.com/story/34994.html

Wintel's Dilemma and Apple's Problem

For the last three weeks I've been talking about the impact the new Sony, Toshiba and IBM cell processor is likely to have on Linux desktop and datacenter computing. The bottom line there is that this thing is fast, inexpensive and deeply reflective of very fundamental IBM ideas about how computing should be managed and delivered. It's going to be a winner, probably the biggest thing to hit computing since IBM's decision to use the Intel 8088 led Bill Gates to drop Xenix in favor of an early CP/M release with kernel separation hacked out.

Sun has the technology to compete. Its throughput-computing initiative -- coupled with some pending surprises on floating point -- give it the hardware cost and performance basis needed to compete on software where it has the best server-to-desktop story in the industry.
No one else does. Microsoft's software can't take x86 beyond some minor hyperthreading on two cores without major reworking -- and Itanium simply doesn't cut it. The Wintel oligopoly could spring a surprise -- a multicore CPU made up from the Risc-like core at Xeon's heart, along with a completely rewritten Longhorn kernel to use it. But no one has reported them stuffing this rabbit into their hat. So, for now at least, they seem pretty much dead ended.

If, as I expect, the Linux community shifts massively to the new processor, Microsoft and its partners in the Wintel oligopoly will face some difficult long-run choices. It's interesting, for example, to wonder how long key players like Intel and Dell can survive as stand-alone businesses once the most innovative developers leave them to Microsoft's exclusive mercy.

Wintel's dilemma is, however, a fairly long-term issue. Much closer at hand is Apple's immediate problem. Just recently Steve Jobs has had to apologize to the Apple community for not being able to deliver on last-year's promise of a 3-Ghz G5 by mid 2004. IBM promised to make that available, but has not done so.
A lot of people have excused this on the grounds that the move to 90-nanometer manufacturing has proven more difficult than anticipated, but I don't believe that. PowerPC does not have the absurd complexities of the x86, and 90-nanometer production should be easily in reach for IBM. The cell processor, furthermore, is confidently planned for mass production at 65-nanometer sizes early next year.

This will get more interesting if, as reported on various sites, such as Tom's Hardware, IBM has been burning the candle at both ends and also will produce a three-way, 3.5-GHz version of the PowerPC for use on Microsoft's Xbox.
Whether that's true or not, however, my belief is that IBM chose not to deliver on its commitment to Apple because doing so would have exacerbated the already embarrassing performance gap between its own server products and the higher end Macs. Right now, for example, Apple's 2-Ghz Xserve is a full generation ahead of IBM's 1.2-GHz p615, but costs about half as much.


Consequences of Apple Decision


Unfortunately this particular consequence of Apple's decision to have IBM partner on the G5 is the least of the company's CPU problems. The bigger issue is that although the new cell processor is a PowerPC derivative and thus broadly compatible with previous Apple CPUs, the attached processors are not compatible with Altivec and neither is the microcode needed to run the thing. Most importantly, however, the graphics and multiprocessor models are totally different.

As a result, it will be relatively easy to port Darwin to the new machine, but extremely difficult to port the Mac OS X shell and almost impossible to achieve backward compatibility without significant compromise along the lines of a "fat binary" kind of solution.
In other words, what seemed like a good idea for Apple at the time, the IBM G5, is about to morph into a classic choice between the rock of yet another CPU transition or the hard place of being left behind by major market CPU performance improvements.

Look at this from IBM's perspective and things couldn't be better. Motorola's microprocessor division -- now Freescale Semiconductor -- is mostly out of the picture, despite having created the PowerPC architecture. Thus, if Apple tries to stay with the PowerPC-Altivec combination, it can either be performance starved out of the market or driven there by the costs of maintaining its own CPU design team and low-volume fabrication services.
If, on the other hand, Apple bites the bullet and transitions to the cell processor, IBM will gain greater control while removing Apple's long-term ability to avoid having people run Mac OS on non-Apple products. Either way, Apple will go away as a competitive threat because the future Mac OS will either be out of the running or running on IBM Linux desktops.


Apple-Sun Partnership


I think there'll be an interesting signal here. If IBM thinks Apple is going to let itself be folded into the cell-processor tent, it will probably allow as many others to clone the new Cell PC as it can make CPU assemblies for. If, on the other hand, IBM thinks Apple plans to hang in there as an independent, it might just treat the Cell PC as its own Mac and keep the hardware proprietary. Notice, in thinking about this, that they don't have to make an immediate decision: There will be CPU assembly shortages for the first six months to a year if not longer.

So what can Apple do? What the company should have done two years ago: Hop into bed with Sun. Despite its current misadventure with Linux, Sun isn't in the generic desktop computer business. The Java desktop is cool, but it's a solution driven by necessity, not excellence. In comparison, putting Mac OS X on the Sunray desktop would be an insanely great solution for Sun while having Sun's sales people push Sparc-based Macs onto corporate desktops would greatly strengthen Apple.
Most importantly, Sparc is an open specification with several fully qualified fabrication facilities. In the long term, Apple wouldn't be trapped again, and in the short term the extra volume would improve prospects for both companies. Strategically, it just doesn't get any better than that.


Some Important Footnotes


I am not suggesting that Sun buy Apple, or Apple buy Sun. Neither company has adequate management bandwidth as things stand. I'm suggesting informed cooperation, not amalgamation.

The transition to Sparc would be easier than the transition to Cell. It might look like the bigger change, but the programming model needed for cell is very different, whereas existing Mac OS software, from any previous generation, need only be recompiled to run on Sparc.
In particular, the graphics libraries delivered with the Cell PC will likely focus on Gnome-KDE compatibility to make porting applications for them easy, but Apple would have to redo its interface-management libraries at the machine level -- something it would not face in a move to Sparc where PostScript display support is well established.

In addition, existing Sun research on compiler automation suggests that multithreaded CPUs like Niagara and Rock could automatically convert PowerPC and even MC68000 executables to Sparc on the fly -- meaning that "fat binaries" would not be needed, although a Mac OS 9.0 compatibility box would probably still make sense.


Sun's Throughput-Computing Initiative


People I greatly respect tell me that Sun's throughput-computing direction isn't suited to workstations like the Mac where single-process execution times are critical to the user experience. The more I study this question, the more I disagree. Fundamentally this issue is about software, not hardware.
Consider, for example, what could be achieved with the shared-memory access and eight-way parallelism inherent in the lightweight process model Sun is building into products like Niagara. This won't matter for applications like Microsoft Word, where the 1.2-GHz nominal rate is far faster than users need anyway, but can make a big difference on jobs like code compilation, JVM operations or image manipulation in something like Adobe's Photoshop.

Given the much higher cache hit rates and better I/O capabilities offered by the relatively low cycle rate, theory suggests that truly compute-intensive workstation software could hit somewhat better than 85 percent system use -- meaning that an eight-way Niagara-1 running at 1.2 Ghz would easily outperform a Pentium 4 at 8 GHz.
Making that happen would, of course, take serious software change, but if the preprocessors now thought to be under development at Sun work as expected, most of that would be automated -- thereby greatly reducing the barriers to effective CPU use on the Mac for PC-oriented developers like Adobe.

Should Apple be thinking different?

jouster
Jul 25, 2004, 03:38 PM
http://www.linuxinsider.com/story/34994.html

[/indent]

Should Apple be thinking different?

I don't have the CPU smarts to address this myself, but I will point out that this article has been extensively rubbished on various forums.

Sun Baked
Jul 25, 2004, 03:42 PM
http://www.linuxinsider.com/story/34994.html

"In other words, what seemed like a good idea for Apple at the time, the IBM G5, is about to morph into a classic choice between the rock of yet another CPU transition or the hard place of being left behind by major market CPU performance improvements"

Should Apple be thinking different?Did they completely forget that Apple hopped on the POWER branch of the tree intead and off PPC branch (sort of), and IBM does have a long-term plan for those POWER processors.

So it doesn't look like the G5 is an instant dead end after the current G5 runs it's course -- because IBM has already announced the GR-UL (aka, Power5-UL) and given an outline for the POWER6.

So it doesn't look like Apple will be left behind by IBMs CPU improvements.

These CPUs do support the majority of the PowerPC ISA, and the addition of Altivec did make the transition rather painless for Apple to this class of CPU -- except for VirtualPC.

Take a midrange/server class chip and drop some of the expensive bus and interconnects (that made it too expensive) and drop it in a desktop and soon into a iMac.

A few years ago the thought of putting a Power4-class CPU in an iMac would have been sort of silly, but with IBM looking to deliver a dual-core PPC970 it doesn't look silly any more.

POWER to the people (http://www-106.ibm.com/developerworks/library/l-powhist/#IDA2ME0C)

StudioGuy
Jul 25, 2004, 04:37 PM
No, XP is just egotistical arrogance from Microsoft.

XP is the traditional Christian abbreviation for "Jesus Christ", from Latin.



Getting quite OT, but I think that is urban legend.
XP is supposedly for "experience".

however, the legend is funnier :D

puckhead193
Jul 25, 2004, 07:30 PM
So in english this means..... G6 or powerbook G5 :p

wdlove
Jul 25, 2004, 07:49 PM
Getting quite OT, but I think that is urban legend.
XP is supposedly for "experience".

however, the legend is funnier :D

This is the first time that I have heard XP being an abbrevation for Jesus Christ. OT is an abbreviation for the Old Testament though.

I'm more interested in knowing if this dual core PPC 970 will be worth waiting till MWSF '05.

StudioGuy
Jul 25, 2004, 08:10 PM
This is the first time that I have heard XP being an abbrevation for Jesus Christ. OT is an abbreviation for the Old Testament though.

I'm more interested in knowing if this dual core PPC 970 will be worth waiting till MWSF '05.

I meant "off topic" :D , but nice transition!

I agree with the worth, but are tired of waiting - time to buy for us if we can afford it. :rolleyes: they'll always be something better making us want to wait...

ffakr
Jul 25, 2004, 11:13 PM
I don't have the CPU smarts to address this myself, but I will point out that this article has been extensively rubbished on various forums.
Well, I don't have the patience to read the whole thing but I skimmed to this bit...
A lot of people have excused this on the grounds that the move to 90-nanometer manufacturing has proven more difficult than anticipated, but I don't believe that. PowerPC does not have the absurd complexities of the x86, and 90-nanometer production should be easily in reach for IBM.
That's enough to tell me the author doesn't have a freaking clue.
The problems with moving to .09 micron have NOTHING to do with the complexity of the instruction set and everything to do with physics. The .09 micron transition has gone poorly for IBM and Intel because, though they expected quite a few problems with current bleed, they didn't expect the level of signal crosstalk that they have been seeing.
I don't give two craps what the author "believe[s]" or doesn't, but he/she shouldn't be passing themselves off as someone worth listening to if they can't bother to do some basic research.

macsrus
Jul 26, 2004, 06:26 AM
[QUOTE=skunk
Should Apple be thinking different?[/QUOTE]



Yeah Right..... Thats all Apple needs to do is to allign itself with a dead end company like Sun.

The Sparc Processor is a slow piece of junk..... And anyone who says otherwise doesnt have a clue what they are talking about...

Sun has consitantly been producing overpriced hardware, And has consistantly been losing customers.

Their hardware provides the worst cost/performance ratio in the industry.
And dont even talk about reliability.... Because i have first hand knowledge of their underhanded dealings with me as a customer...
We had hundreds of their systems and they would crash when the wind blew....
We even had to sign a NDA or they wouldnt fix their problem...
Im so freaking glad that we finally got rid of every SUN piece of junk that we owned.

Frobozz
Jul 26, 2004, 10:21 AM
Try playing any latest game and tell PC gamers that... nuff said

I'd bet a dual 2.5GHz with a Geforce6800 Ultra would be able to keep up nicely, even if it might still be slightly slower. Keep in mind that the Alienware machines we're speaking of are finely tuned for gaming, whereas the PowerMac is finely tuned for creative production, etc.

I think that Macs are highly compeitive on raw power these days. The only thing I can possibly complain about is out of my control-- Half-life 2... will it ever come to the Mac? We know Doom3 is, but my DP 1GHz G4 isn't likely to handle it. :-)

dxp4acu
Jul 26, 2004, 01:47 PM
This is the first time that I have heard XP being an abbrevation for Jesus Christ. OT is an abbreviation for the Old Testament though.

I'm more interested in knowing if this dual core PPC 970 will be worth waiting till MWSF '05.

It's not XP. It is Chi Rho, which is the Christogram (The symbol for Christ). The latin symbols for those two are like an X and a P.

The P is usually in the X. Look it up, pretty cool story. The christogram came to Constantine in a vision, and could have made him a Christian.

http://www.forumancientcoins.com/forvm/Articles/Constantine_Ch_Rho_files/Constantine_Ch_Rho.htm

wdlove
Jul 26, 2004, 01:48 PM
I meant "off topic" :D , but nice transition!

I agree with the worth, but are tired of waiting - time to buy for us if we can afford it. :rolleyes: they'll always be something better making us want to wait...

Definitely the waiting is becoming very tedious. As of now the wait has been over a year.

jbh001
Jul 26, 2004, 02:55 PM
Why not call it the G5^2

MikeBike
Jul 27, 2004, 01:00 AM
nuckinfutz:


Lemme give you a secret to estimating the feasibility of a rumor relating to a large R&D investment: you need to establish that the people spending on R&D are going to recoup that cost. For years now I've used that simple rule to correctly predict the non-arrival of crasy things that everyone else seemed to swallow hook, line and sinker. Apply that rule to the idea of there being both a 970-dual and a Power5-lite. When I apply it, I see that the two chips aim at the same market, harming profit. Note, neither chip is aimed at the market that will be increasingly held by the 970fx, so I see no conflict there.

It is possible that a G5-dual could occupy the market for a year or more before being replaced by a Power5-lite. There are some questions there, I'd wonder if going from true dual core to SMT would be a step backwards, and also I wonder why Apple would be designing for SMT already.


The only problem with your theory is you've forgotten the potential reward for success in this area. The Huge Intel market would eventually be up for grabs. A market IBM once owned... :cool:

MikeBike
Jul 27, 2004, 01:07 AM
Ugh...

Game consoles do not provide a better gaming experience than a PC dammit!

Its the worst Mac cop-out ever!

Games are what keep me locked to PC, thats for sure.

One day my precious... (http://www.apple.com/powerbook/)

You should try a "game" that could pay off for you in the future...
GarageBand for example.
I hear it's a good gig, being a rock and roll star.

ddtlm
Jul 27, 2004, 01:46 AM
MikeBike:

Heh, the only thing people predict the demise of more often then Apple is x86. ;) Won't happen.

MikeBike
Jul 27, 2004, 02:00 AM
MikeBike:

Heh, the only thing people predict the demise of more often then Apple is x86. ;) Won't happen.

I'm not predicting the demise of x86.
Why is AMD chasing after Intel?
Seems to me the potential money pot is one big reason.

Smart people go after the money.
Ibm should be looking at every Apple sale,
even if they lose a potential Linux sale,
as a win. Ibm should be busting their collective butt to make Apple succeed. Linux servers on Power / Apple on PPC / even AMD on x86 put money in IBM's pockets.

Frobozz
Jul 27, 2004, 09:20 AM
http://www.linuxinsider.com/story/34994.html

[/indent]

Should Apple be thinking different?

The article you quote is indeed thought provoking, but I chalk it up to the wishful thinking of one individual. There really is no basis for his prioritization of the cell architecture over the G5. To top that off, IBM is a bit more intelligent than that-- if cell was ever going to be something they could sell more of to Apple, they would have made it that way. IBM is a megacorporation looking for profits above all else.

I think his conjecture that the 90nm wall was totally off base. While it is true that perhaps the die size and etching features of the G5 may be less complex than a Intel x86 chip, that has nothing to do with the problem they cite. I'm not a chip architect, but let's assume my previous statement is indeed true. It doesn't make good business to flat out lie to your investment community about such things as they are probably audited about statements like that. If IBM said they ran into lamination issues or heat dissipation problems (which caused delamination?) then that's 99.9% likely to be the cause. Again, they are a corporation out to make money FIRST, so there's no reason to withhold anything from a buyer.

RIP
Jul 27, 2004, 09:48 AM
Try playing any latest game and tell PC gamers that... nuff said

Games are childs play... nuff said

jouster
Jul 27, 2004, 11:04 AM
Games are childs play... nuff said

Yes they are. Multi-billion dollar per year child's play.

bonk
Jul 27, 2004, 02:45 PM
so a G5 with two of this new chip and a 6800 ultra would be a...



dual dual dual dual G5? ha!



anybody say that yet?

Little Endian
Jul 28, 2004, 09:11 AM
I am quite surprised that no one has really mentioned how in the world IBM and Apple are supposed to get over the heat Problems produced by a Dual Core PowerPC 970 series chip @ 3Ghz per core. Apple has allready seen it fit to put Liquid Cooling in to cool a Single core PowerPC 970 @ 2.5Ghz. Dual Core 3Ghz sounds pretty darn hot to me and I am curious how the heat problem will be overcome. Don't tell me it will be at 65nm because I doubt we will see 65nm until the ending of 2005 at the earliest and this Rumor article states that it would see Production in early 2005 which would mean on a 90nm fab. Remember Dual Core was meant to control PowerConsumption and heat while striving to offer better performance. From sources I have read most CPU manufactures planning to produce Dual Core Versions of their Single Core Chips would actually be clocked slower per core than their Single core counterparts that is where the Power savings and heat is supposed to be reduced yet offer better performance. An example would be The fastest Single Core Athlon 64 might top out at 3Ghz while the Dual Core version will run at 2Ghz per core in "effect" producing 4Ghz combined which would in most cases offer increased performance with less heat and Power Consumption. The same Goes with Intel and their upcoming Dual Core Version of Xeon Processors as is the same for IBM and the 970. Basically what it boils down to is if they can't even give us a Single core 3Ghz chip what makes one think they can Produce a Dual Core 3Ghz chip and have it ready sometime early next year?

AidenShaw
Jul 28, 2004, 09:23 AM
I am quite surprised that no one has really mentioned how in the world IBM and Apple are supposed to get over the heat Problems produced by a Dual Core PowerPC 970 series chip @ 3Ghz per core.

Do remember, though, that one of the reasons for the liquid cooling is that the PPC970fx chip is physically rather small. The problem is both the total amount of heat generated (watts), and the concentration of the heat (watts/mm^2).

A dual core chip would physically be about twice the size of the single core, so the heat density wouldn't be much different from the current chips. While more heat is generated, it isn't a new problem.

Bigger radiators/fins.... :rolleyes:

Also, much of the commentary about dual cores refers to the slowing of the MHz increases. So, the dual core 3.0GHz could be thought of in lieu of a 4GHz single core, therefore producing much lower heat densities than the 4GHz chip.


...[dual core] which would in most cases offer increased performance...

For multi-threaded MP-aware applications, yes. And for running multiple non-MP-aware apps simultaneously.

It would not help the single non-MP-aware app, that would run slower.

I know that you say "most", but the words "many" and "some" would also apply.

A bit of skepticism about the benefits of multi-processing is good to have. You'll win sometimes with this approach (that is, the slower dual-core), but sometimes you'll lose.