PDA

View Full Version : The PowerPC 980 Roadmap


MacRumors
Nov 20, 2003, 02:53 PM
The PowerPC 980 Processor is a rumored chip to be coming from IBM as a followup to the PowerPC 970. Just as the PowerPC 970 is based on the Power4 Processor, the PowerPC 980 is rumored to be based on the upcoming Power5 Processor.

First hints of a new derivative chip came from this extensive roadmap of the PowerPC, Apple, Motorola and IBM's relationships from November 2002. At that time, the PowerPC 970 had been announced by IBM but not yet publicly endorsed by Apple. Apple later announced PowerPC 970 based PowerMacs in June of the following year (2003).

Today, Appleinsider reports that the PowerPC 970 will max out at 2.6GHz with the Power 980 coming due in Q3 of 2004 to bring the PowerMac line up to and beyond 3.0GHz. The 980 is reportedly a 90nm chip and may be branded as the G6. Appleinsider also comments on updates to the subsystems of the upcoming PowerMacs offering significant performance gains.

This "980 @ 3GHz" timeframe was echoed by a detailed report from July 2003. In that report, Apple and IBM were said to be working on parallel development of the Power5 and PowerPC 980. That report also claimed that the 3GHz PowerMac would not arrive by way of the 970 -- which was said to top out at 2.6-2.8GHz, while the 980 was expected to start at 2.6-3GHz (from the July report).

A subsequent report noted that Apple was already given PowerPC 980 samples in October, and also made predictions about 90nm 970 based PowerMac updates in February at 2.5-2.8GHz. The also offered significant details regarding cache, SPEC scores, and bus multipliers on the as-yet-unannounced PowerPC 980.

Without any tangible releases from these rumors, it's hard to gauge their accuracy, though the information appears to be consistent.

yoman
Nov 20, 2003, 02:54 PM
5 GHZ man...According to the Apple Insider article the Power 980 could max out this high.

Edit: just adding explanation to 5GHz remark

machan
Nov 20, 2003, 02:55 PM
I seriously doubt they'll brand anything as "G6" by the middle of next year. Still, this is good news as I've been planning for a year on buying a new tower at the end of next summer.

g30ffr3y
Nov 20, 2003, 02:56 PM
i dont even have a g5 yet... they cant put out a g6... holy crap!!!

Wash!!
Nov 20, 2003, 02:56 PM
As long as they are making faster macs they can called what ever they want..:)

mvc
Nov 20, 2003, 02:57 PM
I'm just Looking forward to Rev.B.

Don't think i can wait for the 980 tho, much as it sounds like a total killer.

Eric_Z
Nov 20, 2003, 02:58 PM
For some reason I just can't connect to Appleinsider atm, anybody else got the same problem? Or is it just my internet connection that's messing with me?

Dont Hurt Me
Nov 20, 2003, 02:59 PM
this reminds me so much of g3 and g4 when they came out.

Trowaman
Nov 20, 2003, 03:00 PM
SWEET JESUS!

Faster faster faster! I wanna go faster! I wonder if they do have a light based processor in the works though . .

mvc
Nov 20, 2003, 03:03 PM
Originally posted by machan
I seriously doubt they'll brand anything as "G6" by the middle of next year…

Absolutely, the G5 brand will be milked for all its worth for a coupla years, until it starts getting "old" or Intel threatens to put out a P6. THEN we'll see the G6.

g30ffr3y
Nov 20, 2003, 03:04 PM
i heard light goes pretty fast...

ennerseed
Nov 20, 2003, 03:05 PM
Appleinsider also reports the new machines may include a Crossbar Switch: "and will use a crossbar switch to speed communication between processor"

I believe (don't know what else it could be) this is like SGI's implementation of a Crossbar Switch which could bring major speed improvements.

From SGI's site: "The unsurpassed crossbar switch design of Octane2 provides excellent levels of interactivity with critical data transfers, such as loading a 3D model from memory to the screen or processing high-definition video images in realtime. The 1.0GB-per-second peak main memory bandwidth and 1.6GB-per-second peak bandwidth between subsystems ensure that you experience smooth, fluid operations while completing even the most complex tasks"

SiliconAddict
Nov 20, 2003, 03:07 PM
Originally posted by Macrumors
The 980 is reportedly a 90nm chip and may be branded as the G6.


Bah! Forget about the G6. Just call it the Gwiz. Too bad Nvidia has the GForce.

ddtlm
Nov 20, 2003, 03:10 PM
Bah, I'm still betting against this. The 130nm 970 can go higher than 2.0ghz and its not unreasonable to expect the 90nm 970 to hit 3.0ghz. IMO, the most reasonable course of events would be faster 130nm 970's early in 2004 followed by 90nm 970's in the summer.

Dont Hurt Me
Nov 20, 2003, 03:12 PM
Its so cool to see this after agonizing over every 67 mghz motorola bump, that is if it happened at all. im sure just as many were waiting for Apple to dump motorola out of the pro towers there are many waiting for Apple to Dump the g4 out of the imac line or at least introduce a new line of G5's for the consumer. the G5 & G6 line is looking so so sweet. 10 times the bus speed of the G4 and double the clock !

daveL
Nov 20, 2003, 03:13 PM
Originally posted by ennerseed
Appleinsider also reports the new machines may include a Crossbar Switch: "and will use a crossbar switch to speed communication between processor"

I believe (don't know what else it could be) this is like SGI's implementation of a Crossbar Switch which could bring major speed improvements.

From SGI's site: "The unsurpassed crossbar switch design of Octane2 provides excellent levels of interactivity with critical data transfers, such as loading a 3D model from memory to the screen or processing high-definition video images in realtime. The 1.0GB-per-second peak main memory bandwidth and 1.6GB-per-second peak bandwidth between subsystems ensure that you experience smooth, fluid operations while completing even the most complex tasks"
There's nothing "new" about a crossbar switch. Sun, SGI and others have used them for years. It's simply a technique used to remove the latency issues associated with multiple subsystems (cpu, i/o, video, etc.) sharing a buss. With a crossbar implementation, you can have multiple point-to-point connections simultaneously; no subsystem has to wait for the buss to become available.

dongmin
Nov 20, 2003, 03:15 PM
As great as the 970 is, it's looking more and more like it was just a stop-gap solution. And it kind of makes me question whether I should get a Power Mac G5 rev. 2 this February like I was originally planning. I know I know, there's always something better around the corner. But the jump from 970 to 980 may be bigger in performance (per clock) than from G4 to 970.

And a year from now, 10.4 or whatever the latest OS update is, should be even more optimized for the G5.

daveL
Nov 20, 2003, 03:18 PM
Originally posted by ddtlm
Bah, I'm still betting against this. The 130nm 970 can go higher than 2.0ghz and its not unreasonable to expect the 90nm 970 to hit 3.0ghz. IMO, the most reasonable course of events would be faster 130nm 970's early in 2004 followed by 90nm 970's in the summer.
The Apple/IBM report, cited above, says the 90nm 970 will top out at 2.6-2.8 GHz and that it will require the 980 at 90 nm to hit 3 GHz.

york2600
Nov 20, 2003, 03:19 PM
Originally posted by ddtlm
Bah, I'm still betting against this. The 130nm 970 can go higher than 2.0ghz and its not unreasonable to expect the 90nm 970 to hit 3.0ghz. IMO, the most reasonable course of events would be faster 130nm 970's early in 2004 followed by 90nm 970's in the summer.

I think what this article is saying is that there is no 90mm 970. They're saying that when IBM hits 90mm it's going to be in a 980.

Wash!!
Nov 20, 2003, 03:19 PM
But then apple came out with the dual 1.8 at the same price a month later, but that is just the way it it.

Dippo
Nov 20, 2003, 03:22 PM
Speaking of all these chip advances...

Hasn't Intel been stuck at 3.2Ghz for some time?

Are they having problems going faster without causing a meltdown?

Stella
Nov 20, 2003, 03:24 PM
I wonder if we will ever see G5 PowerBooks?

Apple may just skip to G5 PowerBooks, if they can keep them cool enough.

leet1
Nov 20, 2003, 03:25 PM
Originally posted by Dippo
Speaking of all these chip advances...

Hasn't Intel been stuck at 3.2Ghz for some time?


They released the p4 extreme, they will be releasing a 64bit chip pretty soon, also, suppose to bump up to 3.4 pretty soon.


Its all about AMD right now though :D with the Opteron.

dongmin
Nov 20, 2003, 03:25 PM
Originally posted by daveL
The Apple/IBM report, cited above, says the 90nm 970 will top out at 2.6-2.8 GHz and that it will require the 980 at 90 nm to hit 3 GHz.

Is it a sure thing that they'll even bother with a 90nm 970? I thought they'd try to eek out another iteration with a 130nm 970 and then go directly to a 90nm 980. I'm assuming that the 980 is more power-efficient than the 970, given that the Power5 is more efficient than the Power4.

So what will make it into the PowerBooks and iMacs? The 90nm 970 or the 90nm 980?

many questions...

ennerseed
Nov 20, 2003, 03:25 PM
Originally posted by daveL
There's nothing "new" about a crossbar switch. Sun, SGI and others have used them for years. It's simply a technique used to remove the latency issues associated with multiple subsystems (cpu, i/o, video, etc.) sharing a buss. With a crossbar implementation, you can have multiple point-to-point connections simultaneously; no subsystem has to wait for the buss to become available.

Do you read what people write before you reply?
But thanks for the added information.

gwuMACaddict
Nov 20, 2003, 03:26 PM
Originally posted by dongmin
As great as the 970 is, it's looking more and more like it was just a stop-gap solution. And it kind of makes me question whether I should get a Power Mac G5 rev. 2 this February like I was originally planning. I know I know, there's always something better around the corner. But the jump from 970 to 980 may be bigger in performance (per clock) than from G4 to 970.

And a year from now, 10.4 or whatever the latest OS update is, should be even more optimized for the G5.


these are my thoughts exactly. with the 980 this close to production, makes me wonder why they brought the 970 out at all. we were all used to waiting around with motorolla anyway


edit: 970 for imac maybe once the 980 is out

Ambrose Chapel
Nov 20, 2003, 03:29 PM
originally posted by leet1
They released the p4 extreme, they will be releasing a 64bit chip pretty soon, also, suppose to bump up to 3.4 pretty soon.


there's also this, which i just saw on macminute:

"Our goal is to hit 4GHz in 2004," said Paul Otellini, Intel's president, during a meeting with financial analysts.

so there's that..

but back OT, all these advances sound great; i just hope apple will quickly filter the G5 down the product line. they may be able to compete in speed at the high end, but they need to expand that.

deputy_doofy
Nov 20, 2003, 03:29 PM
Originally posted by dongmin
As great as the 970 is, it's looking more and more like it was just a stop-gap solution. And it kind of makes me question whether I should get a Power Mac G5 rev. 2 this February like I was originally planning. I know I know, there's always something better around the corner. But the jump from 970 to 980 may be bigger in performance (per clock) than from G4 to 970.

And a year from now, 10.4 or whatever the latest OS update is, should be even more optimized for the G5.

I agree with what you said. I'm no chip expert, but it seems to me that the G4 and the Power4 have been around for the same amount of time, roughly.
The G5 performance is simply where the G4 performance SHOULD be.
We all knew/know the G4 was a better chip than the P3/P4, but only as far as the chip itself goes. The architecture on which the G4 resides was the crap holding back the chip.

I have a feeling that the G6 chip is really gonna smoke. :)

ddtlm
Nov 20, 2003, 03:30 PM
daveL:

The Apple/IBM report, cited above, says the 90nm 970 will top out at 2.6-2.8 GHz and that it will require the 980 at 90 nm to hit 3 GHz.
I might be missing something, but all I saw was rumors.

dongmin:

Is it a sure thing that they'll even bother with a 90nm 970?
Well the Register has an article today that sounds pretty confident that IBM will discuss a 90nm G5 at a conference in February.
http://www.theregister.co.uk/content/39/34108.html

Other than that, I think it makes a lot more sense for IBM to use the 970 design a while, since its not really outclassed yet. It seems reasonable for them to develop the "980" at a slower pace and launch it in 2005.

jkojima
Nov 20, 2003, 03:32 PM
I don't believe that Apple would roll out a "G6" for at least two years. The "G" series marketing campaigns are not just about the core chip, but an entire system architecture, and Apple needs to milk the G5 for a while... not in the least in order to recoup the development costs.

arn
Nov 20, 2003, 03:33 PM
90nm 970 is happening.

IBM has said so

http://www.macrumors.com/pages/2003/11/20031113110834.shtml

arn

ddtlm
Nov 20, 2003, 03:34 PM
gwuMACaddict:

these are my thoughts exactly. with the 980 this close to production, makes me wonder why they brought the 970 out at all
Hold on a sec, as the "980" been confirmed, cause I've been under the impression that so far its just nothing but a lot of rumors. Sure IBM has something coming after the 970, but as far as I know they have provided no names, no dates, no details... nothing.

yoman
Nov 20, 2003, 03:38 PM
Originally posted by Dippo
Speaking of all these chip advances...

Hasn't Intel been stuck at 3.2Ghz for some time?

Are they having problems going faster without causing a meltdown?

I just read at

http://www.macminute.com/

That intel is planning to hit 4Ghz in 2004. I wonder if IBM will be able to catch up.

Omek
Nov 20, 2003, 03:41 PM
I can't believe Intel is now going to make a 64-bit processor right after they said it was an "unecessary" technology. That's such crap. And now since Apple and AMD are selling well, Intel's going to make one.

I hope Intel is smoked next year by the 980.

Dreadnought
Nov 20, 2003, 03:48 PM
This is the same story with the G3 and G4. In 1998 I was waiting for the G4 but it never came... So in 1999 I bought a G3 and a half year later, the first G4 came out. The G3 should have been on 500 Mhz by then, but was only at 400/450 max. The G4 should have started at 500 mhz, but started at 450 and went for a brief period to even 400 (to much problems with the 450). The G3 is still alive and kicking and can go theoretically to 1500 Mhz (production at this moment 1100 Mhz). The G4 should have stopped at 1000 Mhz and the G5 should have been available then, in 2001! Or at least, this was Moto's (road)plan.

And you know what...history repeats! Tommorrow a new lesson, now my children, do your homework and study chapters 1 and 2! :D

spencecb
Nov 20, 2003, 03:53 PM
Originally posted by jkojima
I don't believe that Apple would roll out a "G6" for at least two years. The "G" series marketing campaigns are not just about the core chip, but an entire system architecture, and Apple needs to milk the G5 for a while... not in the least in order to recoup the development costs.

I think that they WILL roll out the G6 within the year. This is because the G4 has been around for quite some time now, and with IBM as their primary partnet now, Apple has the resources to speed up their advancement in chip technology. Therefore, the G5 will soon replace the "consumer" portion of Apple's product line (such as the eMac and iMac), leaving the G4 for the iBook, and introducing the G6 as the Pro user chip, with the PowerMac and the PowerBook...and hopefully, they will introduce new hardware, maybe using a low end G6 or entering the "consumer" line with a high end G5....maybe the target market of the G4 Cube, but this time being more successful with a lower price point

LightFantastik
Nov 20, 2003, 03:58 PM
I think that "fast roadmap"...quickly going from the G5 to G6 is a good thing. It keeps Apple in the news, and shows people that the age of minor speed bumps, ala Motorola, are a thing of the past, and that Apple can definitely keep up with, if not surpass X86 (which of course, WE all know they can...convincing the Intel/AMD/MS faithful is the hard part).

Keep up the good work Apple/IBM !! :)

ITR 81
Nov 20, 2003, 03:58 PM
The 90nm 970 will more then likely end up in the iMac and PB's.

Apple has in the past change the name based on the fact the processor changes along with it. This also means the MB will change as well and probably the ram included.

AMD and Intel will have to shoot for 4GHz by yr end or Apple will catch up by 2005 at cited development rate. They do not want to compete head to head with IBM and Apple for sales because if they do they will start to lose market share.

The 980 is based on the Power5. IBM already has stop gap processor coming out named the Power5+ which is suppose to hold folks over until the Power6 comes out which is being designed now.

Apple will not slow down..if anything they may try to speed things up because I know Jobs wants to eat into the PC market if at all possible.

If the 980 can goto the 5GHz speed then it would suggest the
Power5+ could push to 6 to 7GHz range.
About at 7GHz the Power6 could then come into play taken it from 7GHz on up to probably 10GHz.


If IBM can put out 3 speed boost's a yr that is great!! But to really put it to Intel and AMD they need 4 speed boost's a yr.

SiliconAddict
Nov 20, 2003, 03:59 PM
Originally posted by Dippo
Speaking of all these chip advances...

Hasn't Intel been stuck at 3.2Ghz for some time?

Are they having problems going faster without causing a meltdown?


Nope.....
http://news.com.com/2100-1006_3-5110034.html?tag=nefd_top

The chipmaker, which discussed its plans in a wide-ranging meeting with financial analysts on Thursday, said it aims to boost the performance of a broad range of its products next year, including cranking up its desktop PC processors.

"Our goal is to hit 4GHz in 2004," Intel President Paul Otellini said during a meeting that was Webcast.

Intel is aiming to reach that clock speed with Prescott, an upcoming processor for desktop computers that will be built using a 90-nanometer manufacturing process. (A nanometer is a billionth of a meter.) Prescott is scheduled to ship this quarter to PC makers, Otellini said. But it's not expected to come in desktop PCs until early next year. Right now, Intel's fastest chip is the 3.2GHz Pentium 4.

While a 4GHz processor may seem fast, as least one analyst said the jump isn't a particularly large one for a brand-new processor.

I personally am more interested in seeing what AMD rolls out. Intel's desktop line sucks monkey dung. Their laptop CPU line is another matter.

At any rate its good to see not only some fresh blood in the PowerMac line but there’s some actual circulation there as well. Unlike Moto who was DOA. Call it. Time of death for Moto: June 23, 2003 1PM EST :)

spencecb
Nov 20, 2003, 04:03 PM
I suppose it is possible that the new 980 chip would not be dubbed as a G6....this is simply because time after time again, rumors said that each new itteration of the G4 would be called a G5, because there were significant advances over the previous one, and "now is when the G5 will come to light." But it never did, until IBM took over and Motorola got the boot (thank god). But anyhow, here's hopping that the G6 will be here next year, and it will blow the pants off of the G5 like the G5 did the G4

illumin8
Nov 20, 2003, 04:05 PM
I personally don't think Apple will go to a G6 brand name this quick in the game. They have invested too much marketing money into promoting the G5 brand already and it would leave current power users that just bought their dual 2.0's feeling like they have the short end of the stick.

As far as the 980/Power5 goes though, I think there's a good chance we could also gain multiple cores on it. The Power5 has dual cores, so a dual processor Power5 actually appears to the OS as if it were a quad processor machine. I read this article at Ars Technica today and he speculates on this:

At this point, I could talk about the need for SMT in an Apple system, but I'll just leave off that sort of commentary for now and observe only that Apple's long-standing and ongoing affinity for SMP designs has resulted in two things: 1) a huge potential for wasted execution resources on the current crop of non-SMT-capable G5s and 2) a body of natively-developed and -ported applications that have been subjected to years of pressure to use multithreading wherever possible in order to wring the best performance out of Apple hardware. I think both of these factors will converge to make a SMT a significant improvement for the Mac platform.

You can read the entire article here: http://arstechnica.com/cpu/003/mpf-2003/mpf-2003-1.html

sethypoo
Nov 20, 2003, 04:05 PM
Originally posted by dongmin
As great as the 970 is, it's looking more and more like it was just a stop-gap solution. And it kind of makes me question whether I should get a Power Mac G5 rev. 2 this February like I was originally planning. I know I know, there's always something better around the corner. But the jump from 970 to 980 may be bigger in performance (per clock) than from G4 to 970.

And a year from now, 10.4 or whatever the latest OS update is, should be even more optimized for the G5.

Good idea, I'm waiting too.

A G6 this soon is not likely for Apple. Wasn't the G4 out for a couple of years before the G5?

I will probably purchase my next computer (Mac, of course) once the Rev. B PowerBook G5 gets released in a 12" or maybe 15" format. I love my 12", but I needs to run cooler, have DV out, and have a lighted keyboard.

12" PowerBook G5.....[[shiver]].

:) :rolleyes: :D

x86isslow
Nov 20, 2003, 04:06 PM
anyone hear anything about a speed bump from moto? are we going to see another revision of g4 powerbooks before mwsf04?

i think apple will stick with moto for now, as they need low power chips for ibooks, which just got the G4.

mebbe the new spunoff moto will intro the '7460' 1.8 Ghz G4.. that way, the ibook can compete with the Pentium M 1.8

sethypoo
Nov 20, 2003, 04:07 PM
Oh, and what if Apple doesn't release the G6 until we get to 990? That's always a possibility, they could milk the G5 for all it's worth.

aethier
Nov 20, 2003, 04:07 PM
Originally posted by spencecb
I think that they WILL roll out the G6 within the year. This is because the G4 has been around for quite some time now, and with IBM as their primary partnet now, Apple has the resources to speed up their advancement in chip technology. Therefore, the G5 will soon replace the "consumer" portion of Apple's product line (such as the eMac and iMac), leaving the G4 for the iBook, and introducing the G6 as the Pro user chip, with the PowerMac and the PowerBook...and hopefully, they will introduce new hardware, maybe using a low end G6 or entering the "consumer" line with a high end G5....maybe the target market of the G4 Cube, but this time being more successful with a lower price point


that is a pretty clever idea. if they do that, they will not have to make the proformance of the iMac and emac suffer, the was they did when using the G4 in the powerline, and consumer line. this way they could have g5 imacs, at 2.5 ghz, and powermacs with g6, and like 3.2 or whatever, instead of using the same chip for high and low end, making cut backs to lthe low end, so that the high end doesn't lose sells. for example currently they cannot upgrade the imac to g5, because it wil compete to much with the powermac, but leaving it with a g4, all it is competing with is some crappy dell with a celereon, and even that is "ify" given that intel has the celeron at 2.5 ghz now, and as much of a mac obsessed person i am, i do not believe that a 1.25 ghz imac, could compete with a 2.5 ghz celeron, which the average person will see as faster, and about 1000+ cheaper...

aethier

(sorry for poor spelling)

applekid
Nov 20, 2003, 04:07 PM
The next chip that Apple will use is the Power 980. There, I said it. This sudden rush of MacRumors posting on the 980 seems to only confirm that the chips after the 970's will be 980's for the Mac. I still think the 980 will be branded G5 until there's something more significant than Power 9x0's, IMHO.

ITR 81
Nov 20, 2003, 04:09 PM
But remember Moto was same processor with just slight speed tweaks.

It would be awsome if Apple could get it's processors within 500MHz of Intel and AMD processors because then they would be in direct competition with the PC market. If they can match or surpass the Intel or AMD processors I expect alot folks switching over because at that point speed would no longer be a reason to say a PC is better then a Mac.

dguisinger
Nov 20, 2003, 04:11 PM
Originally posted by spencecb
I suppose it is possible that the new 980 chip would not be dubbed as a G6....this is simply because time after time again, rumors said that each new itteration of the G4 would be called a G5, because there were significant advances over the previous one, and "now is when the G5 will come to light." But it never did, until IBM took over and Motorola got the boot (thank god). But anyhow, here's hopping that the G6 will be here next year, and it will blow the pants off of the G5 like the G5 did the G4

Other than speed there were never significant advances from one G4 to another. Cache doesn't count either, companies change the cache sizes between revisions all the time.

But, say you are adding a second AltiVec unit...or a 3rd, or doubling the pipelines. Now you have a new processor. Development tools must be re-engineered to take advantage of new optimizations. You can't say optimized for G5 revision 2. Thats when a processor name changes.

leet1
Nov 20, 2003, 04:13 PM
Originally posted by ITR 81
It would be awsome if Apple could get it's processors within 500MHz of Intel and AMD processors because then they would be in direct competition with the PC market.


Actually the Opteron is at 2Ghz and performs as good as the 2.0 g5 :D

howard
Nov 20, 2003, 04:17 PM
the g5 is gonna get knocked off the top spot pretty quick ...whatever though, they can name it whatever they want.

it would be nice to do away with the old g4 and g3 names and have g6 in power line and g5 in consumer line in about a year and a half.

ITR 81
Nov 20, 2003, 04:19 PM
Are you sure because as far as I can see AMD 64 is still using PR ratings meaning it has lower clockspeed but can compete at that rating.

Also AMD 64 is just a licensed MIPS 64 architecture from it's RISC processors that are currently in SGI machines.
AMD is great but if not for the MIPS 64 license they would not be competing now with Intel. Intel on the other hand makes everything in-house.

Manatee
Nov 20, 2003, 04:20 PM
All right!!!

PowerBook G5 = <yawn>

PowerBook G6 = Rocks!!!


(but since I may be waiting a few days, maybe I'll upgrade from my Ti to an Al) ; )

Topo
Nov 20, 2003, 04:28 PM
What about the Xserve2? and the plans of releasing 64processors servers. I bet there going to be powered by G6!!!!

ITR 81
Nov 20, 2003, 04:28 PM
A G6 in the middle next yr would be great because it could be used in only Pro user line and then consumer computers can then get the G5. Because like someone already stated they can not do it now due to the fact it would compete directly with the PM's.

a 90nm 970 also pretty much says G5 PB or does it? If the 980 comes out in 90nm form why not then just use them in the PB's and send the G5's to the ibooks which then could compete more so with the dell and gateway craptops.

Stella
Nov 20, 2003, 04:30 PM
Apple will use the latest and greatest processor it can for its Pro line.

Apple will not want to repeat the G4 adventure again - i.e., an image of a slow and outdated processor.

The 980 is the way to go - G6 is its Pro line, with suitable processor speeds to keep consumers happy. G5s will be the consumer line.

I wouldn't be happy to see the G5s being milked for a year in the Pro line, when there is the G6 available and waiting. It doesn't make sense. Apple have competition and have to keep up with processor speeds.

Apple are out to make and sell computers, they need the fastest processors they have in their machines - especially the Pro lines.

Photorun
Nov 20, 2003, 04:35 PM
Originally posted by yoman
I just read at

http://www.macminute.com/

That intel is planning to hit 4Ghz in 2004. I wonder if IBM will be able to catch up.

Okay, for one thing Intel is speculating, I guess so are we with the 970/980. I'd like to shut the yaps of the MHz mhthsters on the peecee side as the next guy. A bigger point is Intel's honcho was referring not to a 64 bit chip but a 32 bit one, so who cares? If we get more 64 bit coding including Apple's next OS next year and have a dual 3.0, we'd still smoke their 4.0 GHz chip, even if they had two of them.

stingerman
Nov 20, 2003, 04:37 PM
I believe that that the confirmed 90NM 970 with PowerTune is for the next generation Powerbooks and iMacs. Apple needs to move beyond the Pentium-M and quickly if it is to maintain its pricing justification against the Pentium-M. The iBook needs to move into the current Powerbooks performance area and be comparable to the Pentium-M's.

The iMac needs to move to the G5 as well and the 90NM 970 would be right for it and its cooling needs.

When Apple moves to the 980, they can then have a clear differentiation between their consumer, SMB and professional lines.

I see an iMac G5, a single processor headless G5 and a dual processor G5/G6 as Apple's key mac desktop product lines. With the eMac for education and low-end sales.

yoman
Nov 20, 2003, 04:42 PM
Originally posted by Photorun
Okay, for one thing Intel is speculating, I guess so are we with the 970/980. I'd like to shut the yaps of the MHz mhthsters on the peecee side as the next guy. A bigger point is Intel's honcho was referring not to a 64 bit chip but a 32 bit one, so who cares? If we get more 64 bit coding including Apple's next OS next year and have a dual 3.0, we'd still smoke their 4.0 GHz chip, even if they had two of them.

I like the idea of smoking Intel. It brings warm feelings to my heart. :D

x86isslow
Nov 20, 2003, 04:50 PM
Originally posted by yoman
I like the idea of smoking Intel. It brings warm feelings to my heart. :D

like toasted rabbit-suit guys?

remember those ads?

tortoise
Nov 20, 2003, 04:51 PM
Originally posted by ITR 81
Also AMD 64 is just a licensed MIPS 64 architecture from it's RISC processors that are currently in SGI machines.
AMD is great but if not for the MIPS 64 license they would not be competing now with Intel. Intel on the other hand makes everything in-house.

AMD has been making very good RISC cores since long before they got into the ia32 ISA compatible game (though hardly anyone seems to remember). The AMD64 ISA has no relation to MIPS64 ISA that I can see; I think you are mixing and matching a bit. AMD cores tend to be extremely efficient, often more so than the RISC cores they compete against (realizing that the AMD cores are RISC, and always have been). AMD64 cores are among the most efficient general purpose cores currently in production and they have a lot of room to grow. Intel, on the other hand, has a lot of problems with their core designs.

The AMD cores are very well engineered and are also somewhat more "general purpose" in design than PPC cores, which tend to have an emphasis on numerical codes that makes them less optimal for many tasks than the AMD core. AMD will have no problem keeping up with IBM, both in terms of clock-per-clock and absolute performance for the foreseeable future, and may lead a bit. Intel is in trouble though. Their core isn't remotely as good as AMDs, and they are having trouble making it scale.

The battles in the future will be between AMD64 and PPC. Intel will be a sideshow.

themadchemist
Nov 20, 2003, 04:52 PM
Originally posted by Dont Hurt Me
this reminds me so much of g3 and g4 when they came out.

Without a doubt--There hasn't been so much buzz & excitement in the Mac community since the early days of the g3 and g4. Everything seemed so promising back then...

I hope that the road of the G5 does not lead to the eventual stagnation and disappointment that the G4 did. In a sense, though, the G5 is more like the G3, in that it "ushered in a new era." In that case, I hope that the G6 doesn't stagnate...Or the G7, or the G8, or the G9, or the GX (pronounced 'ten'!).

ITR 81
Nov 20, 2003, 05:04 PM
Originally posted by tortoise
AMD has been making very good RISC cores since long before they got into the ia32 ISA compatible game (though hardly anyone seems to remember). The AMD64 ISA has no relation to MIPS64 ISA that I can see; I think you are mixing and matching a bit. AMD cores tend to be extremely efficient, often more so than the RISC cores they compete against (realizing that the AMD cores are RISC, and always have been). AMD64 cores are among the most efficient general purpose cores currently in production and they have a lot of room to grow. Intel, on the other hand, has a lot of problems with their core designs.

The AMD cores are very well engineered and are also somewhat more "general purpose" in design than PPC cores, which tend to have an emphasis on numerical codes that makes them less optimal for many tasks than the AMD core. AMD will have no problem keeping up with IBM, both in terms of clock-per-clock and absolute performance for the foreseeable future, and may lead a bit. Intel is in trouble though. Their core isn't remotely as good as AMDs, and they are having trouble making it scale.


Nope I'm not mistaken AMD also licensed the MIPS 32 arch. from them as well. AMD got the license around 2 yrs ago. MIPS showed on there site back then AMD buying the license and AMD also showed a white paper at the time of how they were going to use the new technology license. AMDZone also covered it as well.

I know alot about AMD since I've watched and used them since there old K5 and K6 processors. I still have thunderbird around here somewhere. All of AMD's new 64 processors are based on technology license they bought from MIPS around 2 yrs ago. So I doubt you will be able to find similarities between the new AMD and MIPS because MIPS has change it arch. some since then.

Rincewind42
Nov 20, 2003, 05:21 PM
I don't see any reason why Apple won't market their top chip as a G5 for at least another year and a half. Remember the G3 went through a number of revisions (and even a case change) before being relegated to a consumer only chip. Now if that chip is a 970, 980, or some other derived chip is a moot point. There have been 5 major revisions of the G4 and even more than that of the G3. The G5 will likely still be used as a marketing moniker in 3 or 4 years time, and certainly won't be referencing a 970.

I would expect the 970 and at least two of the revisions after that to be called G5. But who knows what else will come along during that time :D .

lind0834
Nov 20, 2003, 05:22 PM
Instead of calling the 980 the G6, so quickley after they introduced the G5 they should calling it the G5 Extreme.

:D :D :D

Photorun
Nov 20, 2003, 05:42 PM
Originally posted by lind0834
Instead of calling the 980 the G6, so quickley after they introduced the G5 they should calling it the G5 Extreme.

:D :D :D

UGH! No extreme anything please... unless it's "Michael Dell and Bill Gates experienced an extreme and painful death followed by an extreme plummet of their respective companies from the face of the Earth."

ITR 81
Nov 20, 2003, 05:43 PM
Apple may also want to change the name because of all the bad publicity it drew in the UK with the G5 campaign.

ICEBERG
Nov 20, 2003, 05:44 PM
Is this mean they will change the new Style of the machines? I like the idea of the new G6 and a new Style of the Towers.:cool:

MacSlut
Nov 20, 2003, 05:51 PM
All these posts about Apple "milking" the G5 are ridiculous.

Apple didn't invest hardly anything in the G5. Sure some posters here and stuff, but really small potatoes in the grand scheme.

It's not like as if they invested in a brand name that they would be tossing away. I think most people would understand G6>G5 rather than thinking, "I know what the G5 was, but G6...gee, I don't know".

Apple did do some investment into the PowerMac G5 design, but it could be changed over to a PowerMac G6 design fairly easily. Remember the switch from a B&W G3 to a PowerMac G4 was mostly the processor and a few minor design tweaks.

The basic design of the PP601 based PowerMac 7100 was tweaked for the 7500 but then kept basically the same for the 7200, 7300, 7600 and the desktop G3.

Of course the G4 branded processor represented several significant changes such that Apple could've (should've) branded it as a G5 at several points but waited for the major change in case and processor for the G5.

I'd really like to see a BIGGER case for the G6. One of the reasons why I've kept my Dual 1.42 is because I like having 4 internal hard drives and two internal DVD burners. If the G6 was just a little taller than the G5, it could accommodate the extra drives. This would be an easy way to distinguish the design between the two without developing a whole new case.

Also contrary to what some have posted, Apple will (and should) always choose to provide faster and better hardware for people to upgrade to. If the upgrades are significant leaps, people will praise the advance.

tortoise
Nov 20, 2003, 06:00 PM
Originally posted by ITR 81
Nope I'm not mistaken AMD also licensed the MIPS 32 arch. from them as well. AMD got the license around 2 yrs ago. MIPS showed on there site back then AMD buying the license and AMD also showed a white paper at the time of how they were going to use the new technology license. AMDZone also covered it as well.


Actually, you ARE mistaken. While they licensed the MIPS ISAs (not core designs), it was for their other processor lines and unrelated to their desktop computer business. AMD has been making high-performance embedded RISC CPUs for far longer than they have been dabbling in the x86 market, and for a number of different popular ISAs and some original ones (like the AMD 29k RISC processors). High performance embedded systems CPUs based on other ISAs is still a major line of business for them. MIPS32 and MIPS64 are popular embedded target ISAs, and licensing those ISAs supports their embedded CPU business.

Their earlier x86 processors were an IA32 ISA on an AMD core. The Opteron (and related) are an AMD64 ISA on an AMD core, with support for the IA32 ISA. There is no MIPS technology in their consumer CPU line. AMD has a lot of experience designing cores and designing ISAs, with the added experience of making many different popular ISAs run on their cores. AMD CPU cores have a long history that predates their x86 offerings and which were designed by AMD. The only thing special about the AMD64 is that this may be the first time they have marketed their own ISA design since the 29k-series of embedded processors.

You may have K5, a K6, an Opteron, and whatnot. But they all use the x86 (or derivative) ISAs, so they obviously can't be using a MIPS ISA. They make chips with the MIPS ISA, but the aren't for the consumer market and they don't run x86 by definition. It seems that you are confused as to the terminology. An ISA is not a processor design.

yamabushi
Nov 20, 2003, 06:06 PM
The G4 is dead please let it rest in peace. The G5 is good now but will be just okay by next August. The 980 will most likely be branded as a G6 and start in the tower. The G5 will rapidly trickle down into the PB and consumer line. Apple will probably keep G5s and G6s around for a while while IBM makes some improvements to both.

The Fishkill plant was built with the plan to produce processors at 130nm, 90nm, and 65nm. IBM has stated that intial tests for 65nm are going well and that 90nm will be ready for full production very soon. IBM has a strong financial incentive to move the process down to 65nm as soon as possible. I believe that both the 970 and 980 will eventually reach 65nm. That means that the clock speeds for both chips can continue to increase and the G5 should have no trouble replacing the G4 in the consumer line.

yamabushi
Nov 20, 2003, 06:13 PM
Originally posted by ICEBERG
Is this mean they will change the new Style of the machines? I like the idea of the new G6 and a new Style of the Towers.:cool:

A new case style would be nice. A 3rd hard drive bay would be great. I like the idea of giving the outside of the case a semi-heatsink look to it with tiny vertical fins. Sort of a corduroy aluminum texture.

Nermal
Nov 20, 2003, 06:14 PM
They can't release a G6, I don't even have a G4 yet! :(

Actually, I'm planning to get one in about a week. And my copy of Panther arrived yesterday, so that'll keep me happy for a while :)

But I'm sure you're more interested in the 980 than what I've got here at home.

the_mole1314
Nov 20, 2003, 06:24 PM
Let me say it so we can beat it out of ourselves before it gets too big...

The G7 will be Intel based. :p

Ok, now that we have all the Intel fans covered, lets talk Mac.

I personally think it's a waste to call it a G6, but I think that seeing a 970 hit 3ghz is a very real possability right now.

I have to go now guys, my pants are soilded from my excitement! :D

ThomasJefferson
Nov 20, 2003, 06:42 PM
Originally posted by ITR 81
A G6 in the middle next yr would be great because it could be used in only Pro user line and then consumer computers can then get the G5. Because like someone already stated they can not do it now due to the fact it would compete directly with the PM's.

a 90nm 970 also pretty much says G5 PB or does it? If the 980 comes out in 90nm form why not then just use them in the PB's and send the G5's to the ibooks which then could compete more so with the dell and gateway craptops.


O'h if this were true. Happy Dance. And I expect all of the above except the G5 ibook. Me'Thinks Gobi, or something else, due to the melted plastic smell.

MrMacMan
Nov 20, 2003, 06:46 PM
:yawn:

Wake me up when they get a Better name like...

The G-006

To lead up to the G-007!

Mwhahaha.

;)

Processor improvements are good.

Faster Bus?
Faster Clock speed?

Ohh yeah!!

:D

jettredmont
Nov 20, 2003, 06:48 PM
Originally posted by yoman
I just read at

http://www.macminute.com/

That intel is planning to hit 4Ghz in 2004. I wonder if IBM will be able to catch up.

Given that a dual-2.0GHz G5 smokes 3.2GHz P4's by most estimations, I imagine a 3.0GHz G5 (with 1.5GHz FSB) will likewise smoke a 4.0GHz P4.

Will Intel ever deliver the P5? That might change the equation slightly, as the P5 is supposedly more efficient (instructions per cycle in real use) than the P4.

Rocketman
Nov 20, 2003, 07:04 PM
Originally posted by MrMacman
:
Wake me up when they get a Better name like...
The G-006
the G-007!
;)
Processor improvements are good.


Multiple replies

Why G5/G6?

Same reason why they used to have a G3/G4. A commodity line and a premium line.

As each generation is released, the prior one is phased out. G5 kills G3 and G6 kills G4 (Motorola). As we all know Apple/Jobs and Apple's customers themselves are in a BIG HURRY to dump Motorola.

G6 is the "dump Motorola" line. Everyone is in a huge hurry for it.


I stated early that the G5 would be about 2.6 Ghz in Jan-Mar 04. I stated early that the G6 would be what was referred to by Jobs as the 3 Ghz late 04 release.

That much is fairly clear from public statements, IBM releases and chip industry events. What is particularly interesting is that some of the "guidance" that enterprise needs to know what to expect in the fututre, is now being released by Apple in an uncharacteristic way. If the information flow becomes even incrementally more about "future products" then Enterprise will have the comfort level needed for adoption of the product. They also need a ramp-up in real service with the sale, and charging fees for that is just fine.

The G5 should really have just as long of legs as the G4 has, or more. That gives it a multi-year life cycle. The G6 when it is released frankly will require more expensive associated components to really give it the resources it can feed. So we can expect maybe even a higher price point than the G5 at release.

For example the G5 was released as a 64 bit clean 32 bit system. Deja vu.

The G6 is likely to be primarilly 64 bit with good 32 bit backwards compatibility.


Overall this is an astonishing series of events as compared to the past several years. This is beginning to realize the actual early promise of the PowerPC.

What I am hoping is the G6/980 is more scalable to 4x and 8x processor systems so true compute power can fit in 3u or 6u deskside computer ala some of the IBM AS/300's and such. Remember the supersize Apple server?

This will also be a server monster and a cluster and grid monster.

When Internet 2 becomes more reality in 5 years or so, true grid computing (WAN) will become practical.


And who says a processor (G6/980) that was sampling in October 2003 would NOT be released in some astronomical Apple system in March 2004?

Rocketman

ITR 81
Nov 20, 2003, 07:13 PM
Originally posted by tortoise
Actually, you ARE mistaken. While they licensed the MIPS ISAs (not core designs), it was for their other processor lines and unrelated to their desktop computer business. AMD has been making high-performance embedded RISC CPUs for far longer than they have been dabbling in the x86 market, and for a number of different popular ISAs and some original ones (like the AMD 29k RISC processors). High performance embedded systems CPUs based on other ISAs is still a major line of business for them. MIPS32 and MIPS64 are popular embedded target ISAs, and licensing those ISAs supports their embedded CPU business.

Their earlier x86 processors were an IA32 ISA on an AMD core. The Opteron (and related) are an AMD64 ISA on an AMD core, with support for the IA32 ISA. There is no MIPS technology in their consumer CPU line. AMD has a lot of experience designing cores and designing ISAs, with the added experience of making many different popular ISAs run on their cores. AMD CPU cores have a long history that predates their x86 offerings and which were designed by AMD. The only thing special about the AMD64 is that this may be the first time they have marketed their own ISA design since the 29k-series of embedded processors.


AMD bought the license in April 29, 2002 just right after they bought Achelmy from MIPS to compliment it which was bought on Feb. 2002. At that time AMD and MIPS said it was to be used for embedded processors but AMD also stated it could be used in there entire processor line. That being said everyone assumed back in 2002 the AMD 64 was going have instruction sets from this license. I was gone fighting a war for over a yr so I guess I might have missed some of the last details...and yes I know about AMD's history on RISC as my college had old AMD Nx586 machine we use to play with.

ITR 81
Nov 20, 2003, 07:19 PM
Tech Tv also reported today of the news of the 980 90Nm processor coming in March of next yr. at 3GHz.

I wonder if the rumor is just getting around or if news is leaking because of someone from Apple or IBM.

rog
Nov 20, 2003, 07:20 PM
This is great news. Just imagine a year from now, every mac at least 2 GHz and a G5 or better! Keep it coming IBM!

Anyone know of the other improvements in the 980 other than clockspeed? Bigger L2 perhaps?

jettredmont
Nov 20, 2003, 07:27 PM
Apple will rebrand the Motorolla 74xx chips as "G5-Ultra" and the IBM 970 chips as "G5-Ti". Maybe those old 750FX chips IBM still has laying around would sell better if they were "G5-MX" instead?

That way, all the PowerBook users can feel like they're getting a cool new processor ...

Hey, works for nVidia and ATI ...


sounds kinda off-topic, but the point is this: Apple stands by its trademarks and brand names a little more than most companies. I definitely don't see Apple going to G6 soon, not because they don't think the public can figure out that "G6" is kinda like "G5" except "one more", but because when you give the marketplace a whirling target of increasing numbers all they see is a blur, which destroys branding.

Customers haven't even had a full year of G5-desire yet. Too early to bring out the G6.

tortoise
Nov 20, 2003, 07:27 PM
Originally posted by rog
Anyone know of the other improvements in the 980 other than clockspeed? Bigger L2 perhaps?

I'm hoping for a very slick on-chip memory controller. That would rock, and close a couple gaps.

yamabushi
Nov 20, 2003, 08:22 PM
The 980 will likely be able to support a huge amount of L2, like the Power5. However the actual amount of L2 in G6 chips for Apple will probably be quite modest.

roy_g_biv
Nov 20, 2003, 08:26 PM
i remember some readers here disputed branding the 970 as the "G5", given it was not a "5th generation processor" like the G4 was fourth generation powerpc. keeping the G5 moniker for the 980 - the POWER5 derivative - just sounds like it makes sense.

and i don't think a shiny new "G6" is needed to market apple computers. as long as the G5 brand remains competitive with intel or AMD processors in speed and scalability, i don't think consumers will care too much about what the actual G-designation is used. (nor do i think a G5/6 line, or a G5/4 line need be set in stone) maybe apple can take the 90nm 970, stick it in an imac and call it a G-something. G-anything, i don't even care... maybe something funny like a "G-sumer", or maybe a "G-Unit" and target that 18-34 demographic.

but then again, in the end it's just fun to speculate.

jaedreth
Nov 20, 2003, 09:08 PM
Steve said the *G5* will reach 3GHz by the end of next year. Not the G6. Keep in mind, the term G5 is Apple's. Sure, it's the next chip design, but it is not a completely new architecture, it's just the sequel.

So this whole talk of "moving to G6" is silly, and assumptive. It's the same chip despite whatever name. Besides, we want the G5 to show itself as the best out there.

I wouldn't expect Apple to change chip generation monikers until at least Power7, unless IBM comes out with some new technology that redefines all the rules before then. (Which they may.)

Jaedreth

~Shard~
Nov 20, 2003, 10:17 PM
Originally posted by jaedreth
Steve said the *G5* will reach 3GHz by the end of next year. Not the G6. Keep in mind, the term G5 is Apple's. Sure, it's the next chip design, but it is not a completely new architecture, it's just the sequel.

So this whole talk of "moving to G6" is silly, and assumptive. It's the same chip despite whatever name. Besides, we want the G5 to show itself as the best out there.

I wouldn't expect Apple to change chip generation monikers until at least Power7, unless IBM comes out with some new technology that redefines all the rules before then. (Which they may.)

Jaedreth

I agree - until IBM comes out with some revolutionary or noticable advanced architecture, I don't think we will see the "G6" name applied to any machines. I think there is a lot of potential with the 970s and 980s, and the addvancements will be relatively quikc (compared to Motorola at least!), so I think staying with the G5 for the next year or 2 is perfectly acceptable. I hope Apple waits until IBM breaks completely new ground with a chip that will deserve to be called the G6.

Gyroscope
Nov 20, 2003, 10:42 PM
Yeah ,they ll stick to 970 as long as they can milk your money from it. Chip development doesn't come cheap so why would they jump to new design when 970 hasn't been around for more than 4 months or so, to pay for itself. Besides there is no official confirmation about 980??(whatever) existence as of yet.

Sweet dreams.

suzerain
Nov 20, 2003, 11:14 PM
About the Intel 4 Ghz thing. If they can get their chips up to that, great. I'm all for speed, anywhere.

But, they are currently at 3.2 Ghz, and the 970 is at 2.0 Ghz. Let's assume the P4 is faster at some things, and the 970 is faster at other things. But, ultimately, these two chips seem to me to be about even (in reality, the 970 is better in FP ops, and the P4 is better on INT ops...but for simplicity, let's say they are even for now).

So, we have this breakdown of "performance ratings"

P4 @ 3.2 Ghz: 100
970 @ 2.0 Ghz: 100

This translates into a "Ghz efficiency rating" of 31.25 for the P4, and 50 for the 970. In other words, the 970 is approximately 60% more efficient per clock!

If, one year from now, the P4 is at 4.0 and the 970 is at 3.0...well then the same "Ghz efficiency rating" ought to translate into these performance scores:

P4 @ 4.0 Ghz: 125
970 @ 3.0 Ghz: 150

Or, that means at next year's speeds, we have a G5 that is 20% faster than the then-current P4.

Now, go further...

Say instead it is a 980 (G6) and the AppleInsider article is accurate (speed increases of up to 40% per clock over the 970).

Then, the comparison becomes:

P4 @ 4.0 Ghz: 125
G6 @ 3.0 Ghz: 210 (max...probably less)

Now, there are two things to keep in mind. It is possible that the 4.0 Ghz speed Intel is targeting is not for the P4, but for the Prescott chip, which presumably will have much greater efficiency per clock than the P4.

Secondly, AMD's Opteron is really where the performance edge in the PC world is right now, even though Intel has all the mindshare among consumers. So, obviously, for now Apple and IBM ought to be competing with AMD until Intel unveils its "Next Big Thing" (i.e., Prescott).

But, when you start to think about a 3.0 Ghz 980 potentially outperforming a 4.0 Ghz P4 by 60% (!), noting that it is capable of dual-processor configurations, it really makes you feel optimistic about 5.0 Ghz G6s, which would perform with the speed of a P4 running at...are you ready...11 Ghz!!! Is intel going to get to 11 Ghz by early 2006?

Gee, somehow I doubt it. Of course, Prescott will change all this...whenever it comes.

Fukui
Nov 20, 2003, 11:55 PM
Originally posted by tortoise
The AMD cores are very well engineered and are also somewhat more "general purpose" in design than PPC cores, which tend to have an emphasis on numerical codes that makes them less optimal for many tasks than the AMD core.
What are "Numerical Codes" ?

jiggie2g
Nov 21, 2003, 02:00 AM
What I think Apple will most likely do is use these 2 processors 970/980
to be completely independent of those bums at Motorolla (who Cut Apples Legs just as we were leading the Speedrace)Labeling the 980 the "G6" and using it in it's Pro Market(Power Mac and Eventually Power Book) . while the G5/970 becomes a Consumer chip (iMac 1st then/eMac and Eventaully ibook). having these 2 Chips Gives Apple Flexibility so they won't have 2 cripple the consumers machines worring about them eating into Power Mac/Power Book sales. and the best part is that we will Finally Be Equal to or better then Performance wise to windoze machines at Both High and Lowend Markets. Intel knows if IBM can get the G Series Chips within 500mhz or less of thier precious Pentium it will be game over apple will start eating into the windoze market share then they'll be the ones talking about "the MegahertzMyth" . I my self am Waithing for a 2GHZ G5 20in iMac ...maybe b4 Next Christmas...

jiggie2g
Nov 21, 2003, 02:01 AM
Well The G6 is every bit as Different to the G5 as the G3/G4 were to one another . Built on 90nn Fab ,PowerTune(making it speed Scaleable to improve batterlife and power consumption) Improved Altivec , Dual Pipelines ,Independent Bus for Each processor in Dual models(Crossbar Switch) , DDR 2 , so Different Technology . as fot the wait between the G3and G4 you have to remember that at that time 1997-1999 we were still Head Perfomance wise , if anyone remember the days when G3's were 3x the Perfomance of a P2 at equal clock speeds forcing Intel to Bring the P3 out so quickly. except now the situation is reversed now Intel is the Leader and Apples Market Share has tuned to Crap as a Result of Motorola's Incompetence being a Major Factor .so Yeah I do expect to see a G6 by summer or 3rd Quater 2004,we are the ones who are behind and have 2 play catch up. the G5 put us neck and Neck with Intels Best but only on the Highend Market, now all the other Mac lines will have 2 follow suit. Jobs will not be satisfied until he can take on the PC on all ground and equal footing. that way atleast apple has a Chance to get out the 3% market rut thier in. either way it will sure as hell be fun to Watch how it goes down, Let the Speed Race begin...Again..lol P.S. Sorry for the Double Post

Kamu-San
Nov 21, 2003, 03:46 AM
What I'd like is a cheap, PC-technology based clone with, instead of a Pentium, a PPC 970.
I really like Apple stuff and I'll buy it just for OS X, but for a lot of things Linux is sufficient and Apple hardware is too expensive to just run Linux.

I'd love to have a cheap PPC970 based Linux-box at home.

Eric_Z
Nov 21, 2003, 03:55 AM
Originally posted by ennerseed
Appleinsider also reports the new machines may include a Crossbar Switch: "and will use a crossbar switch to speed communication between processor"

I believe (don't know what else it could be) this is like SGI's implementation of a Crossbar Switch which could bring major speed improvements.


Or perhaps something like this (http://e-www.motorola.com/files/sndf/doc/reports_presentations/SNDF2003_DALLAS_A2513.pdf) [MAI Logic, se page 9-10 and 18-19]?

Analog Kid
Nov 21, 2003, 06:15 AM
I can't believe how much bandwidth goes into discussing when Apple will change their labels to say G6...

ktlx
Nov 21, 2003, 06:57 AM
Originally posted by suzerain
About the Intel 4 Ghz thing. If they can get their chips up to that, great. I'm all for speed, anywhere.

But, they are currently at 3.2 Ghz, and the 970 is at 2.0 Ghz. Let's assume the P4 is faster at some things, and the 970 is faster at other things. But, ultimately, these two chips seem to me to be about even (in reality, the 970 is better in FP ops, and the P4 is better on INT ops...but for simplicity, let's say they are even for now).

Since both processors need to be reimplemented at 90nm in order to reach those processor speeds, your comparison is not valid in really any way.

When the processors are reimplemented at 90nm, each design team will make changes to the processor. Who knows what each design team will do? Surely no one on this forum who is at liberty to speak.

For example, let's say Intel decides to just do a shrink with no significant design changes. Let's also say that IBM decides to add SMT and 1MB of Level 2 cache. Then the comparison will be skewed toward favoring a 3Ghz PowerPC.

The converse would also be true even though many on this forum are acting as if IBM will add all sorts of magical, wonderful features that will make its processor ideal at 90nm while Intel will do nothing more than reimplement its existing Pentium 4 at 90nm.

geerlingguy
Nov 21, 2003, 07:41 AM
Originally posted by g30ffr3y
i dont even have a g5 yet... they cant put out a g6... holy crap!!!

If you remember, the G5 was being mentioned about two years ago (just not quite as definite). About a year ago, more definite rumors were coming out, then it was released to the public a few months ago. I'd say we won't see "G6" until 2005.

Hopefully I'll be wrong. :D

DrBoar
Nov 21, 2003, 08:50 AM
The IBM 970 is not a stop gap solution any more than the 604 or G3 was.

Thoughout the Macintosh history starting with the 6800 in 1994 every leading edge cpu -020-040, 601, 604, G3 was replaced in about 2 years time. The G3 replaced the 604 in November 1997 and was replaced by the G4 in August 1999.

The freak here is the G4 that lead the race for 4 years ( I am using the term "leading" very loosly here). We should have had a G5 CPU back in 2001 instead of the Quicksilver G4. So there is big demand to replace the old old G4 so I wellcome the 970 and the 980 and any other IBM CPU that will them go faster:p

psurrena
Nov 21, 2003, 09:18 AM
I'm by no means an expert on chips but shouldn't CISC chips be nearing their end as processors grow from 32 to 64 to 128?

As far as I know and please correct me if I'm wrong:

RISC will process a collection of data as close to 64bits at once. So it will collect 8 pieces of 8bit info, one 32bit and four 8bits, one 64bit and so on.

CISC can only process one 8bit, one 32 bit, one 64bit at a time.

SO! As processors move from 64 to 128 an so on, CISC chips with become overwelmingly complex. While I doubt we will move to 128 in our current form (cisc/risc) I think that 64bit will provide a challenge for Intel/AMD while IBM will finally be in a situation where they can move forward without too much sweat.

Nemesis
Nov 21, 2003, 09:20 AM
[QUOTE]Originally posted by illumin8
Illumin8, tell me more about your 15" AluBook. Is it good? Fast enough? Robust? Any special comments about it? Because I wish to buy one for myself ...

Rincewind42
Nov 21, 2003, 09:40 AM
Originally posted by psurrena
I'm by no means an expert on chips but shouldn't CISC chips be nearing their end as processors grow from 32 to 64 to 128?

As far as I know and please correct me if I'm wrong:

RISC will process a collection of data as close to 64bits at once. So it will collect 8 pieces of 8bit info, one 32bit and four 8bits, one 64bit and so on.

CISC can only process one 8bit, one 32 bit, one 64bit at a time.

SO! As processors move from 64 to 128 an so on, CISC chips with become overwelmingly complex. While I doubt we will move to 128 in our current form (cisc/risc) I think that 64bit will provide a challenge for Intel/AMD while IBM will finally be in a situation where they can move forward without too much sweat.

Hopefully whoever gave you that explanation of the whole thing will never say that to someone else :D .

First, CISC & RISC are largely academic descriptions today. There are no modern chips that are truly CISC or RISC. x86 flies a little closer to CISC and PPC closer to RISC simply due to their design heritages, and even then the x86 only looks like a CISC design at the ISA level. Basically Intel put an x86 translator on top of a RISC core back in the Pentium line and declared a victory for CISC that they have been trying to get away from ever since (hence the doomed Itanium).

In the end, there is no true CISC or RISC anymore. Traditional CISC chips have been taking on traits from RISC designs to get more speed, and RISC chips have been gaining ever larger instruction sets in order to do more advanced things. RISC has increasingly become less about instruction set size, and more about instruction set complexity - it's about the only distinguishing factor between CISC & RISC chips. No matter how large the instruction set gets on a RISC chip, each instruction tends to do relatively little and use very few resources. On a CISC chip the large resource hogging instructions that it accepts may be broken down into simpler less intensive instructions on the fly to be fed into a RISC-like core.

Jon the Heretic
Nov 21, 2003, 01:36 PM
Originally posted by Rincewind42
In the end, there is no true CISC or RISC anymore. Traditional CISC chips have been taking on traits from RISC designs to get more speed, and RISC chips have been gaining ever larger instruction sets in order to do more advanced things. RISC has increasingly become less about instruction set size, and more about instruction set complexity - it's about the only distinguishing factor between CISC & RISC chips. No matter how large the instruction set gets on a RISC chip, each instruction tends to do relatively little and use very few resources. On a CISC chip the large resource hogging instructions that it accepts may be broken down into simpler less intensive instructions on the fly to be fed into a RISC-like core.

I never bought this whole "there is no CISC or RISC" argument that has been popularized by the Wintel crowd for several years now. It is the stuff of FUD, intended only to confuse potential customers of RISC processors that staying with the x86 line is A-OK. If there were so little difference, why is Intel so desperately trying to make the transition to Itanium? (Right, Itanium isn't "RISC", it is "EPIC". Insert hyucks here.)

On today's x86 lines, the need to translate x86 instructions to RISC op-codes is an albatross around that design's neck. No one is writing software which is compiled for the op-codes---they compile it for x86. x86 software must contend with the ridiculous complexity of x86 instructions, as well as the nightmare mess of the x86 register set.

I would only accept the "RISC-like" definition of today's Wintel chips if software was no longer compiled for x86 and could totally sidestep the impossible mess of x86 registers. The outer wrapper of microcode and the type of software (compiled for RISC or for CISC) cannot be ignored in the RISC/CISC equation. If you were to compile for the "op-codes" and skip the whole x86 microcode baggage (the "translation" that is), you will have software that won't even run across the complete x86 family! AMD and Intel use different op-codes, and even different Pentiums use different op-codes from each other. Only x86-level code is *safe* for software compatibility and that means CISC software.

I would like to single out a single sentence that I take serious issue with:
Originally posted by Rincewind42
RISC has increasingly become less about instruction set size, and more about instruction set complexity - it's about the only distinguishing factor between CISC & RISC chips.

There is a common misconception that RISC means fewer instructions; this has never been the case! In fact, even the 601 had as much or more instructions than its CISC counterparts of the time. RISC chips will often have more instructions because they ARE simpler instructions, taking fewer cycles to complete per instruction than RISC chips. You often need more instructions in order to do the work of a fewer very complex instructions. CISC chips have many complex specialized instructions that can take many cycles to complete, thus reducing the size of the compiled code and then need to have a huge set of primitive operations. The philosophy behind RISC rejects the need for complex instructions and has pushed for most instructions to be more primitive and to complete in 1 to 2 cycles. What this means is that RISC programs will be bigger when compiled (more primitive instructions, fewer "compact"/complex ones). And it goes without saying that RISC likes lots of generic registers.

The very acroynm for RISC is the source of this confusion, so I understand why so many people get this wrong (although I suspect some Wintel types get it wrong on purpose...!) It has been suggested that RISC and CISC be renamed to something clearer and truer to their fundamental philosophical differences. For RISC: "Reduce Instruction Set to the Compiler". Alas, I can't remember what was suggested for a new CISC name, but it meant this "let the chip handle the complexity of the instruction set, not the compiler". The point is this: When you compiled for a RISC chip, the compiler is doing the work of figuring how everything is scheduled and which instructions are needed to complete each relatively primitive operation, which typically only takes the chip a couple of clock cycles per instruction; but when you compile for CISC, the compiler is primarily working with those higher-level instructions, many of which may take many cycles to execute by the processor, including the processor's need to further decompose those instructions down and down into smaller and smaller operations. The chip is doing that decomposition---not the compiler! This is very UN-RISC like behavior. Remember that since different members of a CISC family may have different "RISC cores" (and hence, different op-codes), the only safe way to guarantee software compatibility with all members of that family is to compile your code for that high-level (read: complex) instruction set.

Some RISC chips also decompose instructions, true, but they are usually simple instructions to begin with. For a 2-cycle instruction, a later implementation of that chip may decompose it into two 1-cycle opcodes based on testing which indicated this decomposition would yield even further performance advantages. The compiler is still doing most of the "work"; decomposition in this context is more of an optimization by the processor than the type of complex translation needed by CISC chips with their so-called "RISC cores".

It is the software and the role of the compiler that makes a chip RISC or CISC, not just the chip design. And as long as you write CISC code for that outer high-level microcode wrapper, no matter how "RISC-like" the chip "core" may be, it is still a *CISC* chip.

primalman
Nov 21, 2003, 02:54 PM
PPC 601 [G1] - 3/94
PPC 603/604 [G2] - 4-5/95
PPC 750 [G3] - 11/97
PPC 74xx [G4] - 9/99
PPC 970 [G5] - 6/03

[the PPC 602 was a server based chip by IBM that never went into desktop type systems]

www.apple-history.com

airmac
Nov 21, 2003, 03:23 PM
All this talk about g6..i mean..apple redesign the case 3x times with g4 and went from 350 mh z to 1.42 ghz. I can see the g6 in late 2005. But that's just me...

tortoise
Nov 21, 2003, 04:20 PM
Originally posted by Jon the Heretic
On today's x86 lines, the need to translate x86 instructions to RISC op-codes is an albatross around that design's neck.

You've simplified it to the point of totally misrepresenting reality. The part of the x86 ISA that is executed in the core is a simple (RISCy) subset of the x86 ISA, with the more CISCy parts being executed in software (microcode). Believe it or not, modern RISC cores go the opposite direction, combining simple opcode combinations into more functionally complex opcodes in the core.

An important thing to realize is that it is easier to optimize a modern CPU core for performance if you have a "simple CISC" ISA than a RISC ISA, which is one of the reasons early "simple CISC" cores (e.g. the original Pentium) did so well performance-wise versus RISC architectures. It allows you to make assumptions that reduce the complexity of making more high-performance architectures by explicitly eliminating instruction dependencies that may be ambiguous in RISC cores.

In fact, if you look at modern PPC cores, they actually translate basic RISC opcodes to something that looks more like the simple CISC opcodes natively executed on most current x86 cores. The problem is that the assumptions under which these old ISAs were designed no longer hold with modern CPU cores, such that neither classic RISC nor classic CISC is particularly optimal. Classic RISC is too simple to optimize throughput in the core, and classic CISC tends to have clunky register structures that can be difficult to use well.

An ideal ISA for modern processor cores bundles more functionality in its opcodes than a classic RISC ISA, but without all the unnecessary and extraneous ISA baggage, like nasty register models, found in older traditional CISC ISAs. An example of an ISA that is more closely optimized for modern processor cores is the AMD64 ISA, which uses simple CISC opcodes but uses a clean RISC-like register model.

ktlx
Nov 21, 2003, 04:26 PM
Originally posted by Jon the Heretic
I never bought this whole "there is no CISC or RISC" argument that has been popularized by the Wintel crowd for several years now. It is the stuff of FUD, intended only to confuse potential customers of RISC processors that staying with the x86 line is A-OK. If there were so little difference, why is Intel so desperately trying to make the transition to Itanium? (Right, Itanium isn't "RISC", it is "EPIC". Insert hyucks here.)


There is a common misconception that RISC means fewer instructions; this has never been the case! In fact, even the 601 had as much or more instructions than its CISC counterparts of the time.

I hope no one mistakes length for correctness. Very little with this write up is correct. I am only going to attend to the two largest mistakes.

As implemented, there is very little difference in the processor cores of today's CPUs caused by the ISA. Anyone wishing to find out more can easily go to ArsTechnica or any one of a number of other sites talking about the CPU core designs. If you are technically inclined, you can often find the architecture presentations by the designers themselves with a quick Google search.

When talking CISC versus RISC you are only talking about ISAs and then you have to go back to the early to mid 80's to discuss real differences. RISC did not suddenly appear in early 90's with the PowerPCs. The PowerPC came relatively late to the game and I would even argue that it is neither a CISC nor RISC ISA. Because it came late, it had the advantage of picking the best parts of both camps.

In order to understand what the original RISC ISA designers were trying to do, you need to understand the popular CISC ISAs of the day. On a 32-bit processor, you had some 8-bit instructions, some 16-bit instructions, some 32-bit instructions and some multiword instructions. Never mind the complexity of operands. You had all sorts of memory addressing instructions ranging from simple "fetch the 32-bit word at address A and place it in this 32-bit register" to "fetch the 32-bit word at address (A plus four times value of register B) and place it in address (C plus four times value of register D)".

Anyone who has written assembly on a VAX 11/780 immediately recognizes what a CISC ISA is. It has a very powerful set of operations available to it. On the other hand, the level of ISA complexity seems almost insane. You could almost translate FORTRAN 4 instruction per instruction into VAX assembler. :D

The original idea of RISC was to simplify the ISA so that all instructions and operands were of the same size (i.e. 32-bits) and to simplify the memory addressing options. An ideal RISC ISA would have only two memory operations available: store register A at the address in register B and read the address in register C to register D. Both operations would work on 32-bit words. All of your other operations would be ALU such as add, subtract, bitwise-or, rotate right, etc. Each of those instructions would only operate on registers and not be able to access memory.

In order to overcome the slow access to memory due to the simple addressing scheme, lots of registers were needed. Some designs had as many as two orders of magnitude more registers as their CISC contemporaries.

However, neither RISC nor CISC were completely right or completely wrong so today the difference is largely meaningless. Anyone who refuses to believe that can simply look at the SPEC scores and chip design overviews on the Web. Chip performance today has nothing to do with CISC versus RISC ISAs.

Rincewind42
Nov 21, 2003, 06:00 PM
Originally posted by tortoise
In fact, if you look at modern PPC cores, they actually translate basic RISC opcodes to something that looks more like the simple CISC opcodes natively executed on most current x86 cores. The problem is that the assumptions under which these old ISAs were designed no longer hold with modern CPU cores, such that neither classic RISC nor classic CISC is particularly optimal. Classic RISC is too simple to optimize throughput in the core, and classic CISC tends to have clunky register structures that can be difficult to use well.

Actually, the 970 doesn't decompose 99% of it's instruction set. The only instructions that are updated are the load/store multiple word instructions, the load/store with update instructions and perhaps a few others. I suspect that there will continue to be very few decomposed instructions - all of the currently decomposed ones were originally not recommended as they could execute more slowly on some PPCs (not that that had been the case before the 970).

I suspect that we are unlikely to see few if any more instructions broken down like this.

Jon the Heretic
Nov 21, 2003, 09:07 PM
Originally posted by ktlx
I hope no one mistakes length for correctness. Very little with this write up is correct. I am only going to attend to the two largest mistakes.

This hyperbolic comment above is just shy of a personal attack ("very little...is correct" and "two largest mistakes".) Riiighht. Because I am bigger than that, I will ignore such obvious baiting for now. Meanwhile, I do find it curious that you make little attempt to support your over-strong position (which was what again...?), except through oblique irrelevancies about instruction size (as opposed to the more central issue about complexity and the role of the compiler which the original author was responding to) and noting (with a resounding duh!) that the PPC 601, while over a decade old now, isn't the first RISC chip. Gee, thanks for that wonderful but completely irrelevant revelation. It should be noted that when the PPC601 was released people were not dismissing the RISC-CISC distinction as meaningful. That's the point...

(You are slightly less than oblique in one area, although it still doesn't address the original article that he so readily slams. You claim that SPEC scores indicate there is no overhead to the ISA translation (something which I feel is inevitable). Using SPEC scores to make this comparison across processor families can not support this wild claim, as this entails comparing Apples to Oranges which differ along far more variables than just CISC and RISC. A meaningful comparison would be to compare the performance of a program that uses the ISA x86 to the execution of a program that didn't use the ISA and was written directly to use the opcodes, bypassing the microcode. The comparison could be made ON THE SAME CHIP; that would say a lot about the overhead of that ISA and wouldn't be Apple and Oranges...)


It is interesting that others here have alluded to the several year old ArsTechnica article which was one of the first public annunciations that there is no difference between RISC and CISC (<http://arstechnica.com/cpu/4q99/risc-cisc/rvc-1.html>). While I love ArsTechnica, the article and its position are NOT gospel and NOT universally accepted. Wintel lovers do like its conclusions, I have noticed in the years since it was first published, as it suits their positions. I think that is sufficient reason to question its conclusions more criticially...

In fact, quite a few people in the forum that originally responded to the ArsTechnica pronouncement of the "post-RISC era" <http://slashdot.org/articles/99/10/21/0848202.shtml> didn't buy it any more than I did, and they made the exact arguments I made above.

Let's look at a few of these counter point of views:

Originally posted by Anonymous Coward
The author of the article rightly notes that a basic design philosophy difference is where the burden of reducing run-time should be placed. The original RISC philosophy was to place the burden on software--this is especially true of (V)LIW processors--whether the programmer, the compiler, or the set of libraries. CISC (or more rightly "old style") design philosophy sought to place the burden on the hardware.

Originally posted by coats
If you take the point of view that a P6 is a RISC core running an x86 interpreter, then still the user-visible architecture is not RISC. It would only be RISC if you let me program the core directly with its native micro-ops. "Hannibal" still doesn't understand this distinction between architecture and implementation.

Originally posted by hattig
RISC does stand for Reduced Instruction Set Chip, but that doesn't mean less instructions, it means less Instruction Formats. Think of how many different instruction formats x86 has, with varying lengths of instructions, non-orthogonal instructions, etc, compared with the simplified instructions provided by RISC processors, which might have as few as 3 or 4 different instruction formats....The article was silly really, the author didn't look beyond the word 'Reduced' in RISC, thought it meant less instructions, then saw that most RISC chips have tonnes more instructions than most CISC chips, and arrived at the wrong conclusion.