IBM on PowerPC 970MP Power Savings

[MacRumors]

http://www.macrumors.com/images/macrumorsthreadlogo.gif (http://www.macrumors.com)

ZDNet (http://news.zdnet.co.uk/hardware/chips/0,39020354,39233637,00.htm) and eWeek (http://www.eweek.com/article2/0,1895,1877406,00.asp) report on a presentation by Norman Rohrer, and IBM engineer at the Fall Processor Forum in San Jose.

Roher discussed the IBM PowerPC 970MP which is being used by Apple in the latest round of PowerMac revisions. IBM describes the new chip as a "low-power, high-performance" processor.

The 970MP houses two processor cores each with its own CPU, AltiVEC unit and 1MB of cache. The previous PowerPC 970 (G5) processor only had a single core. In addition, the new 970MP can shut down one of the two cores and further reduce power consumption by reducing its frequency. Peak consumption is described as 100W which can be reduced to 40W.

Even with such power savings modes, it seems unlikely that the 970MP will approach the anticipiated plans from Intel (http://www.macrumors.com/pages/2005/08/20050823140931.shtml) to produce lower power consumption chips. Intel's future Performance per Watt roadmaps were cited (http://www.macrumors.com/pages/2005/06/20050606143135.shtml) as reasons for the switch to Intel which is planned to begin in 2006.

[combatcolin]

Cool.

Does the power saving mode of OSX take advantage of this?

Can the OS slow the CPU down and switch off a Core if its not needed?

[arn]

Cool.

Does the power saving mode of OSX take advantage of this?

Can the OS slow the CPU down and switch off a Core if its not needed?

not sure... not sure how useful this is on a desktop machine anyway. Presumably sleep mode shuts down power consumption to almost nothing anyway.

arn

[crazzyeddie]

The ability to reduce heat dissipation from 100W -> 40W is sure impressive though, even if most people will never see their chips consume the full 100W. I wonder if Apple's CHUD tools will eventually support manually switching off one of the cores?

Now the only thing left is the real word benchmarks for these new chips. Lets see how they do compared to their Dual Processor counterparts...

And I'm almost positive that OS X should be able to slow the CPU's down. The 'Automatic' setting on all new Powermacs has taken advantage of the Processor Slewing ability of the G5.

[gekko513]

not sure... not sure how useful this is on a desktop machine anyway. Presumably sleep mode shuts down power consumption to almost nothing anyway.

arn
It's useful because it keeps down the fan noise.

[Lord Blackadder]

100W peak power consumption, whew!

Now do you see why there is no G5 PowerBook? ;) :rolleyes: :D

[emotion]

100W peak power consumption, whew!

Now do you see why there is no G5 PowerBook? ;) :rolleyes: :D

Exactly what I was going to post. This is what IBM meant when they said they had 'low power G5s' in the wings!

[Lacero]

I'm not really excited about this announcement. It's like saying, "Hey, look Timmy! I got you a new pair of artificial legs!", or something like that.

[JoeG4]

Yeah, the CPU control has been around for ages on the G5, it's just getting better :)

The thermal stuff is BS in a way; When comparing "apples to apples", the P4 is a monster. All the current 3.2-3.8 chips are rated up to a max of 130w, and even at the 100w rating that dualcore G5 gets, that's still 30w (about 2 low power G5s) cooler than a P4...

Where Apple plans to "make up for speed" they probably will in code. OS X has been "notoriously sloppy" with that one for years, shoot every new build so far has been "slightly tweaked" to squeeze more performance out since 10.0 days, and you KNOW that wasn't just hardware (If I load 10.0.4 on my dual 800 it'll run WAY slower than 10.4.0!)

Right now the only thing Intel has to winning Apple's favor is that they haven't pulled a "motorola" on Apple (yet). While the proof is in the pudding (P4? Massive decision change.. hey wait first NetBurst is the future and then it's crap?!).. sounds kinda like Apple right now..

But they haven't gotten ditched mid-line. Admittedly, I still don't think anyone at IBM was dumb enough to tell Steve Jobs to tell everyone they'd be able to have 3ghz processors in a year. Not even Intel went from 2-3ghz in 1 year, and they're all.. reputed and stuff.. for being "so fast" right now.

It still might be that the whole thing was a political decision -- sure maklar and rosetta have been around for ages, we knew that anyway; The decision to put it in action was not.

Every step in the "Intel" direction reveals yellow tape hanging loose. Intel people say "Some apple people came over and really liked what they saw", IBM people say "we never saw it coming", and Motorola people? Oh wait, they trade insults with Steve Jobs over the ROKR..

A certain company of the AIM alliance just got caught red handed! ;) Unfortunately, IBM gets to pay for it since they're not exactly off the hook for the 3ghz thing.

That, and apparently Steve Jobs really likes options, even that means being kicked out of Nordstrom's for Ross and Big Lots.

[Bonte]

Intel's future Performance per Watt roadmaps were cited as reasons for Apple's switch to Intel which is planned to begin in 2006.


Apple switcht to Intel because all PC's are x86, consoles are PPC but Apple needed this switch to survive in the long run, late but brilliant move. :D

[longofest]

Dude... I submitted this. No credit?

Edit: thanks arn :)

[Lord Blackadder]

Dude... I submitted this. No credit?

<pats longofest on head> good job, buddy!


The thermal stuff is BS in a way; When comparing "apples to apples", the P4 is a monster. All the current 3.2-3.8 chips are rated up to a max of 130w, and even at the 100w rating that dualcore G5 gets, that's still 30w (about 2 low power G5s) cooler than a P4...

True, but EVERYTHING runs cooler than a P4, so lets not get excited. Suffice to say, the G5 is a hot chip.

[igetbanned]

<pats longofest on head> good job, buddy!




True, but EVERYTHING runs cooler than a P4, so lets not get excited. Suffice to say, the G5 is a hot chip.

Everything except a hideously overclocked AMD.:D

[Steamboatwillie]

You know, all the hype about the switch to Intel, for whatever technical or political reason, does nothing for me. I work on R/S6000 POWER based computers at work running AIX and it's the bees knees lemme tell you. I think that the POWER platform is awesome and I for one will miss it when it's gone. Don't get me wrong, I am all for faster, cheaper, low power, low heat cpus I just like IBM.

[gekko513]

<pats longofest on head> good job, buddy!




True, but EVERYTHING runs cooler than a P4, so lets not get excited. Suffice to say, the G5 is a hot chip.
Dual Core Opteron 95W
Dual Core G5 100W
Dual Core Xeon 150W

I wouldn't say it's very hot. It's comparable to the best competitor.

[JoeG4]

Aint nothing like 'packing a set of G5s though. It's like sitting at the wheel of a modern muscle car or something. Vroom vroom!

[emotion]

You know, all the hype about the switch to Intel, for whatever technical or political reason, does nothing for me. I work on R/S6000 POWER based computers at work running AIX and it's the bees knees lemme tell you. I think that the POWER platform is awesome and I for one will miss it when it's gone. Don't get me wrong, I am all for faster, cheaper, low power, low heat cpus I just like IBM.

Everyone I know hates AIX.

The intel move is initially about laptop cpus not desktop. Give it two years and intel will have dumped all trace of Netburst (P4s/Xeon) then they'll be in better shape for the desktop.

[Lord Blackadder]

Dual Core Opteron 95W
Dual Core G5 100W
Dual Core Xeon 150W

I wouldn't say it's very hot. It's comparable to the best competitor.

I wasn't saying that it's unreasonably hot, but isn't cool either.

Of course, the real comparison will be with the next Intel chips, since that's where we are headed.

[swissmann]

Can anyone compare this to what we are told is coming from Intel?

[Dr. Dastardly]

Really to little to late for IBM. Yeah the quad is impressive but if Apple really stuck around with IBM they would have monster desktops and the absolute most under powered laptops ever. The powerbook line is long in the tooth now just imagine only two or three years down the road with the rate of their updates.

2ghz G5 in 2008!!

[AidenShaw]

(P4? Massive decision change.. hey wait first NetBurst is the future and then it's crap?!).. sounds kinda like Apple right now...
Really does sound like Apple...

68K is the future, then it's crap
601 is the future, then it's crap
603 is the future, then it's crap
G3 is the future, then it's crap
G4 is the future, then it's crap
G5 is the future, then it's crap


The term "future" means "the next couple of years" in the computer industry - for Apple, for Intel, for IBM....

[makman]

Be very wary of what Intel has sold Apple on its roadmap.

There's been a lot of discussion that Paxville, their new dual core Xeon, is absolutely terrible, eating massive power and still behind AMD Opteron:

http://slashdot.org/article.pl?sid=05/10/19/1426240&tid=118

In addition, Intel just tore up much of their roadmap, cancelling Whitefield (Xeon) and introducing yet another delay to Montecito (Itanium):

http://www.theinquirer.net/?article=27192

Finally, just today one of Infoworld's columnists voiced just what he thought about Intel's future roadmap, and it doesn't look pretty:

" Another journalist at the same event posited that AMD’s technological lead over Intel will be short-lived and is calling “game over” once Intel’s new Pentium M-derived cores debut across the product line. With due respect to my colleague, AMD will extend its lead, showing Intel’s reactive strategy for what it is."

http://www.infoworld.com/article/05/10/26/44OPcurve_1.html?source=rss&url=http://www.infoworld.com/article/05/10/26/44OPcurve_1.html

Finally, here's what the market thinks about Intel's prospects in the 3 months:

http://finance.yahoo.com/q/bc?t=3m&s=INTC&l=on&z=m&q=l&c=amd

and that's with massive Intel buyback of stock. Clear institutional distribution.

Don't assume all is fine and dandy at Intel.

[dontmatter]

Yeah, the CPU control has been around for ages on the G5, it's just getting better :)

The thermal stuff is BS in a way; When comparing "apples to apples", the P4 is a monster. All the current 3.2-3.8 chips are rated up to a max of 130w, and even at the 100w rating that dualcore G5 gets, that's still 30w (about 2 low power G5s) cooler than a P4...

Where Apple plans to "make up for speed" they probably will in code. OS X has been "notoriously sloppy" with that one for years, shoot every new build so far has been "slightly tweaked" to squeeze more performance out since 10.0 days, and you KNOW that wasn't just hardware (If I load 10.0.4 on my dual 800 it'll run WAY slower than 10.4.0!)

Right now the only thing Intel has to winning Apple's favor is that they haven't pulled a "motorola" on Apple (yet). While the proof is in the pudding (P4? Massive decision change.. hey wait first NetBurst is the future and then it's crap?!).. sounds kinda like Apple right now..

But they haven't gotten ditched mid-line. Admittedly, I still don't think anyone at IBM was dumb enough to tell Steve Jobs to tell everyone they'd be able to have 3ghz processors in a year. Not even Intel went from 2-3ghz in 1 year, and they're all.. reputed and stuff.. for being "so fast" right now.

It still might be that the whole thing was a political decision -- sure maklar and rosetta have been around for ages, we knew that anyway; The decision to put it in action was not.

Every step in the "Intel" direction reveals yellow tape hanging loose. Intel people say "Some apple people came over and really liked what they saw", IBM people say "we never saw it coming", and Motorola people? Oh wait, they trade insults with Steve Jobs over the ROKR..

A certain company of the AIM alliance just got caught red handed! ;) Unfortunately, IBM gets to pay for it since they're not exactly off the hook for the 3ghz thing.

That, and apparently Steve Jobs really likes options, even that means being kicked out of Nordstrom's for Ross and Big Lots.

IBM is Nordstroms and intel is Big Lots?

Admittedly, I don't claim to be deep in knowledge of processors and their relative differences, but I will say this: however much the quad may rock, and PPC may be a smart arhitecture, apple's got nothin' for laptops now, and even less on the road map. NOTHING.

So, apple can move in the area where the PPC/power architecture does work well-high end workstations and desktops, and sacrafice essentially all of the consumer market, as well as go into direct competition with IBM, which is bad, as apple is dependent on IBM for chips.

Or, apple can move to intel, get a HUGE performance and power gain in laptops and a very nice looking road map, and trust that by the time apple gets to intel powermacs, intel's next generation of desktop chips will be here and will be truly powerful.

[longofest]

Really does sound like Apple...

68K is the future, then it's crap
601 is the future, then it's crap
603 is the future, then it's crap
G3 is the future, then it's crap
G4 is the future, then it's crap
G5 is the future, then it's crap


The term "future" means "the next couple of years" in the computer industry - for Apple, for Intel, for IBM....

Remember that just because a new chip is coming out, or because a tech vender like Apple goes from one chip design to another (ppc -> x86) doesn't make the old immediately "crap." I am getting a Quad, and it will most likely be supplanted by a x86 PowerMac in about a year, but that doesn't mean that it is then crap. It is still what it is. It hasn't magically become slower. It just means that technology has moved on.

It may feel slower if I put the latest and greatest software on it that is meant to use the latest and greatest hardware which I no longer have. For instance, I tried to use PhotoBooth on my Dual 1Ghz G4 with a Geforce 4 Ti (non-core Image compatible). It was sluggish to say the least. But that is because it was using technology that my hardware wasn't ready to fully support.

Believe me, you will all feel better if you stop thinking of your machines as crap once newer machines come out.

[dontmatter]

Remember that just because a new chip is coming out, or because a tech vender like Apple goes from one chip design to another (ppc -> x86) doesn't make the old immediately "crap." I am getting a Quad, and it will most likely be supplanted by a x86 PowerMac in about a year, but that doesn't mean that it is then crap. It is still what it is. It hasn't magically become slower. It just means that technology has moved on.

It may feel slower if I put the latest and greatest software on it that is meant to use the latest and greatest hardware which I no longer have. For instance, I tried to use PhotoBooth on my Dual 1Ghz G4 with a Geforce 4 Ti (non-core Image compatible). It was sluggish to say the least. But that is because it was using technology that my hardware wasn't ready to fully support.

Believe me, you will all feel better if you stop thinking of your machines as crap once newer machines come out.

Well said. And enjoy he quad, that thing's gonna be smokin'!

[combatcolin]

What i would really like in a Power Mac is the ability to run at the slowest speed possible and then ramp up when things get busy and then slow back down again.

Does anyone need multi Ghz processing to run a web browser or a word processor?

Electricty bills are never going to get any cheaper!

[texlaw04]

Remember that just because a new chip is coming out, or because a tech vender like Apple goes from one chip design to another (ppc -> x86) doesn't make the old immediately "crap." I am getting a Quad, and it will most likely be supplanted by a x86 PowerMac in about a year, but that doesn't mean that it is then crap. It is still what it is. It hasn't magically become slower. It just means that technology has moved on.

...

Believe me, you will all feel better if you stop thinking of your machines as crap once newer machines come out.

Thanks for bringing us back to reality! Yes, new is good, but yesterday's machines, if well designed and well-made, are not going to become less productive for the core apps most of us use.

Mobile computing has to be at the DRM standard? Not exactly. We have already seen that the most consistent sellers in the notebook market are mid-range, non-DRM designs with a good balance of performance, weight, battery life, and peripheral support. Why? Because most of us buy a notebook for actual "mobile" computing. Personally, I ditched my Firewire Powerbook [stifles a tear] because it hurt my shoulder to carry it across the UT campus and wasn't supported by the testing software at UT School of Law. The 4.4 lb, magnesium-cased, powder-coated P3 750 Toshiba Portege 4005, a true business-class machine, has been with me for almost 4 years and it continues to perform business-related functions very well, despite being slow by modern standards. Still, does Office 2003 fail to load? No. Have the replacement machines in the sub-5lb market become lighter or gained a larger screen? No. Do I even pretend to do multimedia functions? Not bloody likely. I bought the best machine available (for my intended use) at that time for a fair price and, 4 years later, it's still trucking. And my 23" Cinema consoles me when I'm at home.

Toshiba and Apple...shouldn't there be a law against that marriage? :rolleyes:

[JoeG4]

No future for PPC laptops? You kidding me?

7448s? 8641s? What about IBM's low power G5 that's pretty much ready for production as it is? (I think they're even already shipping, if not about a month or two away).

Having said that, I agree, it still isn't quite the Pentium M offerings, but I find it a tough pill to swallow to switch all the desktops JUST for the laptops.

[plinden]

Dual Core Opteron 95W
Dual Core G5 100W
Dual Core Xeon 150W

I wouldn't say it's very hot. It's comparable to the best competitor.
Don't forget the G5 is considerably smaller than many, if not most, other mainstream CPUs. I'm going on memory now, so I'm probably wrong, but I seem to recall that the G5 is 1/3 the size of the Xeon, so the heat output/square mm is twice as high.

[solvs]

Dude... I submitted this. No credit?
Congratulations! You get a cookie.:p

http://www.morguefile.com/imageData/LOWREZ/cchipcookie002.jpg

But seriously, I liked the G5 and all too... but Apple knows the G4 is going nowhere fast. And the G5, though nice now (I have one... finally) won't help them much in the future. Intel says they have something, I hope they do. The P-M isn't that bad. If the P5, or whatever they call it, is more like it than the P4, we should be ok.

If not, there's always AMD.:D

[nagromme]

No future for PPC laptops? You kidding me?

7448s? 8641s? What about IBM's low power G5 that's pretty much ready for production as it is? (I think they're even already shipping, if not about a month or two away).
In what quantities can Apple COUNT on those chips? This has been a problem with both IBM and Freescale in recent years. It delayed the iMac G5 so long that there WERE no iMacs on sale for months.

And if Apple were planning on jumping on one of those allegedly available-soon PPC chips, they wouldn't have just updated the other specs on the PowerBooks.


Having said that, I agree, it still isn't quite the Pentium M offerings, but I find it a tough pill to swallow to switch all the desktops JUST for the laptops.
Laptops are the most URGENT reason to move to Intel (and laptops are the biggest sellers), but not the only one. Desktops will benefit too.

Remember, this transition is NOT based on Intel vs. PPC today. It's based on FUTURE chips. It's based on Intel vs. PPC next year, and the year after that.

So Intel in a PowerMac today is not needed. The G5 is a great chip. But the time will come in FUTURE when Intel has something better. Apple's getting ready in advance.

[DavidCar]

Something that concerns me is that the Quad tech specs list:

Line voltage: 100 - 125V AC or 200 - 240V AC (wide-range power supply input voltage)
Maximum current: At least 10A (low-voltage range) or 5A (high-voltage range)

What does this mean by "At least?" This already means 1000 to 1250 Watts, "at least."

I hope the power can be throttled down with all the interesting 970mp power options available.

[Fabio_gsilva]

I'm not really excited about this announcement. It's like saying, "Hey, look Timmy! I got you a new pair of artificial legs!", or something like that.

Well, sorry, I don't got it...

[AidenShaw]

Something that concerns me is that the Quad tech specs list:

Line voltage: 100 - 125V AC or 200 - 240V AC (wide-range power supply input voltage)
Maximum current: At least 10A (low-voltage range) or 5A (high-voltage range)

What does this mean by "At least?" This already means 1000 to 1250 Watts, "at least."

I hope the power can be throttled down with all the interesting 970mp power options available.
Also note that the new PM has a different power cord.

They went from the 10 amp (recommended) IEC C14 connector to the 16 amp IEC C20 connector....

http://www.quail.com/iec.cfm

[bigwig]

Remember, this transition is NOT based on Intel vs. PPC today. It's based on FUTURE chips. It's based on Intel vs. PPC next year, and the year after that.

I used to work for a company that relied on Intel's "future" (SGI). Just look at what Intel has done with the Itanium, compare that to their promises, and realize that relying on future projections from Intel is a fool's game.

Furthermore, look at Intel's current roadmap. Low power, low speed, and multi-core. The only reason that Intel has a performance advantage over the G4/G5 is Intel was able to massively increase clock rate to compensate for the x86's terrible ISA and architecture. PPC has always had a significant advantage in power consumption, and with Intel apparently abandoning massive clock rates the x86's advantages all but disappear. A G5 core is much smaller than an x86, allowing for more cores per die, it uses less power per core, and easily matches the low clock rates Intel is talking about (low-to-mid 2GHz). A G5 will blow away Intel's similarly speced units, especially on vector FP.

Since it's IMO obvious that a move by Intel to lower-speed cores only makes PPC more, not less, competitive with x86, why go Intel? Was it IBM's reputed insistence that Apple pony up some development dollars that drove Apple to Intel?

[dongmin]

No future for PPC laptops? You kidding me?

7448s? 8641s? What about IBM's low power G5 that's pretty much ready for production as it is? (I think they're even already shipping, if not about a month or two away).Show me some proof that these aren't vaporware. If the 7448 were shipping as they were supposed to by this time, they would be inside the new PBs. They're pin-compatible, so there's no reason for Apple to not use them. Same old story for Motorola/Freescale.

[manu chao]

What i would really like in a Power Mac is the ability to run at the slowest speed possible and then ramp up when things get busy and then slow back down again.

Does anyone need multi Ghz processing to run a web browser or a word processor?

Electricty bills are never going to get any cheaper!

Apparently that is where the Pentium M really excels.

[DavidCar]

Also note that the new PM has a different power cord.

They went from the 10 amp (recommended) IEC C14 connector to the 16 amp IEC C20 connector....

http://www.quail.com/iec.cfmSo the maximum current is "at least" 10 amps, and at most 16 amps? Then "at least" 1000 to 1250 watts up to 1600 to 2000 watts?

I hope someone does some power consumption tests on these things.

[AidenShaw]

So the maximum current is "at least" 10 amps, and at most 16 amps? Then "at least" 1000 to 1250 watts up to 1600 to 2000 watts?

I hope someone does some power consumption tests on these things.
The previous generation G5 had up to a 600 watt power supply - at full load and with typical efficiency (75%), that's 800 watts or 7.3 amps steady. Typical load would probably be half that, unless you were loaded with disks and high wattage graphics and other cards.

However, the "peak" load is at poweron, when the capacitors fill. (Ever notice the lights dim a bit when you turn it on from cold?)

I didn't see the wattage rating for the dual-dual power supply, but presumably it's a bit larger than 600 watts, and the power-on surge exceeded 10 amps.

Product safety and regulatory groups like UL and similar agencies won't approve a device if the peak current (even for an instant) exceeds the recommended rating of the cord and receptacle. Therefore, Apple moved from the standard 10 amp IEC socket to the 16 amp socket.

This shouldn't mean that the new system draws more than 10 amps continuously - and its typical consumption may even be lower than the previous model.

[zang]

The problem I see that Intel currently has is that PC makers don't innovate. They simply use what's available, rather than help shape the future. IMO, it's been nothing but a hindrance to Intel's development. They have to lead, and yet have no idea what their lead is being created for. Imagine creating an engine for a "generic" car that could be an SUV, a sportscar, compact, without ever having the car companies helping you design it or telling you what it's going to go into and whether they need better economy, more torque, or more speed. That's been happening to Intel for the past 20 years.

Apple sits in the top 10 computer manufacturers in terms of units sold. They're suddenly hopping into a partnership with Intel. Apple knows what it wants it's processors for, and has always tried to build the processor into the machine, rather than build the machine onto the processor. Apple can provide Intel with focus and goals, rather than just asking "so what do you have that's new and backwards compatible?"

Bear in mind that Apple *is* a co-developer of the PPC. They're 1/3 owner of the technology behind it, and have even developed some parts single-handedly (Altivec for instance). Even though Apple never manufactured chips itself, it was always integral in the development of the PPC.

A partnership with Intel makes sense right now. Apple needs a partner who can supply things on time and to it's requirements, and Intel can finally get a partner who wants to forward it's platform and can (and will) help in the development of the processors.

[aegisdesign]

In what quantities can Apple COUNT on those chips? This has been a problem with both IBM and Freescale in recent years. It delayed the iMac G5 so long that there WERE no iMacs on sale for months.

IBM have said in the past that it's also Apple's procurement process which causes problems there. IBM have limited fab space for what amounts to essentially a custom processor and Apple order in small quantities with short lead times to keep inventory low. If IBM are busy making other chips then Apple miss their slot. Intel on the other hand just keep on churning out commodity chips so there's no danger of Apple missing a slot.

And if Apple were planning on jumping on one of those allegedly available-soon PPC chips, they wouldn't have just updated the other specs on the PowerBooks.

Who knows. I imagine they've had the Intrepid2 chipset in the works for a year expecting the 7448 to be ready for it but some delay or technical hitch meant they stuck with the 7447A. Or maybe indeed, quantities weren't available.

Laptops are the most URGENT reason to move to Intel (and laptops are the biggest sellers), but not the only one. Desktops will benefit too.

Laptops are only just the biggest sellers. Are they the highest margin sellers though?

Desktops won't benefit from Intel for quite some time yet based on the projections in the roadmaps for chips that haven't been released yet.

Remember, this transition is NOT based on Intel vs. PPC today. It's based on FUTURE chips. It's based on Intel vs. PPC next year, and the year after that.

Indeed, and unless you have a crystal ball, or you're Steve Jobs and you've been told the chip vendors roadmaps for both PPC and Intel, then predicting what each vendor has in the works is tricky.

So Intel in a PowerMac today is not needed. The G5 is a great chip. But the time will come in FUTURE when Intel has something better. Apple's getting ready in advance.

Actually, I think the transition is more subtle than that. With a switch to Intel the Mac will never be slower than the Wintel machines (except for the odd time when AMD is ahead like now). This is Apple finally saying that they aren't going to gamble on the PowerPC being faster than Intel and they aren't going to invest in R&D to ensure that. They spent a fortune on the G5. It's more cost effective to admit defeat.

It's a great pity. And egg will be on Apple's face if IBM, PA Semi or Freescale (however unlikely) come out with lower power, faster than Intel chips in the next couple of years.


This notion of AMD being hotter is like.... 2 years being out of date. Where have you been those 2 years :D

True except in mobile chips. The Turion just doesn't get close to Pentium M. This transition is all about mobile chips.

[elgruga]

IF IBM do manage to make a fast and cool chip in the next two years, Apple can just buy it, and put it in a box, cant they?
Or are we abandoning PPC so that all the work already done has gone?
Can someone explain what Apple are trying to do?

Is it going to be that all PC software will run on OSX Apple/Intel hard and software?
Are we going to have a simple competitor for Windows in OSX?
Or are we going to just have a whole new set of software being written for OS X on Intel?
Does this mean that software co.'s will have a version for Windows, a version for Apple on PPC, a version for Apple on Intel and a version for Linux etc.?

I dont really get what Apple is doing - are they going to run Intel/OSX and PPC/OSX side by side?

[840quadra]

IF IBM do manage to make a fast and cool chip in the next two years, Apple can just buy it, and put it in a box, cant they?
Or are we abandoning PPC so that all the work already done has gone?
Can someone explain what Apple are trying to do?

Is it going to be that all PC software will run on OSX Apple/Intel hard and software?
Are we going to have a simple competitor for Windows in OSX?
Or are we going to just have a whole new set of software being written for OS X on Intel?
Does this mean that software co.'s will have a version for Windows, a version for Apple on PPC, a version for Apple on Intel and a version for Linux etc.?

I dont really get what Apple is doing - are they going to run Intel/OSX and PPC/OSX side by side?

Many of your questions have been answered more then once on these forums. I would suggest doing a search in regards to "intel".

In short.

1. Apple has publicly stated is roadmap for the future. PPC is being phased out. This may change, or have a back door agenda that isn't known to the public. Apple says that new PPC products will end some time next year.

2. No , Regular PC software will not run on the new intel Macintosh computers. The only exception would be unix based applications that will port, or just work with the Darwin BSD core of OS X.

3. A simple competitor for Windows in OS X? I have no ideah what you are after with this question

4. Look up "Universal Binaries" on the forums, that will answer most of these questions. Otherwise Rosetta will handle legacy programs, excluding classic.

http://forums.macrumors.com/image.php?u=47064&dateline=1127904880&type=profile

[Lacero]

Well, sorry, I don't got it...
Me neither. When I read it back, it probably means Timmy is already crippled, like the PowerPC chips. So if someone gave Timmy an artificial leg to replace his crutches, he's still crippled. So with even these low-power announcements, Timmy isn't too enthusiastic.

[Mr_T]

I have only a layman's understanding of the inner workings of a processor, so this may be terribly ignorant...

I don't see why IBM would be touting this. What's the point of getting a dual core chip and then saying "but wait! it can use much less energy and dissapate less heat when it automatically shuts down one of it's cores and reduces the frequency of the remaining core!" Why should I be excited about a super fast chip which can basically turn itself into a much slower chip? Doesn't that seem like saying, "My car can be much quieter when I only go 5 miles/hour?". Granted I use my Mac more than most folks (number of active programs open at once, heavy photoshop use, etc...), but then again my wife complained about her iMac being a bit slow/unresponsive at times. I fixed that by setting her energy settings to "highest". So again, what's so exciting about basically underclocking your expensive computer?

~ Mr.T

[840quadra]

I have only a layman's understanding of the inner workings of a processor, so this may be terribly ignorant...

I don't see why IBM would be touting this. What's the point of getting a dual core chip and then saying "but wait! it can use much less energy and dissapate less heat when it automatically shuts down one of it's cores and reduces the frequency of the remaining core!" Why should I be excited about a super fast chip which can basically turn itself into a much slower chip? Doesn't that seem like saying, "My car can be much quieter when I only go 5 miles/hour?". Granted I use my Mac more than most folks (number of active programs open at once, heavy photoshop use, etc...), but then again my wife complained about her iMac being a bit slow/unresponsive at times. I fixed that by setting her energy settings to "highest". So again, what's so exciting about basically underclocking your expensive computer?

~ Mr.T


Look at it this way, if you have high energy costs, need to keep systems online 24/7 for use as processing nodes or servers, having a system that will idle down while remainging "fully online" (not sleep) is a good thing.

Some tasks are more disk, memory, or network intensive then others. Having an intelligent system that can turn down the processor speed (a large energy user) in those situations is good, especially when you have allot systems running.

In the data center I work at, we will give customers a price break if they allow such features to be enabled in their system BIOS (talking Rack mount Dell and HP servers).


Also note that the new PM has a different power cord.

They went from the 10 amp (recommended) IEC C14 connector to the 16 amp IEC C20 connector....

http://www.quail.com/iec.cfm

Now I have to borrow power cords from our Dell 6650s as opposed to our HP DL3xx servers :)

http://forums.macrumors.com/image.php?u=47064&dateline=1127904880&type=profile

Me neither. When I read it back, it probably means Timmy is already crippled, like the PowerPC chips. So if someone gave Timmy an artificial leg to replace his crutches, he's still crippled. So with even these low-power announcements, Timmy isn't too enthusiastic.

Sometimes we wonder about you Lacero :) :p

http://forums.macrumors.com/image.php?u=47064&dateline=1127904880&type=profile

[Quartz Extreme]

not sure... not sure how useful this is on a desktop machine anyway. Presumably sleep mode shuts down power consumption to almost nothing anyway.

arn

The goal of this is likely to make the system run cooler/quieter.

Does the power saving mode of OSX take advantage of this?

Can the OS slow the CPU down and switch off a Core if its not needed?

Yes. There are settings in Energy Saver: maximum, reduced and automatic. Maximum keeps all cores running at full clock speed all the time, reduced turns off a core and clocks it down, and automatic switches it between the two dynamically depending on the task.

I would guess that the quad system would turn one core on each chip off when in this reduced power mode, as this architecturally makes sense.

Where this ability to disable a core for power saving will really shine is in the Intel PowerBook of the near future. (Intel's successor to the Pentium M, codenamed Yonah, is a low-power, 64-bit, dual-core chip, due out in mid 2006. It currently runs at 2.16GHz on a 667MHz FSB, and it can process SSE3 instructions and some cool new virtualization technologies. It is also completely new design, built from scratch! Now that is one insanely great chip. :D)

[Mr_T]

Look at it this way, if you have high energy costs, need to keep systems online 24/7 for use as processing nodes or servers, having a system that will idle down while remainging "fully online" (not sleep) is a good thing.

Some tasks are more disk, memory, or network intensive then others. Having an intelligent system that can turn down the processor speed (a large energy user) in those situations is good, especially when you have allot systems running.


Thanks for that explanation. This is probably why the announcement doesn't excite me. I don't think I'll be running any Powermacs as a server (or a node in some rendering farm). I can see why this is a good thing for some though.

~ Mr.T

[solvs]

You -> Stupid

This notion of AMD being hotter is like.... 2 years being out of date. Where have you been those 2 years :D
hideously overclocked AMD
;)

[AidenShaw]

Now I have to borrow power cords from our Dell 6650s as opposed to our HP DL3xx servers :)

http://forums.macrumors.com/image.php?u=47064&dateline=1127904880&type=profile
I'd use the cords from DL58x servers - I prefer HP to Dell ;-)

Intel's successor to the Pentium M, codenamed Yonah, is a low-power, 64-bit, dual-core chip, due out in mid 2006. It currently runs at 2.16GHz on a 667MHz FSB, and it can process SSE3 instructions and some cool new virtualization technologies.
Yonah is currently due around the end of 2005, and is 32-bit. It has SSE3, VT (Vanderpool virtualization technology), and a 667 MHz bus.

The next pass, Merom, is 64-bit and is due mid 2006.

Here's a roadmap table from AnandTech (http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2554):

Pentium M Roadmap
CPU Codename GHz Cores FSB Launch
TBD Merom TBD 2 667 2H'06
x60 Yonah 2.33 2 667 2H'06
x58 Yonah LV 1.83 2 667 2H'06
x50 Yonah 2.16 2 667 Jan'06
x48 Yonah LV 1.66 2 667 Jan'06
x40 Yonah 2 2 667 Jan'06
x38 Yonah LV 1.5 2 667 Jan'06
x30 Yonah 1.83 2 667 Jan'06
x20 Yonah 1.66 2 667 Jan'06
780 Dothan 2.26 1 533 Jul'05
778 Dothan LV 1.6 1 400 Jul'05
773 Dothan ULV 1.3 1 400 Jan'06
766 Yonah 1.83 1 667 2H'06
756 Yonah 1.66 1 667 Jan'06
TBD Yonah ULV 1.2 1 533 Q2'06
TBD Yonah ULV 1.06 2 533 2H'06
TBD Yonah ULV 1.06 1 533 Q2'06

[stockscalper]

I'd use the cords from DL58x servers - I prefer HP to Dell ;-)


Yonah is currently due around the end of 2005, and is 32-bit. It has SSE3, VT (Vanderpool virtualization technology), and a 667 MHz bus.

The next pass, Merom, is 64-bit and is due mid 2006.

Here's a roadmap table from AnandTech (http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2554):

Pentium M Roadmap
CPU Codename GHz Cores FSB Launch
TBD Merom TBD 2 667 2H'06
x60 Yonah 2.33 2 667 2H'06
x58 Yonah LV 1.83 2 667 2H'06
x50 Yonah 2.16 2 667 Jan'06
x48 Yonah LV 1.66 2 667 Jan'06
x40 Yonah 2 2 667 Jan'06
x38 Yonah LV 1.5 2 667 Jan'06
x30 Yonah 1.83 2 667 Jan'06
x20 Yonah 1.66 2 667 Jan'06
780 Dothan 2.26 1 533 Jul'05
778 Dothan LV 1.6 1 400 Jul'05
773 Dothan ULV 1.3 1 400 Jan'06
766 Yonah 1.83 1 667 2H'06
756 Yonah 1.66 1 667 Jan'06
TBD Yonah ULV 1.2 1 533 Q2'06
TBD Yonah ULV 1.06 2 533 2H'06
TBD Yonah ULV 1.06 1 533 Q2'06


Nice long list of vaporware

[AidenShaw]

Nice long list of vaporware
...vaporware that Apple is staking its future on

[Steamboatwillie]

Everyone I know hates AIX.

The intel move is initially about laptop cpus not desktop. Give it two years and intel will have dumped all trace of Netburst (P4s/Xeon) then they'll be in better shape for the desktop.

Funny, I hear the opposite from my peers. I have found AIX to be the best platform for enterprise computing that I have ever had the pleasure of using. Either way I still like IBM. I don't have anything against Intel. I am sure the new Intel Apples will be awesome.

[emotion]

Nice long list of vaporware

What? Those Dothans are out, on time and they're every bit as fast as they said they would be.

Don't confuse The Pentium M line with all this Pentium D/4/Netburst crap they've had out...that stuff ain't going anywhere near a mac.

In any case IBM/Moto have touted their fair share of vapourous hardware over the years and I'm willing to give another set of jokers a chance :) (as are Apple). At least this vapour has a chance at being cheaper because of the volumes made.

To give you an idea of how fast a Dothan is in comparison to G5 and G4 here's a performance list for my main application (Ableton Live). Admittedly the app needs some optimisation for Apple hardware but the results speak for themselves.

http://www.ableton.com/forum/viewtopic.php?t=24715

Should cheer you up a bit.

[emotion]

Funny, I hear the opposite from my peers. I have found AIX to be the best platform for enterprise computing that I have ever had the pleasure of using. Either way I still like IBM. I don't have anything against Intel. I am sure the new Intel Apples will be awesome.

Well, it's what you're used to I guess. I'm not as experienced with AIX (for HPC, which is what I do) as I am with Irix, Unicos and Linux.

[igetbanned]

My post was deleted.

bummer.

[avus]

...vaporware that Apple is staking its future on

At least Apple won't be alone being frustrated with a chip company. Seriously, Apple can have relationships with IBM, AMD, and Intel at the same time - Why not?

[BGil]

Dual Core Opteron 95W
Dual Core G5 100W
Dual Core Xeon 150W

I wouldn't say it's very hot. It's comparable to the best competitor.

Opteron's are available in 55w as well.

The AMD TDP rating is absolute maximum under worst possible conditions (i.e. at 96w the chip or mobo will probably go up in smoke). The G5 rating is expected wattage at full-load and the Intel rating is 75% of maximum.


Seriously, even if you try to massively overclock the AMD proc you will never get it to 95w.

[BGil]

At least Apple won't be alone being frustrated with a chip company. Seriously, Apple can have relationships with IBM, AMD, and Intel at the same time - Why not?

I thinkApple may have a deal with Intel that bars them from working with AMD for at least a few years. I don't think Intel would spend all this money on Apple's transition to X86 if Apple could easily just jump to AMD once the changeover is complete.

I bet the contract is for a decade or more.

[rlwimi]

This is HILARIOUS.

But,but,but Uncle Steve told us that PPC sucks now and this fantastic Intel ROADMAP.

The ROADMAP!!!

Love my iPod Steve, but ya really shouldn't have tried to nickel and dime IBM over the mobile G5 chip...

Have fun with the Intel processor trainwreck! You earned it Stevie Boy! Too bad I won't be along for the ride. I'll be off in the computational promised land called Cell.

[BGil]

Cell? With what OS? On what machine? A PS3?

[AidenShaw]

The AMD TDP rating is absolute maximum under worst possible conditions (i.e. at 96w the chip or mobo will probably go up in smoke). The G5 rating is expected wattage at full-load and the Intel rating is 75% of maximum.
Yes, but the Intel chip is about as likely to hit 100% as an Apple system is to hit the GFLOPS ratings that Apple quotes - in other words 0%.

The absolute maximum assumes all data in L1 and L2 caches, all pipelines and execution units always busy, .... One cache miss or bubble and you're below the max. (Sounds like Apple and Intel are using roughly the same formula.)

In addition, the Intel chips automatically reduce the heat generation if the chip gets too hot - so in effect the Intel chip cannot exceed the capability of whatever heat sink system you put on it.

[BGil]

Yes, but the Intel chip is about as likely to hit 100% as an Apple system is to hit the GFLOPS ratings that Apple quotes - in other words 0%.

The absolute maximum assumes all data in L1 and L2 caches, all pipelines and execution units always busy, .... One cache miss or bubble and you're below the max. (Sounds like Apple and Intel are using roughly the same formula.)

In addition, the Intel chips automatically reduce the heat generation if the chip gets too hot - so in effect the Intel chip cannot exceed the capability of whatever heat sink system you put on it.

Okay but that still doesn't really tell you how much power or heat is being pushed through those processors. The best way to measure the differences would be to use similar configs and test the power draw at the wall. that's what a few PC sites do.
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2353&p=4
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2410&p=2

Intel's stated TDP of the P4 630 (3ghz) is 84w and AMD's stated TDP of a 2.4ghz dual core Athlon X2 4800+ is 110w but the AMD proc uses 52 less watts at full-load. That just shows you how much those TDP numbers aren't directly comparable. Considering AMD sells dual core opterons at 55w max I'd have to say AMD is far ahead of Intel right now and probably ahead of whereever Intel is going to be in the next few years. The Yonah has about a 3 month window then it'll be surpassed by the dual core Turion as well.

[pigbat]

Funny, I hear the opposite from my peers. I have found AIX to be the best platform for enterprise computing that I have ever had the pleasure of using. Either way I still like IBM. I don't have anything against Intel. I am sure the new Intel Apples will be awesome.

I hadn't used AIX in years and had become a hardened Solaris guy until I just took a new job. I am blown away by the pSeries machine where we're running 20+ LPARS of AIX and Linux. AIX has come a long way and while I still prefer the user friendly Solaris administration I'm slowly becoming a believer.

[shadowfax0]

Remember that just because a new chip is coming out, or because a tech vender like Apple goes from one chip design to another (ppc -> x86) doesn't make the old immediately "crap." I am getting a Quad, and it will most likely be supplanted by a x86 PowerMac in about a year, but that doesn't mean that it is then crap. It is still what it is. It hasn't magically become slower. It just means that technology has moved on.

It may feel slower if I put the latest and greatest software on it that is meant to use the latest and greatest hardware which I no longer have. For instance, I tried to use PhotoBooth on my Dual 1Ghz G4 with a Geforce 4 Ti (non-core Image compatible). It was sluggish to say the least. But that is because it was using technology that my hardware wasn't ready to fully support.

Believe me, you will all feel better if you stop thinking of your machines as crap once newer machines come out.


This is a really great point, and I'm glad you brought it up. I'm running a dual 450 MHz G4 with 1.5 GB of RAM, with ~480 GB of storage. To be completely honest, I'm still very happy with it. I'm going to get a new Dual core G5 now; seeing as my computer turns 6 come January. However, this doesn't mean I'm going to get rid of it; it still has a lot of life left in it, and I have grown rather attached to it. Mind you, it doesn't encode video so fast, doesn't go through my lab data very quickly (faster than the machines in the lab however! : ), but it's a solid, reliable machine; I don't think I've turned it off for more than 2 weeks total since I've had it! Anyhow, I plan the same sort of life for my Quad, and can't wait!

[John Rivers]

So now with the new Powermacs released, I'm wondering whether or not to wait until the Intel/x86 updates roll around next year, or just stop hesitating and get a Quad (either way, it'll be the best offered at the time cause it's gotta last me awhilie). I won't officially be enrolled and going to school until next Fall (which is when the Intel's might come out), but there are soooo many projects I would love to work on and figure out right now. So basically: should the Intel architecture be that much faster that it would just be worth waiting?

Also, regarding Intel Powerbooks, how would those possibly handle programs like Final Cut Pro and (more importantly) Motion? Thanks everyone.

[emotion]

In terms of Intel on desktop Macs it isn't looking as good as it once was. I was looking forward to the convergence of IA64 and IA32 (Itanium and Xeon) but it seems that now this is delayed. I think Apple would do well to keep IBM sweet for a little while yet.


http://www.theregister.co.uk/2005/10/29/intel_xeon_2009/

[emotion]

To answer the last posters question. If I was gonna go for a desktop mac that quad would be awesome if your apps are multiprocessor aware.

[AidenShaw]

I was looking forward to the convergence of IA64 and IA32 (Itanium and Xeon) but it seems that now this is delayed.

http://www.theregister.co.uk/2005/10/29/intel_xeon_2009/
I believe that the "convergence" merely meant that Xeon and Itanium processors could share a common chipset and maybe motherboard.

It would not make x86/x64 run IA64 software natively, or vice versa.

Note that "IA32" is the 32-bit architecture - virtually all of the current CPUs except the laptop line are x64 (AKA EM64T) 64-bit architecture.

[emotion]

I believe that the "convergence" merely meant that Xeon and Itanium processors could share a common chipset and maybe motherboard.

It would not make x86/x64 run IA64 software natively, or vice versa.

Note that "IA32" is the 32-bit architecture - virtually all of the current CPUs except the laptop line are x64 (AKA EM64T) 64-bit architecture.

It was not specifically the convergence of ABIs more the fact that it'll all end up as one unified server/high end workstation architecture.

I had hoped that that was waht Apple would pick up after the G5s.

[AidenShaw]

It was not specifically the convergence of ABIs more the fact that it'll all end up as one unified server/high end workstation architecture.

I had hoped that that was waht Apple would pick up after the G5s.
It's not a difference in ABIs, it's completely different ISAs.

There's no reason that IA64 and x64 can't have the same ABI.

[emotion]

It's not a difference in ABIs, it's completely different ISAs.

There's no reason that IA64 and x64 can't have the same ABI.


Yeah true, must reply in less of a hurry :)

When Intel lose netburst and move more toward IA64/EPIC i'll find that more interesting.

[Paul D. Buck]

For those of us that do scientific computing, changing to the Xeon or P4 will mean Apple has no intrinsic advantage over any white box PC.

With the Altivec capability, for example, a 10 hour work unit went to 7 hours (Einstein@Home) ... I run a dual G5 2.0 GHz and it beats my 3.4 GHz dual Xeons for most things still ...

Of course I run 24/7 and have no interest in turning down the processor, but want the fastest throughput ... I will love to see what a 1M cache does compared to the 256M cache I currently have ... alas, no cash at the moment...

[emotion]

For those of us that do scientific computing, changing to the Xeon or P4 will mean Apple has no intrinsic advantage over any white box PC.

With the Altivec capability, for example, a 10 hour work unit went to 7 hours (Einstein@Home) ... I run a dual G5 2.0 GHz and it beats my 3.4 GHz dual Xeons for most things still ...

Of course I run 24/7 and have no interest in turning down the processor, but want the fastest throughput ... I will love to see what a 1M cache does compared to the 256M cache I currently have ... alas, no cash at the moment...

there is no way that apple will go with netburst architecture (well i'd be very surprised).

for scientific computing the itanium architecture is very very nice when your code is suitably dealt with (either by hand or failing that a good compiler).

[nborders]

too little to late.

~n

[RobHague]

http://smileys.smileycentral.com/cat/36/36_1_46.gif

Can someone tell me exactley what Desktop CPU Intel are going to have by 2006? 2007? That will show us why PPC don't "cut it" anymore? I get depressed looking at IBM's announcements because I know we wont be seeing anything from them in upcoming Apple products now.

I was hoping that Apple might do at least one more PPC update to the Power Macs :o maybe getting the Dual-Dual Core's down to the 2.0 and 2.3Ghz area's :D I also heard about a company working on its own PPC processors, Quad-Cores? Dual-Quad Cores :eek: but i guess its just Intel Inside we have to look forward too... it had better be amazing.

Apple switching to Intel. Biggest mistake ever!

http://smileys.smileycentral.com/cat/23/23_28_100.gif