No 3GHz Soon and No G5 Laptop

[Frobozz]

There is no doubt that Apple always uses the same 4 or 5 commercial apps
as their benchmark... there is good reason for this.... because they r the only freaking apps that run on a mac

Lets face it Apple really only markets to other apple users.... PC users dont really switch because they cant run the apps they are used to running...
And there are not really any Macintosh equivalents....

I know I will get flamed for this post, but a walk into any software/computer store will prove me right... 1000s of PC apps.... few or less Mac apps on the shelf.

That being said I still like my MAC...

Quantity versus quality.

I prefer the 10 apps I use most of the time to the 50 dopplegangers that are virtually indistinguishable in the PC world.

In all seriousness... I have _never_ run into a situation where I wanted to do a task and couldn't find a wonderfully suitable application to do it on my Mac. Just because there aren't a thousand copy-cats out there to fill the shelves doesn't mean the ones we have are not top notch.

 

[Frobozz]

Lets try.... Autocad... Matlab... Mapping Software.... Any Shrinkwrap Engineering Software.... Accounting Software... etc..I could go on for hours about the lack of real software for the MAC.....

This isnt really the MACs fault per se its just a fact of its small market share.

Autocad? Autocad!? Please don't meantion that steaming pile of **** again. Ever. Seriously.

For mapping software you can use GRASS. Accounting software... I'd use Quickbooks. You could argue both of these as being top of their class, but they are certainly good apps.

I think the Mac will be hard pressed to fill an accounting firm's office any time soon. I will agree with that. And, to be honest, most Engineers are woefully out of touch with any technology they weren't trained on in school. But in the case of accounting firms, they could honestly use $299 eMachines and have an over-powered system for their needs. I don't see why Apple would even want to compete with that.

 

[macsrus]

The other big question is what the ROI of VATech is. It's been 11 months since they spent the $7M to $10M for System X, and it's not yet in production. That's quite a lot of money for nothing (yet)!

Actually, It doesnt even exist anymore.....

It has been dismantled and sold off.....

Was never stable running applications because of the lack of ECC MEMORY on the nodes.

But it is being replaced by Xserve G5's 2.3 Ghz... with ECC

 

[maverick13]

Lets try.... Autocad... Matlab... Mapping Software.... Any Shrinkwrap Engineering Software.... Accounting Software... etc..I could go on for hours about the lack of real software for the MAC.....

This isnt really the MACs fault per se its just a fact of its small market share.

Matlab exists for OS X,as well as Mathematica .

 

[maverick13]

Not surprising - outside of the Mac rumour sites that VA cluster isn't getting much traction.

It's been good for InfiniBand (I've seen a few places that decided on IB instead of Myrinet based on VATech), and IBM's using it as a reference for the PPC BladeCenter systems.

If Apple were smart, they'd "knife" OS X and push Linux for these clusters. It's a Linux community out there, and trying to force OS X on most of these big clusters will be a lost cause.

OS X will be OK for the small shops without any real staff, and for the content creation folks already using Macs who need a handful of systems for a "render farm". The multi-million dollar clusters, however, are the domain of Linux and UNIX.</QUOTE>

http://www.roanoke.com/news/roanoke\6133.htm
What bothers me in these forums is that anyone posts whatever comes to his mind without realy having a clue whatsoever.

 

[maverick13]

The point is that most of the work in HPC is being done in Linux - and people don't want to have to port their Linux programs to UNIX and BSD variants. (X Window System support is not really an issue - superclusters don't use graphics cards.)

It's a path of least resistance issue - having to port to OS X is harder than porting to PPC Linux (and even harder than to Itanium Linux).

As far as all the "awards" go, let's see what's on the Top500 list next week - particularly at the high end.

"Traction" means more top supercomputers made from PowerMacs, not a bunch of beauty pageant awards. (An "Oscar" is nice, but bucks at the box office are what drives movies.)

Port what??? Have you ever programmed on Linux??
Why do not people think before they post?? It's EXACTLY the same if stick to classic APIs and there is NO REASON not to do so.

 

[AidenShaw]

illogical logic

You mean 30% more CPU's that run at a 50% higher clock rate were slightly slower than the G5.
...
While it's obvious the MHz scale to performance ratio is not linear, a 3GHz G5 class IBM chip will apparently stomp on a lot (all?) of Intel / AMD fare.

Clock rate isn't relevant (remember the MHz myth???). It's 30% more of the fastest Xeons (at the time) vs the fastest PPC970 (still).

Doing something 30% faster is nice, but not "killer". Doing it two or three times faster "kills". (For example, the PPC970 "kills" x86 on BLAST, but not on LINPACKD.)


But IBM has only been able to make 2.0GHz PPC970 chips, so any talk of 3GHz is hot air. IBM's talking about 2.2GHz in the future for the BladeCenter. Sometime this summer (northern hemisphere) Apple has promised to ship an overclocked water-cooled 2.5GHz system.

3 GHz isn't there....

 

[pjkelnhofer]

But IBM has only been able to make 2.0GHz PPC970 chips, so any talk of 3GHz is hot air. IBM's talking about 2.2GHz in the future for the BladeCenter. Sometime this summer (northern hemisphere) Apple has promised to ship an overclocked water-cooled 2.5GHz system.

3 GHz isn't there....

I am so sick of people saying that Apple is overclocking the chips.

Overclock
To run a microprocessor faster than the speed for which it has been tested and approved.

You are saying that Apple is taking chips that IBM is only testing at 2.0GHz and running them at 2.5GHz. Do you have any evidence to support this? How do you know that IBM has not improved the yields on the 970FX to the point where they are getting chips that can run at 2.5GHz (a stable 2.5GHz) in high enough numbers to ship them to Apple as 2.5GHz chips? You are basically accusing Apple of lying to consumer as to what chip is in these computers.
I believe that a PC manufacturer got in a lot of trouble for taking Pentiums and selling them in systems running at higher speeds then Intel had rated them.


edit: typos

 

[jragosta]

Clock rate isn't relevant (remember the MHz myth???). It's 30% more of the fastest Xeons (at the time) vs the fastest PPC970 (still).

Doing something 30% faster is nice, but not "killer". Doing it two or three times faster "kills". (For example, the PPC970 "kills" x86 on BLAST, but not on LINPACKD.)


But IBM has only been able to make 2.0GHz PPC970 chips, so any talk of 3GHz is hot air. IBM's talking about 2.2GHz in the future for the BladeCenter. Sometime this summer (northern hemisphere) Apple has promised to ship an overclocked water-cooled 2.5GHz system.

3 GHz isn't there....

That's silly semantics. In the computer industry, a 30% advantage is pretty significant.

ESPECIALLY when you consider the cost difference. AND when you consider that Apple currently has a system that's 25% higher clock speed than that and Intel has only ramped the clock speed by about 10% - so the advantage would be even greater if you were building a system today.

 

[jragosta]

But IBM has only been able to make 2.0GHz PPC970 chips, so any talk of 3GHz is hot air. IBM's talking about 2.2GHz in the future for the BladeCenter. Sometime this summer (northern hemisphere) Apple has promised to ship an overclocked water-cooled 2.5GHz system.

Thanks for proving once again that you don't know what you're talking about.

Apple is shipping systems with IBM 2.5 GHz parts. Doing what you're accusing them of (shipping 2.0 GHz parts and saying that they're 2.5 GHz) would be blatantly illegal. There's absolutely no reason to believe that.

 

[AidenShaw]

Apple is shipping systems with IBM 2.5 GHz parts.

Speaking of "not knowing what you're talking about":

http://www.apple.com/pr/library/2004/jun/09powermac.html

the dual 2.5 GHz Power Mac G5 model is expected to be available in July

To me, that's *not shipping*. An "expectation" of shipping in July is *not shipping today*. Once you see posts here of "I got my 2.5 today" then you can say it's shipping.

Doing what you're accusing them of (shipping 2.0 GHz parts and saying that they're 2.5 GHz) would be blatantly illegal. There's absolutely no reason to believe that.

Look at that liquid cooling system again.... Tell me that Apple isn't pushing the chip faster than normal.

I'm not a lawyer, but since the IBM CPUs aren't available on the open market - who's to say what is legal and what is not?

If the chip runs at 2.5 GHz, and Apple fully warrants the chip, then what would be inherently illegal? Where is the fraud?

I'm sure that IBM has advised Apple that the chip will run at that speed, as long as the CPU temperature is kept below a certain threshold. Apple built an extreme cooling system to meet those specs. It's well known in overclocking circles that keeping the temperature below the normal limits helps improve stability. (http://www.techworthy.com/PCUpgrade/August2003/Build-Your-Own-Water-Cooled-PC.htm http://www.highspeedpc.com/Merchant2/merchant.mv?Screen=CTGY&Store_Code=HSPC&Category_Code=WC)

Is it an illegal fraud - of course not. Is Apple running the clock faster than they could with normal heat sinks, of course. Apple's just doing what the aftermarket overclockers do - putting in extreme liquid cooling to keep things stable.

 

[jragosta]

Speaking of "not knowing what you're talking about":

http://www.apple.com/pr/library/2004/jun/09powermac.html


To me, that's *not shipping*. An "expectation" of shipping in July is *not shipping today*. Once you see posts here of "I got my 2.5 today" then you can say it's shipping.





Look at that liquid cooling system again.... Tell me that Apple isn't pushing the chip faster than normal.

I'm not a lawyer, but since the IBM CPUs aren't available on the open market - who's to say what is legal and what is not?

If the chip runs at 2.5 GHz, and Apple fully warrants the chip, then what would be inherently illegal? Where is the fraud?

Nice attempt to move the goalposts.

You claimed earlier that Apple was overclocking IBM 2.0 GHz chips. THAT is why I said that you're clueless.

What's legal is simple. If IBM sells them as 2.5 GHz chips, then Apple can claim that they're 2.5 GHz (or slower). If IBM sells them as 2.0 GHz chips, it would be illegal for Apple to claim that they're 2.5 GHz - no matter how much testing they do. It would be an open and shut case of fraud. In fact, one PC vendor (Gateway, IIRC) already lost a similar case so it's not even open for interpretation.

All these accusations that Apple is overclocking IBM chips are nonsense.

Please try to become informed about topics before discussing them. You're only making yourself look foolish.

 

[pjkelnhofer]

I'm sure that IBM has advised Apple that the chip will run at that speed, as long as the CPU temperature is kept below a certain threshold. Apple built an extreme cooling system to meet those specs. It's well known in overclocking circles that keeping the temperature below the normal limits helps improve stability. (http://www.techworthy.com/PCUpgrade/August2003/Build-Your-Own-Water-Cooled-PC.htm http://www.highspeedpc.com/Merchant2/merchant.mv?Screen=CTGY&Store_Code=HSPC&Category_Code=WC)

Keeping temperatures lower improves stability regardless of whether the chip is overclocked, run at its rated speed, or underclocked. There is more physics to chip being able to run at a certain speed than simply keeping it cool. If there weren't I could run an 8086 at 10GHz if I could just keep it cool enough.

Is it an illegal fraud - of course not. Is Apple running the clock faster than they could with normal heat sinks, of course. Apple's just doing what the aftermarket overclockers do - putting in extreme liquid cooling to keep things stable.

It is in fact illegal for a computer manfucturer to sell a chip as faster than the chip manufacturer rated it, and if IBM is rating the chips as 2.5GHz then they are by definition not overclocked.
The fact they are using liquid cooling proves nothing at all. When CPU's went from simply having heatsinks to having heatsinks with fans was it because they were "overclocked"? No it was simply the most effective way to keep the chip within it's nominal temperature range. I could argue that every single chip sold today is overclocked because it needs a fan or a heatsink or any special considerations for cooling.

 

[wdlove]

The lively discussion is very interesting. I agree, just can't imagine Apple doing anything to harm their reputation. That is after all what a company depends on for their survival. IBM is also a well known old line company.
I'm sure that Apple wants to over come the problems of Rev. A.

 

[Soire]

WWDC?

Forgive the posting of a question about WWDC in this thread, but there aren't any other active threads about WWDC predictions (that I could find)

So besides new iMacs and Tiger, what else is a likely bet? The displays or iPods? Is just iMacs and Tiger enough?

What I would like to know is what was released last year at WWDC? I have a hankering to buy a display and an iPod, and I'm hoping that the wait will be worth it, and at least one of the two will be updated by the end of the month.

ps- do the releases all come on the first day of WWDC or are they spread out?

 

[mklos]

Look at that liquid cooling system again.... Tell me that Apple isn't pushing the chip faster than normal.

I'm not a lawyer, but since the IBM CPUs aren't available on the open market - who's to say what is legal and what is not?

If the chip runs at 2.5 GHz, and Apple fully warrants the chip, then what would be inherently illegal? Where is the fraud?

I'm sure that IBM has advised Apple that the chip will run at that speed, as long as the CPU temperature is kept below a certain threshold. Apple built an extreme cooling system to meet those specs. It's well known in overclocking circles that keeping the temperature below the normal limits helps improve stability. (http://www.techworthy.com/PCUpgrade/August2003/Build-Your-Own-Water-Cooled-PC.htm http://www.highspeedpc.com/Merchant2/merchant.mv?Screen=CTGY&Store_Code=HSPC&Category_Code=WC)

Is it an illegal fraud - of course not. Is Apple running the clock faster than they could with normal heat sinks, of course. Apple's just doing what the aftermarket overclockers do - putting in extreme liquid cooling to keep things stable.

I highly doubt Apple is running a 2 GHz G5 overclocked to 2.5 GHz. For one thing, IBM already announced that they have PPC970fx processors running up to 2.5 GHz and below. They probably have liquid cooling on them to keep them cooling quietly. I believe Apple even said that in a press release when they introduced the update. This is one of the reasons why they haven't gotten to 3 GHz yet because of heat issues that probably even liquid cooling wouldn't help.

Got any more bright ideas you think Apple is doing?

 

[KeareB]

What's legal is simple. If IBM sells them as 2.5 GHz chips, then Apple can claim that they're 2.5 GHz (or slower). If IBM sells them as 2.0 GHz chips, it would be illegal for Apple to claim that they're 2.5 GHz - no matter how much testing they do. It would be an open and shut case of fraud. In fact, one PC vendor (Gateway, IIRC) already lost a similar case so it's not even open for interpretation.

All these accusations that Apple is overclocking IBM chips are nonsense.

I'm no chip expert, but I would imagine that chip manufacturers spec chips UNDER CERTAIN CONDITIONS. My guess is that IBM would not spec the 970FX chip at 2.5 GHz if the computer assember told IBM that they planned to use the chip in Iraq in the summer in a sandproof, ventless, box, with no fans -- and IBM knew the computer would be in room without air conditioning with ambient temperatures reaching 130 degrees farenheit.

Wouldn't it be reasonable to assume that chips, prior to the advent of liquid cooling, would be spec'd under what they would have defined as "normal operating conditions" (which probably included something like adequate venting, airflow, and within a normal ambient room temperatures of 40-100 degrees farenheit).

Isn't it possible that IBM placed additional conditions on the certification of these chips as 2.5 GHz? As in, we will spec these at 2.0 GHz under (previous) "normal operating conditions", but we will certify them at 2.5 GHz under super-cooling conditions (i.e., liquid cooling).

In this sense, I don't look at it as "Yes" or "No" - is Apple overclocking? In short, I am assuming that IBM will certify a given chip at different speeds given different expected operating conditions. Isn't this reasonable?

 

[macsrus]

There is more physics to chip being able to run at a certain speed than simply keeping it cool. If there weren't I could run an 8086 at 10GHz if I could just keep it cool enough.

True to an extent.... There are 3 factors that make or prohibit the ability to overclock...
1. Heat you have to get rid of it
2. Clock can be adjusted per the phisics of the process...
(your 8086 to 10GHz is not a valid example because the transistors built into the chip are to far from ea other and theyu cant switch fast enuff.)
Besides even if it was possible to increase the speed to 100GHZ an 8086 would still be slower than a 1GHZ P4
3. voltage. Voltage is always increased on the CPU's to differentiate between noise and High Lows
The voltage increase leads back to Heat.

It is in fact illegal for a computer manfucturer to sell a chip as faster than the chip manufacturer rated it, and if IBM is rating the chips as 2.5GHz then they are by definition not overclocked.

It is illegal to remark a 2.0GHZ CPU as a 2.5 GHZ and then sell
But is not illegal for a OEM such as Apple to sell SYSTEM such as a 2.5 GHZ POWER MAC using a CPU at 2.0 GHZ that they clocked to 2.5
THIS has been done for several years in the PC world... and is done all the time by VIDEO Card Vendors... The OEM provides the warranty on the SYSTEM.

 

[pjkelnhofer]

I'm no chip expert, but I would imagine that chip manufacturers spec chips UNDER CERTAIN CONDITIONS. My guess is that IBM would not spec the 970FX chip at 2.5 GHz if the computer assember told IBM that they planned to use the chip in Iraq in the summer in a sandproof, ventless, box, with no fans -- and IBM knew the computer would be in room without air conditioning with ambient temperatures reaching 130 degrees farenheit.

Wouldn't it be reasonable to assume that chips, prior to the advent of liquid cooling, would be spec'd under what they would have defined as "normal operating conditions" (which probably included something like adequate venting, airflow, and within a normal ambient room temperatures of 40-100 degrees farenheit).

IBM simply states that the chip runs at X speed within a temperature range of Y and Z degrees when they sell it. In fact if you look at the website they will not even make suggestions on how to cool their chips, they leave that entirely up to the end manufacturer.

Isn't it possible that IBM placed additional conditions on the certification of these chips as 2.5 GHz? As in, we will spec these at 2.0 GHz under (previous) "normal operating conditions", but we will certify them at 2.5 GHz under super-cooling conditions (i.e., liquid cooling).

I think people are still confused about what liquid cooling is. This is no way "super-cooling". It is not refrigeration. It is simply using the high thermal mass of water to more efficiently move heat away from the processor. Heat travels in only on direction. When something "cools-off" it is transferring its heat to its surroundings (be that air or another solid object). Water can absorb a lot more heat that air so it "cools" more efficiently.

In this sense, I don't look at it as "Yes" or "No" - is Apple overclocking? In short, I am assuming that IBM will certify a given chip at different speeds given different expected operating conditions. Isn't this reasonable?

The expected operating conditions are the same. With the increased speed comes increased heat and it is Apple responsibility to come up with a way to cool the chip. The 2.5GHz most likely could have been cooled with simply a fan, but the increased noise probably would have been detrimental to sales (I think they may have learned a lesson with the old "Wind Tunnel" G4's).

 

[pjkelnhofer]

True to an extent.... There are 3 factors that make or prohibit the ability to overclock...
1. Heat you have to get rid of it
2. Clock can be adjusted per the phisics of the process...
(your 8086 to 10GHz is not a valid example because the transistors built into the chip are to far from ea other and theyu cant switch fast enuff.)
Besides even if it was possible to increase the speed to 100GHZ an 8086 would still be slower than a 1GHZ P4
3. voltage. Voltage is always increased on the CPU's to differentiate between noise and High Lows
The voltage increase leads back to Heat.

I realize that there is a lot more than than just the heat. A lot of people seem to think the only involved in making a chip faster is keeping it cool. The 8086 was an exageration to prove that there is a lot more physics involved than that.
The main reason why overclocking usually works (for an end user) is that there is a very good chance the chip would have passed at a higher speed initially but was only rated a certain speed by the manufacturer because that is what they were looking for. If Intel needs more 2.4GHz chips to fill an order from Dell, they are not going to bother testing to see if the chips can run at 3.0GHz because it doesn't matter if they can. They simply stamp them 2.4 and move on.

It is illegal to remark a 2.0GHZ CPU as a 2.5 GHZ and then sell
But is not illegal for a OEM such as Apple to sell SYSTEM such as a 2.5 GHZ POWER MAC using a CPU at 2.0 GHZ that they clocked to 2.5
THIS has been done for several years in the PC world... and is done all the time by VIDEO Card Vendors... The OEM provides the warranty on the SYSTEM.

I think you are wrong about this. I don't think Apple is allowed to say that they are selling 2.5GHz chips if that is not in fact what they are. If someone opened up there 2.5GHz G5 and found chips stamped 2.0GHz by IBM I think they would have a pretty strong lawsuit.
I would say that Apple's own website sure makes it sounds like they are getting 2.5GHz chips from IBM.

from the G5 specs
... thanks to IBM’s sophisticated process technology that builds them just 90 anometers wide. Such superior technology developments turbo-charge the G5 processor to speeds of up to 2.5GHz.

 

[jragosta]

I'm no chip expert, but I would imagine that chip manufacturers spec chips UNDER CERTAIN CONDITIONS. My guess is that IBM would not spec the 970FX chip at 2.5 GHz if the computer assember told IBM that they planned to use the chip in Iraq in the summer in a sandproof, ventless, box, with no fans -- and IBM knew the computer would be in room without air conditioning with ambient temperatures reaching 130 degrees farenheit.

Wouldn't it be reasonable to assume that chips, prior to the advent of liquid cooling, would be spec'd under what they would have defined as "normal operating conditions" (which probably included something like adequate venting, airflow, and within a normal ambient room temperatures of 40-100 degrees farenheit).

Isn't it possible that IBM placed additional conditions on the certification of these chips as 2.5 GHz? As in, we will spec these at 2.0 GHz under (previous) "normal operating conditions", but we will certify them at 2.5 GHz under super-cooling conditions (i.e., liquid cooling).

In this sense, I don't look at it as "Yes" or "No" - is Apple overclocking? In short, I am assuming that IBM will certify a given chip at different speeds given different expected operating conditions. Isn't this reasonable?

Even if your speculation were true, the chips are not overclocked.

Several people are accusing Apple of illegal and unethical behavior - with no basis in fact. The chips are rated at 2.5 GHz BY IBM, so they are not overclocked _by definition_. They do require special cooling, no doubt. That doesn't change the fact that they're not overclocked.

 

[macsrus]

I think you are wrong about this. I don't think Apple is allowed to say that they are selling 2.5GHz chips if that is not in fact what they are. If someone opened up there 2.5GHz G5 and found chips stamped 2.0GHz by IBM I think they would have a pretty strong lawsuit.

First off Im not saying that Apple is definatly overclocking the CPUs

Second In my company's negotiations with IBM over a cluster composed of their JS20 Blade Servers(PPC 970). It was brought up as to why Apple was currently selling 2.0 GHZ 970s in the Xserve while IBM was only selling 1.6 GHZ in their Blades.

IBM told us and I quote "that running the CPU at 2.0(these were still 130nm at the time), was more of a risk for thermal problems than IBM was willing to take."
They also said "if Apple wanted to clock theirs to 2.0 and take the chance of burning them up then that was Apples problem."

I would suspect that IBM feels the same way about the current batch of PPC they are selling to Apple.

 

[KeareB]

Even if your speculation were true, the chips are not overclocked.

Several people are accusing Apple of illegal and unethical behavior - with no basis in fact. The chips are rated at 2.5 GHz BY IBM, so they are not overclocked _by definition_. They do require special cooling, no doubt. That doesn't change the fact that they're not overclocked.

I would agree that even if my speculation were true, that they should not be considered overclocked or illegal. It would, however, mean that Apple is pushing the envelope and is able to certify a 2.5GHz machine ONLY because of the introduction of special cooling unit.

Coming at this issue a different way, my question would be whether the heat specs of the current crop of 2.5 GHz 970FX will be the same as the heat specs of the crop of 2.5 GHz 970FX that will be produced 12 months from now? Or, will advances mean that they can produce a cooler 2.5GHz chip that might not require liquid cooling (i.e., could Apple in 12 months time put a 2.5 GHz processor it in the current 2.0 box)?

Fundamentally, what I am trying to get at is whether this (combined with Apple's published performance gains of only 9-11% over the 2.0 chip which is lower than one would expect with a 25% clock jump) is solid evidence that the 970FX is maxed out speed-wise, and whether IBM and Apple know this and will henceforth be putting their next crop of eggs in the 975/980 basket. If this is the case, I'd be more tempted to wait for the first 975/980 machine.

 

[pjkelnhofer]

First off Im not saying that Apple is definatly overclocking the CPUs

Second In my company's negotiations with IBM over a cluster composed of their JS20 Blade Servers(PPC 970). It was brought up as to why Apple was currently selling 2.0 GHZ 970s in the Xserve while IBM was only selling 1.6 GHZ in their Blades.

IBM told us and I quote "that running the CPU at 2.0(these were still 130nm at the time), was more of a risk for thermal problems than IBM was willing to take."
They also said "if Apple wanted to clock theirs to 2.0 and take the chance of burning them up then that was Apples problem."

I would suspect that IBM feels the same way about the current batch of PPC they are selling to Apple.

That is also some one from IBM sales who most certainly does not want you buying XServes instead of Blades.
Also, the XServe never had a 130nm 970 in it, it did not go to G5 until the 970FX was availible and that has been the only chip ever in the G5 XServe.
Finally, I could sell you a server that was running 970's at 1.2GHz and use the same argument against IBM blades, that there is more risk running them at 1.6GHz then my company is willing to take.
I am not saying that the IBM people are lying, but they are salesmen for IBM and as such comments about competitors need to be taken with a grain of salt.

 

[pjkelnhofer]

I would agree that even if my speculation were true, that they should not be considered overclocked or illegal. It would, however, mean that Apple is pushing the envelope and is able to certify a 2.5GHz machine ONLY because of the introduction of special cooling unit.

Like I said early, by the same argument any chip that requires a heatsink or a fan is pushing the envelope since those are also "special cooling units" compared to what cooled processors 10-15 years ago. I think this just a matter of Apple being ahead of the game.

Coming at this issue a different way, my question would be whether the heat specs of the current crop of 2.5 GHz 970FX will be the same as the heat specs of the crop of 2.5 GHz 970FX that will be produced 12 months from now? Or, will advances mean that they can produce a cooler 2.5GHz chip that might not require liquid cooling (i.e., could Apple in 12 months time put a 2.5 GHz processor it in the current 2.0 box)?

You probably could put a 2.5GHz in the current 2.0 box if you didn't mind the fans running non-stop. So long as it is still a 2.5GHz PPC970FX I would think the requirements would stay the same. The improved yield will only result in faster processors, and even then there is a limit to how fast the 970FX will ever go. We may have to wait for the next generation chip from IBM (970GX, 975, 980) what they call it, to keep going faster.

Fundamentally, what I am trying to get at is whether this (combined with Apple's published performance gains of only 9-11% over the 2.0 chip which is lower than one would expect with a 25% clock jump) is solid evidence that the 970FX is maxed out speed-wise, and whether IBM and Apple know this and will henceforth be putting their next crop of eggs in the 975/980 basket. If this is the case, I'd be more tempted to wait for the first 975/980 machine.

I don't think the performance gains prove anything. I would not a expect a 25% faster processor to be 25% faster in real world performance. There are diminishing returns as you simply add clock speed to a chip.