Ars Technica Interviews 970 Designers

11223 writes "John "Hannibal" Stokes has interviewed Pete Sandon, the PowerPC 970's main designer, and David Edelsohn, a compiler writer from IBM, and clarified several points about the 970 regarding group formation, vector issue queues and performance, and more. The interview is a very interesting read for anyone who has been following his earlier articles on the processor that Apple calls the G5."

Comments

Anybody else

[Pxtl]

misenterpret this to mean "ars interviews nine-hundred and seventy different people"?

Re:Anybody else

[cK-Gunslinger]

I did. I just thought Ars was trying to 1-up Anantech on their 13- Motherboard CEOs interview.

The Reality Distortion Field

[Jon Abbott]

One of the best quotes from the interview was from David Edelsohn: "IBM is not gonna try to compete with Apple's reality distortion field :)"

Windows based 970?

[Bill, Shooter of Bul]

I think if you read the end of the article where they are talking about the possibility of straight non apple boxes with 970 inside, you'll notice that they can not reply. Why not? It would be obvious to have linux based servers on top of the platform, so to have no responce to that makes me wonder if they aren't talking to soemone else about something nonobvious. What is the most non obvious step that would really get it in trouble with apple? Another deal with Microsoft. Heck the NT Kernel is portable and is currently being ported to itanium2 and amd64 why not ppc 970? I don't know how closely apple has tied panther to Their chipset, but if it isn't too tight this could mean cheap apple clones( they wouldn't ship with osx, but it could be installed). Now that would kill apple, and as a guy who had advance knowledge of it, I would simply say "No Comment" when asked about non apple based ppc 970 platforms.

Re:Windows based 970?

[pv2b]

Not legally.

You still need a licence to run Mac OS X, and I think it would be trivial for Apple to add a clause (if it's not already there) that would forbid installing the software on non-apple hardware.

There is also a port of Mac OS X for Intel processors being maintained in parallell, mainly because it CAN be done very easilly with minimal effort. Covering all bases...

Remember, Apple is a hardware company, and will protect their core business, which currently is and always has been hardware.

Re:Windows based 970?

[binaryDigit]

Another deal with Microsoft. Heck the NT Kernel is portable and is currently being ported to itanium2 and amd64 why not ppc 970?

It is already ported to PPC (maybe not 970, but PPC) and has been since 3.51. Even with that, I _highly_ doubt that Microsoft will venture down that road again. Back in the late 80's/early 90's when RISC was the "NEXT BIG THING", M$ was hedging their bets by making sure that NT was available for all manner of RISC flavours. Now that RISC is "NOT THE NEXT BIG THING", I really don't think M$ cares anymore, to them it's just another platform they'd have to support and probably wouldn't make any money off of.

Now having it run Linux is a no brainer. IBM is obviously in Linux in a big way, so having some 970 based boxen are obvious. Now having "generic" white box 970's designed to run Linux is a different story. I don't know if this would make sense from a market perspective. Perhaps cheaper commodity based servers? Perhaps giving Dell a run for their money in the Linux market for higher end workstations? Hard to say, esp the latter since IBM is notorious for not wanting to cannibalize their higher end sales by having lower end box's with better price/performance ratios.

BTW, you can kiss off the Apple clone notion. Makes absolutely no sense for Apple unless they can assure themselves of at least doubling Mac market share by such a move. Anything less would mean a repeat of their previous foray, which Stever would never allow to happen.

Re:Windows based 970?

[pv2b]

Apple doesn't want you to buy OS X, they want you to buy a Mac.

I repeat, Apple is a hardware company.

Re:Windows based 970?

[Bill, Shooter of Bul]

Yeah, thats what I'm saying. What would stop people from doing it ( if its technically possible some people will do it regardless of leagality)? Here's a riddle for everyone to think about: If apple's a hardware company what unique piece of hardware does it actually make? The motherboard?
I think apple is really a software company that has managed to force people to buy hardware from it( at high margins) so their software can be run. As obvious as that statement is, I think many people forget it.

Re:Windows based 970?

[cant_get_a_good_nick]

Windows NT 3.51 and 4.0 supported PowerPC. Was dropped because of bad sales. Solaris had a PowerPC port too if you can believe it. This ws all when PowerPC was shiny and new and PowerPC was going to take over the world, giving a consistent platform free of all that x86 cruft. Problem was NT in that day wasn't compatible yet with loads of software, and Windows 3.1 and 95 were very much x86 only, so the software market never followed to PowerPC. Intel threw enough silicon at the problem to make x86 performance acceptable, and the RISC world withered.

The most interesting thing for me with all this "cheap PowerPC" stuff is it seems to be the rebirth of CHRP, which Apple kind of scotched becasue they were fearful of clones back then. Maybe they realize they need to kill some of the "hardware premium price" and get costs more in line with Intel boxes.

Re:Windows based 970?

[pv2b]

People hacking OS X to run on their non-apple machines is not a real threat to Apples.

COMPANIES selling non-Apple machines running OS X are a real threat to Apple.

The legal issues won't stop the first crowd (but then again, Apple won't have lost a great deal), but the people who actually buy computers and work with them as well as Joe home user will not go to any lengths to save a few bucks just to run OS X on a non-Apple box.

That's where Apple gets its money -- and it's pretty well protected.

Apple does make the entire computer, which is much more than a sum of its parts.

I'm not sure how many of the components that go into a car (I'm not a car nut) are actually made by the car company themselves, but let's for the sake of argument say that the car company doesn't make any of the components in the car. But the design of the car and putting the car together is still something the car company does, and that is the value they add.

This is basically what Apple does, to make a product you don't neccecarilly have to make your components yourself.

Re:Windows based 970?

[frankie]

What hardware do they make? Last I checked they have gone the commodity PC hardware route

• Bridge & board: "The chipset is an Apple design"
• Servers
• Peripherals and Standards

User-serviceable parts (RAM, HD, AGP, etc) are commodity, but the hard stuff is designed in Cupertino.

Re:Windows based 970?

[WNight]

What reason is there to expect that EULAs are valid?

They're post-sale contracts. This sort of thing has never been legit.

When they show you the license in the store, and you must overtly agree to it to buy the product, then they may be legal. Until then they're lies.

But, they'll have to be a lot simpler. Judges are already invalidating long small-print contracts for regular consumers. If it takes a law degree to understand, you can't possibly enter into it knowingly. Thus, the company should reasonably know that nobody reads (and hence, nobody agree to) their contracts.

Further, the concept of post-sale restrictions was decided in the early 1900s, with the First-Sale doctrine. Books were being sold with 'contracts' inside the cover limiting resale rights. It didn't work then, it won't work now, even if the many issues keeping EULAs from being valid contracts were fixed.

(Such as, they disclaim consumer rights they aren't allowed to disclaim under the Magnuson-Moss warranty act. Many EULAs disclaim all responsibility even if the product doesn't function at all, etc. Not allowed, and in fact, likely criminal to claim.)

You do everyone a disservice by saying that EULAs might be valid. It's misleading and can be very damaging.

Improvements to GCC?

[Realistic_Dragon]

At one point in the interview it looks like IBM and Apple are working together on GCC improvements and donating the code back to the FSF.

This is a fairly big deal as people have pointed out before that GCC on PPC isn't as hot as it should be, but with that kind of muscle and money behind it it should go forwards by leaps and bounds.

With the new GCC improvements it looks like Linux on those new, remarkably cheap, P970 IBM boxes is going to be a real winner. And AFAIK Gentoo already runs on PPC fine - no one is going to be bitching about compile times with 4 1gig+ CPUs crunching away at it!

Re:Improvements to GCC?

[cactopus]

This box -- i.e. even future inexpensive ones are really designed to be Itanium and Xeon killers in addition to Sparc killers. They're basically going to be priced in the 3G-10G range for quite a lot of 64 bit computing muscle and the ability to run AIX. When you outgrow them you can move to POWER4+ on pSeries quite smoothly. A trusted and proven architecture.

These are basically this generation's Alphas... but with a better market positioning and without Digital/Compaq/HP at the helm. (we all know their pilot is dead at the wheel)

Re:Improvements to GCC?

[Duncan3]

Yes, it is great. Unfortunately in the next couple questions he goes on to say that the underlying engine GCC uses is completely unable to deal with the complexities of the 970 and other deep and wide architectures.

So either GCC needs a serious reworking on a fundamental level, or more likely since it already exists, they will just release a separate compiler that doesn't suck.

I just hope they release the proprietary compiler for OS X sometime before the G5 hits. The 2-3x performance hit of GCC is really starting to hurt Apple, and piss off all the developers. If IBM only releases the good compiler for their OSs and Linux, they are effectively telling people that OS X is not welcome on the 970.

Noone seriously considers using gcc on x86 now that the Intel compiler is free (root beer). And frankly that alone makes AMD chips totally unattractive for a computation farm (that and the nuclear plant you need to power the things).

IBM needs to step up and do the same.

Re:Improvements to GCC?

[Realistic_Dragon]

Yes, it is great. Unfortunately in the next couple questions he goes on to say that the underlying engine GCC uses is completely unable to deal with the complexities of the 970 and other deep and wide architectures.

It has been stated that GCC is attempting to be a 'good' solution for all architectres rather than the 'best' for any one. It's not incapable of adapting to be the best for the P970, but that would require a permanent fork from the general GCC code.

So either GCC needs a serious reworking on a fundamental level, or more likely since it already exists, they will just release a separate compiler that doesn't suck.

The advantage of using GCC is that developers can write with reasonable confidence for any platform with GCC available. I have done some heavy porting work between code written for TC/TCC and code written for GCC and it's no fun at all. By comparison code for GCC for PPC and x86 isn't that bad to chop and change.

Given the nature of Apple (who seem to like being able to use open source apps written for GCC) and IBM (who want Linux - developed for GCC - to compile on their boxes) I think they will prefer to optimise GCC as far as possible for the P970.

Given the competence of IBM and Apple programmers (especially the former) I suspect that they will do a pretty good job.

Re:I can't wait..

[Mononoke]

...to see what this thing will be like running Windows Longhorn woth its cool 3D UI.

You already can, because their R&D dept already released it. It's called Mac OSX.

Re:Altivec execution

[Jon Abbott]

The following was snipped from this message:

"The AltiVec subunits are more independant than in the 7400, i.e. there isn't just a single vector ALU, instead the vector FPU, vector simple IU, and the vector complex IU can now accept AltiVec instructions concurrently (up to two vector instructions per clock); this means technically, the G4e does have 4 AltiVec units, while the MPC7400 has only two, but in practice the G4e merely relaxes some instruction scheduling restrictions that the 7400 has to adhere to."

Re:IBM in the apple.slashdot.org section ?

[valkraider]

Why is Apple+IBM such a new thing?

How about October, 1991?

More info about the PowerPC alliance.

Apple never said that IBMs *technology* was no good. They said that IBM made boring corporate Personal Computers that didn't foster individuality and creativity amongst it's users. I would say that is still correct.