Book: The Race for a New Game Machine

[silhouette]

I finished the book. Took an hour in total.

Don't buy it , its a waste of $23 bucks

The most interesting things is that MS wanted 4 cores but later reduced it to 3 when they found out that weren't going to be able ot make it profitable.

IBM decided to go with MS because sony added another 2 spu units up from the original 1/6 design to 1/8 design and that would cut into the profit margins for ibm who was supposed to produce the chip at their fab.

THe original dual issue that IBM moved out for multithreading was put back in by ms and since they used a common core for everything but vmx units (vector units )

A sony engineer who worked on the original vmx units fixed a bug on the ms verison of the unit which was enhanced over the sony one. The bug didn't affect the sony one. he still fixed it.

OOE was removed because IBM couldn't get all 3 chips (apple , ms , sony) done in the time frame and keep OOE. That pissed off Apple and seems to have set them on the road to Intel

Thats really all the technical stuff. Now if you want to hear about this guys few stressfull years making these chips its a good read. But not so much if you want to learn about the chips .

Thanks eastmen.

It seems MS initially targets a much closer computational power with the original Cell design. A 4 core Xenon would not be very far from a 6 SPU Cell in terms of theoratical flops. Had IBM pulled that off and Sony had not changed the targets, Ms would have almost achieved this.

 

[Carl B]

Had IBM pulled that off and Sony had not changed the targets, Ms would have almost achieved this.

Well it's not that Sony changed the target to 8 so much as the Austin team changed the target to 6 without getting clearance to do so.

 

[eastmen]

Well it's not that Sony changed the target to 8 so much as the Austin team changed the target to 6 without getting clearance to do so.

actually according the book it was sony that changed it one day in 2002 after they were working on it for over a year already.
The design was allways 1/6 @ 4ghz according to the book , it then changed to 1/8@ 4 ghz , then had the improved ms core without the improved vmx units and then i guess at some point it became 1/8 @ 3.2 and then 1/7 @3.2 for yields

 

[Carl B]

actually according the book it was sony that changed it one day in 2002 after they were working on it for over a year already.
The design was allways 1/6 @ 4ghz according to the book , it then changed to 1/8@ 4 ghz , then had the improved ms core without the improved vmx units and then i guess at some point it became 1/8 @ 3.2 and then 1/7 @3.2 for yields

According to the dude that wrote the book, perhaps. But have you read through the link above? As far as that guy was concerned it may always have been 6 SPEs, but there was a disconnect between what the team put in motion and what Kutaragi felt to be the case. Back at HQ, 8 SPEs was the original plan, and part of the reason it went to 8 was simply due to the fact that Kutaragi was made aware that they were working around 6.

EDIT: For context wrt the "Engineers" thread, I recommend also reading the last section of the first post in this one: http://forum.beyond3d.com/showthread.php?t=19139 , which is a Kutaragi Cell interview from 2005. Sheds further light on the arbitrariness of the 8 SPE decision/disconnect between IBM and Kutaragi, and most of us had read it before One posted the "Engineers" thread above, which is why you have contextual color posts like post #29 in the Engineers thread.

 

[eastmen]

According to the dude that wrote the book, perhaps. But have you read through the link above? As far as that guy was concerned it may always have been 6 SPEs, but there was a disconnect between what the team put in motion and what Kutaragi felt to be the case. Back at HQ, 8 SPEs was the original plan, and part of the reason it went to 8 was simply due to the fact that Kutaragi was made aware that they were working around 6.

EDIT: For context wrt the "Engineers" thread, I recommend also reading the last section of the first post in this one: http://forum.beyond3d.com/showthread.php?t=19139 , which is a Kutaragi Cell interview from 2005. Sheds further light on the arbitrariness of the 8 SPE decision/disconnect between IBM and Kutaragi, and most of us had read it before One posted the "Engineers" thread above, which is why you have contextual color posts like post #29 in the Engineers thread.

Just stating what the book said . I really don't know what hte case was. The book says he just came in and said he wanted 8 and that ibm was upset because it would no longer be the money maker they were planing it ot be.

 

[Toaddio]

Am I the only one that remembers original designs having 16 SPUs?

 

[Carl B]

Am I the only one that remembers original designs having 16 SPUs?

You're thinking about the 1 TFLop design from the patent I believe, which had a lot more going on. The STI team entertained the idea of pursuing that vision via an MCM design, but abandoned it early on. That aspect is actually covered in the Engineers thread.

 

[liolio]

I don't know if it will add much hto the discussion (I still haven't read it) but here the link to an interview of David Shippy (IBM technical architect).
http://www.gamasutra.com/php-bin/news_index.php?story=21820

 

[AlNom]

Thanks for the link, lio.

At that time, only the IBM engineers knew about it. They struggled with betraying their partners -- and that was my initial reaction, too."

Ah yes... Code of Ethics and Conflict of Interest... the bane of any serious engineer.

Edit: There's not too much interesting as far as technical details go, and most of what is mentioned was already covered to some extent.

 

[Dominik D]

Link to the full interview is here:
http://www.gamasutra.com/view/feature/3904/processing_the_truth_an_interview_.php

The technical bottom line of this interview is: both consoles share technology behind low-power cores, but architectures are different and MS had no access to details about CELL. The nontechnical bottom line is: book wasn't written with sensationalism in mind, it was supposed to be a tale about engineers who delivered stuff on time.

Good thing Leigh Alexander did this interview. Another great piece of work by her.

 

[taucias]

My guess is that the development team behind the XB360 CPU got plenty of information about the PS3 CPU design and this was used to the advantage of Microsoft. But in terms of using technology and R&D from the CELL, it is obviously incorrect.

It would seem to me that the PPC core was something IBM had already mostly developed ahead of time and that IBM's connection with CELL development was used to leverage a deal with Microsoft on the grounds that they could produce something that would be able to compete with it for a reasonable price.

 

[Carl B]

I don't think that interview shed overly much additional light on the situation outside of what we were already aware, but it did give the authors title/role of a lead architect for the PPU. So clearly a lot of his own experiences would tie quite directly to the cross-developments between both camps, since it was the Power core that was essentially the tech MS gained/had access to. But I'd imagine that the names we normally associate with IBM's Cell team (Hofstee, Minor, Kahle, etc...), probably had little to do with what was going on with XeCPU development, since the R&D going towards the SPU's seemed to occur within a different... I'll say 'realm' for lack of a better word.

 

[Tap In]

hey, that interview has the answer to everyone's question here since 2005. (although we already knew the answer)

But can Shippy's insight on both console's processors finally answer the age-old debate about which console is actually more powerful?
"I'm going to have to answer with an 'it depends,'" laughs Shippy, after a pause. "Again, they're completely different models. So in the PS3, you've got this Cell chip which has massive parallel processing power, the PowerPC core, multiple SPU cores… it's got a GPU that is, in the model here, processing more in the Cell chip and less in the GPU. So that's one processing paradigm -- a heterogeneous paradigm."
"With the Xbox 360, you've got more of a traditional multi-core system, and you've got three PowerPC cores, each of them having dual threads -- so you've got six threads running there, at least in the CPU. Six threads in Xbox 360, and eight or nine threads in the PS3 -- but then you've got to factor in the GPU," Shippy explains. "The GPU is highly sophisticated in the Xbox 360."
He concludes: "At the end of the day, when you put them all together, depending on the software, I think they're pretty equal, even though they're completely different processing models."

 

[3dcgi]

Dean Takahashi published a good 3 page article about the book.

 

[Tap In]

nice review by Dean

this sums it up for me...

The great failure of Ken Kutaragi, the head of Sony’s game business, and his lawyers was that they failed to foresee how their competitor could catch them. IBM failed many years back to acquire all rights to Microsoft’s MS-DOS operating system. Bill Gates turned around and licensed it to all PC makers, creating an IBM PC clone industry that eventually allowed Microsoft and Intel to usurp all power from IBM. In this case, Kutaragi could have asked for exclusivity in the video game business for the PowerPC core, but he was blindsided. Sony’s failure to bar IBM from selling its own chip technology to Microsoft — which launched before Sony could use the chip technology itself — is astounding. This will go down as one of the worst business decisions of all time.


But as Microsoft’s engineers told me for my own book, they asked for a lot of modifications to that chip. They were aware of Sony’s own patents for the Cell processor and knew what they wanted to do that would differentiate the core. So Microsoft’s head engineer, Jeff Andrews, asked for changes and wound up with a very different microprocessor. The core was the same, but there were three cores on Microsoft’s chip and just one on Sony’s. And Microsoft had asked ATI Technologies, now part of Advanced Micro Devices, to fashion its own graphics chip. Microsoft engineers didn’t want anything to do with the Cell architecture or the vector units. They wanted much more powerful vector processing on the main cores. In the end, the Microsoft Waternoose chip was a custom microprocessor.


There was one IBM engineer who had his name both on Cell microprocessor patents as well as Xbox 360 microprocessor patents. But that didn’t mean that Microsoft completely copied Sony’s chips. Microsoft’s final chip had 165 million transistors, while Sony’s Cell had 234 million. You could say that Microsoft stole Sony’s chip, as some superficial book reviews say, but it’s just not true. Shippy agrees that angle is “overplayed” in the press (such as the Wall Street Journal story).
“That wasn’t the reason we wrote the book,” Shippy said. “IBM’s own core technology started even before Cell. We developed high-frequency, circuit design techniques at low power. That enabled us to create these supercomputers on a chip. It’s over sensationalized that Sony funded the Xbox 360 chip.”

 

[Shifty Geezer]

this sums it up for me...

I don't get the whole 'fiasco' here.

Sony’s failure to bar IBM from selling its own chip technology to Microsoft — which launched before Sony could use the chip technology itself — is astounding.

There is no failure. There's nothing particularly exciting about PPC in either console. If Sony had barred MS from using IBM, MS would have brokered a deal with Intel or AMD, and paid to have big vector units if they wanted them. There was no financial advantage to Sony locking out MS from one CPU vendor - it'd just be a waste of effort.

 

[AmFreak]

@ Shifty Geezer

Exactly - it would be another story if Microsoft "got" the SPUs, but the PPC Core is nothing fancy.
But nothing surprising here D. Takahashi is always a little on the Microsoft side

 

[BoardBonobo]

I don't get the whole 'fiasco' here. There is no failure. There's nothing particularly exciting about PPC in either console. If Sony had barred MS from using IBM, MS would have brokered a deal with Intel or AMD, and paid to have big vector units if they wanted them. There was no financial advantage to Sony locking out MS from one CPU vendor - it'd just be a waste of effort.

Maybe it's just that. MS would have to have paid for the extra work to be done. Instead they seem to have benefited from Sonys cash in terms of R&D.

 

[Carl B]

Takahashi is always a little on the Microsoft side

Well, indeed he is a little. The "audio visual freaks" line and the playing up the 'gravity' of Sony's missteps definitely seemed a little exaggerated, but it was a good article as a whole for sure.

Again it sort of went into some things we already knew, but it offered clarity in several regards where we had the bare minimum of information before. We got a clearer sense of timelines and inflection points, and a little third-party confirmation as to the consensus B3D theory of Blu-ray inclusion as the primary launch delayer.

I enjoyed it, exuberant sentences aside. Any one of the three pages wold have been a worthwhile mini-article in and of itself.

 

[one]

In this case, Kutaragi could have asked for exclusivity in the video game business for the PowerPC core, but he was blindsided.

Did Dean Takahashi forget who had designed GC/Wii's CPU?