Jim Keller leaves AMD

[Arachnotronic]

You guys are all ridiculous. Jim Keller is an engineer that likes challenges. He was brought in to design a CPU. He did that. He leaves. That's what he does. By the doom and gloom logic being put out, Apple's SoCs should no longer be competitive because he left.

Yeah, except Apple has hired many of the best and the brightest from many top chip companies. AMD has been suffering from significant brain drain for years.

 

[el etro]

Perhaps, but the "leaving to pursue other opportunities" cliche usually means something is not going well, either technically or politically.

I don't think something gone seroiusly wrong for this happen. IMO he's just came to AMD only to do Zen.

 

[mrmt]

I already answered that.

You can't say anything for sure. Much of the talk is about a cash injection coming soon, why cut down on vital R&D?

I'm not sure if you think Brazos was responsible for that chunk of AMD R&D budget. Brazos was basically a low cost initiative. It wasn't a high cost product like Bulldozer. So if you think Brazos accounted for all the cuts AMD faced in the last few years you are mistkane.

And why do you take this cash injection for granted? All we have is a bunch of "news" sites talking about possible operation but so far nothing got disclosed by any part and neither by the specialized financial press.

And think about it, if this was a planned move by AMD why isn't AMD naming a successor today? Or even better, why didn't AMD disclosed a succession plan for Keller's position? Everything points out for Keller quitting, so he either fell out of favor with the current management because of the initiatives he pursued on his term or he didn't like where the company was heading. Once Zen debuts we'll have a clearer idea of what happened.

I don't think something gone seroiusly wrong for this happen. IMO he's just came to AMD only to do Zen.

Pie-in-the-sky stuff is happening if AMD knew its chief architect was going to leave and didn't prepare for it, e.g, naming a successor and disclosing it to the market.

 

[Arachnotronic]

I don't think something gone seroiusly wrong for this happen. IMO he's just came to AMD only to do Zen.

Very interesting spin. Do you think he let AMD know this a priori so that upper management could ensure an orderly transition to a new chief CPU architect?

 

[exar333]

This was either a massive fail on the PF front communicating the transition, or he was suddenly let go. I am leaning toward the latter.

 

[Arachnotronic]

This was either a massive fail on the PF front communicating the transition, or he was suddenly let go. I am leaning toward the latter.

Why would they let go of their chief CPU architect?

 

[ikachu]

AMD RSUs vest over a 3 year period so I guess he was probably waiting for that date to pass so he'd get all the stock that was part of his signing bonus.

I don't think he would leave if Zen and K12 development was a disaster, but on the other hand I don't think this is a great sign for AMD's roadmap beyond those cores.

 

[Arachnotronic]

AMD RSUs vest over a 3 year period so I guess he was probably waiting for that date to pass so he'd get all the stock that was part of his signing bonus.

I don't think he would leave if Zen and K12 development was a disaster, but on the other hand I don't think this is a great sign for AMD's roadmap beyond those cores.

Maybe not a disaster, but probably not as good as AMD management has been hyping up to investors.

 

[SOFTengCOMPelec]

Trustworthy site, seems to have the announcement as well, now.
(Anandtech).

http://www.anandtech.com/show/9643/jim-keller-leaves-amd

AMD’s history has been well documented, especially given several reorganizations in the early part of this decade along with changes in senior staff and how both its market share in CPU and GPU markets is progressing. Today we have learned that one of those senior staff, the head of the CPU group Jim Keller, is to leave AMD effective September 18th (today). Readers may remember that Jim Keller was a recent re-hire in 2012, tasked with leading AMD's CPU group and helping the company develop new core processor architectures in order to bring AMD's architecture in line the competition.

Jim Keller has worked at AMD before, most notably developing the K7 and K8 processors that formed the basis of much of AMD’s success at the turn of the century. This includes assisting in the generation of the x86-64 instruction set that would form the basis of many of the x86 based computers people used today. At other points in time Jim has also spent several years each at Apple helping design their A4 and A5 SoCs as well as at DEC on Alpha processors, giving him a wide degree of experience in CPU development that AMD has been tapping during his latest tenure there.

As a re-hire at the top of the CPU chain, Keller's latest project at AMD was to develop the next generation of high performance processors for AMD and to build a team around the concept of PC performance. This was announced as a rapid departure from the module design of Bulldozer-based cores sharing parts of a processor and towards a new base architecture called Zen. Other projects in the pipeline at AMD CPU group include ARM-based AMD processors (K12), an ARM counterpart of sorts for Zen that is set to launch later on.

 

[VR Enthusiast]

I'm not sure if you think Brazos was responsible for that chunk of AMD R&D budget. Brazos was basically a low cost initiative. It wasn't a high cost product like Bulldozer. So if you think Brazos accounted for all the cuts AMD faced in the last few years you are mistkane.

Brazos and Merom was mentioned in response to ShintaiDK's request for example's of working A0 chips.

[And why do you take this cash injection for granted? All we have is a bunch of "news" sites talking about possible operation but so far nothing got disclosed by any part and neither by the specialized financial press.

I'm not taking anything for granted. What I'm definitely not doing is assuming (falsely) that AMD cut back on Zen R&D. Anyone can figure out how stupid that would be and that the R&D cuts came from elsewhere.

But now you mention it, yes Brazos's successors - or lack thereof - probably made up for a decent portion of the cuts amongst others. Cutting Zen R&D though? That would be stupid. AMD are a lot of things but they aren't stupid.

And think about it, if this was a planned move by AMD why isn't AMD naming a successor today? Or even better, why didn't AMD disclosed a succession plan for Keller's position? Everything points out for Keller quitting, so he either fell out of favor with the current management because of the initiatives he pursued on his term or he didn't like where the company was heading. Once Zen debuts we'll have a clearer idea of what happened.

It's possible but who can say what the motivations were? Maybe something popped up yesterday, a new opportunity. Maybe he was just sitting around doing nothing, waiting on silicon coming back before leaving.

Is finding a replacement so desperate right now? Zen, Zen+ and Zen++ are probably already designed. You know Papermaster is IBM A-grade chip designer too? He can surely fill in that job in a quiet period for a few weeks.

Pie-in-the-sky stuff is happening if AMD knew its chief architect was going to leave and didn't prepare for it, e.g, naming a successor and disclosing it to the market.

If we're being honest, Keller seems to do this a lot, doesn't he?

 

[pablo87]

#1 epidemic engineer departures, either no replacements or inexperienced ones.
#2 lack of resources
#3 company/ stock has no upside.
#4 no faith in mgmt or BoD
#5 no faith in Global Foundries.

I think without addressing #4 and #5 with a full reset by spinning off GPU to hopefully pay the debt off, only someone who can't find a job elsewhere should work there. I agree that Lisa is next.

 

[mrmt]

I'm not taking anything for granted. What I'm definitely not doing is assuming (falsely) that AMD cut back on Zen R&D. Anyone can figure out how stupid that would be and that the R&D cuts came from elsewhere.

In hindsight stupid decisions were plenty in AMD history, why do you think they wouldn't do it again? Besides, given their financial situation, I don't think they had much choice with regarding of cutting back R&D, Zen or not Zen.

It's possible but who can say what the motivations were? Maybe something popped up yesterday, a new opportunity. Maybe he was just sitting around doing nothing, waiting on silicon coming back before leaving.

If it was planned AMD would have made a nice press release thanking him for all the years he gave to AMD and announcing a replacement for him, even on a temporary basis. If he was fired AMD would at least have named a replacement for him. As it seems Keller seems to have bailed out without prior notice and AMD was left in the cold to find a replacement for him.

Is finding a replacement so desperate right now? Zen, Zen+ and Zen++ are probably already designed. You know Papermaster is IBM A-grade chip designer too? He can surely fill in that job in a quiet period for a few weeks.

If Zen+ and Zen++ were finished AMD would be releasing these chips and not Zen.

 

[jpiniero]

Why would they let go of their chief CPU architect?

Because they have no money? AMD will likely announce more layoffs soon and either he left on his own because of the layoffs or he was part of it. No point in sticking around when they aren't going to last long enough to see his vision fulfilled.

 

[VR Enthusiast]

In hindsight stupid decisions were plenty in AMD history, why do you think they wouldn't do it again? Besides, given their financial situation, I don't think they had much choice with regarding of cutting back R&D, Zen or not Zen.

I think they did have a choice of what to cut back on R&D. What makes you think that they didn't cut back elsewhere? If you (or ShintaiDK) have proof that Zen R&D was cut I would like to see some evidence of it.

If it was planned AMD would have made a nice press release thanking him for all the years he gave to AMD and announcing a replacement for him, even on a temporary basis.

Did Apple do the same when he left for AMD? I didn't see it if so.

If he was fired AMD would at least have named a replacement for him. As it seems Keller seems to have bailed out without prior notice and AMD was left in the cold to find a replacement for him.

Maybe they're not in a hurry? The chip is designed now it's quiet period. A few weeks won't make any difference.

If Zen+ and Zen++ were finished AMD would be releasing these chips and not Zen.

That's not how it works.

 

[mrmt]

I think they did have a choice of what to cut back on R&D. What makes you think that they didn't cut back elsewhere? If you (or ShintaiDK) have proof that Zen R&D was cut I would like to see some evidence of it.

Where supposedly could AMD cut?

Did Apple do the same when he left for AMD? I didn't see it if so.

Maybe because Keller leave wasn't planned?

Maybe they're not in a hurry? The chip is designed now it's quiet period. A few weeks won't make any difference.

Do you even know what quiet period means? They are not in the quiet period, that only happens after registering a document in the SEC, and that will only happen in October, and the replacement of a chief architect certainly doesn't fall into the kind of information the SEC would restrict disclosure.

That's not how it works.

With 1-2 years the design of a MPU isn't finished. The high level design can be, but there's *a lot* of work to do after finishing high level design, work that Keller won't have a hand into it.

 

[Glo.]

I have a distinct impression that people from AMD reading this thread are laughing their ass off.


http://www.bitsandchips.it/52-engli...-amd-zen-is-completed-and-it-will-be-powerful

http://semiaccurate.com/forums/showpost.php?p=234879&postcount=45 He wrote this 2 months ago. 2 months.

 

[mrmt]

I have a distinct impression that people from AMD reading this thread are laughing their ass off.


http://www.bitsandchips.it/52-engli...-amd-zen-is-completed-and-it-will-be-powerful

http://semiaccurate.com/forums/showpost.php?p=234879&postcount=45 He wrote this 2 months ago. 2 months.

Yeah, AMDZone lite FTW.

 

[Phynaz]

I'll bet he didn't know he was leaving AMD when he got up this morning.

 

[5150Joker]

I have a distinct impression that people from AMD reading this thread are laughing their ass off.


http://www.bitsandchips.it/52-engli...-amd-zen-is-completed-and-it-will-be-powerful

http://semiaccurate.com/forums/showpost.php?p=234879&postcount=45 He wrote this 2 months ago. 2 months.

Author: Gian Maria Forni
Expert: Expert markets and CPU
Although a degree in Humanities


A liberal arts major..yeah we'll take his word on it.

 

[Phynaz]

You guys are all ridiculous. Jim Keller is an engineer that likes challenges. He was brought in to design a CPU. He did that. He leaves. That's what he does. By the doom and gloom logic being put out, Apple's SoCs should no longer be competitive because he left.

His role as chief architect on Zen would have been over about two years ago. What was he doing all this time?

 

[VR Enthusiast]

Where supposedly could AMD cut?

I'd supposed in the "essential" stuff like Brazos. Low end GPU too, when was the last time AMD spent R&D there compared to 2012? When did AMD last spend money on SOI? When did AMD spend R&D on a new node period since 2012?

You think that stuff can't account for a 20-30% drop in R&D? I don't mind us going over this properly in another thread but this isn't the thread to discuss it.

Maybe because Keller leave wasn't planned?

I've been saying this all along.

Do you even know what quiet period means? They are not in the quiet period, that only happens after registering a document in the SEC, and that will only happen in October, and the replacement of a chief architect certainly doesn't fall into the kind of information the SEC would restrict disclosure.

You forgot that Papermaster is acting leader already or just didn't pay attention to it?

With 1-2 years the design of a MPU isn't finished. The high level design can be, but there's *a lot* of work to do after finishing high level design, work that Keller won't have a hand into it.

If you can show me a single case of a Jim Keller chip which failed *after* he left...or even before he left...it would make your case stronger.

 

[geoxile]

Most likely he's left because Zen is done and there are revision plans prepared for the following 3 to 5 years if not longer (considering it's AMD, they can't change archs too often), and AMD doesn't have clear plans for CPU archs following that, so they've either cut Mr Keller or he's left as there is nothing significant for him to work on. Most likely he left on his own, the guy's pretty high profile. No doubt he's looking for the next big thing to jump onto. If he doesn't retire he might even return to AMD when and the if the occasion arises.

 

[VR Enthusiast]

I'll bet he didn't know he was leaving AMD when he got up this morning.

How much would you like to bet?

 

[Redentor]

Author: Gian Maria Forni
Expert: Expert markets and CPU
Although a degree in Humanities


A liberal arts major..yeah we'll take his word on it.

Rob Siltanen, the mind behind the ads "Think Different" of Apple, has a Bachelor of Journalism. A liberal arts major has brainwashed the engineers there.

Engineers =/= understand IT world

 

[Phynaz]

How much would you like to bet?

Lol. If this was planned there would have been a succession plan in a press release.

Are there any statements from him like "I love AMD, they are on a great course and now is the right time to try something new"? Of course there isn't, because when you are terminated without notice there isn't time.