Jim Keller Re-Joins AMD as Vice President and Chief Architect for CPU Cores

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
AMD has announced that Jim Keller has re-joined AMD as CVP and Chief Architect for CPU Cores reporting to Mark Papermaster. If The name sounds familiar, that would be because Mr. Keller was lead architect for K8, co-author of the HT specification, co-author of the 64 bit x86 spec and part of the system engineering team that launched K7.

With more than 30 years in the semiconductor industry, Mr. Keller has spent the past several years as the director of platform architecture at Apple and was a member of the team responsible for A4 and A5 mobile processor design. He previously served as the VP of Design for PA Semi; and at Sibyte, he created a scalable family of devices resulting in a successful business for Broadcom.

From AMD:

AMD announced today that Jim Keller, 53, has joined the company as corporate vice president and chief architect of AMD's microprocessor cores, reporting to chief technology officer and senior vice president of technology and engineering Mark Papermaster. In this role, Keller will lead AMD's microprocessor core design efforts aligned with AMD's ambidextrous strategy with a focus on developing both high-performance and low-power processor cores that will be the foundation of AMD's future products.


Mr. Keller's bio:

Jim has been a computer architect for 30 years. He's worked on processors for mobile, PCs, workstations, severs and networking. His passion is innovative yet practical computer design. He was the lead architect for two of the fastest processors, the Alpha 21164 and the 21264, the infrastructure for K8 and 64bit x86, powerful network processors and recently the most successful mobile processors ever. Practical to Jim means designing build-able systems and the teams and processes that go with them.

2008-12 Apple, Director, Platform Architecture Jim was a member of the small team that architected several generations of mobile processors including the A4 and A5 chips. He also worked with the Mac group specifying the compute, power and system needs for two generations of MacBook Airs.

2004-2007 PA Semi, VP engineering
Jim built and led the team to build a powerful networking SOC and it's integrated PowerPC processor. This chip combined low power, high performance out of order issue CPUs, and networking IO to support 10Gig processing and a wide range of control plane functions.

2000-2003 Sibyte, Broadcom, Director, Network Processors Jim was chief architect for a line of MIPS based network processors that supported 1Gig networking interfaces, PCI and other control functions. Jim and his team created a scalable family of devices that created a successful business for Broadcom.

1998-1999 AMD, Senior Fellow
Jim was the lead architect for K8, co-author of the HT specification, co-author of the 64 bit x86 spec and part of the system engineering team that launched K7. With K8 we created a new socket, glue less MP and brought 64 bits to the PC ecosystem.

1982-1998 Digital, Corporate Consulting Engineer I started my career in computer design at Digital. Jim's first project was the extremely successful VAX 8800, a large dual processor mini computer where I designed the cache controller and bus interface system. Jim joined the Semiconductor group in 1989 and was co-architect of the second generation Alpha chip, the 21164, then was co-architect of the third generation Alpha, the 21264 and lead for RTL and verification.

1980-1982 Harris, Design Engineer
Jim joined Harris right out of school as an electrical engineer doing board design, system debug and logic design for networking controllers.


*UPDATE*

The official press release from AMD:


AMD announced today that Jim Keller, 53, has joined the company as corporate vice president and chief architect of AMD's microprocessor cores, reporting to chief technology officer and senior vice president of technology and engineering Mark Papermaster. In this role, Keller will lead AMD's microprocessor core design efforts aligned with AMD's ambidextrous strategy with a focus on developing both high-performance and low-power processor cores that will be the foundation of AMD's future products.

"Jim is one of the most widely respected and sought-after innovators in the industry and a very strong addition to our engineering team," said Papermaster. "He has contributed to processing innovations that have delivered tremendous compute advances for millions of people all over the world, and we expect that his innovative spirit, low-power design expertise, creativity and drive for success will help us shape our future and fuel our growth."

Keller was most recently a director in the platform architecture group at Apple focusing on mobile products, where he architected several generations of mobile processors, including the chip families found in millions of Apple iPads, iPhones, iPods and Apple TVs. Prior to Apple, Keller was vice president of design for P.A. Semi, a fabless semiconductor design firm specializing in low-power mobile processors that was acquired by Apple in 2008. While there, he led the team responsible for building a powerful networking System on a Chip (SoC) and its integrated PowerPC processor. Keller previously worked at SiByte(R) and Broadcom as chief architect for a line of scalable, MIPS-based network processors that supported 1Gig networking interfaces, PCI and other control functions. Before Broadcom, he spent several years at AMD, playing an instrumental role on the design team responsible for the groundbreaking AMD Athlon(TM) 64 and AMD Opteron(TM) 64 processors, which featured the world's first native x86-64 bit architecture.

Keller co-authored the widely adopted HyperTransport specification, as well as the innovative x86-64 processor instruction set, which is used around the world today in hundreds of millions of desktop, notebook and server systems. Jim was a corporate consulting engineer at DEC, and architected two generations of Alpha processors during his tenure there. He holds a Bachelor of Science degree in Electrical Engineering from Penn State University.
 
Last edited:
If The name sounds familiar, that would be because Mr. Keller was lead architect for K8, co-author of the HT specification, co-author of the 64 bit x86 spec and part of the system engineering team that launched K7.

In other words everything great that AMD ever did lol.
 
I wonder if reports come through soon of everything past the 2013 roadmap being trashed?
 
My guess is he made a fortune working for Apple and now he can afford to work on projects he is passionate about.
 
My guess is he made a fortune working for Apple and now he can afford to work on projects he is passionate about.

Yeah and can get back to working in company where what he says carries more weight probably.
 
is this good news?
Well i think they lost one or two architecture folks lately so i don't know what this means for AMD
 
I've gone nearly 4 years with my current CPU and I used to upgrade every year no fail. Chances are I can go another 2 years with it.;)

Exactly. My wife's Q6600(@ 3.6) shows no signs of not being able to run any program or game I throw at it. That was bought in 2008. I suspect another 2 years out of that easily.
 
Sure... In 3~7 yrs knowing AMD's modus operandi....:rolleyes:

Developing a new architecture takes that long, regardles of who you are.

The only reason Intel comes out with refreshes more often is because they have multiple concurrent projects going on.

Look at how long it took Intel to recover from Netburst!
 
Even though I haven't built an AMD system in years this is very good news. The last thing we all need is for AMD to "go away". So here's hoping this turn means good things for AMD. :)
 
Ancient aliens theorists say that this guy is an alien helping our evolution.

e177e98a80.jpg
 
this is one of the most encouraging decisions made since mr read took the big man’s chair.
 
this looks promising

Indeed. Though it will probably be 3 years or more before we see something come out of this given the long development cycle of CPU architectures. Though we may see some nice improvements to AMD's Bulldozer architecture earlier than that in a refresh if he sees something with it that can be done relatively easily to enhance performance.
 
Indeed. Though it will probably be 3 years or more before we see something come out of this given the long development cycle of CPU architectures. Though we may see some nice improvements to AMD's Bulldozer architecture earlier than that in a refresh if he sees something with it that can be done relatively easily to enhance performance.

Seeing Steamroller-based processors won't be out until 2013 at the earliest in Kaveri APUs, my best guess is the one after that-- Excavator. It would be great if he decides to change that completely since nothing is known about it except it's somewhere on AMD's roadmap coming after Steamroller. Though I will be equally surprised if he brings Steamroller-based desktop processors earlier than 2014 with improvements-- big or small.

It's nice to see someone from the "old guard" that made AMD successful in the past come back again. This is going to be interesting to see how this plays out in the years to come.
 
It's nice to see someone from the "old guard" that made AMD successful in the past come back again. This is going to be interesting to see how this plays out in the years to come.

So when will Dirk make his return?:D
 
So when will Dirk make his return?:D

Dirk was better as an engineer and project manager than he was as an executive. I would love to see him come back in that capacity and work some of that K7 magic again, but that position is probably no longer what he is interested in.

Unless - of course - he feels like he's made his CEO millions, and now he just wants to do something he's passionate about. I wonder if they'd have him back if he were interested
 
Indeed. Though it will probably be 3 years or more before we see something come out of this given the long development cycle of CPU architectures. Though we may see some nice improvements to AMD's Bulldozer architecture earlier than that in a refresh if he sees something with it that can be done relatively easily to enhance performance.

Lets hope for that to become reality. I think the previous guys did not know what the hell to do.
My opinion is to bring a new high end part for desktop pcs in mid 2013 and not to wait until the end of 2013. A refresh as you said. AMD needs something to get back on track.
 
Zarathustra[H];1038996255 said:
What about Jerry Sanders? :p

Getting the band back together?

It'd be like a Guns N' Roses reunion where Axl would be Hector Ruiz.
 
Getting the band back together?

It'd be like a Guns N' Roses reunion where Axl would be Hector Ruiz.

I don't know. It would seem to me Hector would be more like Robin Finck in that analogy. Not there from the beginning, and not there now. :p
 
Dear God, don’t bring Hector back. Under his "leadership" AMD went from selling cpus for $1000 a pop with the first 1 ghz slot A cpu, to selling the pos k2 for $35.
 
Exactly. My wife's Q6600(@ 3.6) shows no signs of not being able to run any program or game I throw at it. That was bought in 2008. I suspect another 2 years out of that easily.
I've been toying with the idea of getting rid of my sig rig for some time now, but performance wise I can't find a reason to do so. If I were constantly encoding videos on it, I could see a point in upgrading, but that isn't the case. Right now and for as long as I've owned it, the comptuer does everything I need it do, and very quickly at that. My 3.4Ghz Core i7 desktop at work doesn't feel faster than my Q6600 at home when doing the same tasks since both are running SSDs. That fact alone makes it hard for me to spend several hundred dollars upgrading to the most up to date high end components. ( I figure $500-600 for an i7, well built mobo, and 8-16GB of quality ram)

It's nice to see someone from the "old guard" that made AMD successful in the past come back again. This is going to be interesting to see how this plays out in the years to come.
I feel the same way!
 
I hope AMD finally makes some kick ass CPU's again that don't require three power plants to run them. I'm serious, some good competition would be nice. It would make Intel finally go HAM on making better CPUs again and would allow AMD to make more money, become bigger maybe and consumers would profit from it.
 
In other words everything great that AMD ever did lol.
HAHAHAHAHAHA exactly. Here's hoping that maybe he can lead them back to the glory days of AMD. Now all I need is nVidia to go back to making awesome nForce motherboards.
 
Is AMD's focus still on making powerful CPU though?
 
Is AMD's focus still on making powerful CPU though?

They are all powerful. It's just that some other CPUs will let you finish a 65 minute HD rip/conversion two minutes quicker.

Thats about it.
 
He left apple because he's a processor guy..what can you do at apple bar get bored quite quickly?
 
Zarathustra[H];1038998823 said:
Well, the article did say he had a hand in developing Apples A4 and A5 ARM processors...

Those were ARM cores + PowerVR GPU. Apple didn't really fabricate/design anything so much as license others' work and he's likely the person that slapped it all together.
 
They are all powerful. It's just that some other CPUs will let you finish a 65 minute HD rip/conversion two minutes quicker.

Thats about it.

While using half the power..and it is more than just 2 minutes.
 
Indeed. Though it will probably be 3 years or more before we see something come out of this given the long development cycle of CPU architectures. Though we may see some nice improvements to AMD's Bulldozer architecture earlier than that in a refresh if he sees something with it that can be done relatively easily to enhance performance.

Yeah I was hoping for this as well but Steamroller is something different then just a normal cpu. When expectations run to high the end results tends to be disappointing.
 
Back
Top