• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Rumor: Semi Accurate: AMD kills Wichita and Krishna

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
seriously? wtf amd, guy like that should be a top priority to keep.

kick board members and overpayed top leaders before getting rid of engineers that know how to do designs like that. Those guys are the backbones of the company.

He wasn't fired. Samsung probably made him an offer that AMD was unable to match.
 
He wasn't fired. Samsung probably made him an offer that AMD was unable to match.

Yeah, I could see something like that being the sort of thing that precipitated Rory's decision to let go of the bottom-10% so they had the financial means to do a better job of retaining the top-10%.
 
Reading the laundry list of people leaving AMD recently is quite frightening.


Still, this doesn't mean the end of the world. If we believed a (well-managed) new company with AMD's IP would be more successful than AMD itself, then perhaps this rebirth will be better for the company in the long-run.

That being said, I can't believe they let the chief architect of Bobcat get away from them. Seriously?
 
Yeah, I could see something like that being the sort of thing that precipitated Rory's decision to let go of the bottom-10% so they had the financial means to do a better job of retaining the top-10%.

AMD had money. They have been making money for a while now (since the Meyer era) and this bobcat guy stayed for 8 years during which AMD had some bad moments, but he stayed. So why did he leave in August just months before the layoffs? Plus important people like him in a big company are usually tied down with things like stock options that don't vest for a few years and other "stay here" incentives.

Personally, I don't think there was a monetary offer he couldn't resist. It could be a monetary offer AMD didn't want to match. But personally I think it's likely more a director change he didn't agree with. Or just a chance to go up another rung on the career path.
 
AMD had money. They have been making money for a while now (since the Meyer era) and this bobcat guy stayed for 8 years during which AMD had some bad moments, but he stayed. So why did he leave in August just months before the layoffs? Plus important people like him in a big company are usually tied down with things like stock options that don't vest for a few years and other "stay here" incentives.

But during most of those 8 years, he was never considered important enough to poach.

After he was responsible for the biggest design win AMD has had since the K8, he suddenly became interesting to all the other companies that want good chip design teams.
 
But during most of those 8 years, he was never considered important enough to poach.

After he was responsible for the biggest design win AMD has had since the K8, he suddenly became interesting to all the other companies that want good chip design teams.

There is also the matter of precedence in these sorts of businesses.

Unless you are the CEO, you don't just get to name your price. If one project manager asks for, and gets, $500k for an annual salary while all others only get $250k then there will be repercussions to come as the other $250k managers get wind of the salary gap.

The term is disenfranchised. To maintain a sense of being enfranchised there must be a sense of fairness amongst those individuals within the same tier of an organization.

So when someone asks for an inordinate level of salary bump, the question to management is "are we prepared to increase the compensation package for this job grade across the board?" because in the end that is what will have to be done to maintain harmony across the entire company of project managers.
 
But during most of those 8 years, he was never considered important enough to poach.

After he was responsible for the biggest design win AMD has had since the K8, he suddenly became interesting to all the other companies that want good chip design teams.

You have a point but I don't think you poach people at his level with only money.

It's more likely someone like him decides to leave (and I'd imagine taking his best people with him) because he feels Samsung offers him something more substantial than money.

I mean we are talking about a company that booted a CEO who made the company profitable and stable. Then took 7 (yes only 2 months less than it normally takes to have a baby) months to find a replacement at a critical time just before the company was to launch a new processor. And the new guy has zero experience with semiconductors but rather came from software and (lets face it) low cost manufacturing in China. What would you do if you were, as it seems, a visionary who can conceive a successful new chip design?

Personally, I would have great incentives to walk to where the management seems to have their heads on right, where the company is actually heading somewhere and not stopped on the side of the road trying to figure out where it wants to go. (That's my personal take of where the new CEO is taking AMD, it's basically stopped at the side of the road about to change direction but no one knows to where).

Just want to clarify that while I think it's a bit odd what AMD is doing. I don't really see any plus or minus in the near term for their position. And I am not making any judgement as to their future success cause for all we know the new CEO is Steve Jobs 2.0 and will make things completely turn around. I am just saying that I can see people might decide that there are greener pastures.
 
Last edited:
There is also the matter of precedence in these sorts of businesses.

The term is disenfranchised. To maintain a sense of being enfranchised there must be a sense of fairness amongst those individuals within the same tier of an organization.

True. But frankly, I'd have no problem with giving the one guy who made a winning product with minimal resources a huge jump in compensation. If the others asked for the same level of compensation, I'd just ask the same level of results from them. 🙂

You have a point but I don't think you poach people at his level with only money.

It's more likely someone like him decides to leave (and I'd imagine taking his best people with him) because he feels Samsung offers him something more substantial than money.

Personally, I would have great incentives to walk to where the management seems to have their heads on right, where the company is actually heading somewhere and not stopped on the side of the road trying to figure out where it wants to go. (That's my personal take of where the new CEO is taking AMD, it's basically stopped at the side of the road about to change direction but no one knows to where).

I absolutely agree. Samsung probably asked him to design them a winning chip, with much more resources (and money) than he got at AMD. That, combined with working for the leader of several sectors as opposed to the also-ran of one, was probably enough. Sure would have been enough for me. 🙂

This is a huge problem for AMD. Back during the meteoric stock rise of the K8 era, they had no problem retaining talent. The second the stock cratered, they started bleeding their best and brightest at all levels. Whether it was due to failing stock options or stupid decisions by superiors hardly matters -- what matters is the fact that the people who made AMD's best products are now working at PA Semi, nVidia, Intel, and apparently now at Samsung.
 
There is also the matter of precedence in these sorts of businesses.

Unless you are the CEO, you don't just get to name your price. If one project manager asks for, and gets, $500k for an annual salary while all others only get $250k then there will be repercussions to come as the other $250k managers get wind of the salary gap.

The term is disenfranchised. To maintain a sense of being enfranchised there must be a sense of fairness amongst those individuals within the same tier of an organization.

So when someone asks for an inordinate level of salary bump, the question to management is "are we prepared to increase the compensation package for this job grade across the board?" because in the end that is what will have to be done to maintain harmony across the entire company of project managers.

There are ways for a successful company to compensate someone (stocks for example) without raising their base salary and causing an imbalance. Yes maybe people will find out about it but if it's a one time grant for doing things well it's not going to make sensible people to feel unhappy. It is a capitalist system where people are rewarded for being good at what they do right?
 
True. But frankly, I'd have no problem with giving the one guy who made a winning product with minimal resources a huge jump in compensation. If the others asked for the same level of compensation, I'd just ask the same level of results from them. 🙂

And they will respond with "no problem, just give me a cherry assignment like you gave that guy and I will then have the opportunity to outperform expectation".

Not every ambassador can be the ambassador to France, and not every ambassador can be treated like they made all the difference in the world when the truth is that there is a team involved that enabled the end solution.

There are ways for a successful company to compensate someone (stocks for example) without raising their base salary and causing an imbalance. Yes maybe people will find out about it but if it's a one time grant for doing things well it's not going to make sensible people to feel unhappy. It is a capitalist system where people are rewarded for being good at what they do right?

There is a range of compensation for every job grade, that is true. But not everything is about money.

I have a very good friend who is rather highly placed in a business based out of Connecticut. He and his wife cannot stand the winters. Regardless his value added to the company, the company cannot justify the expense that would be necessary to alter the weather patterns of Connecticut to be to his liking. So he quit to go work in Texas.

Who knows why this guy left AMD and went to Samsung. Only he knows why. It may have been out of frustration for the culture and bureaucracy of the job itself, regardless the stock options tossed at his feet.

I once had a consulting gig that paid $1k/day plus expenses, but it required traveling to Europe for weeks at a time and even though the money was nice the quality of life wasn't. I missed my kids too much, and they missed me. I'd rather live in my van but still see my kids than live in a mansion but feel like Harry Chapin wrote Cat's in the Cradle with me as his muse.
 
Come on beating Intel at anything with limited resources is hardly a cherry assignment. The bobcat is the best CPU out of AMD in a half a decade. If those other team leaders want more money then perform like this guy did. It would of been like ATI in the day letting go of the guy who lead the design team of the ATI9700 pro. AMD should of doubled any offer samsung made and devoted more resources to his future products.
 
No Krishna perhaps? I was so excited, though even if the chief engineer left, I would think the team would've been so far along and informed to finish it without their chief.

Also, I was thinking that maybe, being 4 x86 cores and what not, that Krishna would be stepping the A4 APU line in performance and encroaching on the A6. However, it's time to focus on Trinity based APUs anyways, but judging from BD, Krishna might actually just take a "4 core" Trinity APU to town (in x86 core performance) with superior power management to boot. Maybe AMD didn't want the lower line stepping on the toes of the upper line? I can foresee all sorts of "it would be super cheap to produce and we could mark up the price" arguments here, but in a structured line of products I can still see the need to try and keep things nice and separate.

Also, are the A4 APUs true from-the-fab dual cores, or are they full quads with disabled cores? Strange that the desktop A4s are 2 cores/160 SPs, and the mobile ones are 2 core/240 SPs. Makes me go hmmmmm:hmm:
 
Last edited:
I just think some people would like to cheaply swap out hardware whenever the latest ARM SOC gets released. (For example, according to this Anandtech article....Past TI OMAPs were pin compatible. If other manufacturers do the same thing, upgrade processor/boards should be affordable enough to make)

There was a reason for that pin-compatibility: market dominance and marketshare justified it.

03.jpg


^ This chart does not do the reality justice, go back to 2006 and earlier and TI held ~80% of the chipset marketshare for mobile phones.

Making their chips pin-compatible was a way to lock out the competition because the barrier-to-entry for a newcomer was essentially the design costs of an entirely new platform whereas the existing customer base had low expense to just keep on using TI's solutions.

Fast forward to today, TI has ~20% marketshare. That pin-compatibility provides little benefit to TI in terms of regaining marketshare. Don't be too surprised to see it go away in coming generations.
 
Hmm, so the 28nm bobcat parts are dead? Guess that explains the additional low power 40nm parts that showed up on their roadmaps!
 
The big change is in 2012 and 2013 when they move to GPU-like 20 and 28nm process. There were server roadmaps saying successor to Piledriver would move to 28nm. Either it was a decision to move to a completely different foundry(TSMC) or completely different architectural decision(lower power, more cores).

I don't think 28nm process would give big increase(if any) in transistor drive current, which affects clock speeds. But it may be power efficient to stick more cores in.
 
Only way this would make any sense at all is if Rory has decided to make an all-stop on bobcat product development in order to do a full court press into an ARM product development effort.
 
Only way this would make any sense at all is if Rory has decided to make an all-stop on bobcat product development in order to do a full court press into an ARM product development effort.

That is what I think is happening.
 
Only way this would make any sense at all is if Rory has decided to make an all-stop on bobcat product development in order to do a full court press into an ARM product development effort.

Do we know for sure that there aren't going to be any 28nm BCs, or just that Wichita and Krishna are goners?

Even if AMD decides to move into the ARM market, this doesn't mean they have kill BC development. Unless they're planning on dropping all x86 development. Which would be huge...
 
Do we know for sure that there aren't going to be any 28nm BCs, or just that Wichita and Krishna are goners?

Even if AMD decides to move into the ARM market, this doesn't mean they have kill BC development. Unless they're planning on dropping all x86 development. Which would be huge...

High-end 28nm Bobcat chips are still scheduled for release, replacing the E-350 and E-450 SKUs. Wichita and Krishna were low power SKUs, and they've been been replaced with tweaked 40nm designs, iirc.
 
Only way this would make any sense at all is if Rory has decided to make an all-stop on bobcat product development in order to do a full court press into an ARM product development effort.

So they want to go from the niche they have with Brazos to competing against multiple manufacturers in a new area with ARM?
 
So they want to go from the niche they have with Brazos to competing against multiple manufacturers in a new area with ARM?

If by "they" you mean AMD's board of directors, then yes, so it would appear given the events of the past 10-12 months.

I can't help but to keep coming back to my main point of concern regarding Dirk's departure from AMD - what on earth did the BoD insist he do that was so absurd, so ridiculous, that Dirk would rather turn down millions of dollars in compensation while pursuing the BoD's strategy (even if it was a foolhardy one)?

There's a lot of BS pointless office work I'd be willing to do at my superiors behest if it meant I'd continue seeing 5-figure paychecks being deposited in my checking account every 2 weeks. It would take some rather stupifyingly idiotic planned course of action by my superiors before I'd jump ship simply to avoid having my career stained with their idiocy.

So what was it that the BoD wanted Dirk to do that he simply could not get up every morning to go to work doing despite the millions dangling in front of him?

I've yet to come to a satisfactory answer or conclusion to this singular question. And until we see the other shoe drop, we'll never know.
 
So they want to go from the niche they have with Brazos to competing against multiple manufacturers in a new area with ARM?

Maybe AMD can do something interesting with an Android app store?

http://www.digitimes.com/news/a20111116PD216.html

PC hardware players to phase out from tablet PC market in 2012

Aaron Lee, Taipei; Joseph Tsai, DIGITIMES [Thursday 17 November 2011]

Due to pure PC hardware players such as Hewlett-Packard (HP), Acer, Asustek and Dell not having any advantages to compete in the tablet PC market, sources from upstream supply chain believe these players will gradually phase out from the market with players that have strong content support such as Apple, Amazon and Barnes & Noble, to continue to compete through lowering their hardware prices.

With Amazon offering its Kindle Fire at US$199 and Barnes & Noble to provide its upcoming Nook Simple Touch at a price of US$99, the pure hardware players are unlikely to profit from the market through price competition.

Since Amazon and Barnes & Noble are mainly profiting from their content platforms, not the hardware, the sources believe these hardware devices will eventually be offered for free.

The sources pointed out that although iPad 2 is also seeing strong demand from consumers, sales were lower than those of iPad, indicating that consumers' strong enthusiasm for tablet PCs has already disappeared.
 
Last edited:
Back
Top