Sign in with
Sign up | Sign in

Opinion: 3 Ideas For AMD's Project WIN

By - Source: Tom's Hardware US | B 46 comments

AMD will be announcing its "Project WIN" this Wednesday as a follow-up to a massive layoff move that was announced a few days ago.

This announcement will unveil the board’s and Rory Read’s new business strategy for AMD and reveal how the company will be competing in a quickly reshaping processor market. The content of that announcement will have to be at least as shocking as the clean sweep through its executive ranks that we have been witnessing for almost a year.

No company ever announces layoffs. They are referred to euphemistically as restructuring or operational streamlining and are used to justify an obligation to help an organization remain competitive. Executive changes often require layoffs and replacements. It’s what we expect, and there is no surprise when we receive news of pink slips. We almost feel guilty about drawing a paycheck when a teary-eyed letter from the CEO explains just how much a burden we have become for the bottom line of an organization.

However, as painful as the layoffs at AMD are, I somewhat feel that the description of "restructuring" may be very appropriate. Several key personalities must leave: Carrell Killebrew, who is credited with the creation of Eyefinity; several PR reps; vice president of marketing, Margaret Franco; corporate marketing fellow, John Volkmann; and most notably, corporate vice president of strategy and fellow, Patrick Moorhead. Moorhead’s time at AMD goes back to Jerry Sanders. Moorhead was the last standing figure that represented the old AMD. One would have to be blind not to see that AMD will be changing dramatically.

In his letter to AMD employees, Read justified the layoffs with a need to “rebalance” AMD’s "skillsets.” The terminations will result in operational savings of $118 million in 2012 and frees up money to fund "key growth areas." In the letter, Read wrote that “a lower cost base allows [AMD] to be more competitive today and to invest back into the business to fuel [its] "attack" strategies in … low power, emerging markets and the cloud.” Details are promised to be provided in a “Worldcast” this Wednesday. These details must include three pillars that will have to carry AMD in the near future.

A Clear Message

I have been very critical of Read’s performance during his first 60 days at AMD and have been told that I need to cut him some slack. However, we know that CEOs who are capable of leading a Silicon Valley CPU company are rare because they require a very specific skill set. Therefore, the chances are slim by default that AMD’s board of directors chose Read as the right person. The confidence that was molded into Read’s scripted and repetitive phrases during the third quarter earnings call made me wonder if Read’s performance has been worth his $1 million sign-on bonus combined with his $1 million annual salary plus bonuses.

There have been questions why previous CEO, Dirk Meyer, had to leave. We haven’t been given specific answers yet. At this time, AMD’s future direction is blurry. What we know is that AMD’s board of directors replaced a high-level engineer at the top with a businessman who is much more focused on operational efficiency than technology leadership. This puts AMD much more in line with Intel, which made that transition more than a decade ago and has done well, especially in times when it needed the expertise in reorganization when the company was in trouble back in 2005. However, Intel has always been very clear about what it intends to do in the months and years ahead and has never left any doubt as to what it wants. This approach made it very clear what Intel is, and what Intel is not.

Read will have to mirror this approach. He will have to understand what AMD was, what it is and what it can be. He will have to credibly address current concerns, such as manufacturing errors at GlobalFoundries, and the way in which the company will address such problems in the future. Read’s reference to the "AMDer" feels a bit awkward considering that he has been on the job for only two months and has, according to his notes during the conference call, spent most of the time travelling and talking with customers. He will have to understand AMD’s inside passion and understand how to capture the organization’s spirit without the help of scripted speeches. The expectation from the outside was that a new AMD would have to hit the ground running. This may not have been the case with Read, at least according to perception.

Without a precise message on Wednesday, Read may already be struggling to succeed at AMD. He may have the goal for the company to "step out of the shadows of others," but there is nothing more important than to clearly define the goals and capabilities of a company after such a critical layoff announcement.

Credibility

This may be very subjective, but I sense that AMD has, on some levels, a growing credibility issue. Sure, Fusion has had considerable success, AMD has captured market share from Intel in the mobile segment, and I am hearing very confident notes from AMD staff that Trinity is something to anticipate. However, if we are honest, AMD’s technology is way too average today on too many levels. What AMD needs is a true flagship product in its core CPU space – a technology that carries the industry mindshare for an entire company.

Intel has those flagship products, which may not bring in revenues on a product level, but they pay for themselves 1,000 times over in perception creation and marketing. Nvidia is building those products within its Tegra platform and has clearly stated its intentions to include in each product generation at least one feature that comes as a complete surprise to its rivals – such as the fifth power saving core in Kal-El. Such surprises, if executed correctly, can create product and market leadership.

AMD made an attempt to devise such a product with the most recent FX-8150 but largely failed to capture the perception of a flagship product. The pitch greatly represents a similar approach as the 2006 Quad FX, which consisted of two dual-core CPUs that created one quad-core system. It did not offer any performance advantages over a dual-core system and was extremely power hungry.

AMD will need an unquestioned flagship product to support its overall product credibility.

Tablets and Smartphones

Let’s be realistic. AMD’s opportunity to jump on the smartphone and tablet processor train has come and gone. While we can still discuss whether there is a market for tablets outside the iPad, it is apparent just how much Intel is struggling to make a dent in this market and come up with a product that is competitive with ARM architectures. Despite its huge resources, Intel needs years to come up with even a halfway decent product that is suited for tablets. What does that mean for AMD if the company also needs to compete in traditional markets that are the source of its income? Should AMD attempt to break into the smartphone market with an x86 product and face the same uphill battle that Intel has chosen? In addition, AMD has the aforementioned credibility issue because the company never competed in ultramobile space with the exception of its Geode processors on the very low end. If AMD is basically occupied with battling Intel in traditional x86 markets, how could AMD sustain an onslaught of ARM vendors, including Qualcomm, Samsung and especially Nvidia? Taking its x86 architecture into the ultramobile space is a huge risk that perhaps AMD should not be taking.

Instead, it is much more likely for AMD to join Nvidia & Co. in the ARM camp and line up against Intel. Not only is there a new market that is opening up in ARM-based Windows devices, but AMD already has ties at ARM, and it could use its graphics technology to become a key differentiator in this market. Once again, AMD would clash with Nvidia and compete with Qualcomm, which purchased its Imageon graphics technology in 2008 and now calls it Adreno. However, a clash with Nvidia may be more reasonable than dealing with Intel.

Given the traction of the ultramobile market and the ongoing speculation that Read was hired because of Meyer’s lack of an ultramobile product roadmap, AMD will have to make a spectacular mobile announcement this Wednesday. The integration of ARM into its product line could do it.

Display 46 Comments.
This thread is closed for comments
Top Comments
  • 15 Hide
    geekapproved , November 8, 2011 7:03 PM
    I disagree completely. AMD does NOT need a flagship cpu product.

    They need to team up with Samsung and dominate Intel in everything.
  • 10 Hide
    GreaseMonkey_62 , November 8, 2011 7:23 PM
    If AMD were to create an ARM based mobile processor that integrates their graphics core, than they'll be able to not only beat Intel in that market, they'll have a serious chip that gives all other ARM manufacturers a run for their money. A dual-core Fusion"esq" ARM processor. Yeah I would definitely want a tablet with one of those in it.
Other Comments
  • 0 Hide
    AbdullahG , November 8, 2011 6:42 PM
    AMD should enter the mobile market, as well as develop a flagship product for their CPU market, but is laying off really the best option at this point? Especially figures that have benefit the company?
  • 1 Hide
    beenthere , November 8, 2011 6:50 PM
    Only time will tell if Read can guide AMD to prosperity. While reorganization was likely do, I'm not sure this approach is the ideal means to achieve the desired change? With AMD's limited resources chasing the cellphone and tablet market before getting their CPU Biz in order, will be a very tall and optimistic challenge.
  • 4 Hide
    wiyosaya , November 8, 2011 6:53 PM
    Hopefully, this business guy, Read, will have an understanding that innovation is what is required and throw his support to the creative people at AMD. Perhaps Dirk Meyer was let go because he tried to make product development fit his concepts instead of letting his top engineers create???

    Time will tell.
  • 15 Hide
    geekapproved , November 8, 2011 7:03 PM
    I disagree completely. AMD does NOT need a flagship cpu product.

    They need to team up with Samsung and dominate Intel in everything.
  • 3 Hide
    ScrewySqrl , November 8, 2011 7:19 PM
    perhaps AMD has passed its peak and can no longer compete with Intel. we saw this in the 1990s with Cyrix, which made 386 and 486 chips that could compete with Intel's 80x86 line, but fell down after the 150 MHz point, then vanished.

    AMD bashed Intel aside with the Athlon series in 2005. But ever since every iteration is falling behind. the FX is a terrible chip as a flagship. Trinity is supposed to be better (especially with improvements to integer calculation), but it will, even if fixed, be a generation behind ivy bridge, maybe 2 if it still only compete with i7-920 of so.

    that far behind is a sign of dying.
  • 7 Hide
    one-shot , November 8, 2011 7:20 PM
    AMD needs to be offer great products it can produce many of and remain profitable. Let's hope Project WIN doesn't fail. ( I went there)
  • 7 Hide
    arson94 , November 8, 2011 7:21 PM
    geekapprovedI disagree completely. AMD does NOT need a flagship cpu product.They need to team up with Samsung and dominate Intel in everything.


    And in a hostile fashion...
  • 10 Hide
    GreaseMonkey_62 , November 8, 2011 7:23 PM
    If AMD were to create an ARM based mobile processor that integrates their graphics core, than they'll be able to not only beat Intel in that market, they'll have a serious chip that gives all other ARM manufacturers a run for their money. A dual-core Fusion"esq" ARM processor. Yeah I would definitely want a tablet with one of those in it.
  • 1 Hide
    GreaseMonkey_62 , November 8, 2011 7:26 PM
    Quote:
    perhaps AMD has passed its peak and can no longer compete with Intel. we saw this in the 1990s with Cyrix, which made 386 and 486 chips that could compete with Intel's 80x86 line, but fell down after the 150 MHz point, then vanished.

    AMD bashed Intel aside with the Athlon series in 2005. But ever since every iteration is falling behind. the FX is a terrible chip as a flagship. Trinity is supposed to be better (especially with improvements to integer calculation), but it will, even if fixed, be a generation behind ivy bridge, maybe 2 if it still only compete with i7-920 of so.

    that far behind is a sign of dying.

    AMD's first 64 bit processor certainly put them at the top of the game. It was a brief moment when I thought AMD could start to be the top dog.
  • 2 Hide
    KelvinTy , November 8, 2011 7:44 PM
    Would it make more sense for the higher "management" staffs to have 80% salary as well?
    Decisions were made and they have consequences, and consequences SHOULD not equal to 10%? (Really?) staffs being fired...
    With the new project at hand, I hope AMD good luck, they need it and I certainly don't want Intel NOR Nvidia eating up the whole market.
  • 8 Hide
    Anonymous , November 8, 2011 7:45 PM
    You see. This, for AMD is something similar to 1997 for Apple Computers Inc.
    If Read could be a quarter of S. Jobs maybe AMD will be a winner again. Who knows!
  • 9 Hide
    Device Unknown , November 8, 2011 7:50 PM
    I feel bad for the people who got laid off, but in all honesty, AMD NEEDED a restructuring. It was painfully obvious that the folks they had their were not cutting it. Bulldozer is my proof. So get some fresh blood in, and try again. This time, do BETTER. YES they need a flagship processor that can compete or beat Intel, that goes without saying. I think that their super low voltage systems need to be a side business. Granted the majority of their income came from these low cost APU's. But think of the profit margin if they could release a CPU that could beat Intel at their own game! Or like you mentioned, partner with ARM but still have the ultramobile systems as side business. Their focus needs to be on their amazing GPU's and a processor to match.
    I really do love AMD, I am totally vested in them to do well, and I will support them anyway I can to get them their. even if that means buying a subpar processor which is still 10x faster than I need it for.
  • 8 Hide
    ewood , November 8, 2011 7:55 PM
    flaghip cpu, arm+gpu fusion for laptop market and discreet GPU. three areas they can make money. they have gpu experience and a strong showing in discreet gpu sector. they have ties with ARM= high performance GPU+ARM CPU chip. they still need to up their x86 game though so they can regain the reputation of producing fast CPUs so when people think AMD they think 'fast', not 'budget'.
  • 2 Hide
    Anonymous , November 8, 2011 7:58 PM
    I do not think AMD needs to enter the mobile space, they are at a huge disadvantage to nvidia who has a few years head start. The only thing AMD can bring to the table is graphics, and nvidia already brought that to the table. I personally prefer a solid cpu over a ok cpu + solid graphics, a tablet just is not the form factor for serious gaming, it can be if there were standardized gaming peripheral devices for phones and tablets but right now there is not, and game control are currently based on gyros/accelerometers and touch.

    The difference between a flagship product and a mass market product is not much from an engineering perspective. Flagships have all cores enabled, more cache, and some other minor tweeks WRT to mainstream products. To say that AMD needs a flagship product is the same as saying all of AMDs product line needs to be competitive with Intel's product line, which is the same as saying needs help from God.

    AMD needs to do 1 thing in the next 2 years, call apple everyday and get them to use the fusion APUs in their notebooks and desktops. With this, AMD will double it's CPU sales, and the income from apple will serve as a stable income stream for years to come. If I was apple, I would use the Liano chips, very balanced chip and energy efficient, not cutting edge but nobody buys apple expecting cutting edge hardware, they expect a cutting edge OS and simplicity.
  • 1 Hide
    saturnus , November 8, 2011 8:20 PM
    Although being the dominant processor design, ARM lacks something that AMD has: an x86 license. It is highly possible that AMD will build co-processor that either make on-the-fly x86 recompiling or just do x86 app processing in general for the ARMv8. That will make AMD part of the next generation dominant platform, not only for android/linux but also for the microsoft platform. It will make ARMs dominance extend to both servers and desktops almost immidiately.
  • 0 Hide
    fazers_on_stun , November 8, 2011 8:33 PM
    Hmm, didn't Read state during the Q3 conference call that 'AMD would not be entering the tablet market?

    However, S/A has a similar take to this article - AMD will be entering the ARM mobile market. S/A also mentioned that the AMD layoffs also got rid of some long-standing engineering teams, in particular ALL the fusion related teams not based in Austin. That would have to affect Trinity.

    S/A goes on further stating AMD is making a "monumental mistake" by getting an ARM license and moving towards cellphones/tablets etc - they have zero wireless experience and patents, and would face much greater competition in a market with zero margins.

    If these rumors pan out, it would seem AMD is going to abandon desktop & server in the next couple of years..
  • 1 Hide
    wishmaster12 , November 8, 2011 8:57 PM
    Amd needs new, not just improved. Bulldozer has been in the Amd development room for years, bulldozer would have been fast "a year ago" but to todays standard its just "ok" fast.
  • 0 Hide
    wishmaster12 , November 8, 2011 9:00 PM
    I want to see a threads going into the bulldozez cores at the same speed as amds hypertransport
  • 0 Hide
    Anonymous , November 8, 2011 9:11 PM
    AMD paid McKinsey a lot of money to create "Project Win". Let's hope it was worth it.
  • 0 Hide
    spookyman , November 8, 2011 9:14 PM
    AMD should of concentrate on the APU market.

    I was hoping to see an faster iteration of the 3850 APU. That would of been a better outcome for them over Bulldozer.

    Maybe incorporate the Radeon GPU into a CPU?
Display more comments