AMD guts PR team; move portends more changes

We knew yesterday’s news of mass layoffs at AMD would involve lots of people—roughly 1,400—losing their jobs around the globe, never a happy development. We understood the situation to be wide-ranging when news filtered in of some executive layoffs, including marketing guru Patrick Moorhead and graphics product manager Carrell Killebrew, whose role in the RV870’s development was famously chronicled at AnandTech. What’s become clear in the hours since the announcement is that some parts of the company were definitely hit harder than others.

One of the areas hardest hit is the relatively high-visibility public relations team, which has been reduced to a shell of its former self. We’ve worked with AMD’s PR and reviews team extensively over time, of course, and happen to know quite a few of the people who worked there. All evening and into this morning, the tweets, emails, IMs, and Facebook posts have kept coming in as members of that team have notified folks they are no longer at AMD. From what we can gather, virtually the entire team has been gutted, with only one high-ranking manager from the graphics group confirmed to be remaining. We’ve heard those who were let go had no sense it was coming, no warning that the layoffs would be this deep or affect the entire group.

The timing seems particularly harsh given the fact that most of the individuals involved had just finished working very hard on one of the most difficult tasks in the industry: introducing a grossly underperforming product—the Bulldozer-based FX processors—to the public.

In a particularly prickly bit of irony, some early news stories about the layoffs were laced with disingenuous spin. They laid the blame for AMD’s problems on the health of the overall PC market in the face of rising interest in smart phones and tablets. That interpretation seems rather warped, given that Intel enjoyed record earnings in its latest quarter, as part of a string of record performances, and Microsoft did, too. The data seem to indicate PC growth has slowed among consumers, but growth does continue.

An early AP story that offered the tablets-and-phones spin has since been updated with a more nuanced version, and one senses that the sentiments expressed about the state of the PC market came directly from the top at AMD. Thus, the updated story includes some interesting indicators about what’s next for AMD. Apparently the company’s board has a definite mission for its new CEO:

Read’s job in large part is to help devise a strategy for AMD to penetrate computing markets where it and rival Intel Corp. have been largely absent. The battle has taken on a new dimension as AMD’s and Intel’s market share in PCs has reached a steady balance for years — Intel’s chips are in about 80 percent of the world’s PCs, and AMD’s are in essentially the rest.

Not having much presence in mobile devices has hurt AMD more than Intel because of its smaller size and it was a key reason AMD ousted Read’s predecessor, Dirk Meyer, in January.

The firm’s future is beginning to look like it will be focused on ever-smaller integrated chips, like the Brazos APU, intended for smaller, lower-power computing devices.

The question now is whether AMD will continue to participate robustly in its traditional markets for desktop PCs, workstations, and servers—and if so, whether it intends to field truly competitive products in the high-end and mid-range segments where Intel has been so difficult to defeat in recent years. Whipping the Bulldozer microarchitecture into shape, given its current competitive position, would likely involve a substantial ongoing investment. If there is a major shift to low-power and mobile-oriented devices, the relatively successful desktop graphics business could be another casualty—although that’s just speculation at this point.

Eventually, Read and his rapidly forming new executive team will have to articulate a new strategy to AMD shareholders and the general public. Given everything we’ve seen to date, we expect sweeping changes in AMD’s direction to be announced, once that time comes.

Update 1:55PM: We have confirmed that AMD has retained at least three senior members of its PR staff. However, we’ve also learned the cuts appear to have affected marketing nearly as deeply as PR. We’re still trying to confirm some departures.

Also, according to the purported email about the layoffs sent to AMD employees, new CEO Read plans to "share more insights into our strategy and path forward in my upcoming Worldcast on November 9th."

Comments closed
    • pogsnet1
    • 8 years ago

    It is the new CEO’s decision, so let him manage now and let’s see whats gonna happen next.

    • Abdulahad
    • 8 years ago

    I think AMD has huge problems in product execution and implementation. When they released Phenom, they marketed it as the True Quad Core design as opposed to Intel Core2Quad which was basically 2 Dual Core chips “glued” together, result…. their Phenom was getting raped.With Bulldozer, they marketed it as the First 8-Core chip… but what they do not say is that our 8-Core chip is getting bulldozed by Intel 4-Core.. Forget Intel, in some tests, Phenom II is even faster than Bulldozer!!!

    My point is that if you do not have a good product, what sort of doubtful PR strategies a company would be implementing?? If AMD has poor product offerings, what is the sense of keeping an expensive PR department?? AMD would shave off several hundred millions in its recent workforce reduction, these savings can be redirected to their R&D which has been starving for years(for example).
    New management, new ideas, we shall see how AMD as a whole performs and what type of product releases its going to flood the market with….
    AMD has been trying hard, but loosing hard to the competition. So if new management is taking drastic steps, we should not be surprised.
    Based on Bulldozer performances, I personally do not have any confidence in Trinity as it’s very foundation is from Bulldozer.

    • ronch
    • 8 years ago

    Sometimes, these layoffs are crazy. I mean, who tells the company’s different departments what to do anyway? Top management, that’s who. All the different departments probably have top managers who report directly to the top brass. Top brass decides if a CPU or GPU architecture should be given the green light or not or whether a certain marketing campaign is a go or not. Top brass approves which tools to use. Or whether to add more engineers. Etc. etc. etc. And when things go kablooey, those poor folks down there get fired and top brass get there bonuses. Unless the folks down there are a bunch of slackers, I think it’s just quite unfair.

    Tell someone to make coffee for you, and when you sip and realize you should have asked for tea, you yell at the person and kick him in the pants. Duh.

      • sschaem
      • 8 years ago

      Nha. Top management give goals an direction, the head of various department are in charge of execution.

      The layoff is on failed execution. The CEO that put AMD off track and caused so much havoc within was already layoff.

        • ermo
        • 8 years ago

        Which CEO are you referring to here? Dirk or Hector?

    • jihadjoe
    • 8 years ago

    Firing those PR people is sweet justice.

    You do NOT revive your most premium badge and slap it onto an underperforming product. All the hype they managed to create has resulted in nothing but a huge blowup, and a lot of negative sentiment once reviews went live. IMO they would have done a better ‘job’ not doing anything at all.

      • Yeats
      • 8 years ago

      I feel bad for the “drones” – the people who simply do what they are told. IMO, it’s the decision-makers, the bosses who crafted the message, who deserve to have the ripcords removed form their golden parachutes.

    • clone
    • 8 years ago

    a lot of ppl have been pretty hard on AMD and in all honesty I don’t believe even half of it’s justified.

    one could blame AMD for not becoming Apple but then no one else is either.

    AMD leveraged the farm to become Intel while Intel was being stupid, unfortunately Intel didn’t stay stupid long enough and AMD has been falling behind ever since which is what exactly was destined to happen once Intel stopped being stupid.

    personally I’m glad to see AMD headed in a new direction because the one they were following was killing them slowly, they aren’t big enough to own foundries, they lack the resources to compete on all fronts and they recognize the need to pursue unexplored avenues.

    I wish AMD all the success in the world and feel sorry to see them losing their desktop focus.

    • MrDigi
    • 8 years ago

    Let me guess their upcoming strategy, like many companies in this postion;
    Can’t afford to add new talent and fund projects, so cut least needed 10% employees.
    Next year start adding needed talent and projects.
    Don’t forget to throw in some pay raises for the top management for job well done.

    Apparently prospects for turnaround from Piledriver and GCN products aren’t looking good.

    Alternate POV;
    New products will be so good, they will sell themselves.

      • NeelyCam
      • 8 years ago

      [quote<]Apparently prospects for turnaround from Piledriver and GCN products aren't looking good.[/quote<] I found this from SemiAccurate article comments ([url<]http://semiaccurate.com/2011/11/04/getting-ready-for-trinity-yet/[/url<]): "fred, you are wrong.Trinity is a lot faster than you think.L2 cache in Piledriver is faster than BD(much less cycles), there will be a nice increase in IPC.Dont be surprised if a two module Trinity destroys FX4100 well over 30% in ST benchs." Also, I think I read that those first Trinity benchmarks were from A1 silicon... I think that's pretty damn impressive. To me, Trinity looks to be doing pretty well. Of course, we don't know how good Ivy Bridge is... and if the big money is in laptops, 22nm trigate power efficiency might be critically important, making it difficult for AMD to produce a competitive alternative.

    • Sam125
    • 8 years ago

    I’m assuming that the marketing and PR departments are going to be reformed after the layoffs as AMD really needs a department that highlights the merits of their products and downplays the possible negatives.

    • juampa_valve_rde
    • 8 years ago

    So after shooting on the foot (instead of the face thanks to the decision of skipping bulldozer on 45nm from Dirk Meyer), the current CEO throws away the talent from the graphics division (flawless execution since HD 3870)…

    Hey Mr Read (and dont paid attention) and Mr Papersomething, why dont just sent a knife to every owner of an amd processor/graphics card, or worse the goddam whole platform, with explicits instructions of how to cut the veins and let the blood drain.

    After many years of support to the underdog of the cpu industry i feel busted… yeah probably the knife idea is happier than cut the last remanents of skilled and talented people at a technology company.

    • wingless
    • 8 years ago

    Among the layoffs was Simon Solotko aka 64NOMIS. AMD is going to shit….

    [url<]http://www.xtremesystems.org/forums/showthread.php?276552-A-very-sad-day-for-all-AMD-fans[/url<]

    • Ryu Connor
    • 8 years ago

    So AMD is firing everyone whose vision doesn’t align with the concept that desktops are dead?

    Sounds like the board of directors have been drinking the Apple punch.

      • destroy.all.monsters
      • 8 years ago

      As enthusiasts we should be worried. That board should have been sent packing.

      The space between ARM and intel? AFAICT there is no space between them – just brutal competition in the handheld space.

      AMD doesn’t even seem to be trying to sell its GPUs in huge supercomputers and appear to have ceded that almost entirely to nVidia.

      • Beelzebubba9
      • 8 years ago

      I think it’s more like AMD is firing people involved with their desktop CPU division because they don’t have a competitive product….

      Brazos is great, Llano has at least some reason to recommend it, but Bulldozer? Yeesh.

        • destroy.all.monsters
        • 8 years ago

        No – marketing, pr. and apparently graphics have been hit hard. I don’t see the reasoning in letting Killebrew go.

        Unless they are maybe looking to kill off discrete gpus – which would be suicide on the short term (- on the long term they’d be competing with Imagination Tech).

      • Krogoth
      • 8 years ago

      Desktops market as we know is already on its way out. The demand for more CPU power only resides in the workstation/server markets. It is practically dead in the mainstream market as any value quad-core chip can effortlessly handle any mainstream task you throw at it.

      Intel already made its move by focusing on power efficiency which benefits laptops and portables more than laptops. The coming X78 boards and desktop SB-E chips are just beta-testing Intel’s next generation server/workstation platforms.

      Apple didn’t start anything at all. The end of the desktop market started when portable platforms became powerful enough to handle mundane stuff, while being affordable. At best, Apple just accelerated the process by making tablets desirable to the masses.

    • srg86
    • 8 years ago

    [quote<]The firm's future is beginning to look like it will be focused on ever-smaller integrated chips, like the Brazos APU, intended for smaller, lower-power computing devices.[/quote<] If that's what they are thinking, then they are pretty much doing what Cyrix did after they were bought out by National Semiconductor. In Cyrix's case it was abandon the 6x86MX/MII and the performance race for the MediaGX which was based on the 5x86 (a budget chip that ran in 486 boards). At the time it was for things like set-top-boxes, but eventually they ended up as the Geode GX and LX at AMD. What is being suggested here sounds eerily familiar except that these low power chips would be for mobile devices such as tablets, I guess.

    • brucect
    • 8 years ago

    Bad for those guys losing their job.Bad for us also they will more guys out at the street for looking for a job. They should go to occupy wall street.
    But its good for AMD since they don’t have (except GPU division) competitive product for market I think they can spend that money for R&D.
    BD sucked every point maybe they can get it right with spending for improvements.

    • HighTech4US2
    • 8 years ago

    So where is this LIST of those let go?

    Is JFAMD on it?

      • sweatshopking
      • 8 years ago

      why do you have such a personal vendetta?

        • Waco
        • 8 years ago

        He seems to think that John intentionally misled everyone when he was simply misinformed and didn’t have the authorization to correct his statements.

        The only thing that you could really be a bit ticked about is the WAY that he made those statements. Many times they were downright confrontational when people doubted him.

        Regardless – there’s no reason to be mad at him. It’s not like he said anything he didn’t believe was true.

          • HighTech4US2
          • 8 years ago

          John Fruehe: AMD’s latest and greatest liar

          [url<]http://scalibq.wordpress.com/2010/08/31/john-fruehe-amd%E2%80%99s-latest-and-greatest-liar/[/url<] John Fruehe finally does the sensible thing and comes up with an excuse [url<]http://scalibq.wordpress.com/2011/10/15/john-fruehe-finally-does-the-sensible-thing-and-comes-up-with-an-excuse/[/url<]

            • Yeats
            • 8 years ago

            HighTech4US2’s blog said this:

            “In fact, John even goes as far as claiming that Bulldozer will have at least 17% higher IPC than Deneb. ”

            Here is the specific post from Anandtech cited to support the above statement.

            “{JFAMD} Actually I said 3 things:
            1. Interlagos will perform 50% better than MC in the same thermals
            2. IPC would be higher
            3. Increase for single-threaded workloads will be “a lot” more than 17%

            I have made no statements that I am aware of about IPC with a percentage implied because I don’t know what the IPC is, all I was told is that it would be higher. If I did say IPC would be higher by a percentage, it was a mistake; ocasionally those things happen to us humans.

            The ~17% number (or the 12.5% or 12.8% numbers) are all tied to people trying to reverse engineer the 50% number (performance of 16-core vs. 12-core) for a single thread. I have said repeatedly that trying to pull single thread performance from a statement about a fully loaded and fully utilized processor is not going to be accurate. It is like trying to figure out travel times at 3AM based on rush hour traffic.

            Based on Scali’s treatment recently I just don’t feel like I need to respond to him at all. Let him think whatever he wants, it is more convenient for him that way.”

            • BestJinjo
            • 8 years ago

            You are an idiot. An excuse? A liar? Why do we even have people like you polluting this forum with garbage?

            If you read that post, it’s very obvious at JF-AMD was given information by the engineering team; and that team firmly believed that IPC was going to be improved. JF-AMD went out of his way to try to explain more specific details regarding Bulldozer’s architecture. Unfortunately, the engineers, who were very confident in their ability to increase IPC, didn’t actually make that happen.

            JF-AMD always stated that he “believed IPC will increase”. He never said it’s going to increase without a doubt or with 100% certainty. His mistake was that he trusted the engineers at his firm and passed on that information with more confidence than he should have. However, in no way was he doing this intentionally or lying.

            • HighTech4US2
            • 8 years ago

            JFAMD is an AMD Operative that spreads misinformation on Forums and attacks others who disagree with him.

            There is no place for corporate operatives on public forums they should stick to the company’s web site.

            As for polluting forums that is what JFAMD was doing.

            • Waco
            • 8 years ago

            Stop trolling.

            • Krogoth
            • 8 years ago

            U mad bro?

            • NeelyCam
            • 8 years ago

            Maybe he bought an AM3+ board in anticipation..

            • Waco
            • 8 years ago

            I did and I had an FX-8120 for a few weeks as well. Even I’m not mad at JF…

      • tejas84
      • 8 years ago

      JF-AMD should be released from AMD. He was on the server side and all he did was preside over AMD losing its server market. He did not do any favors for AMD and I bet Rory Read will can him.

        • Yeats
        • 8 years ago

        I don’t think “preside” means what you think it means.

        • beck2448
        • 8 years ago

        AMD has gone in a few years from 20% of the server market, which is also the most profitable CPU division, to less than 5%. That is a disaster that I don’t see them recovering from . Intel is just way out in front and have the cash to stay there indefinitely.

          • BestJinjo
          • 8 years ago

          It wasn’t a few years. It started since Phenom I, which happens to be the first in the line of 2 more CPUs to follow that was uncompetitive with Intel. Of course, it’s ludicrous to blame such a massive market share loss on 1 person, esp. one that had 0 to do with the engineering merits of Phenom I/II and BD (their horrible IPC, high power consumption, inferior memory controller, that reduced memory bandwidth that’s often critical in server applications, and low absolute performance).

        • BestJinjo
        • 8 years ago

        You are a funny guy. Are you implying that 1 person, who isn’t even a high-end executive was somehow responsible for AMD losing server market share for half a decade?

        Obviously this had NOTHING to do with lacklustre performance of Phenom I/II, delays of Bulldozer (obviously related to engineering mistakes and someone very high-up at the top approving the ridiculous decision to choose an 8-core shared design instead of a fast 4-core design)?

        Ya, that’s it buddy. A single guy ruined their server market share, a guy that had nothing to do with making the flops that were Phenom I/II and Bulldozer. We all know you would have convinced the large corporations of this world to pay thousands of dollars for inferior AMD processors because well you are the greatest sales person in the world, right?

      • Vasilyfav
      • 8 years ago

      I sure hope so. His “holier-than-thou” attitude betrayed a typical marketing shill.

      He knew precisely that Bulldozer was going to be a performance flop and continued to post in the vein of “sorry I can’t tell you anything, but please wait for the actual release” on OC.net.

      So many people did wait only to find they bought AM3+ boards in may for nothing.

        • Yeats
        • 8 years ago

        [quote<]His "holier-than-thou" attitude betrayed a typical marketing shill. [/quote<] He is a "typical marketing shill". Betrayal not found.

        • NeelyCam
        • 8 years ago

        [quote<]So many people did wait only to find they bought AM3+ boards in may for nothing.[/quote<] You can't blame JF for those folks' stupidity. I mean, why the hell would you buy a board for a chip that's not out there yet?? (Of course, AMD was really [i<]really[/i<] trying to push people to buy new boards so the "transition is smooth"... I think that kind of marketing is borderline criminal)

          • BestJinjo
          • 8 years ago

          Ya, seriously. There are only 2 people in the world that actually *needed* to upgrade their CPU for faster performance but waited 9 months since Sandy launched in 2011:

          (1) AMD faithful that like the brand and wanted to support it (fair enough)
          (2) Not-so-bright individuals who expected Bulldozer to smoke Sandy Bridge despite full well-knowing that Phenom II had a 35-40% IPC deficit against Sandy. So the idea that an 8-core Bulldozer would have higher IPC than Nehalem would have resulted in an 8-core Nehalem style processor, that well for $250-300 would easily beating Intel’s Core i7-990X. How realistic was such an outcome?

          These people only have themselves to blame, not JF-AMD. Logically, if they REALLY needed to upgrade the CPU, they wouldn’t have waited for 9 months (when SB had such excellent performance) only to blame their own poor decision making on someone else.

          Also, logically, there is no reasonable explanation as to why AMD needed to delay Bulldozer for so long if it had IPC of Nehalem or faster and still came with 8-cores!! Such CPU would have been a beast even at just 3.0ghz, nevermind 4.2ghz.

          • Meadows
          • 8 years ago

          Not necessarily. I only bought mine for example because the previous chipset(s) sucked ass.

            • NeelyCam
            • 8 years ago

            Yes; we already established that you had good reasons to buy one.

            I’m talking about those who bought one just because BD was coming out.

        • Bensam123
        • 8 years ago

        You know what’s cool about the AM sockets? They can still put a x6 core Phenom in it that is still quite competitive even though most programs don’t make use of the two extra cores.

      • destroy.all.monsters
      • 8 years ago

      Should we expect a news report soon about you waiting outside his house with a sniper rifle and dynamite?

        • HighTech4US2
        • 8 years ago

        Should we expect a news report soon about you waiting outside a preschool with some candy and a stuffed toy.

          • Yeats
          • 8 years ago

          Do you see the difference? destroy.all.monsters made a comment based on your obvious hatred for JFAMD; you made a comment about destroy.all.monsters that is baseless and tasteless and not pertinent to anything.

            • HighTech4US2
            • 8 years ago

            I strongly dislike Corporate Operatives who pollute the public forums.

            JFAMD is a Corporate Operative.

            As for hate it seems that both destroy.all.monsters and yourself seem to have that feature enabled.

            • Yeats
            • 8 years ago

            Please explain to me how I have the “hate feature enabled”. I don’t believe I have expressed any opinion regarding you.

            Yes, JFAMD is a “Corporate Operative”. He basically says as much in his every post, as in his sig he acknowledges he is an employee of AMD. This allows those who read his posts to adjust their opinions of his messages based on this. He is an Overt Operative, not a Covert one.

            Your personal crusade against JFAMD has actually attenuated the message you’ve been preaching for a while; Bulldozer is underwhelming, and leading up to launch the information AMD had John Fruehe and others spread about BD performance is false. A question to ask is, “is the messenger responsible for the content of the message?”

            • destroy.all.monsters
            • 8 years ago

            [b<]HATE FEATURE ENABLED![/b<]

          • destroy.all.monsters
          • 8 years ago

          Nah my kids are all grown. Older than you possibly – and thankfully with a great deal more sense.

      • NIKOLAS
      • 8 years ago

      I never thought I would see a bigger idiot than Randy Allen, then along comes John Freuhe aka Randy Allen II.

      • sweatshopking
      • 8 years ago

      You’re an extreme weirdo. Let bygones be bygones. You needn’t get so caught up in the this. You don’t own intel, what the hell do you care? geez, you’re talking about a real person, have some respect. you wouldn’t speak like that to him in person, so you probably shouldn’t online.

        • sschaem
        • 8 years ago

        Reading the various thread where he posted, I think JFAMD pissed off allot of AMD fans, ALLOT.
        JFAMD did a huge disservice to his employer.
        So its kind of natural to think AMD would want to improve their image and let go of negative influence.

        The issue is did he breach his employments contract by misleading people on AMD products as an individual?
        And is he that valuable to turn a blind eye?
        If so, he might just be reprimanded and asked not to “comment on unreleased product”.

          • Waco
          • 8 years ago

          He was cleared to release what he did…but not cleared to tell everyone that he was told incorrect information.

          Why people are pissed AT HIM I don’t understand.

      • ronch
      • 8 years ago

      Is JFAMD John Fruehe of AMD?

      Edit : Oh, ok, he is John Fruehe. Enough said.

      • drbaltazar
      • 8 years ago

      fx 8150 isnt crompehended by ms amd and mobo manufacturer.and sad part is amd did say they wouldnt speak about the fixes since they are at the fringe of tech.we saw just the beginning here.the issue as to do with various front core parking affinity.and the fact that those 8core are supposed to also have the capacity to do 1 big thread of 8 paralle process per clock .sadly till amd supply the fix.cause cleary ms isnt in amd secret.we ll have towait.one thing is sure this processor was released 6 month too early at least

    • Bensam123
    • 8 years ago

    You know, it’s not like there aren’t other chip makers in those segments. AMD will butt heads with ARM if this comes true. And ARM has been in the smaller chip business much longer then AMD has, that’s almost like starting a new in unfamiliar territory, which there is another ‘Intel’ that is firmly entrenched with years of experience for this exact market.

    I applaud the changes in an attempt to put the company down a better path, but this seems largely folly. AMD hasn’t really been exceptionally successful in the low power area to date either. Their current competitor with the Atom kinda sucks and their current processors aren’t as power efficient as even Intels.

    I really think BD should be scrapped and they should continue work on their Phenoms or at least put it on the back burner till it’s up to speed. I don’t even think it’s the energy efficiency problems either. If it was fast enough people would buy it, but not many people want to buy a product that’s about three years behind the time.

      • Anonymous Coward
      • 8 years ago

      The way you talk, you’d think that Phenom was something other than a huge disappointment when it launched!

        • Bensam123
        • 8 years ago

        I am speaking of the Phenom 2, sorry if I didn’t make that distinction.

          • Anonymous Coward
          • 8 years ago

          Of course, but really what is the difference? They didn’t reinvent Phenom #1, they just got on with things and made it work. Why do you expect that it cannot happen the same with Bulldozer?

            • Bensam123
            • 8 years ago

            Phenom 1 was a lot worse off then Phenom 2. In there current form Phenom chips aren’t all that bad. They aren’t as fast as Intels chips, but they’re holding pretty solid ground price/performance in the middle.

            BD has a lot more issues with it then an erratum bug in it’s initial design. It’s hot and slow. The Phenom has two more cores on it, it’s keeping relatively similar performance, and it’s more energy efficient. BD shouldn’t have been released in its current form. Now they actually have to produce the chip, which is pretty bad (alot of people wont buy it) in addition to improving it.

    • ronch
    • 8 years ago

    Sometimes, one has to wonder whether AMD’s greatest adversary is Intel or… itself.

      • Anonymous Coward
      • 8 years ago

      Are you trying to be clever there? Of course their biggest problem is Intel. AMD has no problem making competent products, Intel just makes better ones.

        • kamikaziechameleon
        • 8 years ago

        Well who removed their only competitive offerings from the market? AMD did. The APU has little desktop relevance with the current benchmarks/prices. Their old athalon x4 processors were great bargains at those prices but they are gone now replaced with something that is not as good and costs more money. Intel didn’t do that.

          • Anonymous Coward
          • 8 years ago

          So your best example of AMD being its own worst enemy is that they took a few models of Athlon X4 off the market to make room for a 32nm K10 derivative with an integrated GPU?

          • Yeats
          • 8 years ago

          “Athlon”, not “athalon”. The Athlon X4’s were nice, but fading and barely competitive with Core i3, and not power-efficient. Llano is just as good, if a bit less OC-friendly, and you can get FM1 Athlon X4 631.

        • ronch
        • 8 years ago

        I’m not trying to be clever at all, but if you look at AMD’s recent past carefully you’ll see what I mean.

          • clone
          • 8 years ago

          to understand what AMD is doing and why you don’t look at the recent past, you look back 10 years and then go forward, they’ve made mistakes but they were to be expected, AMD’s biggest problem is that Intel stopped trying to leverage their market dominance and instead started focusing on competition again when they replaced Pentium 4 with Core.

          AMD released a better product in 1999 than Intel, it’s superiority lasted until 2004 at which point Intel launched Core and AMD understandably hasn’t been able to keep up ever since.

          AMD never kept up prior to 1999, they had a brief window where they tried to become Intel, it was a worthy gamble but in the end they didn’t make it and have been struggling not to become the company they were in 1998 before their success.

          AMD is AMD’s biggest problem only if you are blaming AMD for being smaller than Intel.

          • Anonymous Coward
          • 8 years ago

          You must assume that I am not familiar with AMD’s recent past.

      • sschaem
      • 8 years ago

      AMD with a great CEO would be a MIND BLOWING company.

      But I dont think R.Read is that person… Listening to him, he his more about execution.

      Short term (~2 years) this is a perfect fit, as we seen already from the clean house, great move.
      But longer term, I’m not sure he can anticipate trends.

    • NeelyCam
    • 8 years ago

    [url<]http://pastebin.com/FGC10TSB[/url<] [quote<]"A lower cost base allows us to be more competitive today and to invest back into the business to fuel our “attack” strategies in— low power, emerging markets and the cloud."[/quote<] Why does that sound like they are giving up on anything higher-end altogether? I wonder if they'll rip apart the whole roadmap and kill Trinity etc. to focus on Deccan? Drop GloFo and switch to TSMC exclusively?

      • OneArmedScissor
      • 8 years ago

      Yes NeelyCam, they’re going to just cancel everything they already finished.

        • SPOOFE
        • 8 years ago

        Hey, if HP can do it, why not AMD? 😀

          • OneArmedScissor
          • 8 years ago

          I considered that, but as insane as they are, to their credit, HP [i<]tried[/i<] to sell the Touchpad first. They have a history of knee jerk overreactions, but AMD has a history of riding things out and just eating the loss. In 2097, you'll see, it will all have been worth it!

            • NeelyCam
            • 8 years ago

            In 2097 I’ll be in Mars.

            • destroy.all.monsters
            • 8 years ago

            I rather expect to be in an urn. I just hope it’s a tasteful urn that looks good in my so far non-existent grandchildren’s living room.

            Next to a shrubbery.

            • NeelyCam
            • 8 years ago

            I trust medical technology to stop/reverse aging by then. Of course, this will suck for trying to keep population growth in check, so that’s why I’m aiming at Mars.

            • BobbinThreadbare
            • 8 years ago

            You’re going to be inside Mars?

            • NeelyCam
            • 8 years ago

            Yes; it would be too cold on the surface.

            (Translation: English is hard.)

            • destroy.all.monsters
            • 8 years ago

            I’m impressed he’s going to be getting his penetration on some 80 years from now. No viagra needed for NeelyCam!

        • NeelyCam
        • 8 years ago

        BD is finished in more ways than one. Trinity/Deccan aren’t finished.

      • Bensam123
      • 8 years ago

      It sounds exactly like that.

      • Anonymous Coward
      • 8 years ago

      I doubt Trinity is too high end to manage. If they give up on that, they might as well just close the company.

      • tejas84
      • 8 years ago

      accept amd fanboys.. NeelyCam is ALWAYS right!

        • NeelyCam
        • 8 years ago

        +1 !!!

    • Arclight
    • 8 years ago

    [quote<]From what we can gather, virtually the entire team has been gutted, with only one high-ranking manager from the graphics group confirmed to be remaining.[...]there is a major shift to low-power and mobile-oriented devices, the relatively successful desktop graphics business could be another casualty—although that's just speculation at this point.[/quote<] This does not compute. The graphics division has been, to the best of my knowledge, very successful ever since the HD 4000 series...I mean i'd get it if they want to give up the high end and mid end market for the CPU where they haven't been able to do much in many years but on the graphics side of things...it just seems stupid.

      • OneArmedScissor
      • 8 years ago

      That says they only left someone from graphics, not that they threw out just the graphics people. I’m not sure what you’re getting at.

      The reality of what you’re describing is the other way around. While the high end desktop market has vanished, the true high end CPU market, servers, will carry on. AMD can dabble there until the end of time, or the company, whichever comes first.

      The discrete graphics market, however, is doomed. There is quite literally nothing they can do to save it, no matter how good of a job they do. Discrete cards will be relegated to HPC now, which puts them in the same position as server CPUs, but they’re not really going to be “graphics” cards anymore.

      However, mainstream GPU design will carry on, as it is increasingly becoming the most important part of new PC CPUs, and someday that will shift to servers, as well.

        • kc77
        • 8 years ago

        [quote<]The discrete graphics market, however, is doomed. [/quote<] The discrete graphics market isn't going anywhere. It's the low end graphics market that's doomed. But discrete as a whole.....LOL Nope. What will games be developed on SB's IGP? The render farms that make all of the AAA titles are done on discrete. All engineering work on the entire planet is done on discrete. Intel's OpenGL support isn't nearly good enough here. Unless you like waiting a minute or two to redraw Autocad drawings when you zoom out. Graphic design is done on discrete. Then of course you've got instrument and control systems. You aren't going to see 3 monitor and 4 monitor stations utilising IGP's. This is where Matrox still survives. Granted as processes get smaller, more powerful GPU's will find themselves in more SoC's but it's not like the aforementioned systems are standing still. They require more power too. Therefore there will always be some need for discrete GPU's. Composited desktops suck up a lot of the gains. That's before you even begin to talk about GPU Compute which SB doesn't even have. Much less the shader power required for any effect greater than noise reduction on video sources. They too will grow over time. Then of course you've got the R&D expenditures needed in order to make consoles. Consoles wouldn't even be possible at the performance/per unit level we see them without the R&D that gets subsidised through the production of discrete models. The Dreamcast nearly killed Imagination and Sega because of the R&D needed to design a custom chip that wasn't designed before. You need the production of discrete to subsidise these costs unless the manufacturer can make it up in volume and lots of it. Think about the cost of Xbox or PS3 if they utilised completely custom designed chips. This is what nearly sunk the PS3....all of the custom designed processing units that really don't serve a purpose other than to grow some engineer's e-penis. It made development more complex and the cost of the PS3 debut in the 500 -600 dollar range..... at a loss. There's a reason Imagination dropped back to Tablets, and Phones and struggles to make Direct X 11 or fully OpenGL compliant chips. They make a profit but most of it is made on old tech and it's patent portfolio.

      • Silus
      • 8 years ago

      Successful ?
      Check the earnings reports from the last couple of years and you’ll see profit for the graphics division in just a couple of quarters. Not to mention that when profit exists, it’s quite tiny.

      You fail at splitting two distinct things. Your perception as a consumer of a product’s success and (what really matters for a money making company) the profits of the division responsible for said product.

      Buying ATI was AMD’s biggest mistake. Those that say that the graphics division at AMD is what keeps them afloat have no idea of what they’re talking about.

        • OneArmedScissor
        • 8 years ago

        It’s so funny to me how everyone and their dog acts like they just [i<]know[/i<] that AMD's CPU profits are in the toilet because their chips are "too big." They're made out of sand, most of them are the same mass produced design, and they're really not that big. Graphics cards sell in the same price brackets, but use numerous different, equally large or [b<]larger[/b<] chips, plus a PCB, RAM chips, much beefier heatsinks, and come in a ginormous, fancy box full of manuals, software discs, and cables - and all of this has to be farmed out to several third parties instead of being done in house. Now, which of those is more likely to turn a profit?

          • Silus
          • 8 years ago

          What’s really funny is that it was the AMD fanboy camp that started the “die size is the most important factor to consider” for profitability sake, so they now have to eat their BS and shut the hell up.

            • Arclight
            • 8 years ago

            Guys feel free to link me articles showing the graphics division of AMD hemoraging money and not making a profit. I included in my original post the disclaimer “to the best of my knowledge” so feel free to show me the “truth”.

            LE: BTW i got a GTX 560 Ti and my previous card was also an nvidia so the view expressed by me was from “the other camp”.

            • Silus
            • 8 years ago

            Just use TR search for AMD’s earnings or go to AMD’s own reports.

            • Peldor
            • 8 years ago

            On an annual basis the graphics division hasn’t lost money since 2007 (with 2010 being a great year at +$135M), though they did have those abominable ‘on paper’ writedowns from the ATI purchase, and this year is weak too.

            Their net margins were 0.8%(2008), 3% (2009), 9% (2010). This year is back down to 2% so far.

            As far as operating income since the acquisition, they are about at breakeven or slightly ahead after 6 years. Combined with the cost of the acquisition I maintain that it’s a near complete failure.

      • Peldor
      • 8 years ago

      This article says nearly the entire [b<]PR team[/b<] was canned except for one graphics guy. Not that the entire graphics team was canned (though I'm not ruling that out).

      • sschaem
      • 8 years ago

      AMD spend as much as it gets with its GPU division.
      In Q3 AMD sold 403 million worth of discreet GPU related product, 12 million was profit. (That include its xbox, wii royalty)
      (Previous quarter was a 7 million loss)

      nVidia sold 843 million and made 83 million in profit.

      AMD GPU division budget is astronomical compared to nVidia.

      If AMD wiped out completely its discreet GPU business in 2010 it would have only lost 6 million in profit this year.

      So I can see AMD getting out of the discreet gaming GPU at this time, its not profitable, unless they can cut cost to nVidia levels.

        • OneArmedScissor
        • 8 years ago

        [quote<]That include its xbox, wii royalty[/quote<] That's a very important point. All they have to do is pocket the money from the gajillions of those which are sold, but they're still lucky to break even after all is said and done. High end graphics cards are becoming a very dangerous gamble. While the profits on the manufacturing side are undoubtedly higher than the $50-ish cards, which have similar fixed costs, and are being eliminated by integrated graphics, they're still expensive to make, rapidly decline in price, and have huge development costs due to the complexity of creating new chips that are twice as complicated as the last crazy thing. The cost could be offset with higher prices in the past, as we saw with things like the various 8800GTXs, but the higher the prices go, the more the market shrinks, and by the GTX 200s, that strategy had already come back to bite Nvidia. If all you can sell is expensive cards, you're just not going to sell enough of them, and that's the position AMD and Nvidia are both going to be facing in the next year or so.

    • LaChupacabra
    • 8 years ago

    Step 1) sell all substantial company assets (GloFo sale netted AMD a loss of over 5 billion in assets)

    Step 2) start projects only to kill them before they’re halfway done, frustrating employees

    Step 3) remove any and all remaining talent of a once brilliant company

    Step 4) take remaining money out of shareholders pockets and distribute among the brass, never work again.

    I’m not saying this is specifically what is happening. And I don’t work for AMD or have any insider information. But I refuse to believe that anyone running a major company like AMD is really as stupid as their decisions are making them seem. AMD has turned into a hollow shell of perceived value. Hopefully there is some plan besides robbing investors, it would be a shame to see AMD turn into Rambus.

      • NeelyCam
      • 8 years ago

      Maybe Rambus will buy AMD with some loot from a lawsuit somewhere?

        • Silus
        • 8 years ago

        ATIC obviously didn’t want them. They had a chance a while back and it seems they weren’t really interested. Oracle may still do it though. NVIDIA may also bid on the CPU division, but of course cannot touch the graphics division. Same goes for Intel, but in reverse.

        For a company that touted their “vision” so much, they really had no vision at all. AMD has been run by quite incompetent people.

          • NeelyCam
          • 8 years ago

          So… Intel buys the graphics division, and NVidia buys the CPU division?

            • Silus
            • 8 years ago

            If Oracle is a contender, I doubt NVIDIA will have a chance of getting the CPU division. Oracle has far more resources. However, Oracle getting the CPU division might mean bad news for the desktop CPU market. So if we ever come to that, I really hope Oracle isn’t on the buyers list.

            As for Intel, they would definitely be interested in the graphics division (former ATI) and they certainly have enough cash to buy it (they would also pay far less than what AMD did back in 2006 though).

      • sschaem
      • 8 years ago

      How about buying ATI for 6 billion.. selling ATI profitable SoC team with all IP and engineering for 65 millions to Qualcomm.

      Paying 12 millions for Dirk Meyer failures as a one time partying gift.

      Hiring storage specialist with no experience in marking to run the entire company marketing strategy.

      Pay >2 million for executives no matter how the company is doing…

      etc…

      R. Read and Papermaster seem to be on the right track. This company now has a chance.

        • Silus
        • 8 years ago

        Actually only the first (buying ATI) was a big mistake. The others only happened because of that.

          • Krogoth
          • 8 years ago

          Nope, AMD need graphical IP from some place.

          The only problem is that AMD took too long to properly execute the mainstream SoAC concept (It is the next big thing). Intel just beat them to the punch with Sandy Bridge.

            • BestJinjo
            • 8 years ago

            False. Intel didn’t need to buy a dedicated GPU company to develop its graphics. It costs a lot less to hire the most talented engineers in the world to come up with graphics good enough for an APU design than $5.4B.

            AMD’s long-term future plan was to have some kind of graphics to include in their CPU designs, not necessarily to have the best graphics that costs billions to develop to compete with NV.

            Now, despite Fusion CPUs having superior graphics capabilities over HD3000, it still isn’t enough to make AMD highly profitable. Clearly, the $5.4B acquisition was not a great move since it put the company under massive debts, forcing it to pay high interest on that debt, which in turn reduced their ability to properly invest in the best engineers and large enough R&D to actually create a competitive CPU architecture to compete with Intel in the desktop/mobile space.

            • Krogoth
            • 8 years ago

            Intel already had IP and resources for GPUs, so it had no need to acquire another source.

            This wasn’t the case with AMD back in pre-ATI era. They needed to get their hands on graphical IP, otherwise they would be venturing into an IP minefield (Nvidia, ATI, Creative, S3) if they tried to develop their GPU solutions in-house.

            Nvidia was too big, S3 was lackluster. The only strategical choice was ATI. It is too bad that it took long for AMD to execute their plan and they lost on their gamble.

            • clone
            • 8 years ago

            exactly.

          • ludi
          • 8 years ago

          You would be on firmer ground to state that AMD overpaid for ATi. As Krogoth notes, they needed mainstream graphics IP to remain competitive in the 5-10year time frame and ATi was the only viable candidate.

      • liquidsquid
      • 8 years ago

      AKA Kodak.

      Pretty much the same path.

        • clone
        • 8 years ago

        not even remotely, Kodak resisted the transition to digital and they failed because of it none of that has anything to do with the decisions AMD made.

          • willmore
          • 8 years ago

          No, they failed to apply digital imaging technology they had developed outside of the professional/scientific markets. They though film would be good enough for consumers and they had good reason to think so. But, only hindsite is perfect.

            • clone
            • 8 years ago

            congrats that’s exactly what I said and I don’t care why they thought keeping digital out of the hands of the mainstream was a good idea because it was beyond idiotic the moment they convinced themselves it wasn’t.

            to even comment that they had any good reasons to resist digital aside from propping their film sales is a joke, when Digital cameras were costing $1200 – $1000 for 0.8mp – 1.1mp everyone I knew lost all interest in film, go to a birthday party and their were 2 camps, the ones who had digital cameras and the ones who were getting digital cameras.

            Kodak killed itself for obvious reasons, I’m an idiot and even I knew Kodak was going to be on life support a year after the first camera came on the market if they didn’t jump into digital hardcore using film sales to finance the venture. while they collapsed I remember talking to a wedding photographer at length about it and we traded back and forth, she talking quality and how digital was “cheating” while I finished with “the quality will rise, it always does with tech and those who try to stick with film will be amongst the last of the horse whip salesman, no if’s &’s or but’s about it.”

    • sschaem
    • 8 years ago

    “introducing a grossly underperforming product—the Bulldozer-based FX processors—to the public”

    And they have done a HORRIBLE job at it. Its almost as the PR team wanted to destroy AMD.
    That 8ghz guiness record, that “Ready and able” making fun of Intel SandyBridge, terrible, terrible move by the PR team.

    Nigel Dessau, AMD CMO seem to still be on their lame AMD blog page… I can only guess that he will be part of the later layoff schedule ?

    Dessau expertise was at sun with storage equipment and 19 years at IBM as storage specialist, and it shows. AMD need a marketing strategist that understand consumers not just large corporation that buy disk servers.

      • Krogoth
      • 8 years ago

      I still fail to see why people keep harping that Bulldozer has terrible performance. The performance of Bulldozer isn’t what is killing it. Power inefficiency is what killing Bulldozer when you compared it to Sandy Bridge solutions.

      The current chips are really “server” class hardware being shoehorned into desktop line-ups, because AMD has nothing else to do a refresh with. Intel did the same thing back when Netburst was getting pounded by K8s in the form of P4EEs chips.

        • BestJinjo
        • 8 years ago

        You must be dreaming. It takes an FX-8150 overclocked to 4.8ghz just to match a stock 2600k, and still yet, even at 4.8ghz the CPU is going to be completely obsolete against a 2500k in games once more powerful 28nm GPUs arrive:

        [url<]http://www.hardocp.com/article/2011/11/03/amd_fx8150_multigpu_gameplay_performance_review[/url<] If Bulldozer had amazing performance, hardly anyone would make a fuss about its price or power consumption. As it stands, the performance is not even better than 2008 Nehalem architecture. The problem with Bulldozer is primarily that of awful performance tied to a ridiculous price. The power consumption just makes things worse. But look at Fermi architecture. If you have amazing performance, consumers are willing to still pay high price for it in-spite of high power consumption. In Enthusiast market: Price > power consumption Performance > power consumption Core i7 920/930/950/960 and i7-860/870/875/880 had high power consumption in overclocked states too, but still sold like hotcakes.

          • Krogoth
          • 8 years ago

          What are you smoking?

          Bulldozer trades blows with Nehalem-based chips, it has superior performance if the application in question depends on integer performance. It even manages to outpace SB chips a by tiny margin. However, Bulldozers fall a bit short, in applications where FP performance is king. SB chips completely embarrass it.

          The problem is that level of performance is from 2009 and Sandy Bridge already exceed that performance level by a little bit, but with [b<]far lower power consumption [/b<]. ie you don't need an oversized HSF and water-cooler to get exceptional CPU performance. This is what makes Sandy Bridge so much more enticing to PC enthusiasts.

          • clone
          • 8 years ago

          the enthusiast market won’t save AMD….. if all the enthusiasts got together and bought AMD processors AMD wouldn’t last a week longer than if they didn’t.

          the enthusiast market is vapor outside of the enthusiasts.

          their is a PR component that has value in catering to enthusiasts but tbh a better price once it starts shipping in volume will offset that.

          power consumption matters only after price and performance.

          no Core i7’s ever sold like hotcakes because of price, they got a lot of lip service in online forums but never sold in significant volume which is also why their price was so high…. gotta milk the enthusiasts for all they are worth..

        • sschaem
        • 8 years ago

        Yep, bulldozer even with its flaws is a step forward from AMD previous architecture, at low voltage.
        (but then mainly because of the 32nm process)

        The issue you mention come from the fact that its the PR department that his painting this picture of the new FX being in a class of its own for desktop/gammers.
        Have you scanned the AMD FX web page? Painting a car red doen’t make it go faster…

        AMD marketing could have found a better message that actually fit this product capabilities.
        In the long run, Its better to be humble then to lie to your custumers.

        • travbrad
        • 8 years ago

        Yep the power efficiency and the price are really what are killing it, and they are directly linked. Both seem to be the result of an extremely high transistor count. It’s going to be very hard to compete with Intel on price when AMD’s CPU is so much bigger (despite not even having an integrated GPU), especially with Intel moving to 22nm soon.

    • FubbHead
    • 8 years ago

    Maybe they’re good people to work with, but had I been in charge of AMD, I would’ve thrown out that whole department long ago. All these cheesy names and abbrevations they’ve come up with just can’t be taken seriously.

      • sschaem
      • 8 years ago

      The names are the least of this department problems.

      For example for OVER 3 weeks AMd.com front page showcased a video stitching plugin. Where you take a video to make a panoramic image…. COMEON!

      This AMD PR department message for the FX:

      Overclock for a big boost in performance and speed.
      Perform *mega-tasking* and *get pure core performance* with new “Bulldozer” architecture.
      Get an extra burst of raw speed when you need it most with AMD Turbo CORE Technology.
      Push your performance with tuning controls in the easy-to-use AMD OverDrive™ software.
      *Enjoy stable, smooth performance with impressive energy efficiency thanks to a 32nm die.*

      Was this written by a Chinese intern ?

      Edit: this verbatim what AMD still use for their marketing message.

        • tay
        • 8 years ago

        PR was useless and tone-deaf. I think gutting the department was useful. All AMD did was release slides months in advance followed by huge disappointment.

          • Yeats
          • 8 years ago

          I agree.

      • ludi
      • 8 years ago

      Marketing has been AMD’s weak point since the days when most of their publicity came from Jerry Sanders shooting his mouth off in classic YouTube-before-there-was-YouTube fashion. No disrespect to the people who just lost their jobs in a lousy economy, but maybe Read decided that the entire PR culture at AMD needed to be torn up root and branch, and replanted.

      Remains to be seen whether that strategy will succeed, of course.

Pin It on Pinterest

Share This