John Fruehe finally does the sensible thing and comes up with an excuse

Here it is:

So here’s the story on IPC from my perspective.

The original design goal was higher IPC. Back before we had taped out the first processors the discussion came up. I am not allowed to say anything in public that hasn’t been vetted past engineering. I specifically asked the engineering team about IPC and they replied that they expected higher IPC and I was cleared on that statement.

In my estimation, I made the IPC stament on XS and I don’t recall making it other places (but I am sure that my comments were reposted.)

This is not a case of me lying, this is a case of me being wrong. And that happens to all of us. There is no way that I would intentionally say something that I did not believe to be true because a.) the interenet is a permanent record and b.) eventually you’ll have to answer to those comments.

In these different forums that I would frequent, I have made
thousands of posts about the product, always trying to help people understand the technology. Everything was done with the intention of being informative. I never made client performance predictions and was very clear on correcting people who took server benchmarks and tried to correlate client performance, because it just doesn’t work like that.

Since the launch I haven’t really been in any forums because I was in Seattle all week with MSFT. Once the product launched, I started receiving lots of hate mail. If I had a dollar for every message that used the word f*** I would have enough money to buy myself a bulldozer server.

Early on there were a lot of people that said I was crazy for trying to talk to the enthusiast community, but I persisted, the occasional idiot is something that everyone has to put up with from time to time. But the volume of hate messages that I received has unfortunately convinced me that I was wrong.

I was doing this on my own time, it was not part of my job. And because of that, I don’t really feel compelled to take any more abuse.

All of the statements that I have made at the time were statements that I believed to be true. I am no more a liar than Paul Otellini when he said IB would ship in Q4. He is not a liar, he is giving the best information he has at the time and things change. If you are going to extend courtesy to others when their statements don’t pan out, you should do that with everyone.

Thanks to exarkun and some of the others that have had a
reasonable response in all of this. Too many people were waiting to attack, it’s almost like they don’t really care, maybe they just like to fight.

Well, what to make of it? He dug the hole for himself, apparently others (I assume inside the company?) even warned him about it. He went on anyway, then fell into the hole that he dug himself.

I think a man in his position should have known better. It’s the responsibility that comes with such a high-profile position at a high-profile company.

Aside from that, I think it’s a bit too easy to just say “Well, engineering told me that, it’s not my fault”. We have no way to verify if engineering REALLY said that (they could have said something slightly different, and Fruehe may just have misinterpreted it, then ran with it). I personally don’t think that engineering would say that, but that’s obvious, since I’ve said all along that I see some big performance issues with how AMD decided to allocate the execution units in their modules. Therefore I would think that the engineers at AMD would know about this as well, and would not bring forth such information. In fact, I even pointed out some AMD slides that said the opposite:

“Throughput advantages for multi-threaded workloads without significant loss on serial single-threaded workload components”.

As I already said about that remark, at the time:

“Well, that pretty much warns us already that serial single-threaded performance (in other words: IPC) may take a small hit… albeit not a ‘significant’ one.”

The logic being the following: it is a marketing slide. There is no reason to mention negative things such as ‘loss on serial single-threaded workload’ as long as you are confident that the resulting product will perform better anyway. My spider-sense told me that they put this in as a warning, to not get people’s expectations up (which JFAMD then completely undermined).

As old as these slides are, they are still correct today. A small, but not significant loss in serial single-threaded workloads is what we see, compared to Phenom II (I equated that to IPC since it appeared that clockspeeds of Phenom and Bulldozer seemed to be about equal. Which is still the case, although Bulldozer does have more turbo potential. Which does not compensate enough of the IPC to beat Phenom II in single-threaded workloads).

However, even if we assume that they DID say so originally (Fruehe made his posts around August 2010), clearly at some point during development those engineers would have realized that they were running into problems and were going to come up short. Why has Fruehe then insisted for so long that IPC will increase? AMD didn’t find out at launch day that IPC was lower than Phenom II. They must have known this months in advance. But even though leaked benchmarks started appearing, indicating the IPC problem, Fruehe continued to insist that the benchmarks were faulty, reinforcing his earlier IPC statements (so at this point he was indeed perpetuating the wrong information, hence lying).

So, apology not accepted. Firstly, I think Fruehe (and AMD in general, also including people such as Richard Huddy, who have occasionally passed on misinformation on forums, or in interviews) just has to learn the hard way that it was not a smart thing to post such information in public forums in the first place. It’s your own fault, you should have known: “what goes on the Internet, stays on the Internet”. Just like Randy Allen, this will haunt you for the rest of your life.

And secondly, I don’t think Fruehe has done enough to prevent the inevitable outcome. Had he retracted his IPC statement and at least warned the people before launch, then I think people may have been a bit more forgiving. Had he done it far enough in advance, the storm may have blown over by the time of the actual launch, and he might have actually somewhat saved Bulldozer’s reputation (and his own). I suppose he only made it worse by actively debunking leaked benchmarks, and even being insulting in the process (implying that people were paid by Intel to post these things… which is ironic enough, posted by someone paid by AMD). That’s the marketing game, John. You should know how to roll with it.

About these ads
This entry was posted in Hardware news and tagged , , , , , , , , , , . Bookmark the permalink.

30 Responses to John Fruehe finally does the sensible thing and comes up with an excuse

  1. NEWIMPROVED JDWII says:

    Hey i knew you would do a article on this.

    I’d say he probably did just say what he over heard.

    But you would have to be blind to notice when he stop saying IPC will go up and starting saying things like theirs lots of single core processors out already. Those kind of smart ass comments he said that to me. He kept down playing IPC and he kept saying things like “Why are people who care about single threading performance wanting to buy newer processors”

    I do believe he does this on his own time and i’m more then sure Amd does not pay him to do it. He said over and over he enjoys being on forums(he must be drunk with the respect he gets)

    As for me i got a Phenom II x6 1100T and i’m going to overclock it to 4.2Ghz. I’m either going to wait until PD comes out or i’m going with Intel for my next CPU upgrade. Luckily i don’t do much important stuff on my PC.

    • Scali says:

      Well… am I right that you would have gotten a Core i5 2500 or such instead of the Phenom, if you didn’t already have the motherboard?
      And perhaps I am also right to say that you got the motherboard because you were expecting Bulldozer to be better than it was?
      Which might have to do with how positive Fruehe was about Bulldozer?

      In other words, has Fruehe/AMD tricked you into buying a motherboard for Bulldozer, and is that the reason why you now have to go with AMD anyway?

    • Scali says:

      Also, I almost forgot that it was AMD’s own slides that originally led me to doubt Fruehe’s words.
      I’ve added a link to them, and a quote (which was also in the original blog I did in August last year).
      So to me there is no doubt that engineering did NOT tell him IPC would go up. I think what they said is something like: “IPC is going down a bit, but we will compensate that with our clockspeed and turbo, so single-threaded performance wil go up”.
      Fruehe then messed that up, saying that IPC itself goes up… and the final nail in his coffin was that clockspeed/turbo did not compensate enough to even get single-threaded performance above Phenom II-level.

      But he knew, he knew all along. He’s just the most incompetent senior director of marketing I’ve ever seen.

  2. Bonzai says:

    I like it that some people are putting the BIOS to be part of the blame currently on the processor. And Piledriver is probably going to be just as bad if they don’t stop and make any major changes (Which, at this point is probably too late.) Overall, It’ll be a cheap chip to OC in the long run but if people want the overall performer for current and future apps, then it will probably be Intel. This just sucks though. I really wanted it to be worth a damn. Guess we’ll see how things go from here.

    Oh yeah, that guy needs to get out. PR spins can suck it. He down right fucked up. Alot of people in denial with AMD atm. Their GPU line is the one thing going strong and even that has some issues. The 7000 HD series maybe coming out sooner for a recovery on AMD’s part.

    • Scali says:

      Yes, it’s not entirely clear yet where the biggest problem is in Bulldozer.
      I’ve seen people testing with various code, some think something is wrong with the L1/L2 cache, others think the instruction decoders are bottlenecking the cores (which is something I warned about… saying that they went from 3 decoders per core to effectively 2 per core… so if they couldn’t make the decoders faster, this may be a problem. Aside from that, the sharing logic may be broken as well. They do have 4 decoders per module, so running a single thread should mean they have more decoders. But even single-threaded benchmarks such as Cinebench are slow).
      Either way, they sound like rather serious problems, so it might indeed take major changes to fix them, which could take more than a year (by which time Intel has new 6-core CPUs and 22 nm products on the market, so the goalposts will have moved again. Just fixing Bulldozer to be competitive with Sandy Bridge is not enough).

  3. Michael says:

    Good writeup. Keep up the posts.

    I don’t buy for a minute that people in the position that Mr. John are clueless about how desktop is going to perform on a core design that is shared between server and desktop. That’s very disingenuous and insulting. The other thing is making posts like the “Bulldozer Pre-launch FAQ” on overclockers.net. Let’s face it: why would anyone, especially someone in AMD server marketing go to such extremes to put something like that together and then habe the “While I work for AMD, my posts are my own opinions.” disclaimed in his signature. Complete and utter pre launch damage control if you ask me.

    You can pool the wool over a few unsuspecting but not over intelligent people that have watched the circus of technology and strategy blunders at AMD.

  4. NEWIMPOROVED JDWII says:

    “Well… am I right that you would have gotten a Core i5 2500 or such instead of the Phenom, if you didn’t already have the motherboard?
    And perhaps I am also right to say that you got the motherboard because you were expecting Bulldozer to be better than it was?
    Which might have to do with how positive Fruehe was about Bulldozer?

    In other words, has Fruehe/AMD tricked you into buying a motherboard for Bulldozer, and is that the reason why you now have to go with AMD anyway?”

    Yes you are correct if i could start fresh their is know way i would get the Phenom II x6 i would be getting the 2500K and a sabertooth board from Intel(i love the sabertooth). I was expecting bulldozer to be better then this i was not expecting a “Intel Killer” but i was expecting a processor with the single-threading performance of the first gen I7(it can use all the shared resources for just one core after all) with the multithreaded performance around a I72600K(it has 8 cores) maybe just a little better.
    As for JF-AMD
    I really thank Fruehe would be a retard if he said his product sucked but i must admit i noticed a trend at the last 3 months when it comes to JF-AMD and he kept down playing IPC. Which was making me a little nervous about my decision. Its funny how good Amd can make a product look before launch with their screen shots of the design that looks real good on paper. As for JF-AMD he is not the reason i got this board i got it because i wanted a better processor and the fanboy in me at the time told me to. Sounds sad but true i just got caught up with all the hype it wont happen again. I’ll tell you why most people like me got so hyped up about this CPU. Here are my reasons for why i did.

    Chew
    FX name
    Marketing slides explaining the design
    JF-AMD was also part of it. I figured “why would some one put their self out their if their lying”.
    And finely the Tech community

    Please note that i’m new to this stuff and i never even knew anything until i built my PC in 2009 so i never went through the older launches such as the Phenom launch or the Athlon FX launch i just read all the articles about CPU designs and benchmarks from even the 90’s to now. I was to young and i was playing on the gamecube then. I’m so in to Tech now a days i’m 19 and going to college to be a Network administrator(all ready have my NET+ cert). So that’s about me you can be sure i’ll open my ears next time and Liston to some other people without a hidden agenda(as JF-AMD would say the one man with a Agenda)

    • Sune Marcher says:

      When it comes to hardware:

      1) don’t buy into hardware before it exists and been independently tested and benchmarked.
      2) don’t be a fanboy – being pragmatic gives you the best bang for the buck. (I’ve been through both Intel and AMD CPUs, currently on Intel… and it seems my next will be Intel as well. Too bad, we could use a little more competition.)
      3) related to #2, don’t base your identity on hardware, there’s so much more interesting & important things in life.

      Three simple rules to follow that’ll make you a happier person less likely to waste money & regret it :)

      • Scali says:

        Speaking of competition… I’ve always said that buying AMD just to ‘support competition’ may not be the smartest thing to do. Perhaps by now it is becoming clear what I mean by that…
        People who have bought AMD processors since the release of Core2, have basically been rewarding AMD for not being competitive. AMD may still be around, but they release one underperforming, buggy architecture after another. Has it been worth rewarding AMD for their failing Phenoms? They survived long enough to give us… Bulldozer.
        Shall we continue to support the underdog, and reward them for the Bulldozer achievement as well? Who knows what ‘competitive’ CPU they will come up with next.

        It seems that what I said was true: Intel is bleeding AMD dry in a pricewar. AMD can survive, but does not get enough revenue to continue developing competitive products (or competitive fabs for that matter… spinning off Global Foundries may obscure the issue, but it certainly has not solved it).

  5. NEWIMPOROVED JDWII says:

    “(So to me there is no doubt that engineering did NOT tell him IPC would go up. I think what they said is something like: “IPC is going down a bit, but we will compensate that with our clockspeed and turbo, so single-threaded performance wil go up”.)”

    Actually they told Toms hardware this “AMD’s architects say it was their goal to “hold the line” on IPC and create hardware that’d scale to much higher frequencies.”

    So i’m trying to come up with a reason why they would tell JF-AMD something different.

    “Fruehe then messed that up, saying that IPC itself goes up… and the final nail in his coffin was that clockspeed/turbo did not compensate enough to even get single-threaded performance above Phenom II-level.”

    Back in Augest i was talking about IPC and i was saying this should be Amd’s main focus on improving the design. This is what JF-AMD said

    “if you care about single threaded performance, there are plenty of single cores to be bought cheap.

    If you only care about single threaded performance my advice is don’t buy an 8-core processor, they are targeted at people who want to run a bunch of apps at once or want to run multi-threaded apps.

    Why are so many people who only care about single threaded performance in the market for new systems?”

    “I can’t tell you how wrong you are about IPC. IPC is not the golden ring that everyone makes it out to be. IPC, multiplied by clock frequency tells you something, but a singular focus on IPC means nothing.

    What if AMD had 50% more IPC than intel? But clock speed was half? Where does that put you?

    What if we have 4X the IPC an a 500MHz clock speed?

    Obsession on IPC is like thinking you are getting great gas mileage on you car because you can go a long way in between fillups. Then one day you wake up and realize that you have a 25 gallon tank.”

    “Let me explain IPC once and for all.

    If you think IPC is all that matters, consider this:

    I weight 195 pounds.

    Now, there you have a single metric like IPC.

    Am I fat? Am I skinny?

    It is IMPOSSIBLE to tell.

    Because you do not know my height.

    IPC, without clock speed, is meaningless.

    FYI, I am 6′ tall, and am considered skiiny by most, but mr. wii insists that I am overweight. Let’s face it folks, the IPC obsession is meaningless because that metric does not translate in a vacuum.”

    Their and after this point i stopped and just stopped caring about what JF-AMD says he is a smart ass. He was wrong the whole time and he was acting like such a dick when i or some one brought IPC up he was first to say something. Please Note this was around the same time i came back here and said i was sorry because i was starting to find out who was right all along. JF-AMD talks about being picked on in High school then he treats fans like this on his own time.

    • Scali says:

      Well yes. What annoys me most is that firstly, AMD fanboys were all over me, while I was just trying to inject some realism into what JFAMD said (boy don’t they look stupid now)…
      I mean, I can look back at anything I’ve posted on Bulldozer over the course of more than a year, and not be embarrassed about anything I said (unlike JFAMD and many fanboys). Take this thread for example, that’s the main thread between JFAMD and myself: http://forums.anandtech.com/showthread.php?t=2100222
      You may have been fooled at first, but at least you came back to my blog even BEFORE Bulldozer was launched, and said that you started to see the truth in what I was saying.

      And secondly, it’s not the first time AMD does this. Take this for example: http://www.zdnet.com/blog/ou/amd-posts-blatantly-deceptive-benchmarks-on-barcelona/567
      Or the famous “40% faster!” video from Randy Allen: http://www.youtube.com/watch?v=G_n3wvsfq4Y
      So AMD did *exactly* the same with Barcelona a few years ago (and back then I said the same thing: Barcelona is not going to be THAT good).

      And as you know, I’ve also said plenty of things about AMD’s GPU division… who do the same thing as well… Like with their tessellation bottleneck. They just blame it on nVidia paying off developers, and claim that it’s because of small triangles. Where firstly you don’t SEE those small triangles in wireframe mode, and secondly, Beyond3D showed that nVidia’s hardware is actually *worse* than AMD’s hardware when rendering small triangles, so that can’t be the problem: http://scalibq.wordpress.com/2010/11/02/proof-of-what-i-said-in-my-previous-blog-from-an-unexpected-place/

      And there’s plenty of other stuff, such as AMD talking about OpenCL and physics acceleration for years, as if they are valid options, while there are no application out there that use OpenCL, and no games that support GPU physics acceleration on AMD hardware.

      That’s what makes it even worse. It’s not the first time that AMD marketing people have lied. They do it all the time.
      That doesn’t mean I’m anti-AMD though. I buy their hardware if it is good: http://scalibq.wordpress.com/2009/11/30/got-my-radeon-5770/
      And if other companies say weird things in marketing, I call them out as well: http://scalibq.wordpress.com/2011/05/18/oh-dear-intel/
      Or if they have some hardware problems: http://scalibq.wordpress.com/2011/10/04/intel-lynnfield-memory-woes/
      Which is the sad part… Although I am perfectly neutral myself, AMD just does a lot more shady things than the other companies, so people often don’t even see the other stuff I say, and say I’m anti-AMD. I wish that were true. The reality is that AMD is just by far the most shady company.

      • k1net1cs says:

        Love the 5770 myself, almost to the point of thinking it’s the 8800GT of DX11 cards, but hate AMD driver team for their inconsistencies; most of their fixes don’t get carried to later driver versions.
        They even manage to mangle their OpenGL driver for their 10.11 Preview 2 which focuses on bringing more performance to Rage.
        Most of the time, it does bring more performance, but at the same time it breaks compatibility with almost every other OpenGL apps for owners of the 6000 series.

        JFAMD…Just F*cking Around, Mate. Derp.

      • Scali says:

        I use the Radeon 5770 in an old box now… Tried Rage on it… Regular driver I had on it: crash. Installed Rage Performance driver: game worked, but it is so slow that it is unplayable (< 1 fps). Installed the Preview 2 driver, supposed to have more Rage fixes: same thing… Just not playable period.

        On my GTX460 the game worked right away. Didn't need any beta drivers, didn't even need the Rage update that was released.

        Also tried on another old machine with a Radeon X1900 in it… Doesn't work AT ALL, because for some reason OpenGL is broken in the legacy installer. It just does not install OpenGL.

      • k1net1cs says:

        When I bought that 5770 I was rocking a 8800GT until one of Nvidia’s driver fried my card.
        Y’know, the one that for some reason turned or slowed the fan off.
        I was in need of a replacement, and it was the only DX11 card at that time that wasn’t too expensive for my blood.

        Like I said, it’s not a bad card in itself.
        It’s just that AMD’s driver development could be an example in a dictionary entry of the word ‘random’.

      • Scali says:

        Well, I had a few nVidia’s fry without the help of a driver. Just bumpgate I guess.
        I got the 5770 for the same reason: my 8800GTS died, and nVidia didn’t have DX11 yet: http://scalibq.wordpress.com/2009/11/15/disaster-struck%e2%80%a6/
        But I have since upgraded to a GTX460, because of the constant failing on AMD’s driver side (no OpenCL, poor OpenGL, no physics acceleration etc).

        These ones finally make Rage work okay on my 5770: http://www2.ati.com/drivers/hotfix/catalyst_11.10_preview/amd_catalyst_11.10_preview3_win7_vista.exe
        Only took them 4 drivers to get it right. GTX460 worked fine out-of-the-box, with drivers released in August.

      • k1net1cs says:

        Still thinking of getting a replacement for it; it’s between a 560 and 560Ti.

        I just hate AMD’s OpenGL support specifically and its driver support in general.
        They still can’t even get a per-game custom profile system right (user interface-wise) like Nvidia’s; Presets my arse.

  6. NEWIMPOROVED JDWII says:

    “And secondly, it’s not the first time AMD does this. Take this for example: http://www.zdnet.com/blog/ou/amd-posts-blatantly-deceptive-benchmarks-on-barcelona/567
    Or the famous “40% faster!” video from Randy Allen: http://www.youtube.com/watch?v=G_n3wvsfq4Y
    So AMD did *exactly* the same with Barcelona a few years ago (and back then I said the same thing: Barcelona is not going to be THAT good).”

    Amd should do what Intel does ” I remember Intel as being extremely conservative with its performance estimates. In the end, Conroe arrived with a performance boost that was 80 percent higher than the company initially had promised. It was Intel’s conservative communication that caught AMD completely on the wrong foot. Just three months before the launch of Core 2, AMD told me in interviews just how confident they were that the Athlon X2 would keep or immediately regain the performance crown from Intel, should Intel in fact capture it. I was not sure if AMD already knew how strong Conroe was and their answers were marketing bubbles, or if they actually fell into Intel’s trap.

    My point being is it might be better to underestimate then overestimate that way if your wrong its a good thing. Amd is not just pissing of their fans their pissing off their stock holders and Higher-up consumers such as (servers buyers).

    • Scali says:

      What I recall from Conroe is that Intel previewed the chip a few months in advance, and it showed something like 20% better performance than Athlons in some tests.
      AMD fanboys were all over it, claiming the benchmarks were fake and all that.
      When it was released a few months later, it wasn’t 20% better, it was even more than that.

      Intel has traditionally demonstrated CPUs in advance, they are not as secretive as AMD is. You will often see preview articles on new Intel CPUs a few months before release, giving the reviewers actual hands-on experience with a system.
      For example, Anandtech did a preview on Sandy Bridge in August last year: http://www.anandtech.com/show/3871/the-sandy-bridge-preview-three-wins-in-a-row
      Some 4 months before the actual chips were launched: http://www.anandtech.com/show/4083
      Sandy Bridge-E was already covered by Tomshardware last month: http://www.tomshardware.com/reviews/core-i7-3960x-x79-performance,3026.html

      So in general, we know some 3-4 months before launch what a new Intel CPU is capable of. And then AMD tries to tell us that previews only a few weeks before launch are fake and not comparable with release hardware? I don’t buy that.
      Sure, the previewed Intel systems aren’t always perfect (Sandy Bridge had problems with the SATA controller in the chipset), but even then, the previews are still quite representative. If you can’t get a reasonably working system 3-4 months before a product launch, then you’re not going to fix that in those remaining 3-4 months either.

  7. NEWIMPOROVED JDWII says:

    by the way i wanted to ask you did you see a die shot of bulldozer and then compare it to Sandy. On intel their is no wasted space on AMD it looks like 10-20% is wasted space.

    Here is Intel

    http://techgage.com/viewimg/?img=/reviews/intel/sandy_bridge_launch/sandy_bridge_die_shot.jpg&desc=Intel%20Sandy%20Bridge%20Die%20Shot

    And here is Amd

    http://www.xbitlabs.com/news/cpu/display/20100901181239_AMD_Displays_Die_Shot_of_Upcoming_Eight_Core_Orochi_Processor_for_the_First_Time.html

    • Scali says:

      Well, if you look on the MacRumors.com forum, there’s a guy by the name of cmaier, who used to work as a CPU engineer at AMD.
      He has said (months before release) that AMD moved entirely to automated tools for chip layout.
      That is probably what we see here: the die is laid out by an automated tool, which is generally less efficient than hand-optimized layout. He said their automated tools would generally need 20% more die-space and get 20% less performance than hand-optimized designs.
      This could be an indication that either AMD no longer has enough skilled engineers to hand-optimize their chips, or that AMD just can’t afford to do it anymore (lack of time/other resources), or a combination of both.
      Perhaps AMD’s idea was to just get Bulldozer out of the door in its current state, and that Piledriver is a hand-optimized version of it. Would buy them about 8 more months to further optimize the layout.
      Intel was not under any pressure with Sandy Bridge, so what you see is a pretty neatly optimized die, even though it is a brand new architecture, like Bulldozer.
      Some say that you can tell that Intel was not hand-optimizing as much with some Pentium 4 dies, such as Prescott.

  8. NEWIMPROVEDJDWII says:

    yeah i hear a lot of the engineers left the company and these are the engineers that help make the Athlon!!!! WHY WHY

    I’d say its both Money and engineers.

    Amd needs to learn something a lot of people can make a Design on paper but if they don’t have the resources to do it it does’t really matter know does it? For example i want to make a CPU design on paper but i could never make it come to life with out the fabs. I also say HOW CAN INTEL OR AMD hand-optimize and keep up with demand. Remember when the Athlon came out it was better then Intel’s processors in fact it was so good Amd could not make enough of them. Remember back in June with Llano same issue. Amd needs to fix GF. One of the major things holding amd back is GF. they should use TSMC.

  9. Bonzai says:

    Sooooo how about that Ivy Bridge?! :3

    • Scali says:

      Well, this I suppose: http://www.anandtech.com/show/4830/intels-ivy-bridge-architecture-exposed
      “Ivy Bridge will bring about higher clock speeds thanks to its 22nm process, however the gains will likely be minimal at best. Intel hasn’t been too keen on pursuing clock speed for quite some time now. Clock for clock performance will go up by a small amount over Sandy Bridge (4 – 6%), combine that with slightly higher clock speeds and we may see CPU performance gains of around 10% at the same price point with Ivy Bridge. The bigger news will be around power consumption and graphics performance.”

      The slides included in that article already explain exactly where those performance improvements and power savings should be coming from, so unlike AMD, there are no mysteries, no ‘missing secret sauce’.

      I guess we won’t have to be afraid of some “JFINTEL” on numerous forums telling us that all leaked information is wrong, and all benchmarks are done with pre-production hardware, BIOS, drivers etc.
      Instead, I think we’ll see various sites doing a hands-on article with real Ivy Bridge hardware a few months before launch, so we won’t even need to rely on leaks in the first place.

  10. Chad Boga says:

    Hey JDWII,
    It is good too see you are new and improved. ;-)

    Scali, keep up the excellent work man, I know from bitter experience how if one wants to be realistic about CPU performance, you will get branded an “Intel fanatic” by AMD loving idiots who don’t know their ar5e from their elbow.

    • Scali says:

      Yea, when I went up against JFAMD over a year ago, pointing out that some of his claims didn’t even match up with AMD’s own slides, and many other claims had no technical background whatsoever (you remove 1 ALU and 1 AGU, yet your IPC goes up… why?), people thought I was crazy.
      I did the same back in the Barcelona days… but this time I decided to use the power of the intarwebz to make sure that the information sticks, and is easy to find, in a few simple blog entries with some direct links to relevant forum posts and such.
      I had to wait more than a year, but finally the benchmarks started leaking in, and slowly but surely people started to realize what JFAMD had been doing on the forums all this time.

      The irony of it all is that even *I* was still too optimistic about Bulldozer’s performance, die size and power consumption. I’m not the Intel fanatic that people want to see in me. I try to be as neutral as possible, not overly negative.

      A lot of the forum threads discussing my original “Fruehe liar”-blogs are pure comedy gold now. Remember guys, what goes on the internet, stays on the internet!
      And it proves this old blog all over again: http://scalibq.wordpress.com/2009/10/26/are-all-amd-fans-idiots/

  11. NEWIMPROVEDJDWII says:

    “Hey JDWII,
    It is good too see you are new and improved.

    Scali, keep up the excellent work man, I know from bitter experience how if one wants to be realistic about CPU performance, you will get branded an “Intel fanatic” by AMD loving idiots who don’t know their ar5e from their elbow.”

    Hey Chad are you the person from Toms. If so i have to give you credit for not raging on Baron for what he was saying i thought you where going to come back trolling but it seems that was Baron who did that. He sounded like a little school girl when i told him to play nice with people and he replied saying I’ll say what i want when i want.

    Either way i’m still a Amd fan but one with brains. After all its just a company.

  12. NEWIMPROVEDJDWII says:

    “The slides included in that article already explain exactly where those performance improvements and power savings should be coming from, so unlike AMD, there are no mysteries, no ‘missing secret sauce’.I guess we won’t have to be afraid of some “JFINTEL” on numerous forums telling us that all leaked information is wrong, and all benchmarks are done with pre-production hardware, BIOS, drivers etc.
    Instead, I think we’ll see various sites doing a hands-on article with real Ivy Bridge hardware a few months before launch, so we won’t even need to rely on leaks in the first place.”

    Yeah i agree, You know remember what i said earlier how Intel some times underestimates their performance. Its a Win Win kind of idea. Hey that’s what i would do. AMD’s marketing go’s to far.

    My opinion on JF-AMD is he is a marketing guy he knows practically nothing when it comes to engineering. All he says is what they tell him, And maybe a little more :)

    • Scali says:

      I don’t think he knows a lot about marketing either :)
      At the very least he’s damaged his own reputation, and I don’t think his strategy really worked out for AMD either. Nearly all reviews of Bulldozer had quite a negative conclusion, mostly because they were comparing the performance to JFAMD’s predictions.
      If JFAMD had underplayed Bulldozer, the reception would have been much more positive (same psychological effect as with Intel: the CPU is better than expected).

  13. Jigar says:

    I think it happens to very best of us, and hey i still believe that you guys can pull it off, Bulldozer didn’t fail in all department, it does shows the sign of being faster. I think AMD engineers can still do it. :)

  14. T. says:

    It seems that your friend JF-AMD didn’t start lying with BD:

    http://semiaccurate.com/forums/showpost.php?p=7949&postcount=86

    “Our TDP is equivalent to their max power, it is the maximum power the processor could draw, every transistor firing in a worst case scenario.”

    AMD documents say a different thing:

    http://support.amd.com/us/Processor_TechDocs/43374.pdf

    The processor thermal solution should be designed to accommodate thermal design power (TDP) at Tcase,maxTDP is not the maximum power of the processor.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s