Ex AMD designer: Bulldozer to disappoint

[terrace215]

http://forums.macrumors.com/showpost...&postcount=612


On paper bulldozer is a lovely chip. Bulldozer was on the drawing board (people were even working on it) even back when I was there. All I can say is that by the time you see silicon for sale, it will be a lot less impressive, both in its own terms and when compared to what Intel will be offering. I don't really want to reveal what I know about Bulldozer from my time at AMD, nor do I want to go into details of what my friends are telling me.


lots more back and forth on that thread follows and precedes that post...

[Manicdan]

his first sentence says he worked there so long ago that BD wasnt even near finished in design, and somehow has a valid opinion of its performance?

[AliG]

his first sentence says he worked there so long ago that BD wasnt even near finished in design, and somehow has a valid opinion of its performance?

not to mention, if he's an EX employee, don't you think he might be just a tad upset at his past employer for getting laid off (or fired)? I think there's a chance he's still bitter about it and is just trying to slander the product

[thebluemeanie1]

If you read through the thread he seems like he is rather bitter about something.

[terrace215]

his first sentence says he worked there so long ago that BD wasnt even near finished in design, and somehow has a valid opinion of its performance?

I suggest taking 5-10 minutes to read the 20 or so posts and back-and-forth on the whole thread.

He claims to keep in close touch with current employees, as well as the ex C_O (guessing CTO, Fred Weber, from other comments), and to have current info as to how things are progressing.

Who knows. But an interesting data point.

Some additional interesting bits:

What did happen is that management decided there SHOULD BE such cross-engineering ,which meant we had to stop hand-crafting our CPU designs and switch to an SoC design style. This results in giving up a lot of performance, chip area, and efficiency. The reason DEC Alphas were always much faster than anything else is they designed each transistor by hand. Intel and AMD had always done so at least for the critical parts of the chip. That changed before I left - they started to rely on synthesis tools, automatic place and route tools, etc. I had been in charge of our design flow in the years before I left, and I had tested these tools by asking the companies who sold them to design blocks (adders, multipliers, etc.) using their tools. I let them take as long as they wanted. They always came back to me with designs that were 20% bigger, and 20% slower than our hand-crafted designs, and which suffered from electromigration and other problems.

That is now how AMD designs chips. I'm sure it will turn out well for them /sarcasm

---------

I still have more than my fair share of sources. I had lunch with a former C_O (fill in the blank) of AMD two weeks ago. I chat, email, and facebook everyday with guys who still work there.

And, yes, P4 sucked. But that's my point. The only time in its entire history that AMD did well is when Intel had a 3 year period of complete fail. That's over now.

-----------


I'm not sure why people don't understand the difference between hopes and predictions. I've never said I don't root for those guys. All I've said is I predict failure. Every year I root for the NY Mets, but I know that chances are they are going to blow it. In this case I look at past performance, take into account things that I personally saw happen, take into account what my friends who are still there tell me, take into account how much money AMD has in the bank, and I make a prediction. It has nothing to do with me wishing them ill will or hoping that things go well for them.

BTW, you ask how AMD could have competed? Well, for one thing, the could have leveraged K8 and the K8 team's success and design techniques instead of wasting years of time on a project that eventually got cancelled using people that had never achieved any success. It took Intel years to come out with Nehalem, and AMD could have been so far ahead by that point that they'd have enough money in the bank that they wouldn't have to accept a low-ball settlement offer in the antitrust suit and they wouldn't have to sell off their fabs.


(One imagines he refers to the original K9 design that got canned)

[Russian]

From what i recall, they started working on bulldozer a long time ago and chucked the entire design and started anew.... so i think it matters when he left AMD...

[terrace215]

From what i recall, they started working on bulldozer a long time ago and chucked the entire design and started anew.... so i think it matters when he left AMD...

One presumes that if he chats with current AMD employees every day, and goes drinking with them, he's talking about the right product.

[metalop1g]

We need to know if he is biaised.. The question is : does he own an i7?

is it legal for a former employee to say this about a design in progress?

[massman]

This appears just one day before the launch of Thuban. Is it just me, or is this a coincidence?

[Manicdan]

i just started reading it
he uses words like sucking alot, and he even says that when amd was winning the cpu battle, its cause they gave up on clock speeds and went to multi core cause they couldnt compete. and he uses logic like how their stock price was 42$ as the only sign of when they were good (this guy sounds like a 15 year old intern with his posts)

fact is AMD is small, they WILL NOT compete in every market, it costs too much. to survive AMD has built a model that is based around being happy with 20%, and leverage the mainstream as much as possible. the only flaw i see with them is that massive L3 cache that i think they should go with something much denser, even if the perf loss is noticeable, cause its eating up way too many chips and profit.

[Caparroz]

One thing is for sure:

If I were an ex-employee of AMD or any tech company I def. wouldn't be so open on an internet forum about my former job, current contacts, etc.

[terrace215]

One thing is for sure:

If I were an ex-employee of AMD or any tech company I def. wouldn't be so open on an internet forum about my former job, current contacts, etc.

Yeah, it does seem risky. He's not saying WHY he thinks what he does, but he is implying that other current employees are essentially giving him insider info.

JF, call your engineering department, stat!

[Sn0wm@n]

exactly... i smell bull**** ....

[terrace215]

exactly... i smell bull**** ....

Bulldozer? That wouldn't be a good sign for power consumption.

[ElSel10]

Could be fake but we will know one way or another when Bulldozer is released.

[Helmore]

It seems to me that he is mainly complaining about management and the mistakes they made.

AFAIK, the current AMD is pretty much managed by what used to be ATI and the ATI and AMD design teams seem to be cooperating quite nicely at the moment. At least that's what I'm led to believe by people like Dave Baumann (IIRC, could have been someone else) and some of AnandTech's articles.

[terrace215]

It seems to me that he is mainly complaining about management and the mistakes they made.

He's complaining about a lot, but the more relevant parts (to BD, specifically) are:

- too much design automation vs custom layout leading to a loss of performance
- good design folks have left
- remaining folks are telling him BD is not going to be all that great in actual silicon, vs what seems a promising paper design

[Sunfire]

"Maier: I went to Sun to work on what we, at the time, called Ultrasparc V, but I quit after three months and went to AMD. At AMD I worked on K6-II, K6-III, and Opteron/Athlon 64. At AMD, I also had a lot of responsibility and we worked on very small teams. We had something like 18 main people working on the Opteron/Athlon 64. I worked there for 8 or 9 years, and eventually became the Manager for Advanced Development, which is where I started doing software programming as my main daily activity."

http://www.iphonejd.com/iphone_jd/20...pp-author.html

[Xoulz]

http://forums.macrumors.com/showpost...&postcount=612


On paper bulldozer is a lovely chip. Bulldozer was on the drawing board (people were even working on it) even back when I was there. All I can say is that by the time you see silicon for sale, it will be a lot less impressive, both in its own terms and when compared to what Intel will be offering. I don't really want to reveal what I know about Bulldozer from my time at AMD, nor do I want to go into details of what my friends are telling me.


lots more back and forth on that thread follows and precedes that post...

lulz... that cmaier guy is a douche bag! He doesn't even fully understand AMD's road map, let alone Intel's. It's so obvious.

He can't even explain cpu+gpu !!

[superrugal]

On paper bulldozer is a lovely chip. Bulldozer was on the drawing board (people were even working on it) even back when I was there. All I can say is that by the time you see silicon for sale, it will be a lot less impressive,

====================

Really?? Sorry I don't think bulldozer is lovely on paper. LOL

[iMacmatician]

not to mention, if he's an EX employee, don't you think he might be just a tad upset at his past employer for getting laid off (or fired)? I think there's a chance he's still bitter about it and is just trying to slander the product

I think on another post he says he left willingly.

It seems to me that he is mainly complaining about management and the mistakes they made.

AFAIK, the current AMD is pretty much managed by what used to be ATI and the ATI and AMD design teams seem to be cooperating quite nicely at the moment. At least that's what I'm led to believe by people like Dave Baumann (IIRC, could have been someone else) and some of AnandTech's articles.

That's the impression I got from reading those articles. Either that or they are moving in that direction.

Really?? I don't think bulldozer is lovely on paper. LOL

I want a printout of the Bulldozer die.

[ajaidev]

"cmaier: On paper bulldozer........................I don't know any of the people currently working on bobcat, but given the price it is likely to sell for, it's not going to make AMD a lot of money (and it will be competing not only against Intel, but against numerous ARM variations). "

Ya just like Atom did noting for Intel.

The thing is people who work for say Intel/AMD are not all fanboys of the company they work for. If you know what is going on and how they do what they do you may have a problem with their methods be it AMD or Intel.

I wanted to stress one thing about bulldozer it is a CMT and intels next chip will be SMT based now the real deciding matter is the size and performance "single bulldozer module vs high-end sandy bridge "6/8c" single core"

[Hornet331]

meh.. i wont give much about that. Internet is haven for hoaxes...

[god_43]

wtf???? oh come on.....seriously this guy cant even hold a discussion in a professional manner...this is not even news??? its a forum post by some random guy on an internet forum??? bs! Terrance you should have known better than to post this here, it should have gone into the wamps (w/e its called) section. i call for this thread to be either closed or moved.

lol after reading the forum more the guy had this to say.... man this guy is just some intel troll nothing else!

Sure, the playing field hasn't always been fair. But at the same time, in the lawsuit, papers from AMD that said even AMD's head of sales thought intel chips were better also came to light.

In any event, it doesn't matter why. The result is the same.

[terrace215]

I wanted to stress one thing about bulldozer it is a CMT and intels next chip will be SMT based now the real deciding matter is the size and performance "single bulldozer module vs high-end sandy bridge "6/8c" single core"

Well, if you want to compare 2 threads vs 2 threads, sure, but don't neglect POWER in there. Performance without relation to power is not very meaningful.