S/A: "AMD outs bulldozer based orochi die"

[Wreckage]

Intel stopped worrying about Bulldozer on March 6th, 2009.

http://www.tomshardware.com/news/AMD-Bulldozer-Orochi,7201.html

 

[khon]

Does anyone have any guesses or theories on how large the die could be?

Would 150mm2 be out of the question?

I would think so. Intel can't even make a quadcore that size, so I very much doubt AMD can make an octocore that small.

It's probably closer in size to Gulftown, which comes in at around 250mm^2.

 

[ModestGamer]

Truth be told, nobody is laughing. It is as much work trying to deal with the rumors now as it is to just keep telling people they won't see a die shot until launch. I have way too much work to spend my time thinking of ways to drive the other guys crazy.

you guys should be hamming it up a bit. Hell I'd photoshop a mcdonalds in there as a overlay with a sign that says.

All you can eat $4.99 and see if they notice.

You may not be actively doing anything in a delibrate fashion. Its working and people are talking about BD everywhere.

My advice. Put some test data out there or hell even some totally bs benchmarks with scores of 1 or 2 instead of 13,000 etc.

wanna make your competition stumble. Make them crazy trying to figure out what your doing.

 

[ModestGamer]

t
If I were intel I would look back at AMD . AMD already reversed Intel cpus. Intel groupes are Laughing there asses off at guys like informal and frank . for 5 years these 2 AMDzoners been spreading complete 100% lies . But hay If what came out of the oven for AMD doesn't cut it they will throw something together . I do think AMD will have better graphics than intel to start with but thats all they will have. You guys were talking about combiming threads isn't that what Intel macro opps does . Take 2 to make 1 without using 2 cores. Myself I want AMD to beat Intel this round really I do . The pricies of those AMD cpus will scare ya . The best part Intel won't lower their cpus pricies. Doesn't matter to me I will still buy intel . I spent half the day today playing with SB as Bob got a hold of a couple programms for AVX . Man are these low end 1155 nice processsors.

get ready for the hammer to fall then. Bench testing is comming soon and I am fiarly sure AMD brought a bulldozer to push the grandfatherclock off the cliff.

 

[JFAMD]

My advice. Put some test data out there or hell even some totally bs benchmarks with scores of 1 or 2 instead of 13,000 etc.

Not my style. If you put fake data out there you lose credibility. We've always maintained no CPU benchmarks before launch.

We'll probably do a STREAM benchmark to show off some of the memory controller enhancements, but core level benchmarks happen at launch, not before.

 

[tommo123]

re stream, i'm still waiting for something worthwhile to run on the thing same for openGL tbf.

i'd use donalds tools from doom9, but he appears to have been given some free gifts from nvidia to develop his tools to run on their hardware only.

i hope ATi push dev relations like that.

 

[ModestGamer]

Not my style. If you put fake data out there you lose credibility. We've always maintained no CPU benchmarks before launch.

We'll probably do a STREAM benchmark to show off some of the memory controller enhancements, but core level benchmarks happen at launch, not before.

I was talking about a very delibrate and obvious spoof. Maybe a video or something to with alot of psuedo drama about how your right on your schedule to release it and stuff.

It would be funny.

We are all waiting on the performance data. I am before upgrading to either BD or SB.

Compell me to buy another amd product.

 

[tatertot]

While AMD still says only "2011" publicly, Charlie D @ semiaccurate just said he's learned a lot of BD news from sources over the past 2 weeks, and that "sadly most of it is bad."

This, he says is the reason for his estimate of "late, late 2011" for a Bulldozer launch.

http://www.semiaccurate.com/forums/showpost.php?p=65708&postcount=172

http://www.semiaccurate.com/forums/showpost.php?p=65717&postcount=27

With Charlie always having been a strong AMD/ATI advocate, it should be difficult to spin this as "FUD". I suspect we'll see some more public signs of the bad news when AMD holds Analyst Day on November 9. Presumably, the analysts will want something more than a "2011" timeframe for Bulldozer a month before 2011 starts.

 

[JFAMD]

Before anyone sticks a fork in bulldozer, might I remind everyone that currently none of the schedules have changed in the last 60 days, so this new round of pessimistic rumors are pretty much that, rumors.

I am going to be in vegas twice before the end of the year. Maybe I can find a sports book where I can place a bet on bulldozer. Based on the current rumors, I could make a killing off of the under.

 

[cbn]

Deleted.

 

[bryanW1995]

I doubt that they are revese engineering CPUs.

Someone said that Intel probably doesn't care what AMD is doing, but I believe there is enough evidence that they spend a disproportionate amount of time competing with AMD.

I don't think that they spend a disproportionate amount of time competing with amd. They should spend a LOT of time checking you guys out, and vice versa. You are their only real cpu competition. Intel has branched out over the past few years with larrabee and then with their mcafee purchase, but at their heart they are still a cpu manufacturer/seller. If intel puts out another p4-type fiasco and gives you guys the outright lead again for 4 years (or even 1 year) their stock would suffer immensely. maybe not at first, but their asp's when not in a monopoly position would tend to decline even if they were able to continue their monopolistic behavior of the past.

Before anyone sticks a fork in bulldozer, might I remind everyone that currently none of the schedules have changed in the last 60 days, so this new round of pessimistic rumors are pretty much that, rumors.

I am going to be in vegas twice before the end of the year. Maybe I can find a sports book where I can place a bet on bulldozer. Based on the current rumors, I could make a killing off of the under.

I'm very much in the "want BD to rock" category, but it's hard not to be slightly pessimistic when anand and others have taken a "wait and see" approach. obviously they might still be pessimistic after the barcelona launch fiasco and the numerous GF-related delays, but it has been a pretty bleak landscape for the past few years for amd fans. heck, I even told somebody to upgrade today from a q6600 to an i7 750 without even considering the amd equivalent cpu. could amd produce a similar upgrade in performance for similar or less money? sure, but amd just hasn't been on the radar for a while now. Crack that whip on your engineers (or, more likely the GF engineers) and give us an excuse to buy amd again!

 

[ModestGamer]

I don't think that they spend a disproportionate amount of time competing with amd. They should spend a LOT of time checking you guys out, and vice versa. You are their only real cpu competition. Intel has branched out over the past few years with larrabee and then with their mcafee purchase, but at their heart they are still a cpu manufacturer/seller. If intel puts out another p4-type fiasco and gives you guys the outright lead again for 4 years (or even 1 year) their stock would suffer immensely. maybe not at first, but their asp's when not in a monopoly position would tend to decline even if they were able to continue their monopolistic behavior of the past.



I'm very much in the "want BD to rock" category, but it's hard not to be slightly pessimistic when anand and others have taken a "wait and see" approach. obviously they might still be pessimistic after the barcelona launch fiasco and the numerous GF-related delays, but it has been a pretty bleak landscape for the past few years for amd fans. heck, I even told somebody to upgrade today from a q6600 to an i7 750 without even considering the amd equivalent cpu. could amd produce a similar upgrade in performance for similar or less money? sure, but amd just hasn't been on the radar for a while now. Crack that whip on your engineers (or, more likely the GF engineers) and give us an excuse to buy amd again!

I'd like to point out that even though Intel has the top spots in regards to CPU power if you look at the performance matchsup AMD wins against almost everything but i980 stuff. Not to mention the lack of performance of multicore stuff is due to poor software authoring.

 

[extra]

I'd like to point out that even though Intel has the top spots in regards to CPU power if you look at the performance matchsup AMD wins against almost everything but i980 stuff. Not to mention the lack of performance of multicore stuff is due to poor software authoring.

Ehhh I don't think so really. I'd say in most things Intel wins, but AMD isn't nearly as bad as people sometimes say. In most cases, there is no perceptible difference in day to day tasks. My x6 did raw conversion in Canon DPP faster than my i7 (at least it felt like it) but my i7 has higher minimum frame rates in EQ2. W/e.

I think bulldozer will compete well with sandy bridge, but right now I don't see that as being AMD's problem with BD.

I think AMD's problem is friggin *release it already*. If it's late late 2011 it may end up coming up against Ivy Bridge instead. And that...is going to be a whole new can of worms. They need BD out sooner rather than later.

 

[JFAMD]

obviously they might still be pessimistic after the barcelona launch fiasco

Since Barcelona:

Shanghai - 3 months early, higher than expected clock speed
Istanbul - 5 months early, higher than expected clock speed
Magny Cours - 3 months early, higher than expected clock speed

I would say that we have executed.

 

[Sp12]

JF, just a quick architectural question here.

If I have an app that can only use 4 threads, assuming proper software support, can I have it cannibalize the shared parts of each module? What sort of performance benefit would I see from stealing all the of the shared components instead of splitting them with another integer in a quad-threaded environment?

Additionally, what are the chances of the server and/or desktop market moving towards EFI? I know intel has come out and confirmed they'll be moving to EFI for sandy bridge at least on the desktop. I'm really looking forward to 3TB drives . What sort of issues are there that complicate EFI in a server environment (I'm thinking legacy support)?

Edit: guess that was two questions.

 

[Idontcare]

early you say! well that just means you were obviously over-resourced and failed to challenge yourselves enough in setting the project objectives

Next time we want twice the deliverables with half the resource investments

 

[lopri]

If I have an app that can only use 4 threads, assuming proper software support, can I have it cannibalize the shared parts of each module? What sort of performance benefit would I see from stealing all the of the shared components instead of splitting them with another integer in a quad-threaded environment?

I'm curious about that, too. I assume it is ideal to have one thread per module which have two cores, but what happens if the OS schedule 4 threads in such a way that they hop around 4 cores in two modules? What would be the performance penalty under such a scenario?

 

[Nemesis 1]

Since Barcelona:

Shanghai - 3 months early, higher than expected clock speed
Istanbul - 5 months early, higher than expected clock speed
Magny Cours - 3 months early, higher than expected clock speed

I would say that we have executed.

What about the 40% faster than Intel in curtain benchmarks . None of what you posted above has delivered anything but broken promises and now all of the sudden AMD doesn't lie . Your kidding right.

 

[ModestGamer]

Ehhh I don't think so really. I'd say in most things Intel wins, but AMD isn't nearly as bad as people sometimes say. In most cases, there is no perceptible difference in day to day tasks. My x6 did raw conversion in Canon DPP faster than my i7 (at least it felt like it) but my i7 has higher minimum frame rates in EQ2. W/e.

I think bulldozer will compete well with sandy bridge, but right now I don't see that as being AMD's problem with BD.

I think AMD's problem is friggin *release it already*. If it's late late 2011 it may end up coming up against Ivy Bridge instead. And that...is going to be a whole new can of worms. They need BD out sooner rather than later.

So your basing alot of performance on zero analysis and instead on perception ? I'd say I flatly disagree. for every product intel has AMD has a comparable product right now that is as fast for less or equal with the exception of the over $500 market.

If core utilization is lows its not AMD's fualt. Talk to the software authors.

 

[ModestGamer]

What about the 40% faster than Intel in curtain benchmarks . None of what you posted above has delivered anything but broken promises and now all of the sudden AMD doesn't lie . Your kidding right.

WTF are you talking about ?

 

[JFAMD]

He's talking about the intel claim that they would be at 10GHz by 2011.

 

[JFAMD]

JF, just a quick architectural question here.

If I have an app that can only use 4 threads, assuming proper software support, can I have it cannibalize the shared parts of each module? What sort of performance benefit would I see from stealing all the of the shared components instead of splitting them with another integer in a quad-threaded environment?

You should read my blog, it was already answered there.

 

[Idontcare]

JF, just a quick architectural question here.

If I have an app that can only use 4 threads, assuming proper software support, can I have it cannibalize the shared parts of each module? What sort of performance benefit would I see from stealing all the of the shared components instead of splitting them with another integer in a quad-threaded environment?

Additionally, what are the chances of the server and/or desktop market moving towards EFI? I know intel has come out and confirmed they'll be moving to EFI for sandy bridge at least on the desktop. I'm really looking forward to 3TB drives . What sort of issues are there that complicate EFI in a server environment (I'm thinking legacy support)?

Edit: guess that was two questions.

You should checkout his blog, link, he talks about this stuff there. Might be an answer to your question already there.

 

[bryanW1995]

lol.

nemesis was talking about a claim made by amd marketing (don't remember whom it was but I know it wasn't john) that barcelona would beat intel's cpu by 40%. it did this clock/clock in that one isolated benchmark iirc, but amd had clocking issues and some other snafus with barcelona. intel's 10ghz claim was even fluffier, but they've produced since then so it doesn't come up as much as the "40% faster than intel" claim these days. if BD knocks SB around a bit but intel makes ridiculous claims beforehand then the pendulum will again swing the other direction no doubt.

@modest gamer: really? amd's most expensive cpu is the 1090t. intel's equivalently priced cpu is the i7 950. let's see what anandtech bench thinks. hmmm, 24-3 with a few ties. overall say ~ 10% faster for i7 950? and that's with 4 cores + ht instead of 6 cores. also, oc headroom is a bit higher on i7 950 but it's clocked ~ 5 % lower by default, so if anything oc'ing gives the i7 even more of an advantage. so now intel has a CLEAR lead at anthing over $200, not exactly what an amd-leaning enthusiast wants to see/hear.

 

[Nemesis 1]

So your basing alot of performance on zero analysis and instead on perception ? I'd say I flatly disagree. for every product intel has AMD has a comparable product right now that is as fast for less or equal with the exception of the over $500 market.

If core utilization is lows its not AMD's fualt. Talk to the software authors.

Are your really that far out there . AMD has nothing to compete with intel on the desktop . Nothing but price . As far as blaming the software venders get real once they can do it correctly Intel will have all the real core they need to do the Job . AMD64 is a perfect example of a hype lie . The same as offerring more than 6 threads on the desk top will see it slowly but surely. Intel and AMD are both guilty here. But you do have to build the hardware befor the software to realize the gain . So you believe the egg came before the chicken . Not so the chicken had to lay the egg and than mother those eggs and keep them warm so as to hatch. I suppose you will argue the incubator was befor the egg.