• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Rumour: Bulldozer 50% Faster than Core i7 and Phenom II.

Page 73 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.
But Integer is one of the reasons the Core series does so well in alot of apps compared to Phenom. Why would AMD not truly beef those units up?

Well, look at it this way. They did, they just gave it to you in the form of more cores instead of faster cores.
 
Where you got this 3.1 GHz? I can hardly imagine that BD will clock that low! AMD already talked quite early about 3.5+ GHz. I am still predicting 4.5 GHz (including TURBO of course).

There was a slide posted a few pages back that showed the top speed bin to be 3.1GHz with a 4.1GHz turbo. I don't know the validity of that slide, but if accurate I am a little disappointed. If it is accurate though, I expect relatively quick increases in the top speed as the manufacturing technology becomes more mature.
 
I don't believe the graph. Doesn't pass the sniff test which includes the observations you are making here.

Yes - for a supposedly leaked AMD slide - it sure doesn't have AMD's typical style and isn't really polished. The 'balanced computing' star isn't something that an engineering department would typically put on a slide, so that pretty much rules out some lower level engineer having been stuck with making a PPT presentation for his boss.

But, it's May already and we haven't seen anything of significance leaked (not even some specs from a Taiwanese mobo maker) - amazing!
 
AMD FX-Series / A-Series Vantage Benchmarks Leaked

Not sure if people have seen this ? Vague
amdbullperf_6a_dh_fx57.jpg
?
 
@notty22

How well does PCMark scale with core count? From that picture going from an i3 2100 to a 2600k didn't make much of a difference. If it is as poor as I'm thinking, then BD's IPC could be closer to SB that it looks like.
 
Well hopefully BD doesn't turn out like the P4. Perfect in some area's, but quite blah in most area's. Thats what its looking to be like... 🙁
 
Last edited:
There was a slide posted a few pages back that showed the top speed bin to be 3.1GHz with a 4.1GHz turbo. I don't know the validity of that slide, but if accurate I am a little disappointed. If it is accurate though, I expect relatively quick increases in the top speed as the manufacturing technology becomes more mature.

IIRC, those were numbers for the available ES chips. They mean nothing as far as final product clocks are concerned.
 
Well hopefully BD doesn't turn out like the P4. Perfect in some area's, but quite blah in most area's. Thats what its looking to be like... 🙁

do you really think this? AMD wasted years and money to under perform their current arch? LET alone not be competitive with intel?
 
There was a slide posted a few pages back that showed the top speed bin to be 3.1GHz with a 4.1GHz turbo. I don't know the validity of that slide, but if accurate I am a little disappointed. If it is accurate though, I expect relatively quick increases in the top speed as the manufacturing technology becomes more mature.

clock speed is not all that and a bag of chips. Efficient use of arch can make clock speed an afterthought.
 
here is an analogy. Its moving day and i showed up with a 18 wheel moving truck. You have a fast pickup truck. You get there first and have to keep going back. I arrive slower than you with the whole contents of the house. While you race back and forth, I unload all contents and leave. You are not finished moving. I am.
 
clock speed is not all that and a bag of chips. Efficient use of arch can make clock speed an afterthought.

But based on the statements made by AMD, BD is specifically a "speed demon", a chip designed to perform well by reaching high clocks. 3.1GHz base clock for the top bin would be alarming.

But, just to repeat myself: IIRC, those were numbers for the available ES chips. They mean nothing as far as final product clocks are concerned.
 
But based on the statements made by AMD, BD is specifically a "speed demon", a chip designed to perform well by reaching high clocks. 3.1GHz base clock for the top bin would be alarming.

But, just to repeat myself: IIRC, those were numbers for the available ES chips. They mean nothing as far as final product clocks are concerned.

Yes based on what we know. This is going to be a fast moving truck. lol Its design is based on moving more. Multi apps. not necessarily single apps. It is still a truck. NOT a sports car.
 
Last edited:
jf-amd has told people until he is blue in the face. bulldozer is server based. it is highly module. That doesn't mean it's a ghz killer. It is designed with the server market in mind. AMD'S bread and butter. Amd trys to compete in the desktop market. Where they prosper is in the server environment. This is where bulldozer was conceived.
 
Last edited:
jf-amd has told people until he is blue in the face. bulldozer is server based. it is highly module. That doesn't mean it's a ghz killer. It is designed with the server market in mind. AMD'S bread and butter.

Is the server market really AMD's bread and butter? I thought their market share in serves was a mere 7%.

Trefis has the server segment as contributing only 20% to AMD's valuation.

https://www.trefis.com/company?hm=AMD.trefis&from=widget:forecast&ovd_urlid=451774#

AMDTrefis.jpg
 
server is high margins and compared to intel and their marketshare, and the slide you posted. Amd is doing well.
amd was close to a finacial meltdown. compare to intel they are are the only competion in a aggressive market.
 
Last edited:
It's crazy, but notebook processors contribute more to their valuation, a market where AMD is (traditionally, this looks to change in a book way soon) an outright disaster.

Hopefully Bulldozer is a return to the original Opteron (indeed it often sounds like it was designed specifically for the server market) but only time shall tell.
 
Is the server market really AMD's bread and butter? I thought their market share in serves was a mere 7%.

Trefis has the server segment as contributing only 20% to AMD's valuation.

https://www.trefis.com/company?hm=AMD.trefis&from=widget:forecast&ovd_urlid=451774#

AMDTrefis.jpg

Current evaluation - absolutely.

But I do recall when Opterons were in demand and gobbling up marketshare vs old pre-intel core product Xeons with FSB based on netburst.

I recall hearing from AMD insider that over 50% of their CPU profits were coming from Opterons (nothing official or concrete, just small talk with some AMD Marketing contacts) even though volume were relatively tiny compared to desktop/notebook volumes.

Additionally, AMD's marketing focus at the time were heavy with incentives toward Opteron volume which supports how profitable Server products were for them even at small volumes.

I would hazard to guess that this was why BD at inception was focused on Server needs - and if a derivative can be competitive on desktop/mobile, then it would be two birds with 1 architecture... errr.. stone.
 
Current evaluation - absolutely.

But I do recall when Opterons were in demand and gobbling up marketshare vs old pre-intel core product Xeons with FSB based on netburst.

I recall hearing from AMD insider that over 50% of their CPU profits were coming from Opterons (nothing official or concrete, just small talk with some AMD Marketing contacts) even though volume were relatively tiny compared to desktop/notebook volumes.

Additionally, AMD's marketing focus at the time were heavy with incentives toward Opteron volume which supports how profitable Server products were for them even at small volumes.

I would hazard to guess that this was why BD at inception was focused on Server needs - and if a derivative can be competitive on desktop/mobile, then it would be two birds with 1 architecture... errr.. stone.

smart guy. thanks for the perspective🙂
 
But Integer is one of the reasons the Core series does so well in alot of apps compared to Phenom. Why would AMD not truly beef those units up? What is it their banking on?
They are beefing them up. They just aren't beefing them up by making them very wide (and thus being far more complicated to manage), because unlike Intel, AMD doesn't have more money than God.

Better handling of branches, better prefetching, a trace cache, and finally getting around to having a a unified int scheduler aught to be able to do wonders for their int performance, especially in servers. Their prior/current CPUs have been 3 ALU/AGU, and not able to always utilize all three of any one, even if the code would allow for it. Only in the case where you would have several clock cycles in a row where all three ALUs could be used, and there was nothing for the AGU to do, would be significantly superior to having 2 ALUs and 1-2 AGUs on the side, and that's not the most common occurrence (having several clocks in a row where any 3 int instructions can be executed is going to be rare enough, outside of SIMD extensions).

Not a bad idea.
If only AMD had Intel or IBM's pot o' gold to work with...
 
Status
Not open for further replies.
Back
Top