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

[wahdangun]

Also get use to IB hype don't worry tho it won't be 2 years of hype just 6 months if thats a comfort.

Enough . Stay on track .
everthing is here in black and
white . Those who count can
read very well.

did you forget your own word ?

 

[Nemesis 1]

No I am on track I just comparing intels yet unavailable cpu to amds not yet avaiable cpu. I got tell BD has zero chance of beating IB by 50%

 

[Abwx]

308 US$....

 

[moriz]

LOL ya this 60 year old guy has no life. I have lead a very exciting adventured filled life thank you very much.

Yet you behave like a spoiled child who happens to have a shinier toy, ridiculing everyone else who doesn't have it.

Looks like age != maturity.

Please just use the report post function. Much more effective than taking on a member here on your own. Anandtech Moderator - Keysplayr

 

[podspi]

Do they ship to the US? I know a few people who are members of Taobao (Chinese equivalent of Ebay), I should ask them to pick one up for me...

But what if it isn't retail and is one of those bugged B0?

 

[Abwx]

Do they ship to the US? I know a few people who are members of Taobao (Chinese equivalent of Ebay), I should ask them to pick one up for me...

But what if it isn't retail and is one of those bugged B0?

Should be a definitive stepping since it seems
that it will be available only in one month..

Even one month would be great, though.

The processors would feature a stock clock speed of 3.8Ghz which can turbo upto a staggering 4.2Ghz. Maintaining a 125W TDP and would feature 8MB L3 Cache.

http://wccftech.com/2011/05/30/amd-bulldozer-fx8130p-sale-priced-19980-yuan-308/

On the other side, there s news that contradict this
good one, as Xbit say it will be delayed...

http://www.xbitlabs.com/news/cpu/di...lock_Speed_of_FX_Bulldozer_Chips_Sources.html

 

[Nemesis 1]

Yet you behave like a spoiled child who happens to have a shinier toy, ridiculing everyone else who doesn't have it.

Looks like age != maturity.

Ya know I use the internet alot I like to keep up with tech . Thats what brought Zinn2b to these forums all those years ago. I was trying to learn . But there was this group of guys beating up on each other. I thought to myself What the F---.

Than I started reading what both sides were measuring e-penis with . At the time I had just put together my first PC The very box I am still using online . I started with the 2.4 p4C and overcloocked to 3.1ghz . Relized I need to water cool and built my first complete home built system . I ran Air alot more tho . I really at the time didn't like the water setup for the time. But this was the same time hammer came out and the AMD guys were complete asses. The Bench marks I got @3.1ghz were killing the best hammers out there. at the time I believe 2.2 hammer and not a clocker. In media I destroyed it . In game it won but I was way closer than review site I used 2 512mb sticks corsair bh5 @ 2 2 2 2 5 timing Idon't recall the what memory speed I used but it was realy high for those sticks at that setting . I am still using those sticks same setting except I am at lower V and running at 200.

But I bragged about it here and posted my scores and it drove them insane . I bested them in almost ever debate . I have history now to back it up also . But I kept getting band for being right . Other than that first ban . That was questionable but not for the time . GO back read zinn2b post and what he posted . than look at history . I been battling ever since . I finally made peice with NV . Its as the wife told me today your being an ass and you damn well know this thing is over . Ya she is right. SO I going to make peace with AMD but I buy intel . I wish the AMD guys the best.

Please keep this on topic and do not insult another member here.
Anandtech Moderator - Keysplayr

 

[Soleron]

Xbit's comments make no sense. If clockspeed potential is the problem, base speed should be higher than 2.5GHz if Turbo is 3.5GHz.

I'd understand if for thermal reasons base couldn't go above 2.5GHz but not speed pathing reasons.

 

[garagisti]

Mostly all of the information is known about Server BD, or is it(i don't know)?

6200 Series
8, 12, or 16 cores

- Q2 2010 Tapeout
- Q4 2010 Sampling to partners
- Q2 2011 Planned shipping to partners
- Q3 2011 Launch...

ACP: 65W, 80W, 105W (it doesn't specify what is for which model... but 105W for 16 cores is not bad really)
4 Integer issues per cycle
New Instruction sets: SSSE3, SSE 4.1/4.2, AVX, AES, FMA4, XOP, PCLMULQDQ
Power Gating: AMD CoolCore™, C1E, C6

Memory:
Northbridge controller has been redesigned
1600 Mhz DDR3 Support
1.25 LV DDR3 Support
Memory power capping via APML

 

[podspi]

I wonder how long this can go on. Will the supposed launch date come and go without any word from AMD?

It's going to be a real shame if AMD has a competitive architecture but misses out AGAIN because of clocking issues...

 

[386DX]

On the other side, there s news that contradict this
good one, as Xbit say it will be delayed...

http://www.xbitlabs.com/news/cpu/di...lock_Speed_of_FX_Bulldozer_Chips_Sources.html

I think Xbitlabs is probably correct. I been saying ever since the model/price list was leaked that it seemed a bit strange to release two 8 Core Unlocked Processor, with the only difference being the clock speed (pointless since they are both unlocked) and one having 125W TDP instead of 95W TDP. It's a dead giveaway there's problems with clock speed, but AMD knows that reviews are gonna test and compare at stock speed so the BD FX-8130P will look better compared to the i2600k even though the FX-8130P maybe running at 90% of its max potential (max overclock/architecture speed) while the i7 2600k is running at 60% of its max potential. You only have to look at the Phenom II 965 BE release to see this, that came out at 140W compared to 125W for the 945/955 BE, it didn't overclock any better but it looked better on the benchmarks. If BD is delayed till the Fall it would make sense they are waiting on a new stepping so they don't have to put out a 125W BD (the Phenom II 965 BE eventually came out at 125W after new stepping).

 

[RobertPters77]

I think Xbitlabs is probably correct. I been saying ever since the model/price list was leaked that it seemed a bit strange to release two 8 Core Unlocked Processor, with the only difference being the clock speed (pointless since they are both unlocked) and one having 125W TDP instead of 95W TDP. It's a dead giveaway there's problems with clock speed

If that's the case then I'm dumping my Amd stock. I don't need another failure like Phenom was.

 

[guskline]

Been an AMD fan a L O N G time but there is trouble in river city for the Bulldozer. First time in years that I will be looking to Intel for my next build.

 

[Idontcare]

If that's the case then I'm dumping my Amd stock. I don't need another failure like Phenom was.

Long term that might be a poor investment decision. Short term though there may well be opportunity to "accumulate" more positions on the cheap, or take a short postion for the near-term, then convert to more long positions when the market has priced in all their downside expectations.

I wonder how long this can go on. Will the supposed launch date come and go without any word from AMD?

AMD is full of smart people, they won't make a public comment on it until they know for certain whether or not B2 silicon is production worthy.

No point coming out with bad news (delay until B2 stepping) only to then have the possibility of needing to come back and give more bad news (sorry folks, B2 is borked as well, delay until B3 stepping now).

That's what they will communicate to their supply chain partners, but to the end-user it will rightly continue to be "all's well in Dresden".

When did Intel let the end-user know about the SB chipset issue? Before they had a known solution? Or after?

(answer: after)

Been an AMD fan a L O N G time but there is trouble in river city for the Bulldozer. First time in years that I will be looking to Intel for my next build.

Ah come on now, what's a couple more months? It's not really an unforgivable offense, is it?

These chips take 90 days to get into the channel from the time they start the wafers. With that kind of production lead-time it is very easy to get the schedule pushed around by a month or two.

 

[OCGuy]

If what xbit is saying is true, than this "50%" faster thread seems to have fallen flat.....much like the Phenom one.

 

[Tuna-Fish]

If what xbit is saying is true, than this "50%" faster thread seems to have fallen flat.....much like the Phenom one.

They're expecting to hit higher clocks once the bugs are ironed out.

... But that is pretty much exactly what happened with Phenom. They had "good" models that showed that Phenom was a beast at the clocks they were expecting to ship it at -- they just never got the design to hit those clocks.

Will that happen this time? Who knows. All it takes to ruin the clocks of a processor is one badly laid path in it. If there's just a few individual paths that cause the clocks to be so low, it's easily fixable. Or the entire core could be borked. Wouldn't take a position on it.

 

[RussianSensation]

If what xbit is saying is true, than this "50%" faster thread seems to have fallen flat.....much like the Phenom one.

That 50% claim is just as ridiculous as the claims that HD6970 to GTX580 in DX11 was going to be what R300 (9700) was to NV30 (FX5800/5900) in DX9. We know how that turned out. :whiste:

2600k often beats the Core i7 980X. So the 50% claim would imply that an 8-core BD would also be ~50% faster in some benchmarks than a 6-core 12 threaded Core i7 @ 3.3ghz?

 

[Abwx]

while the i7 2600k is running at 60% of its max potential. .

You think that a 2600K would do about 5ghz in a 125W TDP ?...:whiste:

 

[Dresdenboy]

Then I think it's safe to say that what we've got here is failure to communicate. If real code could execute up to X IPC, over a short period, the front end should have >=X IPC (generally to the nearest whole number > X), else it could be a bottleneck to optimized high-ILP code; not that if there are Y execution pipelines, the front end should have Y IPC.

So I found some nice IPC traces. Take this one:

(from "Techniques for Accelerating Microprocessor Simulation", Ramkumar Srinivasan). The simulated OOO CPU has 4-wide decode/issue/commit, 4 ALUs etc. There are other examples with much more IPC variation at shorter time frames.

In this case I'd say that the 4-wide decoding isn't the limiting factor here.

 

[cusideabelincoln]

2600k often beats the Core i7 980X. So the 50% claim would imply that an 8-core BD would also be ~50% faster in some benchmarks than a 6-core 12 threaded Core i7 @ 3.3ghz?

Regardless if the thread title is based off a rumor, this argument is so wrong it hurts. Under most conditions we would expect the 8 core Bulldozer, under the pretenses of this rumor, to be 50% faster than the 4 core 2600k in applications which are heavily multithreaded scenarios. Heavily multithreaded applications are exactly the area where the 980X beat the 2600K, sometimes significantly. So tell me again... what the hell were you thinking when you made this implication? At the right clockspeed 8-core BD could be 50% in select applications than the 2600K, but that does not mean it would be 50% faster than the 980X. Since the 980X is up to 30% faster than the 2600k in select benchmarks from your very own AT review. And those benchmarks are the ones that are biased towards more threads like the ones in 980X and Bulldozer... I think you can put two and two together now.

 

[Abwx]

So it seems that Llano got the full 32nm capacities at the expense of BD...

http://www.hardware.fr/news/11583/computex-bulldozer-bel-bien-retarde.html

 

[Arachnotronic]

Also, isn't it 50% faster than i7 950? Not 50% faster than SNB?

 

[RyanGreener]

So it seems that Llano got the full 32nm capacities at the expense of BD...

http://www.hardware.fr/news/11583/computex-bulldozer-bel-bien-retarde.html

Not surprising since Llano is more mainstream and most likely going to be more profitable.

 

[exar333]

Not surprising since Llano is more mainstream and most likely going to be more profitable.

Not likely.

BD is consumer and server CPUs; the latter is $$$$.

 

[RyanGreener]

Not likely.

BD is consumer and server CPUs; the latter is $$$$.

I thought servers don't need to run at high clock speeds to be effective though? Or does this whole stepping issue stem from the fact that BD isn't energy efficient enough?

Well, I'm not really sure, but I'm pretty sure general consumers are buying Llano way more than Bulldozer. For desktops AND laptops I think it'll be a hit with OEMs or budget conscious consumers.