Bulldozer Review! Legit?

[RobertPters77]

If these are legit. Goodbye amd. Intel and Nvidia won.

 

[piesquared]

Just because its BETA doesn’t make it unreal

Exactly, but it sure gives the IntEl puppets an avenue for attempting to discredit the results.

 

[MentalIlness]

i think the 870 has dynamic turbo on..

which means it only gets 3.6 on 2 cores.. and 3.2ghz on all 4.

I dont remember if they could do 3.6 on all 4 cores with turbo on without overclocking.

You guys been around sandy too long...


So its still not making sense..

Little off topic here, but how exactly is your username pronounced ? Ive always wondered this.

Is it like......egg-o-morla ? Or ?

 

[Arkaign]

Exactly, but it sure gives the IntEl puppets an avenue for attempting to discredit the results.

BF3 will go gold soon, if things stay roughly the same, the benches will have a real point. The results are extremely bizarre to say the least though, even amongst the various AMD vs. AMD and Intel vs. Intel matchups. I expect dramatic changes with final code, more optimized drivers, and reputable sites doing the benching. Also I expect good improvements for the 1GB midrange cards at final.

 

[Arkaign]

Little off topic here, but how exactly is your username pronounced ? Ive always wondered this.

Is it like......egg-o-morla ? Or ?

I would guess : 'aye-go-more-la'

 

[Black96ws6]

Little off topic here, but how exactly is your username pronounced ? Ive always wondered this.

Is it like......egg-o-morla ? Or ?

Close! It's actually pronounced, "One does not simply EYE-GO into MORDOR!"

 

[Olikan]

If these are legit. Goodbye amd. Intel and Nvidia won.

what? how nvidia won?
i really wanna read this

 

[Plimogz]

If these are legit. Goodbye amd. Intel and Nvidia won.

With regards to Intel, these new AMD chips are, in the absolute, going to excel at some (enterprise) workloads, and otherwise be priced as to be reasonable alternatives to Sandy Bridge.

Saying that AMD's going to win this round would be insane, but I don't see them being in a worse position now than they've been for the last couple of years.

And hopefully I'm not just imagining that Piledriver sounds like it might be more focused on improved single thread performance, anyway.

 

[guskline]

It's fair to say that with the 8150FX, AMD is back in the game! When I can, I'm going to snag a 8150FX to put in rig 2 below and see how it matches up against rig 1.

 

[jacktesterson]

AMD will price these right I hope.

I will buy a FX-8150 and update when Piledriver comes out again, without having to get a new motherboard.

One thing AMD has going right....

 

[Ancalagon44]

I think BD will dominate in server spaces. Could provide very cost effective performance. Gaming? Only in some games. In others it looks like it might be terrible. So much for JFAMD saying IPC has improved - obviously went quite a way backwards.

 

[frostedflakes]

Yeah IPC looks pretty underwhelming, but with the longer pipeline I suspected it might be worse than Phenom II in that area. If clocks scale well, though, hopefully that should be able to make up for it. Single threaded performance should be decent overclocked to 5GHz or so. Of course a 2500K or 2600K at 4.5GHz+ is no slouch either.

 

[Elixer]

I think BD will dominate in server spaces. Could provide very cost effective performance. Gaming? Only in some games. In others it looks like it might be terrible. So much for JFAMD saying IPC has improved - obviously went quite a way backwards.

Has it ? He could have been talking about corner cases for all we know.
Legacy x86 vs the newer extensions...

 

[frozentundra123456]

I think BD will dominate in server spaces. Could provide very cost effective performance. Gaming? Only in some games. In others it looks like it might be terrible. So much for JFAMD saying IPC has improved - obviously went quite a way backwards.

I honestly am not criticizing what you said, but I dont understand your statement. I dont know that much about the server area, so I just dont understand why you say Bulldozer will dominate in the server space. Even with the chip optimized for multithreaded workloads, doesnt it still use more power per output produced that Sandy Bridge?? Are you saying that the purchase price of the chip will be cheaper than comparable Intel, or that it will have better compatability with other parts, etc.?

 

[sangyup81]

Yeah IPC looks pretty underwhelming, but with the longer pipeline I suspected it might be worse than Phenom II in that area. If clocks scale well, though, hopefully that should be able to make up for it. Single threaded performance should be decent overclocked to 5GHz or so. Of course a 2500K or 2600K at 4.5GHz+ is no slouch either.

I'm hoping the shared double L2 cache and the shared double FP brings some efficiency to make up for the pipeline.

 

[bryanW1995]

Before we get too worked up, let's look at the 58 combined posts between the 2 guys linking this site. Don't get me wrong, this could be accurate info, but I'll believe it when I see Anandtech and AlienBabelTech writeups.

20% lower IPC than phenom in a single-threaded app seems absurd to me.

 

[StrangerGuy]

I honestly am not criticizing what you said, but I dont understand your statement. I dont know that much about the server area, so I just dont understand why you say Bulldozer will dominate in the server space. Even with the chip optimized for multithreaded workloads, doesnt it still use more power per output produced that Sandy Bridge?? Are you saying that the purchase price of the chip will be cheaper than comparable Intel, or that it will have better compatability with other parts, etc.?

A top-end AMD chip that barely beats a 2500K with 2x the cores at nearly 2x the power draw, worse IPC than PhII and abysmal single-threaded performance despite 4.2GHz turbo. I'm hardly impressed.

Anyway, this proves that the OBR guy was right all along.

 

[sangyup81]

I honestly am not criticizing what you said, but I dont understand your statement. I dont know that much about the server area, so I just dont understand why you say Bulldozer will dominate in the server space. Even with the chip optimized for multithreaded workloads, doesnt it still use more power per output produced that Sandy Bridge?? Are you saying that the purchase price of the chip will be cheaper than comparable Intel, or that it will have better compatability with other parts, etc.?

Check out some of the prices of Xeon chips and then compare them to the prices of AMD's Magny-Cours and Interlagos solutions. Also, this is not really for a few multithreaded applications but for virtualization.

 

[inf64]

This video is from March 8 this year. It was shot at Cebit 2011. It features Macci and John Taylor @ AMD. Mr Taylor said @ 2:20 mark that bulldzoer was designed to deliver 30 to 50% more performance within the same TDP envelope and roughly the same die are versus the cpu it replaces. The video is here.

What I want to know now is in which real world or synthetic benchmarks/workloads is Zambezi going to deliver 30-50% more performance than Thuban and with which magic dust is this going to happen? With the latest numbers it barely beats Thuban in highly MT workloads like cinebench(both old 10 and new 11.5) and handbrake. The difference ranges from tiny to 10 or so %. This is 8T vs 6C case,so best case scenario for Bulldozer. Bulldozer even runs at much higher clocks (both stock and Turbo). Where is 30% difference (just forget 50%)? Oh yes,AES and such are just outliers so those are corner cases.

If it was supposedly designed to deliver 30-50% more performance and if Mr Taylor stated this in the context of very parallel workloads (which is legitimate ) then we can say Bulldozer failed since it can't overall outperform Thuban by more than 20%,let alone 30% or now astronomical 50%. In order to achieve this ,the Bulldozer that John Taylor talked about must be the same one from this slide (and no,this slide was not fake). What happened in the meantime ? How from this 30-50% throughput machine we ended up with barely faster than Thuban? Unless the Bulldozer John Taylor spoke about in the video was expected to run at 4.5-5Ghz stock clock and Turbo to 5.5Ghz ,all within 125W TDP envelope,then something else went wrong and now we get "this" (slower than Thuban at the same clock by 5-15%,depending on the app and barely beating it since it has weaker core scaling and only 1.33x more cores to make up the difference).

 

[grimpr]

CB10:
(5800 / 3800MHz) / (4024 / 4200MHz) = SB has 1.6x IPC per clock than BD.

Ouch.

Come on now, who gives a damn about Cinebench 10, you know it doesnt work anymore.

http://www.youtube.com/watch?v=-M1eX5Vvrg8

 

[classy]

A top-end AMD chip that barely beats a 2500K with 2x the cores at nearly 2x the power draw, worse IPC than PhII and abysmal single-threaded performance despite 4.2GHz turbo. I'm hardly impressed.

Anyway, this proves that the OBR guy was right all along.

LOL

Is there some secret CPUs I don't know about? The 2500K, 2600K, I7 965, and 980X are the fastest cpus Intel makes. Let me repeat that again, those cpus are the FASTEST that Intel makes. So for all the absolute mealy mouth bs being spew, "it barely beats this" and "why did they compare it to this" is just utter nonsense. We will know in just a few days, but truth be told if the latest leaks hold water, AMD will have released cpus that will be among the fastest you can buy on this planet. Some of you guys are just embarrassing yourselves with some of these posts.

If the leaks hold water in the end, then they turned the trick. Damn give some credit where credit is due.

PS: I am typing this on a 2500K

 

[Black96ws6]

LOL

Is there some secret CPUs I don't know about? The 2500K, 2600K, I7 965, and 980X are the fastest cpus Intel makes. Let me repeat that again, those cpus are the FASTEST that Intel makes. So for all the absolute mealy mouth bs being spew, "it barely beats this" and "why did they compare it to this" is just utter nonsense. We will know in just a few days, but truth be told if the latest leaks hold water, AMD will have released cpus that will be among the fastest you can buy on this planet. Some of you guys are just embarrassing yourselves with some of these posts.

If the leaks hold water in the end, then they turned the trick. Damn give some credit where credit is due.

PS: I am typing this on a 2500K

I think the issue is, we're not seeing any real improvements here. The PhenomII X6 1100T is already between a 2500k and 2600k in multi-threaded (scoring a 5.84, see below):

This while clocked @ only 3.3Ghz.

Then you have "8 core" BD, clocked @ 3.6Ghz, and it's barely any faster, despite higher clocks and 2 extra cores.

In addition to this, you have worse single-threaded performance than an 1100T, again, due to the clock speed differences listed above.

That, to me, is a fail. Is that what everyone has been waiting and waiting all this time for? A slightly less IPC PhenomII "X8" that performs slightly worse than a Thuban X6 (considering the higher clock\turbo speeds)?

 

[podspi]

If that is actually BD's ST performance it will be very disappointing. Unless PD is a huge step up, Trinity will not be that compelling of a product, especially against much more powerful mobile IBs...

 

[Remobz]

I think the issue is, we're not seeing any real improvements here. The PhenomII X6 1100T is already between a 2500k and 2600k in multi-threaded (scoring a 5.84, see below):

This while clocked @ only 3.3Ghz.

Then you have "8 core" BD, clocked @ 3.6Ghz, and it's barely any faster, despite higher clocks and 2 extra cores.

In addition to this, you have worse single-threaded performance than an 1100T, again, due to the clock speed differences listed above.

That, to me, is a fail. Is that what everyone has been waiting and waiting all this time for? A slightly less IPC PhenomII "X8" that performs slightly worse than a Thuban X6 (considering the higher clock\turbo speeds)?

I always knew it would be a failure. I only wanted proof before I bought an Intel CPU

 

[inf64]

This was posted by Dresdenboy some time ago:
http://semiaccurate.com/forums/showpost.php?p=132311&postcount=824

Ah, it's N1truX and he posted in my first language..

IIRC he works for some magazine. He had to obscure his informations because of a NDA. He saw BD and made photos (might result in some article) and he heard something interesting about BD's performance directly from John Taylor and other ppl. Further he wrote that if leaked benches are close to the truth (although they seem to be a bit too low to him) any results (esp. latest Sandra leak) might be explained somehow. But the problem is not with Bulldozer (core/whole processor?) but with something else. The solution won't be available until launch which will also mean bad first reviews (at least sub par of what a fixed system would be capable of).

If there is any truth to this then it might be OS scheduling problem. But don't get your hopes high.