Bulldozer Review! Legit?

[LOL_Wut_Axel]

I am just so sick of all the speculation, I can't wait until Anand's review....

Not much speculation going on now, really. You could say that three months ago, but as of now we have a very good idea of how it'll perform. That is to say, good in multi-threaded, but rather horribly in single-threaded.

Not that it was a mystery: you could see that from a mile away if you looked at the clock speeds and the number of integer cores, plus the revisions and delays.

 

[frozentundra123456]

No no. Not even funny. Benchmarking is science. And all the rules that apply to scientific study apply here too.

So far the benches have hard time to fulfil even the basics.

Actually, I thought it was quite humorus and also had a definite grain of truth. Granted benchmarks are real measurements, but they are still subject to the prejudiced interpretation put on them by biased people, both pro-intel and pro-AMD. And when the benchmarks are still in the conjecture stage, it leads to even more outlandish interpretation.

 

[frozentundra123456]

You are a cruel and unkind character with a depressing personality.

But correct.

 

[aigomorla]

AMD has one-fifth the R&D budget of Intel and the fact that they can produce CPUs roughly comparable to Intel (which despite the mighty company it is, must cheat) is remarkable in itself.

http://arstechnica.com/hardware/reviews/2008/07/atom-nano-review.ars/6

im not saying anything bad about amd.

i am saying bad about the idiots who decide to pit the AMD next to a cpu which completely outclasses it, only because it costs 5x more.

 

[Red Squirrel]

I was expecting to see someone reviewing a tractor.

 

[lOl_lol_lOl]

im not saying anything bad about amd.

i am saying bad about the idiots who decide to pit the AMD next to a cpu which completely outclasses it, only because it costs 5x more.

I must agree then.

Instead of limiting yourself to half the CPU/GPU market b/c of fanboyism just to support the pockets of a bunch of shareholders who prolly dont need the money, you should have full power of choice as a consumer. Then the only person who wins is the one who chooses the best perf/price.:thumbsup:

Anyway who are we kidding, 3930K is the tech to get. It'll smoke 8170 and OC to 5 Ghz.

 

[aigomorla]

Then the only person who wins is the one who chooses the best perf/price.:thumbsup:

this is the only logical reason why one gets a computer...

But then again... my computer is also a form of ART, which is why it has 3 independent water loops, cooling the entire system.

 

[lOl_lol_lOl]

this is the only logical reason why one gets a computer...

But then again... my computer is also a form of ART, which is why it has 3 independent water loops, cooling the entire system.

Specs please, why 3 loops?

 

[lOl_lol_lOl]

I was expecting to see someone reviewing a tractor.

Thank God someone else here has some sense. I found a leaked review of the Bulldozer.

http://www.buzzillions.com/reviews/nortrac-bulldozer-30-hp-reviews

The 'catdung mover' review is the one to read.

 

[BlueBlazer]

And:

That POV Ray 3.7 did not mention which beta version. Between Anandtech's Beta 23 (41.7%) and HardwareCanuck's Beta 40 (25.6%), I would assume the latter. However for the ABBYY OCR10, there is one from Tom's Hardware that seems to be multi-threaded in that test (judging by the results), and shows Core i7 2600K to be 35% faster than Core i5 2500K. :hmm:

 

[RussianSensation]

I just wanted to point out an obvious flaw: Don't use any Blizzard games as an unbiased benchmark for CPU or GPU performance. Blizzard have always developed their game engines using intel cpus and nvidia gpus and optimize for both and not amd/ati hardware since warcraft 2.

Ok so what happens if a gamer plays Blizzard games? Should they just ignore that Intel performs faster at the moment?

Where there any Intel 'bias' comments when Athlon XP+ whooped Pentium 4 in games, SuperPi, etc.?
Where they any Intel 'bias' comments when Athlon 64 whooped Pentium 4 "C" in games, SuperPi, etc.?
Where there any Intel 'bias' comments when Athlon X2 whooped Pentium D in games, SuperPi?

I keep hearing the same thing about any benchmark where Intel wins: "Intel cheated! Intel pays to win! Intel sabotages benches! etc. etc."

1) You realize a large portion of people do play Blizzard games such as WOW, and SC2 and will buy Diablo 3. If the game performs better on a certain brand of CPU, those users will buy that brand because those are some of the most popular games on the PC. So their benchmarks are 100% valid in terms of a gamer's purchasing decision. If you mostly play GPU limited games, then variations in CPU performance in games may not matter to you. That's fair.

2) Looking at how poorly Q6600 performs in SC2 vs. Phenom II, how can you say that the game is biased towards Intel. It just performs better on a more modern architecture with faster cache and faster performs per clock - which are currently the staple of Intel Nehalem/Lynnfield and SB architectures. If BD improves performance per clock/per core and has faster cache performance, it will surely be fast in Blizzard games too.

 

[sangyup81]

1. Where there any Intel 'biases' when Athlon XP+ whooped Pentium 4 in games?
2. Where they any Intel 'biases' when Athlon 64 whooped Pentium 4 "C" in games?
3. Where there any Intel 'biases' when Athlon X2 whooped Pentium D in games?

4. You realize a large portion of people play Blizzard games such as WOW, and SC2 and will buy Diablo 3? So in that case, it's not really relevant who optimized the game better. If it performs better on a certain brand of CPU, those users will buy that brand because those are some of the most popular games. So their benchmarks are 100% valid in terms of end of the day purchasing decision.

5. Looking at how poorly Q6600 performs in SC2, you can't tell me the game is biased towards intel code compiler. It just performs better on an architecture with faster cache and faster performs per clock - which are obvious advantages of Intel Nehalem/Lynnfield and SB architectures at the moment.

Compare relative performance in blizzard games to relative performance in other games and even in the P4 days, Intel showed a lot better in blizzard games than they did in other games. Was the Athlon 64 faster? Yes but not by as much as it was in non-blizzard games. No one is accusing blizzard of making games run slower on AMD intentionally but writing it better for Intel and Nvidia isn't much of a stretch

 

[RussianSensation]

So, FX series is a failure because they dont perform better than Intel in Cinebench ?

The main point here is that an 8-core processor should perform extremely well in a program that is highly multi-threaded - such as a rendering benchmark. How can you have Phenom II X6 1100T scoring almost the same @ 3.3ghz as an 8-core BD does @ 3.9ghz? There is no particular bias in Cinebench against AMD since X6 beats 2500k with ease. So you would think that an 8-core BD with performance per core similar to Phenom II and higher clocks would be much faster than their own X6.

Of course this is just 1 benchmark, but excluding AMD marketing slides, all the leaked benches are not showing fireworks in the making.

 

[ocre]

what if these where the only benchmarks that the BD does well in?

 

[sangyup81]

The main point here is that an 8-core processor should perform extremely well in a program that is highly multi-threaded - such as a rendering benchmark. How can you have Phenom II X6 1100T scoring almost the same @ 3.3ghz as an 8-core BD does @ 3.9ghz? There is no particular bias in Cinebench against AMD since X6 beats 2500k with ease. So you would think that an 8-core BD with performance per core similar to Phenom II and higher clocks would be much faster than their own X6.

Of course this is just 1 benchmark, but exclusive AMD marketing slides, all the leaked benches are not showing fireworks in the making.

With the lack of details given in those benchmarks, there are a number of reasons why the system is getting lower scores. At least we can always count on anandtech reviews to be detailed and thorough so let's not make any conclusions before those come out

 

[StrangerGuy]

FX-4110 leaks from China, gathered from XS.org. Basically reconfirms the other leaks we have so far:

 

[frostedflakes]

Think my X4 955 does better in Fritz at stock. :/

edit: Get 15.03x (7213 kn/s) at 3.2GHz

 

[sangyup81]

I expanded all those images hoping to see the memory tab so I can see what speed the memory and the CPU-NB are running at

 

[StrangerGuy]

I expanded all those images hoping to see the memory tab so I can see what speed the memory and the CPU-NB are running at

Added.

 

[sangyup81]

Added.

Still no CPU-NB but thanks for showing the memory speed. At 1873-9-9-9-24-1T, it better not be running at 2 GHz like the Phenom II

 

[HNNstyle]

Added.

Maybe you can explain to a noob like me, how do those benches compare to a comparable intel at the same price?

 

[StrangerGuy]

Maybe you can explain to a noob like me, how do those benches compare to a comparable intel at the same price?

Can't find any prices on 4110, but I'll expect it to be around $120 since the FX-6110 is already $155.

http://www.extremetech.com/computin...cing-revealed-a-lot-cheaper-than-sandy-bridge

Which means i3-2100 will be the competitor.

 

[podspi]

That fritz result is downright disturbing. Regardless of ST performance, MT performance (for equal number of threads) has gone DOWN (per clock).

 

[aphelion02]

Can't find any prices on 4110, but I'll expect is to be around $120 since the FX-6110 is already $155.

http://www.extremetech.com/computin...cing-revealed-a-lot-cheaper-than-sandy-bridge

Which means i3-2100 will be the competitor.

You haven't answered his question...

 

[grimpr]

If BD improves performance per clock/per core and has faster cache performance, it will surely be fast in Blizzard games too.

No it wont, ever, at the best it will be 10% slower as long as Blizzard,Maxon and others use Intels tools and not Microsofts or gnus.

http://www.agner.org/optimize/blog/read.php?i=49