Fudzilla: Bulldozer performance figures are in

[3DVagabond]

So your saying that BD will beat SB in the majority of review benchmarks . Its a possiability only if the reviewes change benchmarks . Anthing more than 4 cores on desktop is a waste to masjority of users. IPC is the real deal always has been always will be . Except to a small amd fanboy base who want a 100 core cpu to run 4 threaded apps . LOL

You say, "Except to a small amd fanboy base who want a 100 core cpu to run 4 threaded apps . LOL". So, I say I use apps that can run as many cores as you can throw at them, does that make me an AMD fanboy?

Believe it or not though, even apps that can run extreme numbers of cores can have aspects that rely on single threaded performance as well. So, single thread performance matters to me too. Am I now an Intel fanboy?

I find the people who want to concentrate on one aspect of performance, the one their favorite brand excels in, and ignore the others are the true fanboys.

 

[Abwx]

Have you forgotten that the same applies to Intel/SB?

Not at all , since Intel s 32nm is already a mature process that has
more or less reached its frequency ceilling.

 

[BlueBlazer]

Granted the price is accurate, it would be the first time that i see someone complaining because a new product has more performance while being much cheaper...

I detect that your sense of logic has gone out the window. "35% more performance" for less?

This post has a lot of words but says nothing. Let me re-cap:

BD speed is unknown
BD release is unknown
You will know when it's available

Awesome. All our questions are cleared-up now...

It seems AMD is "all at sea", even the release date is unknown for someone on the inside.

Anyway, has anyone seen this >> Benchmarks : High-End Desktop Performance: AMD "Bulldozer" CPU/APU

 

[trollolo]

ok nevermind i actually love america:
>> Benchmarks : High-End Desktop Performance: AMD "Bulldozer" CPU/APU

Even with its 8 real cores and improved multi-threading, Zambezi still cannot match Sandy Bridge even in software mode.

oh you

 

[BlueBlazer]

oh you

I didn't type that. Where did "i hate america, look at this link:" come from?

 

[trollolo]

I didn't type that. Where did "i hate america, look at this link:" come from?

i'm saying that the statement "even with 8 cores, it still gets beat out by existing sandy bridge processors" is guaranteed to be an inflammatory statement.

I'd mention that I wouldn't mind that, but then the mods will swoop down on me with the vengeance of a thousand lonely housewives.

 

[Markfw]

I detect that your sense of logic has gone out the window. "35% more performance" for less?

It seems AMD is "all at sea", even the release date is unknown for someone on the inside.

Anyway, has anyone seen this >> Benchmarks : High-End Desktop Performance: AMD "Bulldozer" CPU/APU

And here is a good one. Notice the comments that they used the same benchmarks on the 2600k, as they did before running @3.6. So even if they are real, they are a 2.8 vs a 3.6 ghz processor.

right......

 

[Accord99]

For those SB Sandra scores, the review indicates 3.6GHz Turbo mode is for 1 core only, the 4 core Turbo speed is 3.1GHz. Compared with their 3GHz Turbo-disabled review:

http://www.sisoftware.net/?d=qa&f=cpu_intel_sb

The scores are close enough to confirm this is true. I also tested out several of the Sandra processor benchmarks on a 2600K running strictly at 3.4 GHz and it was consistently ~10% faster than the Sandra review scores with the Turbo enabled SB.

 

[Nemesis 1]

So why does the review show AMD @ 2800(2x1400) 3800(2x1900)
intel @ 3000 (100x30) 3600(100x36)

 

[exar333]

I detect that your sense of logic has gone out the window. "35% more performance" for less?

It seems AMD is "all at sea", even the release date is unknown for someone on the inside.

Anyway, has anyone seen this >> Benchmarks : High-End Desktop Performance: AMD "Bulldozer" CPU/APU

Hmm, very interesting.

The 2600k wipes the floor with the 8-core BD in the review, but that is assmuning it is real (that's definitely tough these days with almost everything fake).

The 8150 would probably trade blows with the 2600k then, if you extrapolated from these scores. Then you OC both, and SB is king again. Not sure if true, but 2x the cores and it still is a loser? :\

 

[LOL_Wut_Axel]

Hmm, very interesting.

The 2600k wipes the floor with the 8-core BD in the review, but that is assmuning it is real (that's definitely tough these days with almost everything fake).

The 8150 would probably trade blows with the 2600k then, if you extrapolated from these scores. Then you OC both, and SB is king again. Not sure if true, but 2x the cores and it still is a loser? :\

I'm gonna assume this is probably fake. If AMD hasn't mustered any IPC increase in comparison to Llano, Bulldozer is a huge fail. The FX-8000s would only match the Core i7s in multi-threaded, and in anything only mildly multi-threaded they'd get their ass kicked.

 

[BlueBlazer]

And here is a good one. Notice the comments that they used the same benchmarks on the 2600k, as they did before running @3.6. So even if they are real, they are a 2.8 vs a 3.6 ghz processor.

Its already stated there, 3.6Ghz is Turbo (for 1 core) while 3GHz for all cores. I think Accord99 and Nemesis_1 nailed it.

I'm gonna assume this is probably fake. If AMD hasn't mustered any IPC increase in comparison to Llano, Bulldozer is a huge fail. The FX-8000s would only match the Core i7s in multi-threaded, and in anything only mildly multi-threaded they'd get their ass kicked.

Look at the source itself (from official Sisoft review). Its not from some blog or forum.

Found these >> Bulldozers in Berlin back room and AMD demo system shows with bulldozers - but only the press. Also Uncertainties Abound: Intel, AMD May Both Delay Next-Generation CPUs.

 

[LOL_Wut_Axel]

Its already stated there, 3.6Ghz is Turbo (for 1 core) while 3GHz for all cores. I think Accord99 and Nemesis_1 nailed it.

Look at the source itself (from official Sisoft review). Its not from some blog or forum.

Found these >> Bulldozers in Berlin back room and AMD demo system shows with bulldozers - but only the press. Also Uncertainties Abound: Intel, AMD May Both Delay Next-Generation CPUs.

I think I'll just wait for the launch. I find it hard to believe Bulldozer has comparable IPC to Llano. Why would anyone buy the FX-4100 over the A8-3870?

 

[grimpr]

And here is a good one. Notice the comments that they used the same benchmarks on the 2600k, as they did before running @3.6. So even if they are real, they are a 2.8 vs a 3.6 ghz processor.

right......

Hans De Vries claims that the Prefetchers are also not enabled on the BD ES chip.

http://semiaccurate.com/forums/showpost.php?p=132211&postcount=788

 

[BlueBlazer]

Hans De Vries claims that the Prefetchers are also not enabled on the BD ES chip.

http://semiaccurate.com/forums/showpost.php?p=132211&postcount=788

If considering the possibilities then its either not enabled, or not working? Already many months has transpired, isn't the platform (CPU, motherboard and BIOS) finalized? :\

I think I'll just wait for the launch.

Hard to comment on that, but I'm seeing many rumors of October at least in quite a number of places. That's way pass the "60-90 days" window. Found this >> AMD OS6272WKTGGGU OPTERON 6200 SERIES PROCESSORS 2.1G 16MB 115WT G34 TRAY - BLT Catalog (look at the ETA date)

 

[Dresdenboy]

If considering the possibilities then its either not enabled, or not working? Already many months has transpired, isn't the platform (CPU, motherboard and BIOS) finalized? :\

Such things have to be enabled by the AGESA code in the board's BIOS. So far coreboot code (AGESA sources) shows different handling of "Early Samples" and final silicon (as came with Gigabyte ROMs made available this week showing a "FX(tm)-" string). And multiple pages of listed CPU/NB config registers should offer enough knobs to reduce performance by tens of percent.

 

[BlueBlazer]

Such things have to be enabled by the AGESA code in the board's BIOS. So far coreboot code (AGESA sources) shows different handling of "Early Samples" and final silicon (as came with Gigabyte ROMs made available this week showing a "FX(tm)-" string).

We already came across some leaked B2 stepping benchmarks earlier (take them with a grain of salt), and B2 stepping was the (final?) one listed in GigaByte ROMs. Perhaps we may see another different B2 stepping on the final CPUs.

And multiple pages of listed CPU/NB config registers should offer enough knobs to reduce performance by tens of percent.

Not sure where's the logic in that, but reducing performance and disabling features (such as cache prefetching) does not allow motherboard manufacturers to tune and optimize their BIOS and hardware settings properly (also in relation to the HyperTransport bus for PCIe and I/O performance). :hmm:

 

[Nemesis 1]

Hans De Vries claims that the Prefetchers are also not enabled on the BD ES chip.

http://semiaccurate.com/forums/showpost.php?p=132211&postcount=788

Good for Hans , But are these early ES BD processors? The clocks seem a bit high to be early es dies.

 

[Nemesis 1]

You say, "Except to a small amd fanboy base who want a 100 core cpu to run 4 threaded apps . LOL". So, I say I use apps that can run as many cores as you can throw at them, does that make me an AMD fanboy?

Believe it or not though, even apps that can run extreme numbers of cores can have aspects that rely on single threaded performance as well. So, single thread performance matters to me too. Am I now an Intel fanboy?

I find the people who want to concentrate on one aspect of performance, the one their favorite brand excels in, and ignore the others are the true fanboys.

As do I. Thats wht BD is up against the wall Intel SB is good at everything

 

[Riek]

Good for Hans , But are these early ES BD processors? The clocks seem a bit high to be early es dies.

more or less the same as the leaked B0 ES clocks. There were some B0 samples running higher clocks.

 

[3DVagabond]

As do I. Thats wht BD is up against the wall Intel SB is good at everything

No argument there. SB is an excellent performer.

I do modeling and rendering. As of right now the best all around performing chip for ~$300 is the 2600K. Put a decent O/C on it and it'll give you performance that wasn't even a consideration for a hobbyist on a budget like me, before it was released. I'm hoping that Bulldozer can up that performance more. The rendering side of my app can run 64 threads/cores on a single workstation. While the editor side of it is only single core, for the most part. It will use additional cores to run scripts and plugins. So, fast single thread performance matters, but rendering is the really time consuming process. Thuban running at ~4GHz is actually an excellent 2nd choice, if you are on an even tighter budget. I'm hoping BD will give ~50% improvement over Thuban at similar clocks (newer design more threads). Then add another 10%-15% in higher clocks. That will make it overall better than the 2600K in my workload where the really time consuming task, rendering, needs as many threads as possible. Even if it's not as fast as SB in single thread operations, it should be plenty fast enough to do the job.

 

[bridito]

If considering the possibilities then its either not enabled, or not working? Already many months has transpired, isn't the platform (CPU, motherboard and BIOS) finalized? :\

Hard to comment on that, but I'm seeing many rumors of October at least in quite a number of places. That's way pass the "60-90 days" window. Found this >> AMD OS6272WKTGGGU OPTERON 6200 SERIES PROCESSORS 2.1G 16MB 115WT G34 TRAY - BLT Catalog (look at the ETA date)

Hmm.... what's that smell... is it a... paper launch?

 

[Abwx]

I detect that your sense of logic has gone out the window. "35% more performance" for less?

It seems AMD is "all at sea", even the release date is unknown for someone on the inside.

Anyway, has anyone seen this >> Benchmarks : High-End Desktop Performance: AMD "Bulldozer" CPU/APU

As usual , you re jumping so fast to conclusions that you systematically
miss the point...

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).

http://semiaccurate.com/forums/showpost.php?p=132311&postcount=824

 

[RussianSensation]

Overclocking is irrelevant for the business.

Ya, I am not talking about business users; I am discussing people who build computers on our forum. You know, overclockers

So Intel cannot launch much higher clockfrequencies either.

22nm Tri-Gate transistor will allow for 37% more performance at the same power envelope OR 50% lower power consumption at the same frequencies (according to Intel). IB is 8 months away.

We have no id how good or bad BD will overclock.

No we don't, but the deeper the architecture, the harder it is for it to scale. So if BD has high IPC (i.e., deep pipeline), it will overclock similar to SB. Also, given that they are both manufactured on 32nm, I seriously doubt BD will clock to 5.5-6.0ghz, not at least until the move to 28/22nm.

For reference, if BD core reaches nehelam ipc it will be faster in 3-4 threads while trading blows in 1-2 threads.

If BD 1 core = 1st gen i7 in IPC, then a 3.6ghz 8 core FX-8150 = 8 Core 3.6ghz Core i5 760, correct? That would be an incredible processor if true for just $300!

But let's look at what you are really saying:

BD 1 core = IPC of 1st generation i5/i7 ==> So that means 2x the performance of an i5 760 @ 3.6ghz for $300 in 8 threaded app?

2500k/2600k is 15% faster on average over i5 760. So that means an i5 760 @ 3.6ghz ~ 3.13g i5 2500k

With the same IPC as a 1st generation i5/i7 AMD would have an 8 core BD 8150 series which is "equivalent" to 8 core 3.13g i5 2500k......and that's without 4.2ghz Turbo Boost on BD?????????? So with Turbo Mode, we would have an even faster CPU than an 8 core 3.13ghz i5 2500k. Ya OK!!!!!!!!!!!!!!!!!

Care to share why AMD wouldn't release such an incredible processor at just 2.8-3.0ghz? Why in the world did they delay an 8 core i5 3.0ghz for 8 months? Cuz let me tell if AMD has an 8 core i5 2500k @ 3.13ghz + additional Turbo for $300-$350 ready for us this September, we'll have a lineup to each MicroCenter store of about 10,000 people on launch day.

AMD is positioning their 8-core BD chips vs. 2500k/2600k. If BD had IPC similar to i7 920/860/760, then they could have easily been selling their FX-4100 3.0ghz chips for the last 8 months @ $200 in place of the slow Phenom II. It just doesn't make sense to me. There is just no way I believe that the FX-8150 4.2ghz = i5 760 with 8 cores @ 4.2ghz. That would be insanely fast.

 

[grimpr]

As usual , you re jumping so fast to conclusions that you systematically
miss the point...




http://semiaccurate.com/forums/showpost.php?p=132311&postcount=824

BD will come October with final release bios, multithreaded Catalysts and one month later pair with the GCNs 7900 Radeons?, man it would be awesome!