AMD Cayman info (or rumor)

[Solus Corvus]

I ask the same, since if we are going for the only metric we can extrapolate quickly, the Flops, its nowehere near 50%.. From 2.7 Teraflops to 3.0 Teraflops is only 11%....

That's the point I was trying to make a while back though. Cypress has strong theoretical performance, but significantly less real performance. If efficiency can be improved, real performance could increase disproportionately to paper power.

antilles is shaping up to be a chuck norris approved gpu

I hate this meme so much. Mr Norris is a loser.

[zerazax]

Yes, hence I don't get the 50% comment

Also, looks like the folks over at bbs expreview are back to commenting on the 6970
Link

NV took the 6950 test 1536SP 10% worse than the 580 ha ha ~! ~

SIMD Core groups as less able to estimate the calculated out, CHH is also a bunch of fools for the NV to get what 1536SP data in happy, NV was a child when you (like you guys like the fool?), That he look, made several false data, NV will be believed?

If true, AMD really pulled a RV770 again and sent out fake info/leaks

[zerazax]

ugh double post

[generics_user]

so no new info; we already had the 1920SP rumor a month ago....

6970 is going to be at least 50% faster than 6870 or there would be no room for 6950.

480x4 vs. 320x5 SPs + 2times higher tesselation performance + more vram support the 50% claim + higher clockspeeds (6990 clock speeds are the same as 5970 at the same board power, but 6970 TDP is supposed to be higher than 5870 TDP so it should have higher stock clocks)

looks like a winner, but still 3 weeks until release....

[Heinz68]

If the slide is not fake it's going to be nice upgrade option for me. Neither GTX 580 or HD 6970 is an upgrade for me.

Going by the limited info on the slide I estimate the HD 6990 should be about 30% faster than my HD 5970 and if the rumors are true it should be much faster in tesselation plus extra 2GB GDDR 5 is also good upgrade.

[Heinz68]

I ask the same, since if we are going for the only metric we can extrapolate quickly, the Flops, its nowehere near 50%.. From 2.7 Teraflops to 3.0 Teraflops is only 11%....

How do you know the HD 6970 is 3 TeraFLOPS. It doesn't show on this slide which somebody claimed is a fake anyway.

http://bbs.expreview.com/viewthread....om=recommend_f

[zerazax]

How do you know the HD 6970 is 3 TeraFLOPS. It doesn't show on this slide which somebody claimed is a fake anyway.

6990 is 6gflops so Cayman xt is prolly at least half that

[ColonelCain]

Fancy that... semi-concrete info.

[Nintendork]

Should be more than 3TFlops:

Teraflops:
Core frequency(higher in single core Cayman) * SP * 2

900*1920*2 --> 3.45TFlops "HD6970" (barts frequency as reference)

[Heinz68]

6990 is 6gflops so Cayman xt is prolly at least half that

At least half is not neceseraly 3TeraFLOPS which Pica084 was using in his calculation.

The HD 5970 has "Processing power (single precision): 4.64 TeraFLOPS" and the HD 5870 is 2.7TeraFLOPS. LINK

In order to get within the 300W load power AMD needed to reduce the clock and voltage on the 5970. Since the HD 6990 slide shows "Load/Idle board power 300W/30W" some similar adjustment was most likely done.

HD 5970 performs about same as HD 5850 Crossfire unless overclocked to at least the HD 5870 levels.

[saaya]

Not really Saaya its 4d shaders therefore its 480x4 for Cayman vs 320x5 for Cypress = 50% increase.

I would estimate performance will be around 40% higher on average with even more notable differences in tesselation intensive benchmarks.

why estimate if the 6000 series is already out and has been benched by dozens of sites? compare reviews and you will see that sp for sp the 6000 series is ~10% faster than the 5000 series.

Well can't say if this slide is fake or not, but I guess you completely overlooked 50% more SIMD engines than HD5870 and double poly/clock.

On paper at least it should eat GTX580 alive.

i dont know how much the increase in poly/clock will help... but i doubt itll help much besides in tesselation benchmarks and new games that come out in 2011... and there itll only matter with high tesselation settings and i have yet to see a game or let alone demo where tesselation actually makes me want to upgrade :P
sure, it MIGHT be faster than the 580, but from what ive seen so far, theres nothing hinting at it... if the 1920sp number is true, it will trade blows with the 580 at best, and probably be a tad slower.

You didn't read about Barts or you're ignoring it.
Barts 6870: 1120SP with full efficiency
Cypress 5870: 1600SP but only 1280SP being used most of the time
Cayman is not a 5D thing. Let's make a guess

Cypress 5870: 1280SP effectively used
Cayman 6870: 1920SP fully used (or 2400SP if we apply Cypress SP inefficiency)

lets make a guess... OR... OOOOORRR we look at real world benchmarks
go and continue doing your 4d 5d math all you want, actual performance boost of 4d over 5d is around 10%, sp for sp at the same clock

You disappoint me.

which means you have high expectations and think a lot of me? thanks
i shant be dissapointing you

Try not just counting on SPs amount, but also the SIMDs. Not saying that AMD mArch has been limited in math calculating or shader power, but with this quite revolutionized, new mArch, it might offer better utilization & upped efficiency compared to the current one applied in Evergreen family & Barts chips.

revolutionized? so far it sounds more like a tweaked rv800?

Say 30 SIMDs are divided into 2 "GPC", each with its own setup & rasteriser engine, then each "GPC" with 15 SIMD will be divided into 3 arrays of 5 SIMDs. Each array has 16 TMUs totalling 48 TMUs/"GPC". As a whole, 2 "GPC" with 48 TMUs each will make a 96 TMUs chip. Just a very, very raw speculation of mine, please don't take it too seriously.

why would they only do that for the 6900 series though and not 6800? sure, 6900 might be a different arch, but... that would be a bit weird wouldnt it? would be cool, but its more wishful thinking and looking for a reason HOW the 6900 could beat the 580 if you ask me :P

id love it... i just dont think its very realistic

Also, Nvidia is apparently readying a dual GF100 card, so AMD will be facing competition.

dual gpu card formula:
if ( perf/W perfMAX - perf/W perfMIN ) > ( perf/W SLI - perf/W single) release

basically, if you can get more perf out of 300W by using 2 gpus, then yeah... but sli doesnt scale that well, so using two gpus is always slower than using one gpu at higher clocks.

[tbone8ty]

that Antilles board is going to run so hot, its huuuge

[japamd]

why estimate if the 6000 series is already out and has been benched by dozens of sites? compare reviews and you will see that sp for sp the 6000 series is ~10% faster than the 5000 series.

But Barts parts are still VLIW 5, no one knows yet how good (or bad) the VLIW 4 will perform. That 10% improvement seen on benchmarks are due to internal optimizations (Techreport states AMD resized some queue and buffers).

Having twice the triangle setup rate/clock and 50% more ALU power, plus the internal optimizations seen on Barts, those improvements alone should give a nice boost over Cypress IMO.

[LightSpeed]

6990 is 6gflops so Cayman xt is prolly at least half that

At least, but im sure more. The 6970 will surely be clocked higher

[zerazax]

At least half is not neceseraly 3TeraFLOPS which Pica084 was using in his calculation.

The HD 5970 has "Processing power (single precision): 4.64 TeraFLOPS" and the HD 5870 is 2.7TeraFLOPS. LINK

In order to get within the 300W load power AMD needed to reduce the clock and voltage on the 5970. Since the HD 6990 slide shows "Load/Idle board power 300W/30W" some similar adjustment was most likely done.

HD 5970 performs about same as HD 5850 Crossfire unless overclocked to at least the HD 5870 levels.

At least, but im sure more. The 6970 will surely be clocked higher

Hence the at least

[Heinz68]

Hence the at least

I don't care about your at least. I was wondering about Pica084 calculation based on 3TeraFLOPS and wanted to know from where he got the info.

At least in this case means nothing.

[zerazax]

I don't care about your at least. I was wondering about Pica084 calculation based on 3TeraFLOPS and wanted to know from where he got the info.

At least in this case means nothing.

/sigh

He was taking it from dividing the 6 TFlops, nothing more/less - you're reading too much into it

[Mechanical Man]

lets make a guess... OR... OOOOORRR we look at real world benchmarks
go and continue doing your 4d 5d math all you want, actual performance boost of 4d over 5d is around 10%, sp for sp at the same clock

Please show those real world tests showing 4sp config performance. All that can be said based on Barts is that performance is atleast 10% sp per sp as othwise there would be no point doing that big arch change(basically making new shaders).

why would they only do that for the 6900 series though and not 6800? sure, 6900 might be a different arch, but... that would be a bit weird wouldnt it? would be cool, but its more wishful thinking and looking for a reason HOW the 6900 could beat the 580 if you ask me :P
.

Maybe it has something to do with where the best efficiency lies at what core config. It could be that 5VLIW sweetspot is ~1,2k sp and after that it declines. And 4VLIW sweet spot is after that. Who knows? Im just saying that there is too much variables that are left unanswered that one could really say about beating or not beating 580GTX.

[Heinz68]

/sigh

He was taking it from dividing the 6 TFlops, nothing more/less - you're reading too much into it

Exactly that's what I was thinking except that 6990 is not 6970x2 same as 5970 is not 5870x2, that's why such division doesn't make a sense same like calculating performance based on the division doesn't make a sense.

I just wanted to point out his 11% is misleading info unless he knows something else than dividing the 6990 6 TeraFLOPS.

PS
I think it should be me to post the "/sigh"

[zerazax]

You're misinterpreting what I wrote - I said it has to be at least 3 TFlops if the 6990 slide is real. He's taking the 11% based on 2.7 -> 3 TFlops, which is pointless given that going from 5-VLIW to 4D is going to affect real world performance more than theoretical throughput

[SimBy]

Finally some real slides.

http://www.frazpc.pl/artykuly/652601...n-HD-6950.html

[Iconyu]

Finally some real slides.

http://www.frazpc.pl/artykuly/652601...n-HD-6950.html

Thanks that explains a lot, all the ROP's have been beefed up, and so have the shaders. It's really is a totally different beast from Barts.

[Lanek]

Hé Hé, i like this preview, it just say enough, but not too much ... anyway, this time tesselation increase is explained ..

[zerazax]

Yep... almost always 2x faster than Cypress and up to 3x faster

edit: looks like slide 72 still needs to be filled in.

What a tease

[spursindonesia]

revolutionized? so far it sounds more like a tweaked rv800?

why would they only do that for the 6900 series though and not 6800? sure, 6900 might be a different arch, but... that would be a bit weird wouldnt it? would be cool, but its more wishful thinking and looking for a reason HOW the 6900 could beat the 580 if you ask me :P

Well, if we always simplify stuffs, then Fermi mArch can also be called a tweaked G80. But we're not simpletons, are we ?

IIRC, it was stated in B3D forum by an AMD rep there that while officially lumped into Northern Islands family, Bart & Cayman have different product development process, since both are targeting very different market. With Bart, basically a tweaked & optimised Evergreen mArch as you've mentioned, AMD try to address its glaring weakness in Evergreen line up, the sweetspot segments, while Cayman seems like the true new gen in this family, as an enthusiast products, where AMD can be more adventurous & creative without worrying about time to market too much.

that Antilles board is going to run so hot, its huuuge

It seems around Hemlock size, i don't think it's gonna be any bigger. The cooling solution & VRM design might have been changed & improved, IMHO.