New rumor about ATI Southern Islands

http://news.ati-forum.de/index.php/n...fikkartenserie

Quote:

---

The new graphics card generation, the direct successor of the current HD5000er series, is due out later this year, in this regard we have received new information from an industry-related source. So shall have already come in the middle of next October, the first offshoot of the new HD6000er series and the code name "bear beard."

The small branch is located in view of the performance between a current HD5770 and HD5870. In addition comes an eight-layer PCBs, 1GB GDDR5 memory with 5Gbps and 2x6pin power connectors are used. The TDP of the new pixels accelerator should be slightly higher than 150W and when connections are 2xDVI, HDMI, 2xMini-DP, with the support of Eyefinity 4, available. In addition, the chip of the "beard" Pinkomaptibel with the PCB is the HD5800, which will drive down production costs significantly. The above specifications refer to "Bart XT.

More information is available to us, relate to the GPU "Cayment" which will appear in November and October. This information confirmed our of 16 July this year. This GPU will have a 10-layer PCBs and large 1GB GDDR5 memory with 6Gbps. The graphics card needs to power a 6pin and 8pin power connector, the TDP comes to less than 300W. The connections are 2xDVI, HDMI, and 2xMini-DP.

Also the code name for the upcoming dual-graphics card of the new generation is already known to us. This carries the code name "Antilles" and will be AMD's new spearhead. We keep you informed.

---

LOL lost in translation! I like the name Bear beard tho! :up: For those who doesn't know, the german word for beard is Bart.

Makes my english look fantastic!:rofl:

Finger's up tho, keep'em comin':up:

im so lost,
perf between 5770 and 5870 (might aswell be 5850)
power consumption close to a 5850 aswell

what the point? its going to be cheaper to produce is all that i see, so did watts per mm2 go up even higher somehow??!?!?!

It looks like a GTX460 killer to me

Quote:

---

The small branch is located in view of the performance between a current HD5770 and HD5870. In addition comes an eight-layer PCBs, 1GB GDDR5 memory with 5Gbps and 2x6pin power connectors are used. The TDP of the new pixels accelerator should be slightly higher than 150W and when connections are 2xDVI, HDMI, 2xMini-DP, with the support of Eyefinity 4, available. In addition, the chip of the "beard" Pinkomaptibel with the PCB is the HD5800, which will drive down production costs significantly. The above specifications refer to "Bart XT.

---

This sounds like either the HD6770 or the HD6850, if ATI keeps its current naming scheme. Most likely a 6850 since it uses two 6pins. The perf difference between a 5850 and a 5870 is roughly 10 - 15 %, so we'd only be seeing that much of an improvement for the HD6xxx series.

Quote:

---

More information is available to us, relate to the GPU "Cayment" which will appear in November and October. This information confirmed our of 16 July this year. This GPU will have a 10-layer PCBs and large 1GB GDDR5 memory with 6Gbps. The graphics card needs to power a 6pin and 8pin power connector, the TDP comes to less than 300W. The connections are 2xDVI, HDMI, and 2xMini-DP.

---

That sounds like the HD6870 then. 1500MHz mem clock and 6+8 pins.

Doesn't really make sense, if the performance increase was only about 15% I don't think ATI would be bother releasing it as a next gen product.

Looks interesting. Basically they are going for 6770/6870/6970.

1200mhz mem clock ? wtf.

6/4 = 1.5 hey !!!

Quote:

---

Originally Posted by madcho

1200mhz mem clock ? wtf.

6/4 = 1.5 hey !!!

---

my bad :p:

Quote:

---

Originally Posted by Frontl1ne

Doesn't really make sense, if the performance increase was only about 15% I don't think ATI would be bother releasing it as a next gen product.

---

What else would you expect? The new generation is still based on 40 nm process, so there isn't much they can do. They have the ~same transistor budget, they have to base their architecture on Evergreen and they only had approx 12 months to come this far. So really, what else can people expect?

They need to do whatever they can to minimize the costs, and if it is working under the limitations of pin compatibility with current gen, then let it be so. The chips will be bigger than current ones for sure.

What I believe is that this news is just to mislead the competition. Now it just gives an impression that it's nothing interesting, just a similar performing updated silicon with possibly lower price. But given the old rumours of focusing on efficiency of the chip, given that the 40 nm process should be pretty much mastered(I know.. :)) by TSMC and those of AMD who work with them, they could achieve somewhat better perf/mm² for the uarch and somewhat better clockspeeds while keeping same yields.

I have hard time believing in "just" an updated silicon to minimize manufacturing costs. Then again, there isn't much what Nvidia could come up with in 40nm so AMD isn't really in position where it would desperately need more speed.. So the next real round will be fought at 28nm and thats where both companies focus.

hmm, may i speculate ? Bart would be a 1280 SP (320*4) 16 ROP 128 bit chip 240-260 mm^2 performance ~HD 5850, Cayman would be a 1920 SP (480*4) 32 ROP 256 bit chip 390-420 mm^2 performance ~GF100 512 SP+, and dual Cayman power consumption would be very, very close to PCIE standard limit of 300 w, unlike HD 5970 TDP which is rather conservatively rated, or would simply be underclocked further compared to its older dual chips kamerad. This speculation is reasonably explaining the rumored TDP of the chips reported in the article.

Wasnt the HD4xxx series surrounded by the same rumours?

Oh it will only have 640 stream processors as the flag ship as its double what they have been doing the past two generations

HD2900 = 320
HD3870 = 320
Assumed HD4870 = 640
Actual HD4870 = 800



So i say lets wait a little bit longer as i can see there being either a memory boost to the chip e.g 384 or maybe a return to 512 ring bus?

Or a dramatic improvement in tessalation in a response to nvidias domination

was the assumption for the 4870 to go with 1+3, and they did 1+4?

theres really only 2 things to look at
perf per watt = efficiency
perf per mm2 = costs

it seems that only 1 of those is really changing by much. which isnt bad considering its on the same process and we cant really expect too much with perf/watt (unless its like nvidia where they chopped off the server stuff to shave power for the gf104)

Quote:

---

Originally Posted by Panther_Seraphi

Wasnt the HD4xxx series surrounded by the same rumours?

Oh it will only have 640 stream processors as the flag ship as its double what they have been doing the past two generations

HD2900 = 320
HD3870 = 320
Assumed HD4870 = 640
Actual HD4870 = 800



So i say lets wait a little bit longer as i can see there being either a memory boost to the chip e.g 384 or maybe a return to 512 ring bus?

Or a dramatic improvement in tessalation in a response to nvidias domination

---

Yes 3870 and 4870 were both in 55nm but then again 38xx was a copy of 29xx and there was not lot changed it was a simple die shrunk where as 48xx was a redesign.

As it is the 58xx are not copies of the 48xx and this means that the 58xx were also redesigns, the 68xx has to be radically redesigned from the 58xx to achieve any spectacular results but at the same nm it will be rather hard to do.

I agree that tessellation might be improved quite a bit.

Quote:

---

Originally Posted by Calmatory

What else would you expect? The new generation is still based on 40 nm process, so there isn't much they can do. They have the ~same transistor budget, they have to base their architecture on Evergreen and they only had approx 12 months to come this far. So really, what else can people expect?

They need to do whatever they can to minimize the costs, and if it is working under the limitations of pin compatibility with current gen, then let it be so. The chips will be bigger than current ones for sure.

What I believe is that this news is just to mislead the competition. Now it just gives an impression that it's nothing interesting, just a similar performing updated silicon with possibly lower price. But given the old rumours of focusing on efficiency of the chip, given that the 40 nm process should be pretty much mastered(I know.. :)) by TSMC and those of AMD who work with them, they could achieve somewhat better perf/mm² for the uarch and somewhat better clockspeeds while keeping same yields.

I have hard time believing in "just" an updated silicon to minimize manufacturing costs. Then again, there isn't much what Nvidia could come up with in 40nm so AMD isn't really in position where it would desperately need more speed.. So the next real round will be fought at 28nm and thats where both companies focus.

---

not all transistors are created equal. there is a lot of room for improvements.

Quote:

---

Originally Posted by ajaidev

Yes 3870 and 4870 were both in 55nm but then again 38xx was a copy of 29xx and there was not lot changed it was a simple die shrunk where as 48xx was a redesign.

---

I thought that the HD2-4xxx were basic reiterations with improvements and only the HD5xxx was the next complete redesign?

2900 was a big change
3870 was a shrink
4870 was similar with more SPs packed together
5800s were just doubled

in theory it sounds like SI is about the same as 3870 to 4870 as far as architecture changes, not sure if the improvements will be as great, but just amount of things changing it seems like a similar comparison.

if the TDP of Cayment is correct and the performance scales fairly well per watt then it will be very powerful

but also very hot and noisy :P

Quote:

---

Originally Posted by Florinmocanu

Looks interesting. Basically they are going for 6770/6870/6970.

---

And not a very big performance boost compared to the cards they will replace. For a new generation, HD6000 really seems like it will be underwhelming. The good part is that hopefully they will drive down prices of HD5000 cards.

Well ill be looking forward to seeing what teh HD6xxx is like. I can see them being like the HD3xxx was to the HD2xxx a short but sweet update

FWIW
HD2900 released May 14 2007
HD 3870 released Nov 19 2007

that price drop was from process changes, not architecture.

AMD already can drop prices for 5800s by a massive amount, but have decided not too.
if this new arch is going to reduce the chip size but leave the same performance, I WILL NOT BE EXPECTING A PRICE DROP, unless nvidia lowers first, why would AMD care to get more market share when they can just make killer margins.

Hope it has 3 timers like the Sapphire 5770 Flex.

Quote:

---

Originally Posted by PaganII

Hope it has 3 timers like the Sapphire 5770 Flex.

---

with the desire for multi monitor support, but lack of adoption, i think that would be a great idea to include it globally

Quote:

---

Originally Posted by Panther_Seraphi

I thought that the HD2-4xxx were basic reiterations with improvements and only the HD5xxx was the next complete redesign?

---

Quote:

---

Originally Posted by Manicdan

2900 was a big change
3870 was a shrink
4870 was similar with more SPs packed together
5800s were just doubled

in theory it sounds like SI is about the same as 3870 to 4870 as far as architecture changes, not sure if the improvements will be as great, but just amount of things changing it seems like a similar comparison.

---

By redesign i mean space redesign not architecture. The 48xx and 58xx are reiterations of 38xx with added features but they were more extensive space redesigns than the 28xx to 38xx.

To achieve stealer performance compared to the 58xx the GPU has to be space redesigned again and in limited space "due to 55nm" a dynamic addition cant be made.

5970's has 3200 shaders and TDP is <300w so this new wonder single chip that is suppose to have 300w TDP seems to be well using different type of shaders because combining 3000-3200 shaders under on chip on 40nm will be very hard to do.

Well this appears to explain alot but I am sure that the ATi fans wont be happy by what this means. It appears we are having a de-evolution of GFX cards from higher complexity cards to lower cost cards. Case in point is highlighted by this info.

Example:

GTX480
--->GTX460 (lower cost, equivalent peformance high end is GF104X2)

HD5870
--->HD6770 (lower cost, equivalent peformance high end is a dual card)

The real question now is this, will their card beat nVidia's 384 shader GF104?

the 40nm chips were all about per, not price, a 4870x2 is not much different from a 5870, except when the 5870 came out it cost about as much as 4870x2 and consumed less power

the end of 40nm cards better have come down in price

I would buy any GPU codenamed Bear Beard

Are there any games coming out before 28 nm which wouldn't be "EPICLY DESTROYED IN FPS" by the current gen high end cards? Are there many of those in target audience who have FPS issues with current gen cards?

..of course even if there were no new games released since Q3 people would buy new hardware "Heck 2500 FPS is way better than 2300 FPS, makes you jump higher, hit better and finally gets rid of stuttering".

multi monitors might do the trick

Right. I'd bet tere are way less people using multiple monitors for gaming than there are people using CF. And CF is "UTTERLY OWNED IN MARKET SHARE" by SLI based on Steam hardware survey. Niche markets. Very niche markets, which pretty much mean nothing in the big picture.

Quote:

---

Originally Posted by Calmatory

Right. I'd bet tere are way less people using multiple monitors for gaming than there are people using CF. And CF is "UTTERLY OWNED IN MARKET SHARE" by SLI based on Steam hardware survey. Niche markets. Very niche markets, which pretty much mean nothing in the big picture.

---

I wouldn't base my opinion on Steam, which seems to bundled a lot with the Way its Meant to be Payed games.

Furthermore if my graphic card would have enough power to run the newest games with AA on 3 displays, then I would consider that over multi-GPU setups.

where does it show CF/SLI on steam hardware survey?

And eye6 was in very few samples, and very late, i waited a month then bought a 5870 "classic", no i'm gonna buy 6870 eye6 for a tri monitor setup.

I think many of 4870 owners fall in love of eyefinity when it was released, but hardware in VGA and display was not really ready and not easily found on the webshops. Maybe yes in USA, but not in Europe.

In 2 months, eyefinity setup should start better.

Quote:

---

Originally Posted by Manicdan

where does it show CF/SLI on steam hardware survey?

---

It is in STATS --> http://store.steampowered.com/hwsurvey

Edit: could not find it???

Quote:

---

Originally Posted by PaganII

It is in STATS --> http://store.steampowered.com/hwsurvey

Edit: could not find it???

---

sorry but i dont see the word STATS jumping out anywhere, and i dont see multi gpu configurations listed anywhere

Quote:

---

Originally Posted by Manicdan

sorry but i dont see the word STATS jumping out anywhere, and i dont see multi gpu configurations listed anywhere

---

http://store.steampowered.com/hwsurvey/?platform=pc

ctrl+f "Multi-GPU Systems (PC)" :yepp:

now its clear, you have to be on the Windows version to see it

Copy & Paste from Steam Survey


Multi-GPU Systems (PC)
NVIDIA SLI (2 GPUs)
91.06%
-1.81%

NVIDIA SLI (2 GPUs)
91.06%
-1.81%

ATI Crossfire (2 GPUs)
5.60%
+1.16%

NVIDIA SLI (4 GPUs)
1.81%
+0.30%

NVIDIA SLI (3 GPUs)
1.54%
+0.36%

would be really nice if they had % of people using multi gpu setups, thats what i was hoping to see, and that ratio just looks way too off. how is there 0% using 3 or 4 ati cards, i wonder if it counts X2 cards the same way it counts crossfire

also notice that ATI went up 1.16% in the last month, which means it got 25% more duel gpu setups last month alone (sounds really fishy)

durp, i read this article wrong.

Quote:

---

Originally Posted by gurusan

I would buy any GPU codenamed Bear Beard

---

:rofl: hahahahh

Quote:

---

Originally Posted by Manicdan

that price drop was from process changes, not architecture.

AMD already can drop prices for 5800s by a massive amount, but have decided not too.
if this new arch is going to reduce the chip size but leave the same performance, I WILL NOT BE EXPECTING A PRICE DROP, unless nvidia lowers first, why would AMD care to get more market share when they can just make killer margins.

---

Either you disagree with the article or you are saying that AMD will release a product with higher TDP but same performance.

That's just incoherent.

Quote:

---

Originally Posted by Dimitriman

Either you disagree with the article or you are saying that AMD will release a product with higher TDP but same performance.

That's just incoherent.

---

read my first post
i believe that AMD will release a product with the same perf, and the same power draw as a 5850ish, but it will be cheaper for them to make, and that i believe we will not see much of that savings passed on to us "just because" i think it will be some high price until nvidia responds

(i didnt mention TDP in that post, where did that assumption come from?)

Quote:

---

Originally Posted by Manicdan

read my first post
i believe that AMD will release a product with the same perf, and the same power draw as a 5850ish, but it will be cheaper for them to make, and that i believe we will not see much of that savings passed on to us "just because" i think it will be some high price until nvidia responds

(i didnt mention TDP in that post, where did that assumption come from?)

---

I see

the article mentions a slight bump in TDP for the new parts and meanwhile you said you expected similar performance.

I think if the replacement to the 5970 will touch 300W its going to bring a big punch in performance for those extra 50W-75W.

But I think mostly they will scale down 5870 with some tweaks on Tesselation to the 3rd fastest and release a top bin chip with higher clocks as the new 6870. I can imagine that the new 6970 will probably be a dual 5870 with minor tweaks also.

Quote:

---

Originally Posted by Manicdan

read my first post
i believe that AMD will release a product with the same perf, and the same power draw as a 5850ish, but it will be cheaper for them to make, and that i believe we will not see much of that savings passed on to us "just because" i think it will be some high price until nvidia responds

(i didnt mention TDP in that post, where did that assumption come from?)

---

I'm pretty sure the new parts will be faster as they intend to use more transistors. The thing is they might not be faster in ways you'd expect, they could improve crossfire, tessellation and tweak little things.
But whatever happens I think AMD will be pushing limits rather than cutting costs as these GPU's have to hold out for until for at least 6 months and they shouldn't leave nvidia that much breathing room.

As price drops on the 5870, I imagine ATI be will switching production soon to their new cards. So 5870's will be dwindling stock, and in my opinion, it wouldn't be safe to make enough to justify a price drop. They can just slowly sell what they have sitting around as they prepare for a hard launch.

im only basing theses estimates by the horribly translated article which says between 5770 and 5870, which is a very wide margin

Quote:

---

Originally Posted by Iconyu

I'm pretty sure the new parts will be faster as they intend to use more transistors. The thing is they might not be faster in ways you'd expect, they could improve crossfire, tessellation and tweak little things..

---

This.


Its been stated since day one Southern Islands was a stop gap between Evergreen and Northern Islands. It's been said that it would take the core from evergreen and match it with the uncore of northern islands. So based on that alone, I think most of the performance improvements will be in other areas other than just raw power. Logically it makes sense.

Maybe this analogy - kinda like the same CPU core getting a better chipset with more features later on its life.

...and some recent news claimed that SI is NI done on 40nm.

Why would AMD put lots of effort on yet another 40 nm product line, when their current full line up goes unmatched at some segments and when 28nm process isn't that far away?

So I see this last iteration very boring, with no reason why this would be nothing but a refresh to remain more competitive. Aimed for those who don't have 5xxx series cards yet.

So 150 W < Barts < 225 W and 225 W ≤ Cayman < 300 W. In that case I doubt Antilles would be 2 Caymans.

Also it looks like Cayman will have 256-bit bus instead of the possibility of 384-bit bus as previously rumored.

From some really quick and dirty calculations, 5770 to 5870 is 2.0x the SPUs with 1.74x the TDP. That gives ≈1380 SPUs for 162 W Barts (I know the SPUs will probably be arranged differently, but…) and ≈2440 SPUs for 250 W Cayman. The close "nice" numbers given a 3+1 or 4+0 setup would be 1280 and 2560 SPUs. But then Cayman may be huge, and there would be no point to a dual-Barts Antilles (unless Cayman clocks were horrible). I don't think the above progression would continue as well with big GPUs, so maybe Cayman will have 1920 SPUs.

If so, the gap between 1280 and 1920 is less than the gap between 800 and 1600 so that with the 256-bit bus rumor for Barts could mean AMD is trying to push them closer together than Juniper and Cypress are.

Now none of this takes into account the other architectural changes of Southern Islands.

Quote:

---

Originally Posted by gurusan

I would buy any GPU codenamed Bear Beard

---

:clap::rofl::rofl::clap:

Quote:

---

Originally Posted by Calmatory

...and some recent news claimed that SI is NI done on 40nm.

Why would AMD put lots of effort on yet another 40 nm product line, when their current full line up goes unmatched at some segments and when 28nm process isn't that far away?

So I see this last iteration very boring, with no reason why this would be nothing but a refresh to remain more competitive. Aimed for those who don't have 5xxx series cards yet.

---

Right now it looks like nvidia have a decent plan for the next few months, they are farming what I imagine to be 475 cores and selling the 460 cores cheap to maintain market share.

ATI have to do something, as a full shader dual 460 card would eat the 5970 alive and such a plan easily falls in line with the path nvidia choose with the 295. A tweaked card can defend against it, and deny nvidia revenue. But that means at least changing the uncore.

Also ATI would not bother making a new dual card unless the refresh changes mean they'll gain some of the performance they lost from the shaders in this generation i.e. 5870 has the silicon but not the speed of crossfire 4890's, while the 5970 (stand alone) and 5870 in crossfire seem to be under performing considering the brute force behind them.

as time goes on the 5800s will get weaker looking with the lackluster dx11 perf
but currently i dont think the 460s are really scaring them, otherwise we would have seen a real price drop (talking 10% at once, not 2% per month for what, 1 month so far, lol, 2010 sucked for gpus)

Quote:

---

Originally Posted by Manicdan

as time goes on the 5800s will get weaker looking with the lackluster dx11 perf
but currently i dont think the 460s are really scaring them, otherwise we would have seen a real price drop (talking 10% at once, not 2% per month for what, 1 month so far, lol, 2010 sucked for gpus)

---

Yeah I'm dreaming that the new single gpu ATI card or new Nvidia card that could actually beat the 5970 at 2560x1600.

Am I mistaken or are we really gonna care about more power when DX11 games are still just starting to show up AND the current gen is actually plenty for what's out there? What's the popular standard out there? 1920x1080, not 2560?

If this is true then we (I) should be able to use 2 x dvi + 1 x hdmi no?

For those of you that missed it...

Quote:

---

Originally Posted by spursindonesia

hmm, may i speculate ? Bart would be a 1280 SP (320*4) 16 ROP 128 bit chip 240-260 mm^2 performance ~HD 5850, Cayman would be a 1920 SP (480*4) 32 ROP 256 bit chip 390-420 mm^2 performance ~GF100 512 SP+, and dual Cayman power consumption would be very, very close to PCIE standard limit of 300 w, unlike HD 5970 TDP which is rather conservatively rated, or would simply be underclocked further compared to its older dual chips kamerad. This speculation is reasonably explaining the rumored TDP of the chips reported in the article.

---

Sounds spot on to me other than the fact that Barts should be 256bit since Turks is the 128bit part.
http://www.4gamer.net/games/110/G011...093/TN/002.jpg
The above was provided by neliz.

Quote:

---

Originally Posted by Tenknics

This.


Its been stated since day one Southern Islands was a stop gap between Evergreen and Northern Islands. It's been said that it would take the core from evergreen and match it with the uncore of northern islands. So based on that alone, I think most of the performance improvements will be in other areas other than just raw power. Logically it makes sense.

Maybe this analogy - kinda like the same CPU core getting a better chipset with more features later on its life.

---

correct.

not only that, but AMD is not aiming at ultra performance (80~120fps).

I suspect these new (half-breed?) cards, might not be any faster than their brethren, but their impedance, or minimum frames per second is constantly higher. Trading off pure performance for intelligent/efficient performance.
(?)

So according to that, only HD5470-HD5670 replacements in Q4/10?

Quote:

---

Originally Posted by Calmatory

So according to that, only HD5470-HD5670 replacements in Q4/10?

---

This roadmap is sh*t from a guy that set his name bottom right.

The true can't be near this. Why ? Just because Dirk Meyer, the CEO said the full new line up will be delivered to market before the end of the year.

So it's very bad and old speculation.

Current best is the ATI forum.de that offer the best idea on what's going next. The roadmap from Honma is just a joke

Quote:

---

Originally Posted by Xoulz

correct.

not only that, but AMD is not aiming at ultra performance (80~120fps).

I suspect these new (half-breed?) cards, might not be any faster than their brethren, but their impedance, or minimum frames per second is constantly higher. Trading off pure performance for intelligent/efficient performance.
(?)

---

LOL fanboy comment ...


seriously these threads make me lol

Quote:

---

Originally Posted by LordEC911

For those of you that missed it...



Sounds spot on to me other than the fact that Barts should be 256bit since Turks is the 128bit part.
http://www.4gamer.net/games/110/G011...093/TN/002.jpg
The above was provided by neliz.

---

Yeah the buswidth and the ROP seems quite puzzling to me up to this moment. I think if ATi push for 384 bit for Cayman & 256 bit for Bart, they won't be pad limited, but the die size might grow excessively since 384 bit means 48 ROP & 256 bit 32 ROP consecutively. But with the slowing of GDDR speed increasing pace & the added die size from core improvement using the same process node, it's quite reasonable for them using the larger buswidth & the associated added ROP, if the chip performance justifies that. OTOH, that would mean the end of sweetspot strategy, or not ? :confused: aarrggh, i should stop speculating by now, a layman shouldn't think too much out of his league.

Quote:

---

Originally Posted by spursindonesia

Yeah the buswidth and the ROP seems quite puzzling to me up to this moment. I think if ATi push for 384 bit for Cayman & 256 bit for Bart, they won't be pad limited, but the die size might grow excessively since 384 bit means 48 ROP & 256 bit 32 ROP consecutively. But with the slowing of GDDR speed increasing pace & the added die size from core improvement using the same process node, it's quite reasonable for them using the larger buswidth & the associated added ROP, if the chip performance justifies that. OTOH, that would mean the end of sweetspot strategy, or not ? :confused: aarrggh, i should stop speculating by now, a layman shouldn't think too much out of his league.

---

I assume the "sweet spot strategy" may only apply to a new process. If you want to increase performance on the same process they will have to go a bit bigger. If they want a noticeable effect. I am sure you will see that sweet spot next for 28nm.

Quote:

---

Originally Posted by sn0wm@n

lol fanboy comment ...


Seriously these threads make me lol

---

+500

Quote:

---

Originally Posted by Manicdan

theres really only 2 things to look at
perf per watt = efficiency
perf per mm2 = costs

it seems that only 1 of those is really changing by much. which isnt bad considering its on the same process and we cant really expect too much with perf/watt (unless its like nvidia where they chopped off the server stuff to shave power for the gf104)

---

very good post :toast:

Quote:

---

Originally Posted by spursindonesia

Yeah the buswidth and the ROP seems quite puzzling to me up to this moment. I think if ATi push for 384 bit for Cayman & 256 bit for Bart, they won't be pad limited, but the die size might grow excessively since 384 bit means 48 ROP & 256 bit 32 ROP consecutively. But with the slowing of GDDR speed increasing pace & the added die size from core improvement using the same process node, it's quite reasonable for them using the larger buswidth & the associated added ROP, if the chip performance justifies that. OTOH, that would mean the end of sweetspot strategy, or not ? :confused: aarrggh, i should stop speculating by now, a layman shouldn't think too much out of his league.

---

how could it be 384bit if its pin to pin compatible with the 5000 series? maybe not pin to pin but i thought pcbs could be reused, cutting down ttm big time and reducing costs?

Quote:

---

Originally Posted by saaya

how could it be 384bit if its pin to pin compatible with the 5000 series? maybe not pin to pin but i thought pcbs could be reused, cutting down ttm big time and reducing costs?

---

Perhaps it's not Cayman which is pin compatible with Cypress, but Bart instead. A 256 bit 32 ROP 1280 SP (320*4) Cayman might end up around 270-290 mm^2, with perhaps ~HD 5850-5870 performance, it should be a pretty potent competitor for GF 104 derivatives chip in US$ 180-220 price segment.

I see Bart as a HD 4770 vs HD 4850, but instead using new process node to gain added efficiency (perf/watt or die area), it uses new mArch development such as rumored shader/SIMD rearrangement of the current ATi's GPU mArch. Pretty similar to nVidia's approach in GF 104 development that succesfully beats its bigger brother in perf/watt & die area while still using the same proces node.

makes sense, and ati had more time to tweak their design than nvidia...
and they have more experience with tweaking 40nm and getting the most out of every transistor than nvidia as well it seems...

but looking at the specs...
1280sps will be enough to reach 5850s 1440sp performance?
that would be a performance boost of around 12%?
thats not too much... you could get that with clockspeeds alone :D
765->850mhz...
so then whats the point of rolling out new silicon?
cause its a 10% smaller chip?
idk... this sounds weird... that wouldnt really make sense...

if those die sizes are correct, then performance HAS to be higher... otherwise it wouldnt make sense to roll out new silicon...
it has to be at least as fast as a 5870... at least...

think about it, a 512sp 480 is how much faster than a 5870?
bart should have roughly the same perf boost as cayman, so it wouldnt be going from 5830 to 5850, thats a rather small boost... it should go from 5830 to 5870+

also consider the lengths of ATI cards...i have a feeling we will be seeing short cards like the 460's

Quote:

---

Originally Posted by spursindonesia

Perhaps it's not Cayman which is pin compatible with Cypress, but Bart instead. A 256 bit 32 ROP 1280 SP (320*4) Cayman might end up around 270-290 mm^2, with perhaps ~HD 5850-5870 performance, it should be a pretty potent competitor for GF 104 derivatives chip in US$ 180-220 price segment.

I see Bart as a HD 4770 vs HD 4850, but instead using new process node to gain added efficiency (perf/watt or die area), it uses new mArch development such as rumored shader/SIMD rearrangement of the current ATi's GPU mArch. Pretty similar to nVidia's approach in GF 104 development that succesfully beats its bigger brother in perf/watt & die area while still using the same proces node.

---

Yeah, it sounds plausible spursindonesia :yepp:

Bart will use 5870 PCB and is pin compatible if you read corectly ATI forums.de

Quote:

---

Originally Posted by saaya

but looking at the specs...
1280sps will be enough to reach 5850s 1440sp performance?
that would be a performance boost of around 12%?
thats not too much... you could get that with clockspeeds alone :D
765->850mhz...
so then whats the point of rolling out new silicon?
cause its a 10% smaller chip?
idk... this sounds weird... that wouldnt really make sense...

---

From what it sounds like and if the changes are true, a 1280SP Barts sound be performing around a 5850-5870. A 1280SP 4d design should perform very similar to a 1600SP 5d design, since they supposedly aren't often utilizing the 5th SP in gaming situations and that they are touting "efficiency" as their main goal.

Neliz was saying something quite awhile ago about the 5850 replacement performing around a 5870. Cayman will be taking on the 512SP GF100 with the cutdown/salvage part being slightly ahead of the current 480SP GF100. Think about it, you need around a 20-25% performance increase over 5870 to match/beat GTX480 in most situations. Cayman needs a 10-20% increase over that to match/beat a 512SP GF100, if it stays around 700-750mhz.

So all that math above comes out to a fully specced Cayman needing to be roughly 35-40% faster to fall within those guidelines, which is what I was originally speculating awhile ago.

On a side note, the info from the roadmap is a good 4 years old but the codenames are correct. Also, the dates are there for mass availability, do not base launch/release dates off the roadmap. Most early roadmaps like these usually give a quarter or so for leeway.

Quote:

---

Originally Posted by saaya

makes sense, and ati had more time to tweak their design than nvidia...
and they have more experience with tweaking 40nm and getting the most out of every transistor than nvidia as well it seems...

but looking at the specs...
1280sps will be enough to reach 5850s 1440sp performance?
that would be a performance boost of around 12%?
thats not too much... you could get that with clockspeeds alone :D
765->850mhz...
so then whats the point of rolling out new silicon?
cause its a 10% smaller chip?
idk... this sounds weird... that wouldnt really make sense...

if those die sizes are correct, then performance HAS to be higher... otherwise it wouldnt make sense to roll out new silicon...
it has to be at least as fast as a 5870... at least...

think about it, a 512sp 480 is how much faster than a 5870?
bart should have roughly the same perf boost as cayman, so it wouldnt be going from 5830 to 5850, thats a rather small boost... it should go from 5830 to 5870+

---

I dunno Saaya, i think every mm^2 counts especially if the chip is gonna be used in sub-200 US$ segment, and considering supply problem that has been plaguing ATi for since the launch of Evergreen family, this will help alleviating that problem somehow.

Regarding performance, i think ~HD 5850 can be counted as being conservative, with 320*4 config, if this rearrangement is successfull, we can expect ~HD 5870 like performance in many occasions, and good performance is not only determined by max & avg FPS, i expect some decent improvements in min FPS & a competitive tesselation performance in these upcoming chips from ATi.

Well, i guess these speculations are rather leaning on an optimist side, but considering what ATi has achieved since HD 2900 debacle, the optimism is quite warranted. In this scenario, nVidia's only competitive in the mainstream segment (GF 104 is pretty good chip, especially if the full spec one can be produced in good quantity), but would have much less profit & incentive to play the price war game in maintaining market share.

http://forum.beyond3d.com/showpost.p...postcount=1318

Quote:

---

According to the latest Catalyst 10.8 Antilles is Cayman based.

By the way, the driver reveals also the relations between marketing and engineering Codenames and all of the new "Southern Islands" GPUs have the prefix "NI" to it, e.g. "NI Cayman", "NI Barts", or "NI Caicos". So it appears SI is really NI in 40nm.
And some additional thing is that the mobile versions will have Pro, XT, and LP versions, whereas one has XT, Pro and sometimes LE versions for the desktop. And also the mobile chips sometimes appear as Gemini (X2?) variants.

---

Quote:

---

223,CAYMAN GL XT (6700),NI CAYMAN
224,CAYMAN GL XT (6701),NI CAYMAN
225,CAYMAN GL XT (6702),NI CAYMAN
226,CAYMAN GL XT (6703),NI CAYMAN
227,CAYMAN GL PRO (6704),NI CAYMAN
228,CAYMAN GL PRO (6705),NI CAYMAN
229,CAYMAN GL (6706),NI CAYMAN
230,CAYMAN GL LE (6707),NI CAYMAN
231,CAYMAN GL (6708),NI CAYMAN
232,CAYMAN GL (6709),NI CAYMAN
233,CAYMAN XT (6718),NI CAYMAN
234,CAYMAN PRO (6719),NI CAYMAN
235,ANTILLES PRO (671C),NI CAYMAN
236,ANTILLES XT (671D),NI CAYMAN
237,BLACKCOMB XT/PRO (6720),NI BLACKCOMB
238,BLACKCOMB LP (6721),NI BLACKCOMB
239,BLACKCOMB XT/PRO Gemini (6724),NI BLACKCOMB
240,BLACKCOMB LP Gemini (6725),NI BLACKCOMB
241,BARTS GL XT (6728),NI BARTS
242,BARTS GL PRO (6729),NI BARTS
243,BARTS XT (6738),NI BARTS
244,BARTS PRO (6739),NI BARTS
245,WHISTLER XT (6740),NI WHISTLER
246,WHISTLER PRO/LP (6741),NI WHISTLER
247,WHISTLER XT/PRO Gemini (6744),NI WHISTLER
248,WHISTLER LP Gemini (6745),NI WHISTLER
249,ONEGA (6750),NI TURKS
250,TURKS XT (6758),NI TURKS
251,TURKS PRO (6759),NI TURKS
252,SEYMOUR XT/PRO (6760),NI SEYMOUR
253,SEYMOUR LP (6761),NI SEYMOUR
254,SEYMOUR XT/PRO Gemini (6764),NI SEYMOUR
255,SEYMOUR LP Gemini (6765),NI SEYMOUR
256,CAICOS GL PRO (6768),NI CAICOS
257,CASPIAN PRO (6770),NI CAICOS
258,CAICOS PRO (6779),NI CAICOS

---

Quote:

---

Originally Posted by tbone8ty

also consider the lengths of ATI cards...i have a feeling we will be seeing short cards like the 460's

---

Id really like that. GTX 460 / HD 5770 are about as large as I want my cards to be.

I dont know if its just me, but when it comes to graphics cards, I am not a size queen.

Quote:

---

Originally Posted by mindfury

http://forum.beyond3d.com/showpost.p...postcount=1318

Quote:

---

According to the latest Catalyst 10.8 Antilles is Cayman based.

By the way, the driver reveals also the relations between marketing and engineering Codenames and all of the new "Southern Islands" GPUs have the prefix "NI" to it, e.g. "NI Cayman", "NI Barts", or "NI Caicos". So it appears SI is really NI in 40nm.
And some additional thing is that the mobile versions will have Pro, XT, and LP versions, whereas one has XT, Pro and sometimes LE versions for the desktop. And also the mobile chips sometimes appear as Gemini (X2?) variants.

---

Quote:

---

223,CAYMAN GL XT (6700),NI CAYMAN
224,CAYMAN GL XT (6701),NI CAYMAN
225,CAYMAN GL XT (6702),NI CAYMAN
226,CAYMAN GL XT (6703),NI CAYMAN
227,CAYMAN GL PRO (6704),NI CAYMAN
228,CAYMAN GL PRO (6705),NI CAYMAN
229,CAYMAN GL (6706),NI CAYMAN
230,CAYMAN GL LE (6707),NI CAYMAN
231,CAYMAN GL (6708),NI CAYMAN
232,CAYMAN GL (6709),NI CAYMAN
233,CAYMAN XT (6718),NI CAYMAN
234,CAYMAN PRO (6719),NI CAYMAN
235,ANTILLES PRO (671C),NI CAYMAN
236,ANTILLES XT (671D),NI CAYMAN
237,BLACKCOMB XT/PRO (6720),NI BLACKCOMB
238,BLACKCOMB LP (6721),NI BLACKCOMB
239,BLACKCOMB XT/PRO Gemini (6724),NI BLACKCOMB
240,BLACKCOMB LP Gemini (6725),NI BLACKCOMB
241,BARTS GL XT (6728),NI BARTS
242,BARTS GL PRO (6729),NI BARTS
243,BARTS XT (6738),NI BARTS
244,BARTS PRO (6739),NI BARTS
245,WHISTLER XT (6740),NI WHISTLER
246,WHISTLER PRO/LP (6741),NI WHISTLER
247,WHISTLER XT/PRO Gemini (6744),NI WHISTLER
248,WHISTLER LP Gemini (6745),NI WHISTLER
249,ONEGA (6750),NI TURKS
250,TURKS XT (6758),NI TURKS
251,TURKS PRO (6759),NI TURKS
252,SEYMOUR XT/PRO (6760),NI SEYMOUR
253,SEYMOUR LP (6761),NI SEYMOUR
254,SEYMOUR XT/PRO Gemini (6764),NI SEYMOUR
255,SEYMOUR LP Gemini (6765),NI SEYMOUR
256,CAICOS GL PRO (6768),NI CAICOS
257,CASPIAN PRO (6770),NI CAICOS
258,CAICOS PRO (6779),NI CAICOS

---

---

Well, those are all islands in the northern hemisphere...

Which means that it's NI that's coming first, not SI. We were never sure which one is going to come first.

Quote:

---

Originally Posted by LordEC911

From what it sounds like and if the changes are true, a 1280SP Barts sound be performing around a 5850-5870. A 1280SP 4d design should perform very similar to a 1600SP 5d design, since they supposedly aren't often utilizing the 5th SP in gaming situations and that they are touting "efficiency" as their main goal.

---

so their chips are more efficient than nvidias, but 20% of their sps are actually idle all the time? idk... sounds too good to be true if you ask me :D
but i hope its true...

Quote:

---

Originally Posted by LordEC911

Neliz was saying something quite awhile ago about the 5850 replacement performing around a 5870. Cayman will be taking on the 512SP GF100 with the cutdown/salvage part being slightly ahead of the current 480SP GF100. Think about it, you need around a 20-25% performance increase over 5870 to match/beat GTX480 in most situations. Cayman needs a 10-20% increase over that to match/beat a 512SP GF100, if it stays around 700-750mhz.

So all that math above comes out to a fully specced Cayman needing to be roughly 35-40% faster to fall within those guidelines, which is what I was originally speculating awhile ago.

---

yeah but for a 40% boost you need more than a 40% spec boost... im a bit sceptical about them getting there... i think they will fall short of beating a 512sp gf100, thats why theyre refreshing the 5970 so soon :D
as for bart... sounds interesting for sure... i wouldnt mind a 5870 for 200$... 460? whats a 460? :D

Quote:

---

Originally Posted by LordEC911

On a side note, the info from the roadmap is a good 4 years old but the codenames are correct. Also, the dates are there for mass availability, do not base launch/release dates off the roadmap. Most early roadmaps like these usually give a quarter or so for leeway.

---

well i dont think anybody is in a hurry to upgrade... actually im curious if there will ever been a real push for upgrades again like weve seen it in the past, since there is no game on the horizon thats going to bring the current hw to its knees... and you need at LEAST 1 really good game that does that to create demand... :shrug:

Quote:

---

Originally Posted by spursindonesia

I dunno Saaya, i think every mm^2 counts especially if the chip is gonna be used in sub-200 US$ segment, and considering supply problem that has been plaguing ATi for since the launch of Evergreen family, this will help alleviating that problem somehow.

Regarding performance, i think ~HD 5850 can be counted as being conservative, with 320*4 config, if this rearrangement is successfull, we can expect ~HD 5870 like performance in many occasions, and good performance is not only determined by max & avg FPS, i expect some decent improvements in min FPS & a competitive tesselation performance in these upcoming chips from ATi.

Well, i guess these speculations are rather leaning on an optimist side, but considering what ATi has achieved since HD 2900 debacle, the optimism is quite warranted. In this scenario, nVidia's only competitive in the mainstream segment (GF 104 is pretty good chip, especially if the full spec one can be produced in good quantity), but would have much less profit & incentive to play the price war game in maintaining market share.

---

i think your right... thx a lot for sharing your thoughts :toast:
and yes, every mm2 counts but i dont think anybody would launch a new part if all it does is safe them around 10% die space...
thats not feasible i think...

Quote:

---

Originally Posted by LordEC911

Neliz was saying something quite awhile ago about the 5850 replacement performing around a 5870. Cayman will be taking on the 512SP GF100 with the cutdown/salvage part being slightly ahead of the current 480SP GF100. Think about it, you need around a 20-25% performance increase over 5870 to match/beat GTX480 in most situations. Cayman needs a 10-20% increase over that to match/beat a 512SP GF100, if it stays around 700-750mhz.

---

How would 6% increase in shaders get nVidia 10-20% performance with a full GF100?

I'd say 30% over 5870 is enough to match a full GTX480.

At the very most..

On average, A 480 is not 20-25% faster than a 5870.

ATI Southern Islands codenames spotted in Catalyst 10.8

http://vr-zone.com/articles/ati-sout...10.8/9684.html

Quote:

---

Originally Posted by mAJORD

At the very most..

On average, A 480 is not 20-25% faster than a 5870.

---

lets see...
http://www.xbitlabs.com/articles/vid..._17.html#sect0
let me do the math...

1920x1080 4aa/16af
17.57%

at 2560x1600 the lead is bigger, but many games arent really playable at that res, even with a 480...

Quote:

---

Originally Posted by Gilgamesh

ATI Southern Islands codenames spotted in Catalyst 10.8

http://vr-zone.com/articles/ati-sout...10.8/9684.html

---

somebody posted about that in this thread before it was on vrzone :P
and if its southern island then why is it all called NI? :D

Quote:

---

Originally Posted by saaya

so their chips are more efficient than nvidias, but 20% of their sps are actually idle all the time? idk... sounds too good to be true if you ask me :D
but i hope its true...

---

it is true, when people ran furmark with the 4800s and flipped out when they got really high temperature and power draw, my response was compare the framerate, it was SO much higher than the competitions. Watts per Frame was really good.

those 5 shader pairs are rarely used in games, some know how to push them (mass effect is one of those games, found out my card sucked playing that game cause it pushed it harder than other games i played)

also those smaller SPs are just that, smaller, the die space cost to add that 5th is not taking up 20% of the chips space, some of us did some fun math of comparing the performance gains by going from 5 to 4 and the size increase that would come with it based on different ratios.

i wonder what would have happened if 4870 was only 4 deep instead, maybe it wouldnt have overheated in furmark, maybe it would have stomped all other chips.

Quote:

---

Originally Posted by blindbox

Which means that it's NI that's coming first, not SI. We were never sure which one is going to come first.

---

Another point of view is that if SI really is just NI but on 40nm the drivers would be exactly the same, or ATI might just have decided it was easier to keep the name NI so they wouldn't have to pull out a atlas for a few months.

Quote:

---

Originally Posted by -Boris-

How would 6% increase in shaders get nVidia 10-20% performance with a full GF100?

I'd say 30% over 5870 is enough to match a full GTX480.

---

If they up the clocks at all, notice how I mentioned 750mhz...

Quote:

---

Originally Posted by mAJORD

At the very most..

On average, A 480 is not 20-25% faster than a 5870.

---

No, but that's what it might take to handily beat the GTX480 in "most situations."

Quote:

---

Originally Posted by Iconyu

Another point of view is that if SI really is just NI but on 40nm the drivers would be exactly the same, or ATI might just have decided it was easier to keep the name NI so they wouldn't have to pull out a atlas for a few months.

---

its just a name tag, a quick ref... its very possible they just didnt change it or dont care about it...

and if its not, it could just mean that the drivers treats them as NI, even though they share more blocks with the current chips than the actual NI chips... but to the driver it doesnt matter, so the driver treats them as NI...

in the end... does it really matter? :D
if they rock, they rock, and nobody will care if its because a certain building block is new or the same as in the current chips... :D

Quote:

---

Originally Posted by LordEC911

No, but that's what it might take to handily beat the GTX480 in "most situations."

---

true, if its 20-25% faster then even all the TWIMTBP titles running at hundreds of fps wont make a difference and itll come out on top :D

Real or fake?

http://we.pcinlife.com/thread-1498296-1-1.html

http://image163.poco.cn/mypoco/mypho...6234917087.jpg

Quote:

---

Originally Posted by mindfury

Real or fake?

http://we.pcinlife.com/thread-1498296-1-1.html

http://image163.poco.cn/mypoco/mypho...6234917087.jpg

---

haha, they erased deviceid while same info is in GPU field.

so its - 233,CAYMAN XT (6718),NI CAYMAN

why they always erase dev-id, there is nothing secret there....:shrug:

im sure were gonna hear that the gpuz dosnt have the details in their database and it wouldnt show up right anyway

@Mindfury, hmmm, Cayman XT aka HD 6870 scores ~12.000 at Vantage Xtreme setting, nice. It has 850 MHz core & 1.6 GHz mem using 256 bit buswidth, so no 384 bit as earlier speculated, but the mem clock is pretty high isn't it ? If the card is using a 6 Gbps GDDR5 chips, that means already OCed clock on its default setting, CMIIW.

WOW, if that score is true, nVidia will be in the world of hurts then. Around 20% faster than GTX 480 & 35% faster than its predecessor, and not that far away from the current king of the hill dual GPU card, pretty darn impressive !

Somehow ... I don't think so.

Hmm that score somehow looks fishy..

Cheat engine to edit the gpu-z details and a screenshot.

2 minutes to fake

isnt it w1zzard who puts the gpu info into gpu-z?

Well, ATi did pull a nice core performance improvement using the same process node in the past, RV 670 to RV 770 comes to mind, so while the score seems very optimistic & some call unrealistic, IMHO it's not outside the realm of possibility.

What does a GTX480 score? Around ~x10,000?

Quote:

---

Originally Posted by LordEC911

What does a GTX480 score? Around ~x10,000?

---

This is from XBitLabs latest article regarding GTX 460 SLI vs HD 5970:

http://www.xbitlabs.com/images/video.../vdm_total.png

GeForce GTX 460 SLI vs. Radeon HD 5970: Two Against One (page 11)

Honestly that score isn't so out there to be unbelieveable or anything. Really anything less would be a let down for a new generation of gpus. Entirely possible. If anything we should hope it is even higher.

That score should be where its at considering how much time they had to work on this. Anything less than 20% lead over a GTX 480 is failure imo. In other words if you have a 5 series now just skip it and wait for 28nm.

definately a 460 Killer!!

Quote:

---

Originally Posted by Johnny87au

definately a 460 Killer!!

---

That would be Bart/Burt, not Cayman.

Quote:

---

Originally Posted by W1zzard

nothing i can see in the gpuz part of the screenshot that suggests it's fake.

---

http://forums.techpowerup.com/showthread.php?t=129668