AMD shows 28nm wafer GlobalFoundries to meet new card?

Quote:

---

AMD split came from the foundry, today once again demonstrated the GlobalFoundries wafer 28nm process, and not a simple repetition of half a year ago.

2009年6月初，GlobalFoundries刚刚宣布纽约工厂Fab 2即将正式动工，就拿出了45nm、32nm、28nm等多种工艺的晶圆，其中最先进的28nm还支持SR AM Bulk测试品，没有实际用途。 Early June 2009, GlobalFoundries just announced that Fab 2 facility in New York will soon commence work on the come up with 45nm, 32nm, 28nm wafer process etc., which also supports the most advanced 28nm test SRAM Bulk goods, there is no practical purpose.

---

http://translate.google.com/translat...&sl=auto&tl=en


translation is not that great but, it looks like later this year 28nm products will be launched...yay!

So it possible to see HD6000 before xmas, great.

Even a HD5000 improved chip.


---
About the 890GX, is it the same RV620 or a RV710?

This chinese article is translated from semiaccurate....

http://www.semiaccurate.com/2010/01/...wafer-spotted/

Quote:

---

Originally Posted by Nintendork

About the 890GX, is it the same RV620 or a RV710?

---

Since it's capable of hybrid crossfire with cedar, it should be a "RV810".

translation sux!

　　

Quote:

---

GlobalFoundries表示，28nm工艺进展顺利，今年下半年就会正式投产

---

This sentence says that 28nm process is in good progress, second half of this year would be in mass production

Quote:

---

Originally Posted by haylui

　　

This sentence says that 28nm process is in good progress, second half of this year would be in mass production

---

Amazing! Really good news! :D

Lower numbers are always exciting. :) Makes sense for 6Ks due to the timing.

This is ?

excellent news

looking foward to new GPUs :up:

I believe i saw a roadmap somewhere that put the 5890 on 28nm process- Should be the refresh for later this year.

That die looks more like the small gpus you guys mentioned, but it is good to see that they have the process working well enough for production, at least.

God damn it... I was about to hit the buy button on 5870 though 28NM looks even better.
I'll never be able to buy a new Video card at this rate.

Quote:

---

Originally Posted by Kurz

God damn it... I was about to hit the buy button on 5870 though 28NM looks even better.
I'll never be able to buy a new Video card at this rate.

---

You'll be waiting a long time if you hold off.

---

ATI historically has tried new processes on small GPUs like when the 4770 came out. It would make sense that that's the case here as well.

Quote:

---

Originally Posted by Particle

You'll be waiting a long time if you hold off.

---

ATI historically has tried new processes on small GPUs like when the 4770 came out. It would make sense that that's the case here as well.

---

not sure they could do that this time. they already use 128bit on their mid range cards, they cant get that much lower. if anything they might ONLY make mobile gpus on 28nm to test it out, where 64bit super slow gddr5 wont be missed as much.

Seems ATi is preparing Anti-Fermi weapon, maybe will not win by a large margin to fermi (all depends how "overpowered" fermi is going to be) then wham suddenly the marketing team comes in action and says "hey we have 28nm they don't and at a better price too!" and price is the key to success...

Given that the fab process is going well; what does this mean for Dozer? Does this mean Bulldozer is now going to be made on a 28nm fab later this year?

28nm is bulk i think.

Bulldozer will be on 32nm HKMG SOI. AMD and Intel don't make CPUs on half nodes. 45 > 32 > 22

Quote:

---

Originally Posted by Particle

Bulldozer will be on 32nm HKMG SOI. AMD and Intel don't make CPUs on half nodes. 45 > 32 > 22

---

If BUlldozer die is too big, I think AMD will do...........:yepp:
28nm SOI+high-k meta gate.:ROTF:

what's the harm of doing bulldozer on 28nm?

I hope 6K comes out before Christmas, amazing GPU for me and high sales for AMD.

It just won't happen. I can't recall them ever making a CPU on a half node. It's not like GF can just pick up and start doing 28nm HKMG SOI because they have bulk. Nor can AMD simply decide they want to move their design to bulk if it was developed for SOI. It isn't that simple.

Quote:

---

Originally Posted by Helloworld_98

what's the harm of doing bulldozer on 28nm?

---

To build a 28nm product you need to start with a 28nm toolkit and 28nm components.

You can't just shrink a 32nm design to 28nm like you shrink the font in a document.

Quote:

---

Originally Posted by Particle

It just won't happen. I can't recall them ever making a CPU on a half node. It's not like GF can just pick up and start doing 28nm HKMG SOI because they have bulk. Nor can AMD simply decide they want to move their design to bulk if it was developed for SOI. It isn't that simple.

---

But the graphics part of Fusion would be shrink from bulk to SOI, right?

Quote:

---

Originally Posted by Particle

ATI historically has tried new processes on small GPUs like when the 4770 came out. It would make sense that that's the case here as well.

---

The 3870 was the first product on 55nm, was it not?

--Matt

Quote:

---

Originally Posted by mattkosem

The 3870 was the first product on 55nm, was it not?

---

3870 was a midrange GPU.

Quote:

---

Originally Posted by trinibwoy

3870 was a midrange GPU.

---

I believe you're mistaken. 3870 was top end when it was released, and remained there until the 3870 X2 was released.

--Matt

The RV670 was presented like AMD's first piece of silicon from their "sweet point" strategy as a term for small sized die. RV770 was slightly behind this line. RV870 is so much behind AMD had to try 40 nm on something else.

Quote:

---

Originally Posted by mattkosem

The 3870 was the first product on 55nm, was it not?

--Matt

---

as far as i recall yes, it was out very shortly after the 2900

Quote:

---

Originally Posted by trinibwoy

3870 was a midrange GPU.

---

midrange to what? performance was very close to the 2900, but used half the power. the 4870 did beat it, and was bigger and used more power, but came out months later

Quote:

---

Originally Posted by Manicdan

as far as i recall yes, it was out very shortly after the 2900

midrange to what? performance was very close to the 2900, but used half the power. the 4870 did beat it, and was bigger and used more power, but came out months later

---

The 3870 came out 6 months after the 2900, and was a refresh of the R600 on 55nm. Then the 4800's came about 6-8 months after that. Ati was on a pretty consistent release cycle when the process technology was ahead of them...now it seems they're ahead of it.

Quote:

---

Originally Posted by Manicdan

not sure they could do that this time. they already use 128bit on their mid range cards, they cant get that much lower. if anything they might ONLY make mobile gpus on 28nm to test it out, where 64bit super slow gddr5 wont be missed as much.

---

Remember, they didn't exactly go with the largest bus possible for the die size. A 28nm shrink of Cypress they are probably aiming for ~200mm2 and have some room to play with some specs while being able to fit a 256bit bus on there.
28nm Juniper is looking closer to ~100mm2, so they might have to throw some clusters on there to get a 128bit bus.

Well, they did at least architecture progress. NVIDIA just added more and more shaders and shrinked. Now they are supposed to release new architecture, but oh, it's problem, they forgot how to do it after those long years or what :D

On the other hand, AMD does weird things too, they try new things with every new line and sometimes just throw them away right in the another new line. Like they've been just experimenting.

As for 28 nm, I don't think AMD will shrink. They don't do this, they'll more probably release new cards. It's wasting time when you only shrink, I think they learned from NVIDIA. I'd say second half of 2010 is long enough after HD 5000 lanch when you compare it with HD 3000/HD 4000/HD 5000.

Quote:

---

Originally Posted by Behemot

Well, they did at least architecture progress. NVIDIA just added more and more shaders and shrinked. Now they are supposed to release new architecture, but oh, it's problem, they forgot how to do it after those long years or what :D

---

I don't agree. I think it was more like AMD who did that - take 320 sp's, first shrink it, then shrink + 800, then shrink + 1600.

Nvidia didn't even bother adding SP's to 8800 GTX until the GT200 series. They just shrank once. Of course, that they didn't add shaders doesn't mean they improved a lot in architecture :D

would a shrink of r800 really give them any information on how build the r900 at 28nm?

Quote:

---

Originally Posted by annihilat0r

I don't agree. I think it was more like AMD who did that - take 320 sp's, first shrink it, then shrink + 800, then shrink + 1600.

Nvidia didn't even bother adding SP's to 8800 GTX until the GT200 series. They just shrank once. Of course, that they didn't add shaders doesn't mean they improved a lot in architecture :D

---

not really

amd added direct x 10.1; 11 in the progress; made huge improvements in opencl performance; added additional dispay support; redesigned the UVD and power savings; added GDDR5

AMD: DX 10 > 10.1 > 11 (2900 - 3870/4870 - 5870)
Nvidia: DX 10 > 10 > 10 (8800-9800-280)

the last real core improvement with nvidia was G80-G92 with much faster CUDA performance (G80 sucks in CUDA)

i just wish that they finally come out with fermi but considering the fact that there were no news about it for almost 2 weeks things dont look too good...

Quote:

---

Originally Posted by Manicdan

would a shrink of r800 really give them any information on how build the r900 at 28nm?

---

Depends on the architecture and it also depends on if they are refreshing at the end of the year @ 28nm and then release a new architecture in 2011. It could go either way.

The real question is, do they really want to try and release a completely new architecture on a new process node with no small test GPUs first?

Quote:

---

Originally Posted by mattkosem

I believe you're mistaken. 3870 was top end when it was released, and remained there until the 3870 X2 was released.

--Matt

---

The 3870 was never a high end card, it might have been to ATI, but the rest of the enthusiast world did not consider it a high end card. The 8800GT was beating it hands down in every game tested. 8800GT was a $200 card at the time. The 3800 still struggled with AA.

^ regardless of what it was...ati produced the 3870 as their high end card..just like with their cpus now..just because they arent beating intel's comp..doesnt mean the black edition aren't their high end cpu's..

Quote:

---

Originally Posted by LordEC911

Depends on the architecture and it also depends on if they are refreshing at the end of the year @ 28nm and then release a new architecture in 2011. It could go either way.

The real question is, do they really want to try and release a completely new architecture on a new process node with no small test GPUs first?

---

and from a new plant, its a new everything, except the word ATI on the front

Quote:

---

Originally Posted by HelixPC;

The 3870 was never a high end card, it might have been to ATI, but the rest of the enthusiast world did not consider it a high end card. The 8800GT was beating it hands down in every game tested. 8800GT was a $200 card at the time. The 3800 still struggled with AA.

---

It was indeed high end in the ATI lineup, replacing the 2900XT. The status of nvidias products at the time is irelevant. The 5870 doesn change the position of the gtx285 in nvidia's lineup, does it? Nor does the i7 on the phenom II line. Each manufacturer has their own product line, and at the time the 3870 was at the top of ATI's.

--Matt

Quote:

---

Originally Posted by mattkosem

It was indeed high end in the ATI lineup, replacing the 2900XT. The status of nvidias products at the time is irelevant.

--Matt

---

That's awfully presumptuous. The decision of what product to test on a new process could have been any of the following:

1) Mainstream by own company standards (internal product selection)
2) Mainstream by global standards (market-based product selection if they're aiming for a certain kind of product range)
3) Mainstream by die size (technical product selection)

Nobody knows which of these is true, and #3 is the most likely.

Try it on something small/easier and also cheap. Then, sell it as a market mainstream so it actually makes you some money that late in the product generation cycle. That's what I would do in that position as it makes a lot of sense both from a business and technical point of view compared to other options. It would hardly make sense to shrink a massive chip to a new process late in the cycle unless they're still selling them by the truckload. Mainstream parts move a lot more units.

Quote:

---

Originally Posted by Particle

That's awfully presumptuous. The decision of what product to test on a new process could have been any of the following:

1) Mainstream by own company standards (internal product selection)
2) Mainstream by global standards (market-based product selection if they're aiming for a certain kind of product range)
3) Mainstream by die size (technical product selection)

Nobody knows which of these is true, and #3 is the most likely.

Try it on something small/easier and also cheap. Then, sell it as a market mainstream so it actually makes you some money that late in the product generation cycle. That's what I would do in that position as it makes a lot of sense both from a business and technical point of view compared to other options. It would hardly make sense to shrink a massive chip to a new process late in the cycle unless they're still selling them by the truckload. Mainstream parts move a lot more units.

---

....i do not see how what you quoted relates to your post? did you miss quote another forum member perhaps?

Quote:

---

Originally Posted by Particle

That's awfully presumptuous. The decision of what product to test on a new process could have been any of the following:

1) Mainstream by own company standards (internal product selection)
2) Mainstream by global standards (market-based product selection if they're aiming for a certain kind of product range)
3) Mainstream by die size (technical product selection)

Nobody knows which of these is true, and #3 is the most likely.

Try it on something small/easier and also cheap. Then, sell it as a market mainstream so it actually makes you some money that late in the product generation cycle. That's what I would do in that position as it makes a lot of sense both from a business and technical point of view compared to other options. It would hardly make sense to shrink a massive chip to a new process late in the cycle unless they're still selling them by the truckload. Mainstream parts move a lot more units.

---

I think you read more than I said. I never made any claims about their selection process. I merely pointed out the fact that the first product on the 55mm node was, at the time, the highest end product on the ATI product line.

The section that you quoted was in response to someone stating that the 3870 was not a high-end card, which was obviously not the case in their product line. Am I missing something?

--Matt

Quote:

---

Originally Posted by mattkosem

I merely pointed out the fact that the first product on the 55mm node was, at the time, the highest end product on the ATI product line.

---

Yeah, but it doesn't matter if it was high-end or whatever. As I pointed, it was part of their "sweet point" strategy, it was small (and 55nm proces had no such problems like 40nm), so no reason why to manufacture RV620 or RV610 sooner than RV670.

Not mentioning they needed something to beat NVIDIA at those times. HD 2900 XT was not bad compared to GTS 320/640 MB, the power consumption was pretty much the same, but you know, less is always better. For that reason RV670 came first.

those wafer shots show a test chip, im pretty sure...
its not a gpu or cpu, its way to heterogenous for that...

I don't honestly understand how there can be any confusion regarding my post, but if nobody else seems to get it I'll just leave it alone.

I might thing that its something that AMD had global foundries do because Intel just released the 32nm and then AMD can go and say we have 28nm in your face Intel. They had one wafer made remember I doubt anything on that wafer is working, it is the first one. and I am also pretty sure that they are releasing bulldozer the new architecture on 32nm and this just says that they really have better R&D now and that they are truly gunning for their old crown again. this means that the released from AMD will be getting faster and faster but also better and better almost equaling Intels speed. But one thing I think AMD should focus on improving is their manufacturing processes quality. Intel still has the best manufacturing process.

Bulldozer will be 32nm SOI, this has nothing to do with AMD shoving 28 nm waffers in Intel's face. stop thinking in a childish way.


28nm will not be adressed to CPU's, after 32nm we will probably see GF deveoping 22nm, otherwise we cannot imagine AMD being competitive agaisnt Ivy Bridge (refresh of Sandy Bridge in 22nm).

PS: Intel showed 22nm waffers as well, so the smaller these companies go, the better the products we will buy will be.

This is something I did not know so now I say thank you.

RV670@55nm, WAS NOT shrink of R600@80nm! It was new design. RV740@40nm WAS NOT shrink of RV770@55nm it was new design. Yes NEW DESIGN did keep same architecture philosophy, but that's about only thing same.

Beside CPUs, AMD do have APUs ;)

Quote:

---

Originally Posted by =SOC= Admiral

I might thing that its something that AMD had global foundries do because Intel just released the 32nm and then AMD can go and say we have 28nm in your face Intel. They had one wafer made remember I doubt anything on that wafer is working, it is the first one. and I am also pretty sure that they are releasing bulldozer the new architecture on 32nm and this just says that they really have better R&D now and that they are truly gunning for their old crown again. this means that the released from AMD will be getting faster and faster but also better and better almost equaling Intels speed. But one thing I think AMD should focus on improving is their manufacturing processes quality. Intel still has the best manufacturing process.

---

amd showed 28nm SRAM test wafers serveral months ago so i'm pretty sure that they can produce working ICs by now :)

they could produce slow IC's with low yields now for low power. making working SRAM is much much easier than a cpu or gpu. intel demoed 22nm sram at idf and its not going to be ramped until 2H 2011.

Obviously not SRAM, most likely an FPGA from experimental customers like Altera or Xilinx.

Should probably be go for GPU in 6-9 months, depending on how long they've come. GF/AMD seem to be promising extremely loose schedules for whatever reason lately...

On the one hand you have GlobalFoundaries never having done such a small scale process before for a gpu and on the other hand TSMC screwed up 40mm.

So which way does ATi jump?

Andy

just cause GF has never built chips, does not mean they hired people who never have
are the people working for GF full of experience?

Quote:

---

Originally Posted by generics_user

amd showed 28nm SRAM test wafers serveral months ago so i'm pretty sure that they can produce working ICs by now :)

---

I come to this site to learn more so the more I read the more I learn. And thank you.

Quote:

---

Originally Posted by mattkosem

The 3870 was the first product on 55nm, was it not?

--Matt

---

Technically yes. But it was just a tweaked 65nm process used on the 2600 and 2400 series not a genuine. This is really more of a way of one upping Intel's 65, 45 and 32nm with 55nm, 40nm and 28nm processes than a genuine node change from improved tech. It's marketing really

Quote:

---

Originally Posted by trinibwoy

3870 was a midrange GPU.

---

In what universe?

its price, die size and power consumption are typical of a midrange gpu. AMD clearly stated that they compete in the high end with their dual gpu card. this strategy became hard to notice when rv770 was so close to gt200.

Vrrrrrrrrrrroooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooommmmmmmmmmm mmm

Quote:

---

Originally Posted by Jamesrt2004

Vrrrrrrrrrrroooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooommmmmmmmmmm mmm

---

Best post ever! :rofl:

im having problem seeing amd doing a half node on a cpu.... so this might be a test waffer for ic's ..... remember they need to be ahead of the tech to make sure their future cpu or gpu design actually works ok etc..... bring on the old amd :D

Good news, but sounds a bit a bit too late in my calender. Does this mean that ATI is not going to counter the Fermi lunch?

I suspect this could become a good excuse to just look at Fermi without even a refresh?. I hope not.

Quote:

---

Originally Posted by Sam_oslo

Good news, but sounds a bit a bit too late in my calender. Does this mean that ATI is not going to counter the Fermi lunch?

I suspect this could become a good excuse to just look at Fermi without even a refresh?. I hope not.

---

ATI is on TSMC right? TSMC's 40nm process is yeilding around 30% I believe.

If ATI simply switched to GF on 40nm I'm sure they could squeeze way more perf. per watt and overall overclockability from their cards. Just my opinion there.

I don't know what's going on, really. But in good old days, ATI and Nvidia had a good fight, with countering each other's GPU almost simultaneity by lunching close to each other.

Nvidia had it's own "excuses" and didn't counter the 5x-series. I hope ATI is not going to use this as a "excuse" to jump over countering the Fermi (at least with a refresh). In case it is gonna to be bad news for Fermi-prices.

If ATI doesn't release a refresh to counter NVIDIA's Fermi, I just hope they get their next generation out of the door later on this year. Because a 28nm card does sound nice.

A 28nm card does sound nice, for sure. But that sounds a bit too late for making any news now, or be interesting for the purpose that I had in mind.

Be realistic. AMD will probably release new line this year/begining of 2011, but still on 40 nm. 28nm process is too way ahead, if something, we can expect another card like HD 4770 for testing. AMD will roll out on 28nm only if Fermi sucked too much thus no hurry needed and than we would see 28nm cards between H1 and H2 2011.

AMD is most likely going to have some form of refresh on 40nm to counter Fermi (depending on performance) and then make a test chip on 28nm before moving on.

The R800-series is the last of the R600-derived architecture supposedly, and so whatever comes next will be a new architecture, which gives AMD even more incentive to test out the 28nm process first on an existing architecture before moving on. RV740 probably saved AMD a ton of money on the transition to 40nm

am i the only one who see's this as only news that ati is allready doing test on a 28nm waffer..... and btw amd/ati will never do the 28nm without doing 32nm before .... the smaller the fab process the more painful they are... so it wouldnt make sense to forget a full node and go straight to a half node process etc......

Quote:

---

Originally Posted by Sn0wm@n

am i the only one who see's this as only news that ati is allready doing test on a 28nm waffer..... and btw amd/ati will never do the 28nm without doing 32nm before .... the smaller the fab process the more painful they are... so it wouldnt make sense to forget a full node and go straight to a half node process etc......

---

GloFo's 32nm SOI tech is only for CPUs. For GPUs they'll have 28nm bulk:

http://img.hexus.net/v2/channel/news...slide1-big.jpg

how fast in general can they go from risk production to early production silicon... like say first test phase before they finalise the tweaking of a process ????


6 months ??? or even more

Quote:

---

Originally Posted by Jamesrt2004

Vrrrrrrrrrrroooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooommmmmmmmmmm mmm

---

Talking about Nvidia 59xx or ATi 29xx?? :D