Power and Innovation to Drive High-End GPUs in 2009

[btarunr]

The year 2008 so far, has been very eventful for the graphics card market. A reinvigorated GPU lineup by ATI, brought in some fierce competition to NVIDIA, which resulted in a tug-of-war with pricing graphics cards in the market, with either company refusing to lose on grounds of pricing. This event, coupled with the announcement of several game titles by game publishers, resulted in bumper-sales of graphics cards, giving the present state of the global economy little or no relevance.

The months to come hold the same amount of importance for both AMD and NVIDIA, where the next round of competition begins with successors to current high-end products being slated. NVIDIA is expected to continue with its monolithic high transistor-count GPU design methodology, while AMD could bring in a little change to the way it uses two efficient GPUs to build powerful products.



The change AMD is planning on is using a Multi-Chip Module (MCM), a design similar to the one on Intel microprocessors, where Intel has been making some of its quad-core processors using two dual-core dice on an MCM. The module facilitates faster interconnection between the GPU cores, than if the cores were to communicate over a PCI-Express bridge chip, or using internal CrossFire ports. This would result in a single GPU with two dice, in essence, a dual-core GPU. The RV870 would be the first to support an MCM design, with unit cores featuring 25% more stream-processors than its predecessor, while upping the rated shader compute power to 1.5 TFLOP/s. In effect, it could feature a 512-bit GDDR5 memory bus, support DirectX 11, and OpenCL (Compute language, a GPGPU API). According to this projection, AMD could be out with a DirectX 11 GPU at least six months ahead of NVIDIA, although it remains to be seen if there are any games that come out that soon to take advantage of the technology. Between Q3 and Q4 2009, AMD could plan on a dual-MCM setup, bringing together two such cores with two unit cores each.

On to the green camp, and NVIDIA seems to have concrete plans to respond to AMD, with the GeForce GTX 270, and GeForce GTX 290 planned for this quarter. The two new SKUs are based on a 55nm silicon fabricated G200, the G200b, also known as GT206. Within this quarter itself, NVIDIA plans to prepare a dual-GPU card based on G200b cores, to lead its front. In Q1 2009, the company is planning its transit to the 40nm silicon process, with GT212. There’s little known about this chip, while GT216 in Q2 shows NVIDIA incorporating GDDR5 memory across a wide memory bus. This will also be the time when NVIDIA plans a dual-GPU card out of these chips. Finally in Q4, comes GT300, the next generation GPU from NVIDIA that brings in support for DirectX 11.

Year 2009 along with the rest of 2008, promise to be times when both graphics giants take computational power to the next level. It remains to be seen, if NVIDIA is able to deliver the punch, in order to rebuild whatever ground it lost to AMD, or if AMD goes for the kill with its innovation. Either way, these are good times for the buyers.

[Octavean]

Wow,....

ATI will have DX11 support in ~Q1 to ~Q2 whereas nVidia wont have DX11 support until ~Q3 to ~Q4. Night and day!

I may have to go ATI for my next card

[InnocentCriminal]

Magic words '512bit GDDR5'. I hope that means a 512bit memory bus instead of the current 256bit .

[btarunr]

The MCM package effectively has a 512-bit interface, and gets treated as a 'GPU', though it has two dice, with each die having 256-bit connections.

[InnocentCriminal]

Good!

[phanbuey]

Originally Posted by btarunr

The MCM package effectively has a 512-bit interface, and gets treated as a 'GPU', though it has two dice, with each die having 256-bit connections.

good call.

[JAKra]

This would result in a single GPU with two dice, in essence, a dual-core GPU. The RV870 would be the first to support an MCM design, with unit cores featuring 25% more stream-processors than its predecessor, while upping the rated shader compute power to 1.5 TFLOP/s.

If one die has 1.5 than the MCM would have 3 TFLOP/s presumably HD 5870 and the dual MCM (HD 5870X2?) would have theoretically 6 TFLOP/s...now that's something.

I might be a bit confused about this MCM stuff, but if I understood right than Litt'l Dragon is one die aprox. 205mm ² on 40nm with 960shaders...that sounds resonable for a die, but if you put it on MCM package...than double that in a dual config...it's pretty obvious who wins the next round...for me that is.

[King Wookie]

Sounds most interesting. At least good competition benefits us all.

[newtekie1]

Originally Posted by Octavean

Wow,....

ATI will have DX11 support in ~Q1 to ~Q2 whereas nVidia wont have DX11 support until ~Q3 to ~Q4. Night and day!

I may have to go ATI for my next card

What does it matter? We are talking about ATi having DX11 6-9 months before nVidia, when we won't even really see DX11 games until a good deal after both have adopted it.

[btarunr]

Originally Posted by newtekie1

What does it matter? We are talking about ATi having DX11 6-9 months before nVidia, when we won't even really see DX11 games a good deal after both have adopted it.

Exactly, Just like flaunting DirectX 10.1 support isn't really what sells current ATI cards.

[phanbuey]

I wanna see benchmarks. This "theorhetical computational power" stuff is baloney.

[Animalpak]

Damn a new X2 with doubled 40nm GPU's in Q1 2009 incoming !!!

Oh snap !

[jbunch07]

Seems like we already had a talk about the 512bit bus issues on here. We'll see how it turns out.

[Selene]

I need/want a new card, but ever time i get ready to buy i read this stuff and it makes me want to wait.

[InnocentCriminal]

Buy a 4850 then sell it when the 5850 is out.

[newbielives]

I can't wait for that card that slaps Crysis around like if it was its little bitch.

[[I.R.A]_FBi]

if you wrote that urself bta much kudos

[btarunr]

Thank you. It's the news team's first editorial

More to come soon.

[[I.R.A]_FBi]

Originally Posted by btarunr

Thank you. It's the news team's first editorial

More to come soon.

Looking forward to seeing them

[lemonadesoda]

Did I understand the *nice* chart correctly?


Are those columns in chronological order, ie, GT200b, GTX270 and GTX290 is NOT on 55nm?

And GT212 is on 55nm but we must wait until Q2 and GT216 for 40nm? GT212 will then be updated to GT212GX2 which is on 40nm?

Sorry for the detailed questions, but important: I want a silent workstation with CUDA, and therefore want a 55nm or better.

[btarunr]

G200b is a 55nm GPU, the contents of a column aren't in chronological order. GTX 270 and GTX 290 are based on the G200b aka GT206. After they're both out, expect a G200b GX2.

GT212/216 are both 40nm GPUs, after they're both out, you'll have a GT212 X2.

[lemonadesoda]

OK, thanks for clearing that up.

[Swansen]

Originally Posted by Selene

I need/want a new card, but ever time i get ready to buy i read this stuff and it makes me want to wait.

lol, i know it right, i have the exact same problem myself.

[Darkrealms]

Wow, both in 40nm first half of next year. I hope Nvidia can keep their OC ability on the 40nm chips ; )
200bx2's now I want to see those benches!

Good to see that first half of next year both companies should be neck to neck with tech releases. I hope that will boad well for AMD/ATI. They really need a boost.

I'm sure I'll be able to wait till Q4 next year with my GTX260 for the 300s to come out : )

[El Fiendo]

I'm afraid nVidia's naming scheme has just managed to cause an aneurysm. I was completely lost 3 times over reading those names and I've always been proud of how I could follow that stuff. Ah well back to studying card line-ups for me.