So, do we know anything about RV670 yet?

Nice :D

Though I'm in "wait until the drivers are sorted" and "wait for the D3D10 performance" mode before I spend money. Which I think is about another 12-18 months, when I expect to build an entirely new system.

Jawed

lol, you remind of my self. I had a Geforce 2MX all the way to the point I decided to get a x1800xt. About every 4 years.... I'm going to try to shorten that up a bit now... but sometimes this hobby conflicts with my music hobby.
 
I an not sold on R680 being 2x RV670 at all. Rxxx codenames are for the chips, right? Using R680 as name for a PCB with two RV670s would be a big departure from previous nomenclature.
 
I an not sold on R680 being 2x RV670 at all. Rxxx codenames are for the chips, right? Using R680 as name for a PCB with two RV670s would be a big departure from previous nomenclature.

Agreed, I think we will be getting a big high end chip sooner or later.
 
I an not sold on R680 being 2x RV670 at all. Rxxx codenames are for the chips, right? Using R680 as name for a PCB with two RV670s would be a big departure from previous nomenclature.

Maybe two RV670/680 on one package and a reusing of 2900XT-PCB with little modifications. ;)
 
I an not sold on R680 being 2x RV670 at all. Rxxx codenames are for the chips, right? Using R680 as name for a PCB with two RV670s would be a big departure from previous nomenclature.

That's what I've been saying all along. Only way I would "eat" it would be 2 dies on 1 chip, but still...
 
Agreed, I think we will be getting a big high end chip sooner or later.
To me it seems ATI GPUs peaked with R600. Henceforth everything is multi-chip for enthusiast class performance.

I'm still boggled at how quickly ATI has passed through the 65nm node. It seems that it'll be the shortest duration of any node ever. About 8 months?

Jawed
 
Two independent display controllers
Drive two displays simultaneously with independent resolutions, refresh rates, color controls and video overlays for each display

I'm really digging this!:smile:

I wonder if you can independently driver each monitor with a seperate VGA?
 
ATI RADEON™ HD 3870


666 million transistors on 55nm fabrication process
PCI Express 2.0 x16 bus interface
256-bit GDDR4 memory interface
Ring Bus Memory Controller
Fully distributed design with 512-bit internal ring bus for memory reads and writes
Microsoft® DirectX® 10.1 support
Shader Model 4.1
32-bit floating point texture filtering
Indexed cube map arrays
Independent blend modes per render target
Pixel coverage sample masking
Read/write multi-sample surfaces with shaders
Gather4 texture fetching
Unified Superscalar Shader Architecture
320 stream processing units


does that mean anything?
 
ATI RADEON™ HD 3870


666 million transistors on 55nm fabrication process
PCI Express 2.0 x16 bus interface
256-bit GDDR4 memory interface
Ring Bus Memory Controller
Fully distributed design with 512-bit internal ring bus for memory reads and writes
Microsoft® DirectX® 10.1 support
Shader Model 4.1
32-bit floating point texture filtering
Indexed cube map arrays
Independent blend modes per render target
Pixel coverage sample masking
Read/write multi-sample surfaces with shaders
Gather4 texture fetching
Unified Superscalar Shader Architecture
320 stream processing units


does that mean anything?

I think it was one of the features in the DX10.1 document. Time for some more performance leaks soon I hope.
 
Back
Top