NVIDIA GeForce 8800 Ultra vs ATI Radeon HD 2900 Pro
NVIDIA GeForce 8800 Ultra
$829
ATI Radeon HD 2900 Pro
$250
|
128 Shaders
768MB GDDR3
612MHz
|
320 Shaders
512MB GDDR3
600MHz
|
|
Peak AI Performance
0.38 TOPS
INT32
|
Peak AI Performance
380 GFLOPS
FP32
|
|
FP32
380 GFLOPS
|
FP32
380 GFLOPS
|
|
Form Factor
PCIe Card
2.0-Slots
|
Form Factor
PCIe Card
2.0-Slots
|
|
TDP
175W
|
TDP
200W
|
|
Power Connectors
2x 6-Pin
-
|
Power Connectors
-
1x 8-Pin
|
Peak AI Performance
|
|
GeForce 8800 Ultra - 0.38 TOPS INT32
x1
Radeon HD 2900 Pro - 380 GFLOPS FP32
x1
FP32
|
|
GeForce 8800 Ultra - 380 GFLOPS FP32
x1
Radeon HD 2900 Pro - 380 GFLOPS FP32
x1
|
|
GeForce 8800 Ultra - 103.7GB/s
x1.01
Radeon HD 2900 Pro - 102.4GB/s
x1
|
|
GeForce 8800 Ultra - 175W
x1
Radeon HD 2900 Pro - 200W
x1.14
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
0%
|
0%
|
|
Manufacturer
NVIDIA
|
Manufacturer
ATI
|
|
Chip Designer
NVIDIA
|
Chip Designer
ATI
|
|
Architecture
Tesla
|
Architecture
TeraScale 1
|
|
Family
GeForce 8000
|
Family
Radeon HD 2000
|
|
Codename
NV50
G80
Variant
G80-450-A3
|
Codename
Pele
R600
Variant
R600 Pro
|
|
Market Segment
Desktop
|
Market Segment
Desktop
|
|
Release Date
5/2/2007
|
Release Date
9/25/2007
|
|
Foundry
TSMC
|
Foundry
TSMC
|
|
Fabrication Node
90nm
|
Fabrication Node
80nm
|
|
Die Size
484 mm²
|
Die Size
420 mm²
|
|
Transistor Count
681 Million
|
Transistor Count
720 Million
|
|
Transistor Density
1.41M/mm²
|
Transistor Density
1.71M/mm²
|
|
Form
PCIe Card
|
Form
PCIe Card
|
|
Shading Units
128 Shaders
|
Shading Units
320 Shaders
|
|
Texture Mapping Units
32 TMUs
|
Texture Mapping Units
16 TMUs
|
|
Render Output Units
24 ROPs
|
Render Output Units
16 ROPs
|
|
Streaming Multiprocessors
16 SMs
|
-
-
|
|
-
-
|
Compute Units
4 CUs
|
|
1500MHz Shader
612MHz
|
-
600MHz
|
|
Peak AI Performance
0.38 TOPS
INT32
|
Peak AI Performance
380 GFLOPS
FP32
|
|
FP32
380 GFLOPS
|
FP32
380 GFLOPS
|
|
INT32
0.38 TOPS
|
-
-
|
|
Pixel Fillrate
14.688 GPixel/s
|
Pixel Fillrate
9.6 GPixel/s
|
|
Texture Fillrate
19.584 GTexel/s
|
Texture Fillrate
9.6 GTexel/s
|
|
L1
Unknown
|
L1
Unknown
|
|
L2
96KB Shared
|
L2
256KB Shared
|
|
768MB
GDDR3
|
512MB
GDDR3
|
|
Bus Width
384Bit
|
Bus Width
512Bit
|
|
Clock
1080MHz
Transfer Rate
2.2GT/s
Bandwidth
103.7GB/s
|
Clock
800MHz
Transfer Rate
1.6GT/s
Bandwidth
102.4GB/s
|
|
TDP
175W
|
TDP
200W
|
|
1x S-Video
2x DVI-I Dual-Link
|
1x S-Video
2x DVI-I Dual-Link
|
|
Max Resolution
2560x1600
|
Max Resolution
2560x1600
|
|
Max Resolution Refresh Rate
60Hz
|
Max Resolution Refresh Rate
60Hz
|
|
Variable Refresh Rate
|
Variable Refresh Rate
|
|
Display Stream Compression (DSC)
Not Supported
|
Display Stream Compression (DSC)
Not Supported
|
|
Multi Monitor Support
2
|
Multi Monitor Support
2
|
|
No Encoders
|
No Encoders
|
|
No Decoders
|
Model
UVD 1.0
|
|
-
-
|
Codec
VC-1
AVC (H.264)
|
|
Direct X
11
Direct 3D
10_0
|
Direct X
10
Direct 3D
10_0
|
|
OpenGL
3.3
OpenCL
1.1
|
OpenGL
3.3
OpenCL
1.1
|
|
Shader Model
4.0
CUDA
1.0
-
-
PureVideo HD
VP1
|
Shader Model
4.0
-
-
GFX
3
-
-
|
|
1x Fan
|
1x Fan
|
|
Power Connectors
2x 6-Pin
-
|
Power Connectors
-
1x 8-Pin
|
|
Slots Required
2.0
PCIe Version
1.0
PCIe Lanes
16
|
Slots Required
2.0
PCIe Version
1.0
PCIe Lanes
16
|
|
Multi GPU Support
Supported
Type
4-way SLI
|
Multi GPU Support
Supported
Type
4-way CrossFireX
|
|
Height
140 mm (5.51 in)
Width
270 mm (10.63 in)
Depth
37 mm (1.46 in)
|
-
-
-
-
-
-
|
Copy Link