NVIDIA GeForce RTX 4090 D vs AMD Radeon RX 6900 XT
NVIDIA GeForce RTX 4090 D
$1,599
AMD Radeon RX 6900 XT
$999
|
14592 Shaders
24GB GDDR6X
2520MHz
|
5120 Shaders
16GB GDDR6
2250MHz
|
|
Peak AI Performance
2.35 POPS
INT4 Tensor Sparse
|
Peak AI Performance
46.08 TFLOPS
FP16
|
|
FP32
73.54 TFLOPS
|
FP32
23.04 TFLOPS
|
|
FP16
73.54 TFLOPS
|
FP16
46.08 TFLOPS
|
|
Form Factor
PCIe Card
3.0-Slots
|
Form Factor
PCIe Card
2.5-Slots
|
|
TDP
450W
|
TDP
300W
|
|
Power Connectors
-
-
-
1x 16-Pin 12VHPWR
|
Power Connectors
-
2x 8-Pin
-
-
|
|
GB6 OpenCL N/A
0%
|
|
|
GB6 Metal N/A
0%
|
|
|
GB6 Vulkan N/A
0%
|
|
GB5 OpenCL N/A
0%
|
GB5 OpenCL 165,115
54%
|
|
GB5 CUDA N/A
0%
|
GB5 CUDA N/A
0%
|
|
GB5 Metal N/A
0%
|
GB5 Metal 162,550
90%
|
|
GB5 Vulkan N/A
0%
|
GB5 Vulkan 114,040
56%
|
|
OCT 2020.1 N/A
0%
|
OCT 2020.1 N/A
0%
|
|
OCT Metal N/A
0%
|
OCT Metal 390
66%
|
|
Peak AI
Performance
2.35 POPS
INT4 Tensor Sparse
|
Peak AI
Performance
46.08 TFLOPS
FP16
|
|
-
-
-
|
-
-
-
|
|
FP8
-
588.35 TFLOPS
Tensor (FP16 Accumulate)
1.18 PFLOPS
Tensor (FP16 Accumulate) Sparse
294.18 TFLOPS
Tensor (FP32 Accumulate)
588.35 TFLOPS
Tensor (FP32 Accumulate) Sparse
|
-
-
-
-
-
-
|
|
FP16
73.54 TFLOPS
294.18 TFLOPS
Tensor (FP16 Accumulate)
588.35 TFLOPS
Tensor (FP16 Accumulate) Sparse
147.09 TFLOPS
Tensor (FP32 Accumulate)
294.17 TFLOPS
Tensor (FP32 Accumulate) Sparse
|
FP16
46.08 TFLOPS
-
-
-
-
|
|
FP32
73.54 TFLOPS
-
-
|
FP32
23.04 TFLOPS
-
-
|
|
FP64
1.15 TFLOPS
-
|
FP64
1.44 TFLOPS
-
|
|
BF16
73.54 TFLOPS
147.09 TFLOPS
Tensor
294.17 TFLOPS
Tensor Sparse
|
-
-
-
-
|
|
TF32
73.54 TFLOPS
Tensor
147.09 TFLOPS
Tensor Sparse
|
-
-
-
|
|
INT4
1.18 POPS
Tensor
2.35 POPS
Tensor Sparse
|
-
-
-
|
|
INT8
-
588.35 TOPS
Tensor
1.18 POPS
Tensor Sparse
|
-
-
-
-
|
|
INT32
36.77 TOPS
|
-
-
|
|
Ray Tracing
170 TOPS
|
-
-
|
|
Pixel
Fillrate
443.52 GPixel/s
|
Pixel
Fillrate
288 GPixel/s
|
|
-
-
|
-
-
|
|
Texture
Fillrate
1149.12 GTexel/s
|
Texture
Fillrate
720 GTexel/s
|
|
Manufacturer
NVIDIA
|
Manufacturer
AMD
|
|
Chip Designer
NVIDIA
|
Chip Designer
AMD
|
|
Architecture
Ada Lovelace
|
Architecture
RDNA 2
|
|
Family
GeForce 40
|
Family
Radeon RX 6000
|
|
Codename
NV182
AD102
Variant
AD102-250-A1
|
Codename
Sienna Cichlid
Navi 21
Variant
Navi 21 XTX
|
|
Market Segment
Desktop
|
Market Segment
Desktop
|
|
Release Date
12/28/2023
|
Release Date
12/8/2020
|
|
Foundry
TSMC
-
|
Foundry
TSMC
-
|
|
Fabrication Node
4N
-
|
Fabrication Node
N7
-
|
|
Die Size
608 mm²
-
|
Die Size
520 mm²
-
|
|
Transistor Count
76.3 Billion
-
|
Transistor Count
26.8 Billion
-
|
|
Transistor Density
125.41M/mm²
-
|
Transistor Density
51.56M/mm²
-
|
|
Form
PCIe Card
|
Form
PCIe Card
|
|
Shading Units
14592 Shaders
-
|
Shading Units
5120 Shaders
-
|
|
Texture Mapping Units
456 TMUs
|
Texture Mapping Units
320 TMUs
|
|
Render Output Units
176 ROPs
|
Render Output Units
128 ROPs
|
|
Tensor Cores
456 T-Cores
|
-
-
|
|
Ray-Tracing Cores
114 RT-Cores
|
Ray-Tracing Cores
80 RT-Cores
|
|
Streaming Multiprocessors
114 SMs
|
-
-
|
|
-
-
|
Compute Units
80 CUs
|
|
-
-
|
-
-
|
|
Graphics Processing Clusters
10 GPCs
|
Graphics Processing Clusters
4 GPCs
|
|
-
-
2230MHz Base
2520MHz
|
-
-
1825MHz Base
2250MHz
|
|
-
-
|
L0
32KB/WGP
|
|
L1
64KB/SM Tex
128KB/SM
-
-
|
L1
-
-
-
128KB/Array
|
|
L2
72MB Shared
|
L2
4MB Shared
|
|
-
-
-
|
L3
128MB Shared
1.66TB/s
|
|
24GB
GDDR6X
-
|
16GB
GDDR6
-
|
|
Bus Width
384Bit
|
Bus Width
256Bit
|
|
Clock
1313MHz
Transfer Rate
21GT/s
Bandwidth
1008GB/s
|
Clock
2000MHz
Transfer Rate
16GT/s
Bandwidth
512GB/s
|
|
-
-
-
-
-
-
-
-
-
|
-
-
-
-
-
-
-
-
-
|
|
TDP
450W
|
TDP
300W
|
|
Temp
90°C Max
|
Temp
110°C Max
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
3x DisplayPort 1.4
-
-
-
-
-
-
-
-
-
-
-
1x HDMI 2.1
-
-
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
2x DisplayPort 1.4
-
-
-
-
-
-
-
-
-
-
-
1x HDMI 2.1
1x USB-C + DP
-
|
|
Max Resolution
7680x4320
|
Max Resolution
7680x4320
|
|
Max Resolution Refresh Rate
60Hz
|
Max Resolution Refresh Rate
120Hz
|
|
Variable Refresh Rate
G-Sync
FreeSync
-
|
Variable Refresh Rate
-
FreeSync
-
|
|
Display Stream Compression (DSC)
Supported
|
Display Stream Compression (DSC)
Supported
|
|
Multi Monitor Support
4
|
Multi Monitor Support
4
|
|
Content Protection
HDCP 2.3
|
Content Protection
HDCP 2.3
|
|
Model
2x NVENC 8
|
Model
VCN 3.0
|
|
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
|
|
Model
NVDEC 5
|
Model
VCN 3.0
|
|
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
|
Direct X
12
Direct 3D
12_3
|
Direct X
12
Direct 3D
12_2
|
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
|
OpenGL
4.6
OpenCL
2.1
Vulkan
1.2
|
|
Shader Model
6.6
CUDA
8.9
-
-
PureVideo HD
VP12
VDPAU
Feature Set L
|
Shader Model
6.5
-
-
GFX
10.3
-
-
-
-
|
|
-
-
-
2x Fans
|
-
-
-
3x Fans
|
|
Power Connectors
-
-
-
-
-
-
1x 16-Pin 12VHPWR
|
Power Connectors
-
-
-
2x 8-Pin
-
-
-
|
|
Slots Required
3.0
PCIe Version
4.0
PCIe Lanes
16
|
Slots Required
2.5
PCIe Version
4.0
PCIe Lanes
16
|
|
-
-
-
-
|
Multi GPU Support
Supported
Type
Bridgeless
|
|
Height
137 mm (5.39 in)
Width
304 mm (11.97 in)
Depth
61 mm (2.4 in)
|
Height
120 mm (4.72 in)
Width
267 mm (10.51 in)
Depth
50 mm (1.97 in)
|
Copy Link