NVIDIA GeForce RTX 3050 vs ATI Rage 128 PCI

NVIDIA GeForce RTX 3050 $249
ATI Rage 128 PCI
2560 Shaders
8GB GDDR6
1777MHz
2 Shaders
16MB SGRAM
103MHz
Peak AI Performance
291.14 TOPS
INT4 Tensor Sparse
Peak AI Performance
-
FP32
9.1 TFLOPS
-
-
FP16
9.1 TFLOPS
-
-
Form Factor
PCIe Card
1.9-Slots
Form Factor
PCI Card
1.0-Slots
TDP
115W
TDP
Unknown
Power Connectors
-
-
1x 12-Pin
-
-
-
-
-
-
GB6 OpenCL N/A
0%
GB6 OpenCL N/A
0%
GB6 Metal N/A
0%
GB6 Metal N/A
0%
GB6 Vulkan N/A
0%
GB5 OpenCL 71,805
24%
GB5 OpenCL N/A
0%
GB5 CUDA 73,995
21%
GB5 CUDA N/A
0%
GB5 Metal N/A
0%
GB5 Metal N/A
0%
GB5 Vulkan 57,700
28%
GB5 Vulkan N/A
0%
OCT 2020.1 N/A
0%
OCT 2020.1 N/A
0%
OCT Metal N/A
0%
OCT Metal N/A
0%
Manufacturer
NVIDIA
Manufacturer
ATI
Chip Designer
NVIDIA
Chip Designer
ATI
Architecture
Ampere
Architecture
Rage 4
Family
GeForce 30
Family
Rage
Codename
NV177
GA107
Variant
GA107-150-A1
Codename
-
Rage 128
Variant
Rage 128
Market Segment
Desktop
Market Segment
Desktop
Release Date
12/16/2022
Release Date
8/1/1998
Foundry
Samsung
-
Foundry
TSMC
-
Fabrication Node
8N
-
Fabrication Node
250nm
-
Die Size
200 mm²
-
Die Size
89 mm²
-
Transistor Count
8.7 Billion
-
Transistor Count
17 Million
-
Transistor Density
43.50M/mm²
-
Transistor Density
0.19M/mm²
-
Form
PCIe Card
Form
PCI Card
Shading Units
2560 Shaders
-
Shading Units
2 Shaders
-
Texture Mapping Units
80 TMUs
Texture Mapping Units
2 TMUs
Render Output Units
32 ROPs
Render Output Units
2 ROPs
Tensor Cores
80 T-Cores
-
-
Ray-Tracing Cores
20 RT-Cores
-
-
Streaming Multiprocessors
20 SMs
-
-
-
-
-
-
-
-
-
-
Graphics Processing Clusters
2 GPCs
-
-
-
-
1552MHz Base
1777MHz
-
-
-
103MHz
Peak AI Performance
291.14 TOPS
INT4 Tensor Sparse
Peak AI Performance
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
FP16
9.1 TFLOPS
36.39 TFLOPS Tensor (FP16 Accumulate)
72.79 TFLOPS Tensor (FP16 Accumulate) Sparse
18.2 TFLOPS Tensor (FP32 Accumulate)
36.39 TFLOPS Tensor (FP32 Accumulate) Sparse
-
-
-
-
-
-
FP32
9.1 TFLOPS
-
-
-
-
-
-
FP64
140 GFLOPS
-
-
-
-
BF16
9.1 TFLOPS
18.2 TFLOPS Tensor
36.39 TFLOPS Tensor Sparse
-
-
-
-
TF32
9.1 TFLOPS Tensor
18.2 TFLOPS Tensor Sparse
-
-
-
INT4
145.57 TOPS Tensor
291.14 TOPS Tensor Sparse
-
-
-
INT8
-
72.79 TOPS Tensor
145.57 TOPS Tensor Sparse
-
-
-
-
INT32
4.55 TOPS
-
-
Ray Tracing
17.8 TOPS
-
-
Pixel Fillrate
56.864 GPixel/s
Pixel Fillrate
0.206 GPixel/s
-
-
-
-
Texture Fillrate
142.16 GTexel/s
Texture Fillrate
0.206 GTexel/s
-
-
-
-
L1
64KB/SM Tex
128KB/SM
-
-
L1
-
-
-
Unknown
L2
2MB Shared
L2
Unknown
-
-
-
-
-
-
8GB
GDDR6
-
16MB
SGRAM
-
Bus Width
128Bit
Bus Width
128Bit
Clock
1750MHz
Transfer Rate
14GT/s
Bandwidth
224GB/s
Clock
103MHz
Transfer Rate
103MT/s
Bandwidth
1.6GB/s
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
TDP
115W
TDP
Unknown
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
3x DisplayPort 1.4
-
-
-
-
-
-
-
-
-
-
-
1x HDMI 2.1
-
-
1x VGA
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Max Resolution
7680x4320
Max Resolution
640x480
Max Resolution Refresh Rate
60Hz
Max Resolution Refresh Rate
60Hz
Variable Refresh Rate
G-Sync
FreeSync
-
Variable Refresh Rate
-
-
-
Display Stream Compression (DSC)
Supported
Display Stream Compression (DSC)
Not Supported
Multi Monitor Support
3
Multi Monitor Support
1
Content Protection
HDCP 2.3
-
-
Model
NVENC 7
No Encoders
-
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Model
NVDEC 5
No Decoders
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Direct X
12
Direct 3D
12_2
Direct X
6
Direct 3D
6.0
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2
OpenGL
1.2
-
-
-
-
Shader Model
6.6
CUDA
8.6
-
-
PureVideo HD
VP11
VDPAU
Feature Set K
-
-
-
-
-
-
-
-
-
-
-
-
-
2x Fans
-
-
-
-
Power Connectors
-
-
-
-
-
1x 12-Pin
-
-
-
-
-
-
-
-
-
Slots Required
1.9
PCIe Version
4.0
PCIe Lanes
8
Slots Required
1.0
-
-
-
-
-
-
-
-
-
-
-
-
Height
112 mm (4.41 in)
Width
242 mm (9.53 in)
Depth
38 mm (1.5 in)
-
-
-
-
-
-
Change Comparison