NVIDIA GeForce RTX 2070 Max-Q vs AMD Radeon RX M580X
NVIDIA GeForce RTX 2070 Max-Q
AMD Radeon RX M580X
|
2304 Shaders
8GB GDDR6
1185MHz
|
2304 Shaders
8GB GDDR5
1077MHz
|
|
Peak AI Performance
174.74 TOPS
INT4 Tensor
|
Peak AI Performance
4.96 TFLOPS
FP16
|
|
FP32
5.46 TFLOPS
|
FP32
4.96 TFLOPS
|
|
FP16
10.92 TFLOPS
|
FP16
4.96 TFLOPS
|
|
Form Factor
Soldered
|
Form Factor
Soldered
|
|
TDP
80W
|
TDP
100W
|
Peak AI Performance
|
|
GeForce RTX 2070 Max-Q - 174.74 TOPS INT4 Tensor
x35.23
Radeon RX M580X - 4.96 TFLOPS FP16
x1
FP32
|
|
GeForce RTX 2070 Max-Q - 5.46 TFLOPS FP32
x1.1
Radeon RX M580X - 4.96 TFLOPS FP32
x1
FP16
|
|
GeForce RTX 2070 Max-Q - 10.92 TFLOPS FP16
x2.2
Radeon RX M580X - 4.96 TFLOPS FP16
x1
|
|
GeForce RTX 2070 Max-Q - 384GB/s
x1.5
Radeon RX M580X - 256GB/s
x1
|
|
GeForce RTX 2070 Max-Q - 80W
x1
Radeon RX M580X - 100W
x1.25
|
GB6 OpenCL N/A
0%
|
|
|
0%
|
0%
|
|
0%
|
0%
|
|
GB5 OpenCL 73,970
24%
|
GB5 OpenCL N/A
0%
|
|
GB5 CUDA 85,865
24%
|
GB5 CUDA N/A
0%
|
|
0%
|
0%
|
|
GB5 Vulkan 63,555
31%
|
GB5 Vulkan N/A
0%
|
|
OCT 2020.1 185
24%
|
OCT 2020.1 N/A
0%
|
|
0%
|
0%
|
|
Manufacturer
NVIDIA
|
Manufacturer
AMD
|
|
Chip Designer
NVIDIA
|
Chip Designer
AMD
|
|
Architecture
Turing
|
Architecture
GCN 4
|
|
Family
GeForce 20
|
Family
Radeon RX 500
|
|
Codename
NV166
TU106
Variant
N18E-G2
|
Codename
Ellesmere
Polaris 20
Variant
Polaris 20 XTM
|
|
Market Segment
Laptop
|
Market Segment
Laptop
|
|
Release Date
1/29/2019
|
Release Date
4/11/2018
|
|
Foundry
TSMC
|
Foundry
GlobalFoundries
|
|
Fabrication Node
12FFN
|
Fabrication Node
14LPP
|
|
Die Size
445 mm²
|
Die Size
232 mm²
|
|
Transistor Count
10.8 Billion
|
Transistor Count
5.7 Billion
|
|
Transistor Density
24.27M/mm²
|
Transistor Density
24.57M/mm²
|
|
Form
Soldered
|
Form
Soldered
|
|
Shading Units
2304 Shaders
|
Shading Units
2304 Shaders
|
|
Texture Mapping Units
144 TMUs
|
Texture Mapping Units
144 TMUs
|
|
Render Output Units
64 ROPs
|
Render Output Units
32 ROPs
|
|
Tensor Cores
288 T-Cores
|
-
-
|
|
Ray-Tracing Cores
36 RT-Cores
|
-
-
|
|
Streaming Multiprocessors
36 SMs
|
-
-
|
|
-
-
|
Compute Units
36 CUs
|
|
Graphics Processing Clusters
3 GPCs
|
-
-
|
|
885MHz Base
1185MHz
|
1000MHz Base
1077MHz
|
|
Peak AI Performance
174.74 TOPS
INT4 Tensor
|
Peak AI Performance
4.96 TFLOPS
FP16
|
|
FP16
10.92 TFLOPS
43.68 TFLOPS Tensor (FP16 Accumulate)
21.84 TFLOPS Tensor (FP32 Accumulate)
|
FP16
4.96 TFLOPS
-
-
|
|
FP32
5.46 TFLOPS
|
FP32
4.96 TFLOPS
|
|
FP64
170 GFLOPS
|
FP64
310 GFLOPS
|
|
BF16
21.84 TFLOPS Tensor
|
-
-
|
|
TF32
21.84 TFLOPS Tensor
|
-
-
|
|
INT4
174.74 TOPS Tensor
|
-
-
|
|
INT8
87.37 TOPS Tensor
|
-
-
|
|
Ray Tracing
16.5 TOPS
|
-
-
|
|
Pixel Fillrate
75.84 GPixel/s
|
Pixel Fillrate
34.464 GPixel/s
|
|
Texture Fillrate
170.64 GTexel/s
|
Texture Fillrate
155.088 GTexel/s
|
|
L1
32KB/SM Tex
64KB/SM
-
|
L1
-
-
16KB/CU
|
|
L2
4MB Shared
|
L2
2MB Shared
|
|
8GB
GDDR6
|
8GB
GDDR5
|
|
Bus Width
256Bit
|
Bus Width
256Bit
|
|
Clock
1500MHz
Transfer Rate
12GT/s
Bandwidth
384GB/s
|
Clock
2000MHz
Transfer Rate
8GT/s
Bandwidth
256GB/s
|
|
TDP
80W
|
TDP
100W
|
|
Max Resolution
7680x4320
|
Max Resolution
5120x2880
|
|
Max Resolution Refresh Rate
60Hz
|
Max Resolution Refresh Rate
60Hz
|
|
Variable Refresh Rate
G-Sync
FreeSync
|
Variable Refresh Rate
-
FreeSync
|
|
Display Stream Compression (DSC)
Supported
|
Display Stream Compression (DSC)
Not Supported
|
|
Multi Monitor Support
3
|
Multi Monitor Support
3
|
|
Content Protection
HDCP 2.2
|
Content Protection
HDCP 2.2
|
|
Model
NVENC 7
|
Model
VCE 3.0
|
|
Codec
AVC (H.264)
HEVC (H.265)
|
Codec
AVC (H.264)
HEVC (H.265)
|
|
Model
NVDEC 4
|
Model
UVD 6.3
|
|
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
AVC (H.264)
HEVC (H.265)
|
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
-
AVC (H.264)
HEVC (H.265)
|
|
Direct X
12
Direct 3D
12_2
|
Direct X
12
Direct 3D
12_0
|
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2
|
OpenGL
4.6
OpenCL
2.1
Vulkan
1.3
|
|
Shader Model
6.6
CUDA
7.5
-
-
PureVideo HD
VP10
VDPAU
Feature Set J
|
Shader Model
6.7
-
-
GFX
8
-
-
-
-
|
|
Not a Card
|
Not a Card
|
|
PCIe Version
3.0
PCIe Lanes
16
|
PCIe Version
3.0
PCIe Lanes
16
|
Copy Link