GPU Comparison
RADEON
AMD Radeon RX Vega M GH
CORE STATE
Polaris 22
VRAM
4 GB
CLOCK SPEED
1190 MHz
TDP
100 W
BUS WIDTH
1024 bit
ARCHITECTURE
GCN 4.0
PROCESS
14 nm
LAUNCH DATE
2018
VS
GEFORCE
RTX A5000
CORE STATE
GA102
VRAM
24 GB
CLOCK SPEED
1695 MHz
TDP
230 W
BUS WIDTH
384 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
2021
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
3dmark_3dmark_steel_nomad_dx12
passmark_directx_10
passmark_directx_11
passmark_directx_12
passmark_directx_9
passmark_g2d
passmark_g3d
passmark_gpu_compute
DETAILED SPECIFICATIONS
SPECIFICATION
RX Vega M GH
RTX A5000
Core Specs
Shading Units
1,536
8,192
+433.3%
Shaders
1,536
8,192
+433.3%
TMUs
96
256
+166.7%
ROPs
64
96
+50.0%
Compute Units
24
—
SM Count
—
64
Clocks
Base Clock
1063 MHz
1170 MHz
Boost Clock
1190 MHz
1695 MHz
Memory Clock
800 MHz
1600 Mbps effective
2000 MHz
16 Gbps effective
Memory
Memory Size
4 GB
24 GB
VRAM (MB)
4,096
24,576
+500.0%
Memory Type
HBM2
GDDR6
Memory Bus
1024 bit
384 bit
Bandwidth
204.8 GB/s
768.0 GB/s
Cache
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
1024 KB
6 MB
Performance
Pixel Rate
76.16 GPixel/s
162.7 GPixel/s
Texture Rate
114.2 GTexel/s
433.9 GTexel/s
FP32 (TFLOPS)
3.656 TFLOPS
27.77 TFLOPS
FP64 (TFLOPS)
228.5 GFLOPS (1:16)
433.9 GFLOPS (1:64)
FP16 (TFLOPS)
3.656 TFLOPS (1:1)
27.77 TFLOPS (1:1)
AI/RT
RT Cores
—
64
Tensor Cores
—
256
Power
TDP
100 W
230 W
TDP (W)
100
230
+130.0%
Suggested PSU
—
550 W
Power Connectors
—
1x 8-pin
Architecture
Architecture
GCN 4.0
Ampere
GPU Name
Polaris 22
GA102
Generation
Vega
(Vega M)
Workstation Ampere
(Ax000)
Process Size
14 nm
8 nm
Transistors
5,000 million
28,300 million
Die Size
208 mm²
628 mm²
Foundry
GlobalFoundries
Samsung
Density
24.0M / mm²
45.1M / mm²
API Support
DirectX
12 (12_0)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.3
1.4
OpenCL
2.1
3.0
CUDA
—
8.6
Shader Model
6.7
6.8
Physical
Slot Width
IGP
Dual-slot
Length
—
267 mm
10.5 inches
Height
—
112 mm
4.4 inches
Outputs
Portable Device Dependent
4x DisplayPort 1.4a
Bus Interface
IGP
PCIe 4.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
—
Quadro Turing
Successor
—
Workstation Ada