GPU Comparison
RADEON
AMD Radeon R9 M395X
CORE STATE
Amethyst
VRAM
8 GB
CLOCK SPEED
—
TDP
75 W
BUS WIDTH
256 bit
ARCHITECTURE
GCN 3.0
PROCESS
28 nm
LAUNCH DATE
2015
VS
GEFORCE
H200 NVL
CORE STATE
GH100
VRAM
141 GB
CLOCK SPEED
1785 MHz
TDP
600 W
BUS WIDTH
6144 bit
ARCHITECTURE
Hopper
PROCESS
5 nm
LAUNCH DATE
2024
PERFORMANCE BENCHMARKS
geekbench_metal
geekbench_opencl
DETAILED SPECIFICATIONS
SPECIFICATION
R9 M395X
H200 NVL
Core Specs
Shading Units
2,048
16,896
+725.0%
Shaders
2,048
16,896
+725.0%
TMUs
128
528
+312.5%
ROPs
32
24
-25.0%
Compute Units
32
—
SM Count
—
132
Clocks
Base Clock
—
1365 MHz
Boost Clock
—
1785 MHz
GPU Clock
723 MHz
—
Memory Clock
1250 MHz
5 Gbps effective
1593 MHz
6.4 Gbps effective
Memory
Memory Size
8 GB
141 GB
VRAM (MB)
8,192
144,384
+1662.5%
Memory Type
GDDR5
HBM3e
Memory Bus
256 bit
6144 bit
Bandwidth
160.0 GB/s
4.89 TB/s
Cache
L1 Cache
16 KB (per CU)
256 KB (per SM)
L2 Cache
512 KB
50 MB
Performance
Pixel Rate
23.14 GPixel/s
42.84 GPixel/s
Texture Rate
92.54 GTexel/s
942.5 GTexel/s
FP32 (TFLOPS)
2.961 TFLOPS
60.32 TFLOPS
FP64 (TFLOPS)
185.1 GFLOPS (1:16)
30.16 TFLOPS (1:2)
FP16 (TFLOPS)
2.961 TFLOPS (1:1)
241.3 TFLOPS (4:1)
AI/RT
Tensor Cores
—
528
Power
TDP
75 W
600 W
TDP (W)
75
600
+700.0%
Suggested PSU
—
1000 W
Power Connectors
None
8-pin EPS
Architecture
Architecture
GCN 3.0
Hopper
GPU Name
Amethyst
GH100
Generation
Gem System
(R9 M300)
Server Hopper
(Hxx)
Process Size
28 nm
5 nm
Transistors
5,000 million
80,000 million
Die Size
366 mm²
814 mm²
Foundry
TSMC
TSMC
Density
13.7M / mm²
98.3M / mm²
API Support
DirectX
12 (12_0)
—
OpenGL
4.6
—
Vulkan
1.2.170
—
OpenCL
2.1
3.0
CUDA
—
9.0
Shader Model
6.5
—
Physical
Slot Width
MXM Module
Dual-slot
Length
—
267 mm
10.5 inches
Height
—
111 mm
4.4 inches
Outputs
Portable Device Dependent
No outputs
Bus Interface
PCIe 3.0 x16
PCIe 5.0 x16
Other
Production
End-of-life
Active
Predecessor
Solar System
Server Ada
Successor
Polaris Mobile
Server Blackwell