GPU Comparison
RADEON
AMD Radeon R7 Graphics
CORE STATE
Spectre Lite
VRAM
System Shared
CLOCK SPEED
—
TDP
25 W
BUS WIDTH
System Shared
ARCHITECTURE
GCN 2.0
PROCESS
28 nm
LAUNCH DATE
2014
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_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
R7 Graphics
H200 NVL
Core Specs
Shading Units
384
16,896
+4300.0%
Shaders
384
16,896
+4300.0%
TMUs
24
528
+2100.0%
ROPs
8
24
+200.0%
Compute Units
6
—
SM Count
—
132
Clocks
Base Clock
—
1365 MHz
Boost Clock
—
1785 MHz
GPU Clock
720 MHz
—
Memory Clock
System Shared
1593 MHz
6.4 Gbps effective
Memory
Memory Size
System Shared
141 GB
VRAM (MB)
—
144,384
Memory Type
System Shared
HBM3e
Memory Bus
System Shared
6144 bit
Bandwidth
System Dependent
4.89 TB/s
Cache
L1 Cache
—
256 KB (per SM)
L2 Cache
—
50 MB
Performance
Pixel Rate
5.760 GPixel/s
42.84 GPixel/s
Texture Rate
17.28 GTexel/s
942.5 GTexel/s
FP32 (TFLOPS)
553.0 GFLOPS
60.32 TFLOPS
FP64 (TFLOPS)
34.56 GFLOPS (1:16)
30.16 TFLOPS (1:2)
FP16 (TFLOPS)
—
241.3 TFLOPS (4:1)
AI/RT
Tensor Cores
—
528
Power
TDP
25 W
600 W
TDP (W)
25
600
+2300.0%
Suggested PSU
—
1000 W
Power Connectors
—
8-pin EPS
Architecture
Architecture
GCN 2.0
Hopper
GPU Name
Spectre Lite
GH100
Generation
GCN 2.0 IGP
(Kaveri)
Server Hopper
(Hxx)
Process Size
28 nm
5 nm
Transistors
2,410 million
80,000 million
Die Size
245 mm²
814 mm²
Foundry
GlobalFoundries
TSMC
Density
9.8M / 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
IGP
Dual-slot
Length
—
267 mm
10.5 inches
Height
—
111 mm
4.4 inches
Outputs
Motherboard Dependent
No outputs
Bus Interface
IGP
PCIe 5.0 x16
Other
Production
End-of-life
Active
Predecessor
TeraScale 3 IGP
Server Ada
Successor
GCN 3.0 IGP
Server Blackwell