GPU Comparison
RADEON
AMD Radeon Vega 8
CORE STATE
Raven
VRAM
System Shared
CLOCK SPEED
1100 MHz
TDP
25 W
BUS WIDTH
System Shared
ARCHITECTURE
GCN 5.0
PROCESS
14 nm
LAUNCH DATE
2018
VS
GEFORCE
B200
CORE STATE
GB100
VRAM
90 GB
CLOCK SPEED
1965 MHz
TDP
1000 W
BUS WIDTH
4096 bit
ARCHITECTURE
Blackwell
PROCESS
5 nm
LAUNCH DATE
—
PERFORMANCE BENCHMARKS
geekbench_metal
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
Vega 8
B200
Core Specs
Shading Units
512
18,944
+3600.0%
Shaders
512
18,944
+3600.0%
TMUs
32
592
+1750.0%
ROPs
8
24
+200.0%
Compute Units
8
—
SM Count
—
148
Clocks
Base Clock
300 MHz
700 MHz
Boost Clock
1100 MHz
1965 MHz
Memory Clock
System Shared
2000 MHz
8 Gbps effective
Memory
Memory Size
System Shared
90 GB
VRAM (MB)
—
92,160
Memory Type
System Shared
HBM3e
Memory Bus
System Shared
4096 bit
Bandwidth
System Dependent
4.10 TB/s
Cache
L1 Cache
—
256 KB (per SM)
L2 Cache
—
50 MB
Performance
Pixel Rate
8.800 GPixel/s
47.16 GPixel/s
Texture Rate
35.20 GTexel/s
1,163.3 GTexel/s
FP32 (TFLOPS)
1,126.4 GFLOPS
74.45 TFLOPS
FP64 (TFLOPS)
70.40 GFLOPS (1:16)
37.22 TFLOPS (1:2)
FP16 (TFLOPS)
2.253 TFLOPS (2:1)
1,191.2 TFLOPS (16:1)
AI/RT
Tensor Cores
—
592
Power
TDP
25 W
1000 W
TDP (W)
25
1,000
+3900.0%
Suggested PSU
—
1400 W
Power Connectors
None
—
Architecture
Architecture
GCN 5.0
Blackwell
GPU Name
Raven
GB100
Generation
Vega IGP
(Raven Ridge)
Server Blackwell
(Bxx)
Process Size
14 nm
5 nm
Transistors
4,940 million
104,000 million
Die Size
210 mm²
—
Foundry
GlobalFoundries
TSMC
Density
23.5M / mm²
—
API Support
DirectX
12 (12_1)
—
OpenGL
4.6
—
Vulkan
1.3
—
OpenCL
2.1
3.0
CUDA
—
10.0
Shader Model
6.7
—
Physical
Slot Width
IGP
SXM Module
Outputs
Motherboard Dependent
No outputs
Bus Interface
IGP
PCIe 5.0 x16
Other
Production
End-of-life
Active
Predecessor
GCN 3.0 IGP
Server Hopper
Successor
Vega II IGP
Server Rubin