ATI FireGL V8650 vs NVIDIA B100
We compared two Desktop platform GPUs: 2GB VRAM FireGL V8650 and 96GB VRAM B100 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA B100 's Advantages
Released 17 years and 3 months late
Boost Clock1837MHz
More VRAM (96GB vs 2GB)
Larger VRAM bandwidth (4.10TB/s vs 111.1GB/s)
16576 additional rendering cores
Score
Benchmark
FP32 (float)
FireGL V8650
0.44 TFLOPS
B100
+14009%
62.08 TFLOPS
Graphics Card
Aug 2007
Release Date
Nov 2024
FireGL
Generation
Server Blackwell
Desktop
Type
Desktop
PCIe 1.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
-
Base Clock
1665 MHz
-
Boost Clock
1837 MHz
868 MHz
Memory Clock
2000 MHz
Memory
2GB
Memory Size
96GB
GDDR4
Memory Type
HBM3e
512bit
Memory Bus
4096bit
111.1GB/s
Bandwidth
4.10TB/s
Render Config
4
Compute Units
-
-
SM Count
132
320
Shading Units
16896
16
TMUs
528
16
ROPs
24
-
Tensor Cores
528
-
-
-
-
L1 Cache
256 KB (per SM)
256 KB
L2 Cache
50 MB
-
-
-
Theoretical Performance
11.01 GPixel/s
Pixel Rate
44.09 GPixel/s
11.01 GTexel/s
Texture Rate
969.9 GTexel/s
-
FP16 (half)
248.3 TFLOPS
440.3 GFLOPS
FP32 (float)
62.08 TFLOPS
-
FP64 (double)
31.04 TFLOPS
Board Design
Unknown
TDP
1000W
200 W
Suggested PSU
1400 W
2x DVI
1x S-Video
Outputs
No outputs
1x 6-pin + 1x 8-pin
Power Connectors
-
Graphics Processor
R600
GPU Name
GB102
R600 V8650
(215RGIDKA13FG)
GPU Variant
-
TeraScale
Architecture
Blackwell
TSMC
Foundry
TSMC
80 nm
Process Size
5 nm
0.72 billion
Transistors
104 billion
420 mm²
Die Size
Unknown
Graphics Features
10.0 (10_0)
DirectX
N/A
3.3 (full)
4.0 (partial)
OpenGL
N/A
N/A
OpenCL
3.0
N/A
Vulkan
N/A
-
CUDA
10.1
4.0
Shader Model
N/A
