AMD Radeon Instinct MI8 vs NVIDIA L40G
We compared two Professional market GPUs: 4GB VRAM Radeon Instinct MI8 and 24GB VRAM L40G to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
AMD Radeon Instinct MI8 's Advantages
Lower TDP (175W vs 300W)
NVIDIA L40G 's Advantages
Released 5 years and 10 months late
Boost Clock2475MHz
More VRAM (24GB vs 4GB)
Larger VRAM bandwidth (864.0GB/s vs 512.0GB/s)
14080 additional rendering cores
Score
Benchmark
FP32 (float)
Radeon Instinct MI8
8.192 TFLOPS
L40G
+998%
89.97 TFLOPS
Graphics Card
Dec 2016
Release Date
Oct 2022
Radeon Instinct
Generation
Tesla Ada
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
-
Base Clock
1005 MHz
-
Boost Clock
2475 MHz
500 MHz
Memory Clock
2250 MHz
Memory
4GB
Memory Size
24GB
HBM
Memory Type
GDDR6
4096bit
Memory Bus
384bit
512.0GB/s
Bandwidth
864.0GB/s
Render Config
64
Compute Units
-
-
SM Count
142
4096
Shading Units
18176
256
TMUs
568
64
ROPs
192
-
Tensor Cores
568
-
RT Cores
142
16 KB (per CU)
L1 Cache
128 KB (per SM)
2 MB
L2 Cache
48 MB
-
-
-
Theoretical Performance
64.00 GPixel/s
Pixel Rate
475.2 GPixel/s
256.0 GTexel/s
Texture Rate
1406 GTexel/s
8.192 TFLOPS
FP16 (half)
89.97 TFLOPS
8.192 TFLOPS
FP32 (float)
89.97 TFLOPS
512.0 GFLOPS
FP64 (double)
1406 GFLOPS
Board Design
175W
TDP
300W
450 W
Suggested PSU
700 W
No outputs
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
1x 8-pin
Power Connectors
1x 16-pin
Graphics Processor
Fiji
GPU Name
AD102
Fiji XT CA
(215-0862120)
GPU Variant
AD102-???-A1
GCN 3.0
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
28 nm
Process Size
4 nm
8.9 billion
Transistors
76.3 billion
596 mm²
Die Size
608 mm²
Graphics Features
12 (12_0)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
2.1
OpenCL
3.0
1.2.170
Vulkan
1.3
-
CUDA
8.9
6.5
Shader Model
6.6
