AMD FireStream 9170 vs NVIDIA L20
We compared two Desktop platform GPUs: 2GB VRAM FireStream 9170 and 48GB VRAM L20 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
AMD FireStream 9170 's Advantages
Lower TDP (105W vs 275W)
NVIDIA L20 's Advantages
Released 16 years late
Boost Clock2520MHz
More VRAM (48GB vs 2GB)
Larger VRAM bandwidth (864.0GB/s vs 51.33GB/s)
11456 additional rendering cores
Score
Benchmark
FP32 (float)
FireStream 9170
0.497 TFLOPS
L20
+11841%
59.35 TFLOPS
Graphics Card
Nov 2007
Release Date
Nov 2023
FireStream
Generation
Tesla Ada
Desktop
Type
Desktop
PCIe 2.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
-
Base Clock
1440 MHz
-
Boost Clock
2520 MHz
802 MHz
Memory Clock
2250 MHz
Memory
2GB
Memory Size
48GB
GDDR3
Memory Type
GDDR6
256bit
Memory Bus
384bit
51.33GB/s
Bandwidth
864.0GB/s
Render Config
4
Compute Units
-
-
SM Count
92
320
Shading Units
11776
16
TMUs
368
16
ROPs
128
-
Tensor Cores
368
-
RT Cores
92
-
L1 Cache
128 KB (per SM)
256 KB
L2 Cache
96 MB
-
-
-
Theoretical Performance
12.43 GPixel/s
Pixel Rate
322.6 GPixel/s
12.43 GTexel/s
Texture Rate
927.4 GTexel/s
-
FP16 (half)
59.35 TFLOPS
497.3 GFLOPS
FP32 (float)
59.35 TFLOPS
99.46 GFLOPS
FP64 (double)
927.4 GFLOPS
Board Design
105W
TDP
275W
300 W
Suggested PSU
600 W
2x DVI
1x S-Video
Outputs
4x DisplayPort 1.4a
1x 6-pin
Power Connectors
1x 16-pin
Graphics Processor
RV670
GPU Name
AD102
-
GPU Variant
AD102
TeraScale
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
55 nm
Process Size
5 nm
0.666 billion
Transistors
76.3 billion
192 mm²
Die Size
609 mm²
Graphics Features
10.1 (10_1)
DirectX
12 Ultimate (12_2)
3.3 (full)
4.0 (partial)
OpenGL
4.6
N/A
OpenCL
3.0
N/A
Vulkan
1.3
-
CUDA
8.9
4.1
Shader Model
6.7
