Intel Data Center GPU Flex 170 vs NVIDIA Tesla P40
We compared two Professional market GPUs: 16GB VRAM Data Center GPU Flex 170 and 24GB VRAM Tesla P40 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
Intel Data Center GPU Flex 170 's Advantages
Released 5 years and 11 months late
Boost Clock has increased by 34% (2050MHz vs 1531MHz)
Larger VRAM bandwidth (512.0GB/s vs 347.1GB/s)
256 additional rendering cores
Lower TDP (150W vs 250W)
NVIDIA Tesla P40 's Advantages
More VRAM (24GB vs 16GB)
Score
Benchmark
FP32 (float)
Data Center GPU Flex 170
+42%
16.79 TFLOPS
Tesla P40
11.76 TFLOPS
Graphics Card
Aug 2022
Release Date
Sep 2016
Data Center GPU
Generation
Tesla Pascal
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
1950 MHz
Base Clock
1303 MHz
2050 MHz
Boost Clock
1531 MHz
2000 MHz
Memory Clock
1808 MHz
Memory
16GB
Memory Size
24GB
GDDR6
Memory Type
GDDR5
256bit
Memory Bus
384bit
512.0GB/s
Bandwidth
347.1GB/s
Render Config
-
-
-
-
SM Count
30
4096
Shading Units
3840
256
TMUs
240
128
ROPs
96
-
-
-
32
RT Cores
-
-
L1 Cache
48 KB (per SM)
16 MB
L2 Cache
3 MB
-
-
-
Theoretical Performance
262.4 GPixel/s
Pixel Rate
147.0 GPixel/s
524.8 GTexel/s
Texture Rate
367.4 GTexel/s
33.59 TFLOPS
FP16 (half)
183.7 GFLOPS
16.79 TFLOPS
FP32 (float)
11.76 TFLOPS
-
FP64 (double)
367.4 GFLOPS
Board Design
150W
TDP
250W
450 W
Suggested PSU
600 W
1x HDMI 2.1
3x DisplayPort 2.0
Outputs
No outputs
1x 8-pin
Power Connectors
8-pin EPS
Graphics Processor
DG2-512
GPU Name
GP102
ACM-G10
GPU Variant
-
Generation 12.7
Architecture
Pascal
TSMC
Foundry
TSMC
6 nm
Process Size
16 nm
21.7 billion
Transistors
11.8 billion
406 mm²
Die Size
471 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 (12_1)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
-
CUDA
6.1
6.6
Shader Model
6.7
