NVIDIA GeForce RTX 4070 10 GB vs NVIDIA P102 100
We compared two Desktop platform GPUs: 10GB VRAM GeForce RTX 4070 10 GB and 5GB VRAM P102 100 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 4070 10 GB 's Advantages
Boost Clock has increased by 50% (2520MHz vs 1683MHz)
More VRAM (10GB vs 5GB)
3968 additional rendering cores
Lower TDP (220W vs 250W)
NVIDIA P102 100 's Advantages
Larger VRAM bandwidth (440.3GB/s vs 420.2GB/s)
Score
Benchmark
FP32 (float)
GeForce RTX 4070 10 GB
+235%
36.13 TFLOPS
P102 100
10.77 TFLOPS
Graphics Card
Unknown
Release Date
Feb 2018
GeForce 40
Generation
Mining GPUs
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 3.0 x4
Clock Speeds
2355 MHz
Base Clock
1582 MHz
2520 MHz
Boost Clock
1683 MHz
1313 MHz
Memory Clock
1376 MHz
Memory
10GB
Memory Size
5GB
GDDR6X
Memory Type
GDDR5X
160bit
Memory Bus
320bit
420.2GB/s
Bandwidth
440.3GB/s
Render Config
-
-
-
56
SM Count
25
7168
Shading Units
3200
224
TMUs
200
80
ROPs
80
224
Tensor Cores
-
56
RT Cores
-
128 KB (per SM)
L1 Cache
48 KB (per SM)
48 MB
L2 Cache
0 MB
-
-
-
Theoretical Performance
201.6 GPixel/s
Pixel Rate
134.6 GPixel/s
564.5 GTexel/s
Texture Rate
336.6 GTexel/s
36.13 TFLOPS
FP16 (half)
168.3 GFLOPS
36.13 TFLOPS
FP32 (float)
10.77 TFLOPS
564.5 GFLOPS
FP64 (double)
336.6 GFLOPS
Board Design
220W
TDP
250W
550 W
Suggested PSU
600 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
No outputs
1x 16-pin
Power Connectors
2x 8-pin
Graphics Processor
AD104
GPU Name
GP102
AD104-275-A1
GPU Variant
GP102-100-A1
Ada Lovelace
Architecture
Pascal
TSMC
Foundry
TSMC
5 nm
Process Size
16 nm
35.8 billion
Transistors
11.8 billion
294 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
8.9
CUDA
6.1
6.7
Shader Model
6.4
