NVIDIA TITAN X Pascal vs NVIDIA GeForce RTX 3070
We compared two Desktop platform GPUs: 12GB VRAM TITAN X Pascal and 8GB VRAM GeForce RTX 3070 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA TITAN X Pascal 's Advantages
More VRAM (12GB vs 8GB)
Larger VRAM bandwidth (480.4GB/s vs 448.0GB/s)
NVIDIA GeForce RTX 3070 's Advantages
Released 4 years and 1 months late
Boost Clock has increased by 13% (1725MHz vs 1531MHz)
2304 additional rendering cores
Lower TDP (220W vs 250W)
Score
Benchmark
FP32 (float)
TITAN X Pascal
10.97 TFLOPS
GeForce RTX 3070
+85%
20.31 TFLOPS
3DMark Time Spy
TITAN X Pascal
9844
GeForce RTX 3070
+38%
13641
3DMark Time Spy Extreme
TITAN X Pascal
4812
GeForce RTX 3070
+40%
6762
Shadow of the Tomb Raider 2160p
TITAN X Pascal
42
GeForce RTX 3070
+26%
53
Shadow of the Tomb Raider 1440p
TITAN X Pascal
78
GeForce RTX 3070
+21%
95
Shadow of the Tomb Raider 1080p
TITAN X Pascal
128
GeForce RTX 3070
+12%
144
GTA 5 2160p
TITAN X Pascal
+44%
98
GeForce RTX 3070
68
GTA 5 1440p
TITAN X Pascal
104
GeForce RTX 3070
105
GTA 5 1080p
TITAN X Pascal
+18%
188
GeForce RTX 3070
159
Graphics Card
Aug 2016
Release Date
Sep 2020
GeForce 10
Generation
GeForce 30
Desktop
Type
Desktop
PCIe 3.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1417 MHz
Base Clock
1500 MHz
1531 MHz
Boost Clock
1725 MHz
1251 MHz
Memory Clock
1750 MHz
Memory
12GB
Memory Size
8GB
GDDR5X
Memory Type
GDDR6
384bit
Memory Bus
256bit
480.4GB/s
Bandwidth
448.0GB/s
Render Config
-
-
-
28
SM Count
46
3584
Shading Units
5888
224
TMUs
184
96
ROPs
96
-
Tensor Cores
184
-
RT Cores
46
48 KB (per SM)
L1 Cache
128 KB (per SM)
3 MB
L2 Cache
4 MB
-
-
-
Theoretical Performance
147.0 GPixel/s
Pixel Rate
165.6 GPixel/s
342.9 GTexel/s
Texture Rate
317.4 GTexel/s
171.5 GFLOPS
FP16 (half)
20.31 TFLOPS
10.97 TFLOPS
FP32 (float)
20.31 TFLOPS
342.9 GFLOPS
FP64 (double)
317.4 GFLOPS
Board Design
250W
TDP
220W
600 W
Suggested PSU
550 W
1x DVI
1x HDMI 2.0
3x DisplayPort 1.4a
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
1x 6-pin + 1x 8-pin
Power Connectors
1x 12-pin
Graphics Processor
GP102
GPU Name
GA104
GP102-400-A1
GPU Variant
GA104-300-A1
Pascal
Architecture
Ampere
TSMC
Foundry
Samsung
16 nm
Process Size
8 nm
11.8 billion
Transistors
17.4 billion
471 mm²
Die Size
392 mm²
Graphics Features
12 (12_1)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
6.1
CUDA
8.6
6.4
Shader Model
6.6
