NVIDIA GeForce RTX 3070 Ti GA102 vs NVIDIA GeForce GT 640 OEM Rebrand
We compared two Desktop platform GPUs: 8GB VRAM GeForce RTX 3070 Ti GA102 and 1536MB VRAM GeForce GT 640 OEM Rebrand to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 3070 Ti GA102 's Advantages
Released 10 years and 6 months late
Boost Clock1770MHz
More VRAM (8GB vs 1536GB)
Larger VRAM bandwidth (608.3GB/s vs 38.40GB/s)
6000 additional rendering cores
NVIDIA GeForce GT 640 OEM Rebrand 's Advantages
Lower TDP (75W vs 290W)
Score
Benchmark
FP32 (float)
GeForce RTX 3070 Ti GA102
+5140%
21.75 TFLOPS
GeForce GT 640 OEM Rebrand
0.415 TFLOPS
Graphics Card
Oct 2022
Release Date
Apr 2012
GeForce 30
Generation
GeForce 600
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 2.0 x16
Clock Speeds
1575 MHz
Base Clock
-
1770 MHz
Boost Clock
-
1188 MHz
Memory Clock
800 MHz
Memory
8GB
Memory Size
1536MB
GDDR6X
Memory Type
DDR3
256bit
Memory Bus
192bit
608.3GB/s
Bandwidth
38.40GB/s
Render Config
-
-
-
48
SM Count
3
6144
Shading Units
144
192
TMUs
24
96
ROPs
24
192
Tensor Cores
-
48
RT Cores
-
128 KB (per SM)
L1 Cache
64 KB (per SM)
4 MB
L2 Cache
384 KB
-
-
-
Theoretical Performance
169.9 GPixel/s
Pixel Rate
4.320 GPixel/s
339.8 GTexel/s
Texture Rate
17.28 GTexel/s
21.75 TFLOPS
FP16 (half)
-
21.75 TFLOPS
FP32 (float)
414.7 GFLOPS
339.8 GFLOPS
FP64 (double)
34.56 GFLOPS
Board Design
290W
TDP
75W
600 W
Suggested PSU
250 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
2x DVI
1x mini-HDMI 1.3a
1x 12-pin
Power Connectors
None
Graphics Processor
GA102
GPU Name
GF116
GA102-150-A1
GPU Variant
GF116-150-A1
Ampere
Architecture
Fermi 2.0
Samsung
Foundry
TSMC
8 nm
Process Size
40 nm
28.3 billion
Transistors
1.17 billion
628 mm²
Die Size
238 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 (11_0)
4.6
OpenGL
4.6
3.0
OpenCL
1.1
1.3
Vulkan
N/A
8.6
CUDA
2.1
6.6
Shader Model
5.1
