NVIDIA GeForce RTX 3080 Ti vs NVIDIA GeForce RTX 4070 Ti
We compared two Desktop platform GPUs: 12GB VRAM GeForce RTX 3080 Ti and 12GB VRAM GeForce RTX 4070 Ti to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 3080 Ti 's Advantages
Larger VRAM bandwidth (912.4GB/s vs 504.2GB/s)
2560 additional rendering cores
NVIDIA GeForce RTX 4070 Ti 's Advantages
Released 1 years and 8 months late
Boost Clock has increased by 57% (2610MHz vs 1665MHz)
Lower TDP (285W vs 350W)
Score
Benchmark
FP32 (float)
GeForce RTX 3080 Ti
34.1 TFLOPS
GeForce RTX 4070 Ti
+17%
40.09 TFLOPS
3DMark Time Spy
GeForce RTX 3080 Ti
19629
GeForce RTX 4070 Ti
+16%
22771
3DMark Time Spy Extreme
GeForce RTX 3080 Ti
10032
GeForce RTX 4070 Ti
+8%
10894
3DMark Speed Way
GeForce RTX 3080 Ti
5330
GeForce RTX 4070 Ti
+3%
5508
Shadow of the Tomb Raider 2160p
GeForce RTX 3080 Ti
92
GeForce RTX 4070 Ti
+10%
102
Shadow of the Tomb Raider 1440p
GeForce RTX 3080 Ti
153
GeForce RTX 4070 Ti
+30%
200
Shadow of the Tomb Raider 1080p
GeForce RTX 3080 Ti
199
GeForce RTX 4070 Ti
+45%
290
Cyberpunk 2077 2160p
GeForce RTX 3080 Ti
+13%
66
GeForce RTX 4070 Ti
58
Cyberpunk 2077 1440p
GeForce RTX 3080 Ti
77
GeForce RTX 4070 Ti
+49%
115
Cyberpunk 2077 1080p
GeForce RTX 3080 Ti
112
GeForce RTX 4070 Ti
+42%
160
Battlefield 5 2160p
GeForce RTX 3080 Ti
114
GeForce RTX 4070 Ti
+9%
125
Battlefield 5 1440p
GeForce RTX 3080 Ti
187
GeForce RTX 4070 Ti
+6%
200
Battlefield 5 1080p
GeForce RTX 3080 Ti
199
GeForce RTX 4070 Ti
200
GTA 5 2160p
GeForce RTX 3080 Ti
100
GeForce RTX 4070 Ti
+50%
150
GTA 5 1440p
GeForce RTX 3080 Ti
150
GeForce RTX 4070 Ti
+13%
170
GTA 5 1080p
GeForce RTX 3080 Ti
186
GeForce RTX 4070 Ti
+2%
190
Graphics Card
May 2021
Release Date
Jan 2023
GeForce 30
Generation
GeForce 40
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1365 MHz
Base Clock
2310 MHz
1665 MHz
Boost Clock
2610 MHz
1188 MHz
Memory Clock
1313 MHz
Memory
12GB
Memory Size
12GB
GDDR6X
Memory Type
GDDR6X
384bit
Memory Bus
192bit
912.4GB/s
Bandwidth
504.2GB/s
Render Config
-
-
-
80
SM Count
60
10240
Shading Units
7680
320
TMUs
240
112
ROPs
80
320
Tensor Cores
240
80
RT Cores
60
128 KB (per SM)
L1 Cache
128 KB (per SM)
6 MB
L2 Cache
48 MB
-
-
-
Theoretical Performance
186.5 GPixel/s
Pixel Rate
208.8 GPixel/s
532.8 GTexel/s
Texture Rate
626.4 GTexel/s
34.10 TFLOPS
FP16 (half)
40.09 TFLOPS
34.10 TFLOPS
FP32 (float)
40.09 TFLOPS
532.8 GFLOPS
FP64 (double)
626.4 GFLOPS
Board Design
350W
TDP
285W
750 W
Suggested PSU
600 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
1x 12-pin
Power Connectors
1x 16-pin
Graphics Processor
GA102
GPU Name
AD104
GA102-225-A1
GPU Variant
AD104-400-A1
Ampere
Architecture
Ada Lovelace
Samsung
Foundry
TSMC
8 nm
Process Size
4 nm
28.3 billion
Transistors
35.8 billion
628 mm²
Die Size
295 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
8.6
CUDA
8.9
6.6
Shader Model
6.7
