Radeon RX 7900 XT vs GeForce RTX 3070 Ti
We compared two Desktop platform GPUs: 20GB VRAM Radeon RX 7900 XT and 8GB VRAM GeForce RTX 3070 Ti to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
Radeon RX 7900 XT 's Advantages
Released 1 years and 6 months late
Boost Clock has increased by 35% (2394MHz vs 1770MHz)
More VRAM (20GB vs 8GB)
Larger VRAM bandwidth (800.0GB/s vs 608.3GB/s)
GeForce RTX 3070 Ti 's Advantages
768 additional rendering cores
Lower TDP (290W vs 300W)
Score
Benchmark
FP32 (float)
Radeon RX 7900 XT
+136%
51480
GeForce RTX 3070 Ti
21750
3DMark Time Spy
Radeon RX 7900 XT
+80%
26829
GeForce RTX 3070 Ti
14872
3DMark Time Spy Extreme
Radeon RX 7900 XT
+77%
13130
GeForce RTX 3070 Ti
7416
3DMark Speed Way
Radeon RX 7900 XT
+42%
5364
GeForce RTX 3070 Ti
3761
Shadow of the Tomb Raider 2160p
Radeon RX 7900 XT
+45%
102
GeForce RTX 3070 Ti
70
Shadow of the Tomb Raider 1440p
Radeon RX 7900 XT
+64%
205
GeForce RTX 3070 Ti
125
Shadow of the Tomb Raider 1080p
Radeon RX 7900 XT
+67%
299
GeForce RTX 3070 Ti
178
Cyberpunk 2077 2160p
Radeon RX 7900 XT
50
GeForce RTX 3070 Ti
+6%
53
Cyberpunk 2077 1440p
Radeon RX 7900 XT
+44%
94
GeForce RTX 3070 Ti
65
Cyberpunk 2077 1080p
Radeon RX 7900 XT
+64%
158
GeForce RTX 3070 Ti
96
GTA 5 2160p
Radeon RX 7900 XT
+123%
170
GeForce RTX 3070 Ti
76
GTA 5 1440p
Radeon RX 7900 XT
+49%
170
GeForce RTX 3070 Ti
114
GTA 5 1080p
Radeon RX 7900 XT
+15%
190
GeForce RTX 3070 Ti
164
Horizon Forbidden West 1080p
Radeon RX 7900 XT
+55%
131
GeForce RTX 3070 Ti
84
Horizon Forbidden West 1440p
Radeon RX 7900 XT
+64%
105
GeForce RTX 3070 Ti
64
Horizon Forbidden West 2160p
Radeon RX 7900 XT
+61%
68
GeForce RTX 3070 Ti
42
Graphics Card
Nov 2022
Release Date
May 2021
Navi III
Generation
GeForce 30
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1387 MHz
Base Clock
1575 MHz
2394 MHz
Boost Clock
1770 MHz
2500 MHz
Memory Clock
1188 MHz
Memory
20GB
Memory Size
8GB
GDDR6
Memory Type
GDDR6X
320bit
Memory Bus
256bit
800.0GB/s
Bandwidth
608.3GB/s
Render Config
84
Compute Units
-
-
SM Count
48
5376
Shading Units
6144
336
TMUs
192
192
ROPs
96
-
Tensor Cores
192
84
RT Cores
48
256 KB per Array
L1 Cache
128 KB (per SM)
6 MB
L2 Cache
4 MB
80 MB
L3 Cache
-
Theoretical Performance
459.6 GPixel/s
Pixel Rate
169.9 GPixel/s
804.4 GTexel/s
Texture Rate
339.8 GTexel/s
103.0 TFLOPS
FP16 (half)
21.75 TFLOPS
51.48 TFLOPS
FP32 (float)
21.75 TFLOPS
1.609 TFLOPS
FP64 (double)
339.8 GFLOPS
Board Design
300W
TDP
290W
700 W
Suggested PSU
600 W
1x HDMI 2.1a
2x DisplayPort 2.1
1x USB Type-C
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
2x 8-pin
Power Connectors
1x 12-pin
Graphics Processor
Navi 31
GPU Name
GA104
Navi 31 XT
(215-145000156)
GPU Variant
GA104-400-A1
RDNA 3.0
Architecture
Ampere
TSMC
Foundry
Samsung
5 nm
Process Size
8 nm
57.7 billion
Transistors
17.4 billion
529 mm²
Die Size
392 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
2.2
OpenCL
3.0
1.3
Vulkan
1.3
-
CUDA
8.6
6.7
Shader Model
6.6