Nvidia RTX 4070 vs Nvidia GTX 1080 Ti
The Nvidia RTX 4070 is a next-generation graphics card based on the Ada Lovelace architecture, featuring 5888 CUDA cores, 12GB of GDDR6X memory, and a 192-bit memory bus. It supports ray tracing, DLSS 3, and other advanced features. In comparison, the Nvidia GTX 1080 Ti is an older graphics card based on the Pascal architecture, featuring 3584 CUDA cores, 11GB of GDDR5X memory, and a 352-bit memory bus. It does not support ray tracing or DLSS. Overall, the RTX 4070 offers superior performance, features, and efficiency compared to the GTX 1080 Ti.
Nvidia RTX 4070
Pros:
More powerful, especially in ray tracing and DLSS games
Has more features, such as ray tracing, DLSS 3, and Reflex
Higher memory capacity and bandwidth
More power efficient
Cons:
More expensive than the GTX 1080 Ti
May require a more powerful power supply
Nvidia GTX 1080 Ti
Pros:
More affordable
Still a powerful graphics card that can play most games
Lower power consumption and does not require a particularly powerful power supply
Cons:
Not as powerful as the RTX 4070, especially in ray tracing and DLSS games
Lacks modern features such as ray tracing and DLSS
Lower memory capacity and bandwidth
Graphics core
Release Date | April 12, 2023 | March 10, 2017 |
Architecture | Ada Lovelace | Pascal |
Applicable Type | Desktop Graphics Card | Desktop Graphics Card |
Number of Streaming Multiprocessors | 58 | 28 |
Bus Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Memory
Memory Type | GDDR6X | GDDR5X |
Memory Capacity | 12GB | 11GB |
Memory Bus Width | 192-bit | 352-bit |
Memory Speed | 21Gbps | 11Gbps |
Clock speed
Base Clock Speed | 1905MHz | 1480MHz |
Boost Clock Speed | 2475MHz | 1582MHz |
Render Config
Raster Operations Units | 128 | 88 |
Texture Units | 192 | 224 |
RT Cores | 48 | |
Tensor Cores | 384 | |
CUDA Cores | 3584 |
Theoretical Performance
FP32 Performance | 30.3 TFLOPS | 11.3 TFLOPS |
Ray Tracing Performance | 60 TFLOPS | |
Tensor Performance | 120 TFLOPS |
Graphics Processor
Manufacturing Process | TSMC 4N | TSMC 16nm FinFET |
Number of Transistors | 45.9 billion | 12 billion |
Die Size | 295mm² | 471mm² |
Board Design
Length | 285mm | 269mm |
Width | 112mm | 111mm |
Thickness | Dual-slot | Dual-slot |
Power Connectors | One 16-pin and one 8-pin | One 8-pin and one 6-pin |
TDP | 285W | 250W |
Benchmark Comparison
The Nvidia RTX 4070 is the latest generation of Ampere architecture graphics card, while the Nvidia GTX 1080 Ti is the previous generation of Pascal architecture graphics card. There are performance gaps between the two, especially with the RTX 4070 having a significant advantage in ray tracing and DLSS.
Game | Resolution | RTX 4070 | GTX 1080 Ti | Performance Increase |
Assassin's Creed Valhalla | 1440p | 108 FPS | 64 FPS | 69% |
Battlefield 2042 | 1440p | 122 FPS | 72 FPS | 70% |
Control | 1440p | 101 FPS | 60 FPS | 68% |
Shadow of the Tomb Raider | 1440p | 132 FPS | 77 FPS | 71% |
Cyberpunk 2077 | 1440p | 87 FPS | 49 FPS | 78% |
Ray Tracing:
The RTX 4070 supports second-generation ray tracing technology, and its performance in ray tracing games is significantly better than the GTX 1080 Ti. Here is some ray tracing test data for some games:
Game | Resolution | RTX 4070 | GTX 1080 Ti | Performance Increase |
Battlefield 2042 | 1440p | 64 FPS | 37 FPS | 73% |
Control | 1440p | 57 FPS | 33 FPS | 73% |
Shadow of the Tomb Raider | 1440p | 62 FPS | 36 FPS | 72% |
Cyberpunk 2077 | 1440p | 47 FPS | 26 FPS | 81% |
DLSS:
The RTX 4070 supports DLSS 3.0 technology, which can significantly improve game frame rates while maintaining image quality. Here is some test data for some games using DLSS:
Game | Resolution | DLSS | RTX 4070 | GTX 1080 Ti | Performance Increase |
Assassin's Creed Valhalla | 1440p | Quality | 165 FPS | 108 FPS | 53% |
Battlefield 2042 | 1440p | Quality | 180 FPS | 122 FPS | 47% |
Control | 1440p | Quality | 150 FPS | 101 FPS | 49% |
Shadow of the Tomb Raider | 1440p | Quality | 190 FPS | 132 FPS | 44% |
Cyberpunk 2077 | 1440p | Quality | 125 FPS | 87 FPS | 44% |
Power Consumption:
The power consumption of the RTX 4070 is slightly higher than that of the GTX 1080 Ti, but it is still within the acceptable range. Here is the power consumption test data for the two graphics cards:
Graphics Card | Power Consumption |
RTX 4070 | 285W |
GTX 1080 Ti | 250W |
Conclusion:
The Nvidia RTX 4070 outperforms the Nvidia GTX 1080 Ti in terms of performance, ray tracing, DLSS, and power consumption. If you are looking for a powerful graphics card that supports the latest technologies, the RTX 4070 is a good option.