GeForce RTX 5070 Ti vs 4070 SUPER
VS
Nvidia GeForce RTX 5070 Ti
GB203-200-A1
We compared two discrete desktop gaming GPUs: the GeForce RTX 5070 Ti 16 GB with 70 pipelines and 8960 shaders against the 1 year older RTX 4070 SUPER 12 GB that utilizes 56 pipelines and 7168 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Review
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score
Value for money (Beta)
Key differences
Reasons to consider the GeForce RTX 5070 Ti
- Performs better (up to 39%) in 3DMark Steel Nomad Lite
- Manufactured using a more efficient 4 nm process technology
- 24% higher maximum theoretical performance (43.9 vs 35.5 TFLOPS)
- Includes 4 GB more video memory
- Has 78% higher memory bandwidth: 896 vs 504.2 GB/s
- Supports Nvidia DLSS 4 technology
- Has 25% more shading units (8960 vs 7168)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 229 164 155 121 |
- |
| The Witcher 3 | 244 215 155 83 |
- |
| Counter-Strike 2 | 298 221 172 86 |
- |
| Far Cry 6 | 178 158 146 89 |
- |
| Hogwarts Legacy | 160 131 107 58 |
- |
| CoD: Modern Warfare III | 212 211 163 115 |
- |
| Ghost of Tsushima | 144 122 109 70 |
- |
| Cyberpunk 2077 | 193 176 122 55 |
- |
| Shadow of the Tomb Raider | 279 265 207 106 |
- |
| 1080p High | 215 | - |
| 1080p Ultra | 185 | - |
| 1440p Ultra | 148 | - |
| 4K Ultra | 87 | - |
| Margin of Error | Low | High |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 27839 | 20879 |
| Solar Bay | 137852 | 98188 |
| Port Royal | 19713 | 13123 |
| Fire Strike | 67490 | 50092 |
| Wild Life Extreme | 58033 | 41625 |
| Night Raid | 188802 | 164614 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 297.9 img/sec | 359.5 img/sec |
| Face Detection | 218.6 img/sec | 210 img/sec |
| Horizon Detection | 10.1 Gpixels/sec | 7.61 Gpixels/sec |
| Edge Detection | 13.9 Gpixels/sec | 10.4 Gpixels/sec |
| Gaussian Blur | 14.8 Gpixels/sec | 11.3 Gpixels/sec |
| Feature Matching | 2.03 Gpixels/sec | 1.97 Gpixels/sec |
| Stereo Matching | 1290 Gpixels/sec | 969.3 Gpixels/sec |
| Particle Physics | 35438.3 FPS | 29061.4 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1323 | 1193 |
| DirectX 11 | 303 FPS | 271 FPS |
| DirectX 12 | 126 FPS | 110 FPS |
| GPU Compute | 17781 Ops/s | 16925 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 5070 Ti
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-08-28 (NOX) | Cinebench 2024 | 88980 |
GeForce RTX 4070 SUPER
| Date | Benchmark | Result |
|---|---|---|
| 📘 2025-12-08 (Rickey) | Cinebench 2024 | 19110 |
| 📘 2025-11-28 (Caleb) | Geekbench 6 OpenCL | 212148 |
| 📘 2025-11-28 (Caleb) | Cinebench 2024 | 20703 |
Artificial Intelligence Tests
Performance in machine learning and artificial intelligence tasksGeekBench 6 ML
GB6 ML Single Precision
GB6 ML Half Precision
GB6 ML Quantized
| Image Classification (SP) | 14913 | 12146 |
| Image Segmentation (HP) | 45199 | 23932 |
| Image Super Resolution (Q) | 39360 | 36321 |
| Face Detection (HP) | 84188 | 52836 |
| Pose Estimation (Q) | 161400 | 179788 |
| Text Classification (SP) | 4010 | 3371 |
| Machine Translation (HP) | 7973 | 5017 |
| Object Detection (SP) | 18934 | 15164 |
| Depth Estimation (Q) | 67761 | 62500 |
| Style Transfer (SP) | 335879 | 357489 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 5070 Ti and 4070 SUPERGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | January 7, 2025 | January 17, 2024 |
| Launch price (MSRP) | $749 | $599 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | Mid-range |
| Architecture | Blackwell 2.0 | Ada Lovelace |
| GPU Codename | GB203-200-A1 | AD104 |
| Rival Equivalent | - Radeon RX 9070 XT | - Radeon RX 7900 XT |
| Recommended CPU | - Intel Core Ultra 5 245K or above | - Intel Core i7 14700K or above |
Desktop GPU rating (8th and 19th place)
Graphics Processing Unit
| Base Clock | 2295 MHz | 1980 MHz |
| Boost Clock | 2452 MHz | 2475 MHz |
| Shading Units | 8960 | 7168 |
| Texture Mapping Units (TMUs) | 280 | 224 |
| Render Output Units (ROPs) | 96 | 80 |
| Compute Units (Pipelines) | 70 | 56 |
| Tensor Cores | 280 | 224 |
| Ray-tracing Cores | 70 | 56 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 48MB shared | 48MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 235 GPixel/s | 198 GPixel/s |
| Texture Fill Rate | 687 GTexel/s | 554 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | PCIe 4.0 x16 |
| TGP | 300 W | 220 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 4 nm | 5 nm |
| Die Size | 378 mm² | 294 mm² |
| Transistor Count | 45.6 billion | 35 billion |
| Transistor Density | 120.63 MTr/mm² | 119.05 MTr/mm² |
Memory
| Memory Type | GDDR7 | GDDR6X |
| Memory Size | 16 GB | 12 GB |
| Memory Clock | 1750 MHz | 1313 MHz |
| Effective Memory Speed | 28000 Mbps | 21000 Mbps |
| Bus | 256-bit | 192-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12.2 | 12 |
| Vulkan | 1.4 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12.0 | 8.9 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 3 |
| DisplayPort | 2.1b | 1.4a |
Cast your vote
Total votes: 156
