GeForce RTX 3080 Ti vs GeForce RTX 4090
[ VERDICT ]
Winner: GeForce RTX 4090 (89/100)
GeForce RTX 4090 leads overall with a score of 89/100 compared to GeForce RTX 3080 Ti (59/100), offering superior measured performance and technical specifications.
Benchmark Performance Comparison
Normalized Lab Scores| Benchmark | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Efficiency | 35.0 Poor | 43.0 Poor WINNER |
| Gaming | 60.0 Average | 93.0 Excellent WINNER |
Detailed Technical Specifications
3D Mark
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Fire Strike | 47869 | 72432 |
| Night Raid | 145258 | 195813 |
| Port Royal | 13265 | 26127 |
| Solar Bay | 96626 | 187525 |
| Time Spy | 19633 | 36299 |
| Wild Life Extreme | 42724 | 85218 |
API
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| CUDA | 8.6 | 8.9 |
| DLSS | DLSS 2 | DLSS 3 |
| DirectX | 12 | 12 |
| DisplayPort | 1.4a | 1.4a |
| OpenCL | 3.0 | 3.0 |
| OpenGL | 4.6 | 4.6 |
| Ray Tracing | Yes | Yes |
| Vulkan | 1.3 | 1.3 |
Games
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| 1080p High | 179 | 255 |
| 1080p Ultra | 151 | 217 |
| 1440p Ultra | 118 | 177 |
| 4K Ultra | 69 | 106 |
| CoD: Modern Warfare III | 16915812182 | 266263219151 |
| Counter-Strike 2 | 27720515582 | 33923818890 |
| Cyberpunk 2077 | 1461288640 | 22620814366 |
| Far Cry 6 | 15314212375 | 203179170104 |
| Forza Horizon 5 | 18612511790 | 275199193155 |
| Ghost of Tsushima | 114968251 | 18815213791 |
| Hogwarts Legacy | 1311088446 | 20016313274 |
| Margin of Error | Low | Low |
| Shadow of the Tomb Raider | 23522016589 | 302284222118 |
| The Witcher 3 | 20417312567 | 292268191101 |
GeekBench 6 ML
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Backend | DirectML | DirectML |
| Depth Estimation (Q) | 36756 | 75599 |
| Face Detection (HP) | 49813 | 78111 |
| Framework | ONNX | ONNX |
| Image Classification (SP) | 9747 | 15600 |
| Image Segmentation (HP) | 26654 | 36657 |
| Image Super Resolution (Q) | 21475 | 48114 |
| Machine Translation (HP) | 4757 | 6349 |
| Object Detection (SP) | 12924 | 21186 |
| Pose Estimation (Q) | 135370 | 289639 |
| Style Transfer (SP) | 274032 | 601428 |
| Text Classification (SP) | 2796 | 4074 |
GeekBench 6 OpenCL
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| API | OpenCL | OpenCL |
| Background Blur | 239.8 img/sec | 318.9 img/sec |
| Edge Detection | 14.8 Gpixels/sec | 21 Gpixels/sec |
| Face Detection | 156.8 img/sec | 222.8 img/sec |
| Feature Matching | 1.53 Gpixels/sec | 2.38 Gpixels/sec |
| Gaussian Blur | 11.9 Gpixels/sec | 23.9 Gpixels/sec |
| Horizon Detection | 10.8 Gpixels/sec | 14.2 Gpixels/sec |
| Particle Physics | 26054.6 FPS | 48090.4 FPS |
| Stereo Matching | 889.6 Gpixels/sec | 1950 Gpixels/sec |
General
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| AMD Equivalent | -Radeon RX 6900 XT | — |
| Architecture | Ampere | Ada Lovelace |
| Build | Discrete | Discrete |
| Case | Desktop | Desktop |
| GPU Codename | GA102 | AD102 |
| Launch price (MSRP) | $1199 | $1599 |
| Purpose | Gaming | Gaming |
| Recommended CPU | -Intel Core i9 12900Kor above | -Intel Core i9 14900Kor above |
| Released | January 3, 2021 | September 20, 2022 |
| Segment | High-end | High-end |
| Vendor | Nvidia | Nvidia |
| Successor | — | -GeForce RTX 5090 |
Graphics Processing Unit
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Base Clock | 1365 MHz | 2235 MHz |
| Boost Clock | 1665 MHz | 2520 MHz |
| Compute Units (Pipelines) | 80 | 128 |
| Instructions Per Cycle | 2 IPC | 2 IPC |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 6MB shared | 72MB shared |
| Ray-tracing Cores | 80 | 128 |
| Render Output Units (ROPs) | 112 | 176 |
| Shading Units | 10240 | 16384 |
| Tensor Cores | 320 | 512 |
| Texture Mapping Units (TMUs) | 320 | 512 |
Memory
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Bus | 384-bit | 384-bit |
| ECC | No | No |
| Effective Memory Speed | 19000 Mbps | 21000 Mbps |
| Memory Clock | 1188 MHz | 2625 MHz |
| Memory Size | 12 GB | 24 GB |
| Memory Type | GDDR6X | GDDR6X |
Passmark Graphics
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| DirectX 11 | 221 FPS | 323 FPS |
| DirectX 12 | 109 FPS | 151 FPS |
| G2D Mark | 1095 | 1303 |
| GPU Compute | 14943 Ops/s | 26022 Ops/s |
Physical
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Die Size | 628 mm² | 609 mm² |
| Fabrication Process | 8 nm | 5 nm |
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| Manufacturing | Samsung | TSMC |
| TGP | 350 W | 450 W |
| Transistor Count | 28 billion | 76.3 billion |
| Transistor Density | 44.59 MTr/mm² | 125.29 MTr/mm² |
Raw Performance
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Pixel Fill Rate | 186 GPixel/s | 444 GPixel/s |
| Texture Fill Rate | 533 GTexel/s | 1290 GTexel/s |
Recent User Tests
| Specification | GeForce RTX 3080 Ti | GeForce RTX 4090 |
|---|---|---|
| Date | Benchmark | Benchmark |
| 📘 2025-11-01(cancel) | Cinebench 2024 | — |
| 📘 2026-04-03(afroman420IU) | — | Cinebench 2024 |
Looking for More Comparisons?
Browse all available head-to-head comparisons or select new devices in the comparison hub.