Below is a comparison of the NVIDIA RTX 3090 vs the NVIDIA RTX 4090. Here you can see the huge leaps in computing power between generations. One thing is for sure, these cards just keep getting better! Check out the specs below!
NVIDIA RTX 3090
| PU Name | GA102 |
| GPU Variant | GA102-300-A1 |
| Architecture | Ampere |
| Foundry | Samsung |
| Process Size | 8 nm |
| Transistors | 28,300 million |
| Die Size | 628 mm² |
| Base Clock | 1395 MHz |
| Boost Clock | 1695 MHz |
| Memory Clock | 1219 MHz, 19.5 Gbps effective |
| Slot Width | Triple-slot |
| Length | 336 mm |
| Width | 140 mm |
| Height | 61 mm |
| TDP | 350 W |
| Suggested PSU | 750 W |
| Outputs | 1x HDMI 2.1, 3x DisplayPort 1.4a |
| Power Connectors | 1x 12-pin |
| Board Number | PG132 SKU 10 |
| Release Date | Sep 1st, 2020 |
| Availability | Sep 24th, 2020 |
| Generation | GeForce 30 |
| Predecessor | GeForce 20 |
| Successor | GeForce 40 |
| Production | Active |
| Launch Price | 1,499 USD |
| Bus Interface | PCIe 4.0 x16 |
| Pixel Rate | 189.8 GPixel/s |
| Texture Rate |
556.0 GTexel/s
|
| FP16 (half) performance |
35.58 TFLOPS (1:1)
|
| FP32 (float) performance |
35.58 TFLOPS
|
| FP64 (double) performance |
556.0 GFLOPS (1:64)
|
| Memory Size | 24 GB |
| Memory Type | GDDR6X |
| Memory Bus | 384 bit |
| Bandwidth |
936.2 GB/s
|
| DirectX |
12 Ultimate (12_2)
|
| OpenGL | 4.6 |
| OpenCL | 3 |
| Vulkan | 1.3 |
| CUDA | 8.6 |
| Shader Model | 6.7 |
| Shading Units | 10496 |
| TMUs | 328 |
| ROPs | 112 |
| SM Count | 82 |
| Tensor Cores | 328 |
| RT Cores | 82 |
| L1 Cache |
128 KB (per SM)
|
| L2 Cache | 6 MB |
NVIDIA RTX 4090
| Texture Rate |
1,290 GTexel/s
|
| FP16 (half) performance |
82.58 TFLOPS (1:1)
|
| FP32 (float) performance |
82.58 TFLOPS
|
| FP64 (double) performance |
1,290 GFLOPS (1:64)
|
| Memory Size | 24 GB |
| Memory Type | GDDR6X |
| Memory Bus | 384 bit |
| Bandwidth |
1,008 GB/s
|
| DirectX |
12 Ultimate (12_2)
|
| OpenGL | 4.6 |
| OpenCL | 3 |
| Vulkan | 1.3 |
| CUDA | 8.9 |
| Shader Model | 6.7 |
| Shading Units | 16384 |
| TMUs | 512 |
| ROPs | 176 |
| SM Count | 128 |
| Tensor Cores | 512 |
| RT Cores | 128 |
| L1 Cache |
128 KB (per SM)
|
| L2 Cache | 72 MB |
| GPU Name | AD102 |
| GPU Variant | AD102-300-A1 |
| Architecture | Ada Lovelace |
| Foundry | TSMC |
| Process Size | 4 nm |
| Transistors | 76,300 million |
| Die Size | 608 mm² |
| Base Clock | 2235 MHz |
| Boost Clock | 2520 MHz |
| Memory Clock | 1313 MHz, 21 Gbps effective |
| Slot Width | Triple-slot |
| Length | 304 mm |
| Width | 137 mm |
| Height | 61 mm |
| TDP | 450 W |
| Suggested PSU | 850 W |
| Outputs | 1x HDMI 2.1, 3x DisplayPort 1.4a |
| Power Connectors | 1x 16-pin |
| Board Number | PG139 SKU 330 |
| Release Date | Sep 20th, 2022 |
| Availability | Oct 12th, 2022 |
| Generation | GeForce 40 |
| Predecessor | GeForce 30 |
| Production | Active |
| Launch Price | 1,599 USD |
| Bus Interface | PCIe 4.0 x16 |
| Pixel Rate | 443.5 GPixel/s |
We hope you enjoyed our comparison of the NVIDIA RTX 4090 vs the NVIDIA RTX 3090. More to come soon, let us know what you thought! Thank you!
Recent Comments