ATI Rage Fury MAXX vs NVIDIA GeForce RTX 3050 8 GB
Graphics Processor
| GPU model | Rage 4 | GA106 |
|---|---|---|
| GPU variant | 215R4GAUC21 | GA106-150-A1 |
| Architecture | Rage 4 | Ampere |
| Foundry | TSMC | Samsung |
| Manufacturing process | 250 nm | 8 nm |
| Number of transistors | 8 million | 13,250 million |
| Die size | 89 mm² | 276 mm² |
Graphics Card
| Release date | Oct 1st, 1999 | 2022 |
|---|---|---|
| Generation | Rage 4 | GeForce 30 |
| Production | End-of-life | Unreleased |
| Interface | AGP 4x | PCIe 4.0 x16 |
| Predecessor | Rage 3 | GeForce 20 |
| Successor | Rage 6 | — |
| gpu.details.availability | — | 2022 |
Clocks
| GPU clock | 125 MHz | — |
|---|---|---|
| Memory clock | 143 MHz | 1750 MHz 14 Gbps effective |
| Base clock | — | 1552 MHz |
| Boost clock | — | 1777 MHz |
Memory Configuration
| Memory size | 32 MB | 8 GB |
|---|---|---|
| Memory type | SDR | GDDR6 |
| Memory bus width | 128 bit | 128 bit |
| Bandwidth | 2.288 GB/s | 224.0 GB/s |
Render Configuration
| Pixel shaders | 2 | — |
|---|---|---|
| Vertex shaders | 0 | — |
| TMUs | 2 | 80 |
| ROPs | 2 | 48 |
| Shading units | — | 2560 |
| SM count | — | 20 |
| Tensor cores | — | 80 |
| RT cores | — | 20 |
| Cache L1 | — | 128 KB (per SM) |
| Cache L2 | — | 2 MB |
Performance
| Pixel rate | 250.0 MPixel/s | 85.30 GPixel/s |
|---|---|---|
| Texture rate | 250.0 MTexel/s | 142.2 GTexel/s |
| FP16 (half) performance | — | 9.098 TFLOPS (1:1) |
| FP32 (float) performance | — | 9.098 TFLOPS |
| FP64 (double) performance | — | 142.2 GFLOPS (1:64) |
Dimensions & Outputs
| Slot width | Single-slot | Dual-slot |
|---|---|---|
| TDP | unknown | 90 W |
| Suggested PSU | 200 W | 250 W |
| Outputs | 1x VGA | 1x HDMI3x DisplayPort |
| Power connectors | None | 1x 6-pin |
| Board number | 673 | — |
| Length | — | 242 mm 9.5 inches |
| Width | — | 112 mm 4.4 inches |
API Support & Features
| DirectX | 6.0 | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 1.2 | 4.6 |
| OpenCL | — | 3.0 |
| Vulkan | — | 1.2 |
| Pixel shader | — | — |
| Vertex shader | — | — |
| CUDA | — | 8.6 |
| Shader model | — | 6.6 |
