| 规格 | NVIDIA RTX 6000 Ada Generation | NVIDIA GeForce RTX 4090 |
|---|
| 架构 | NVIDIA Ada Lovelace architecture | Ada Lovelace (AD102-300, 128 SM) |
| 显存 | 48GB GDDR6 | 24GB GDDR6X |
| 显存ECC | Yes | — |
| 显存位宽 | 384-bit | 384bit |
| 显存带宽 | 960 GB/s | 1008GB/s (21Gbps x 384bit) |
| CUDA核心 | 18,176 | 16384 (128 SM) / Tensor Core 512 (四代) / RT Core 128 (三代) |
| Tensor Core | 568 | — |
| RT Core | 142 | — |
| FP32算力 | 91.1 TFLOPS | — |
| RT Core性能 | 210.6 TFLOPS | — |
| Tensor性能 | 1457.0 TFLOPS | — |
| 功耗 | Total board power: 300 W | 450W TGP (平均游戏315W, 建议系统电源850W) |
| 互联 | PCIe 4.0 x16 | PCIe Gen4 x16 (不支持NVLink) |
| 散热 | Active | — |
| 形态 | dual slot, full height (4.4in H x 10.5in L) | FE三槽 304x137mm, 供电 3x8pin转接或12VHPWR(450W Gen5) |
| 显示输出 | 4x DisplayPort 1.4a | HDMI 2.1a x1 + DisplayPort 1.4a x3, 4K240/8K60(DSC) |
| 供电接口 | 1x PCIe CEM5 16-pin | — |
| 编解码引擎 | 3x encode, 3x decode (+AV1 encode and decode) | — |
| NVLink | No | — |
| 算力 | — | FP32 82.6 TFLOPS (官方发布口径, 现页四舍五入标83) / Tensor 1321 AI TOPS (FP8稀疏) / RT 191 TFLOPS |
| 发布年 | — | 2022 (2022-09-20 GTC发布, 2022-10-12开售, MSRP $1599) |
| 频率 | — | Base 2.23GHz / Boost 2.52GHz |
| 工艺 | — | TSMC 4N, 763亿晶体管, 608.5mm² (AD102 满血144SM, 4090启用128) |
| 媒体引擎 | — | 2x 第八代NVENC / 1x 第五代NVDEC, AV1编解码 |
| CUDA算力等级 | — | Compute Capability 8.9 |