Nvidia simply revealed the complete potential of Ada Lovelace, and the RTX 4090 Ti GPU might boast a 70% greater CUDA core rely than RTX 3090 Ti. That similar improve applies to the AD102 chip’s ray tracing and Tensor cores, however its enhance clock speeds match the just lately introduced GeForce RTX 4090.
During an Nvidia digital press occasion, the corporate revealed max RTX 4000 GPU specs, highlighting the variations between Lovelace and Ampere. While the RTX 4090 isn’t geared up with a full AD102 chip, souped-up variants just like the RTX 4090 Ti would possibly pack all of the accessible 18,434 CUDA cores, 144 third-gen RT cores, and 576 fourth-generation Tensor cores.
| Nvidia AD102 GPU | |
| VRAM (GDDR6X) | 24GB |
|---|---|
| Boost clock | 2.52GHz |
| Base clock | 2.23GHz |
| CUDA cores | 18,434 |
| Ray tracing cores | 144 (Third gen) |
| Tensor Cores | 576 (foruth gen) |
| Bus Width | 384-bit |
Comparatively, the complete Ada Lovelace GPU options 71% extra cores than its Ampere GA102 predecessor. However, choosing an RTX 4090 Ti seemingly gained’t get you quicker speeds over and above the RTX 4090, as each include a 2.5GHz enhance clock ceiling.
On Tuesday, Nvidia formally introduced the RTX 4080 and RTX 4090, two next-gen graphics playing cards that wield the brand new Ada Lovelace structure. In addition, the inexperienced crew unveiled a brand new model of its AI upscaler, however you’ll solely be capable to use DLSS 3.0 on RTX 4000 GPUs.
Nvidia hasn’t introduced a greatest graphics card contender kitted out with a full AD102 GPU, nevertheless it’ll doubtlessly arrive later within the RTX 4000 collection life cycle. There’s additionally nonetheless no signal of both the RTX 4060 or RTX 4070, however two RTX 4080 fashions are set to reach in November. The 12GB VRAM flavour is already inflicting a stir amongst fans, because it falls behind its 16GB counterpart by way of specs and efficiency.
