IT House News on July 16th, Nvidia’s new GeForce flagship based on the Ada Lovelace architecture may still be a few months away, but the GPU ID information leaked last week has proven useful to third-party application developers.
Now, AIDA64 has announced support for the just-announced Nvidia GeForce GTX 1630, AMD Radeon Pro W6600 (Navi 23), and the Hopper H100 data center accelerator card with 96GB of video memory.
In addition, AIDA64 6.70.6033 has been applied to the unreleased NVIDIA GeForce RTX 4090 (AD102), AMD 600 series motherboard platform (AM5).
However, the AIDA64 release notes don’t explain the RTX 4090’s specific GPU information, only listing it as an AD102-based product, but that’s no secret.
From the previous information, the NVIDIA GeForce RTX 4090 is based on the GA102-300 GPU and will be released this fall. It will be the next-generation flagship product of NVIDIA, with 16,384 CUDA cores, 24GB GDDR6X memory, and a TDP of up to 450W.
In addition, the RTX 4090 should arrive this fall as the first GPU in the new lineup of the RTX 40 series, followed by the RTX 4080 and RTX 4070, which may explain why AIDA64 currently only lists one SKU.
IT House has reported that the whistleblower kopite7kimi once said that Nvidia will release the RTX 40 series graphics cards in September or October this year, and the first model is expected to be the RTX 4090. Regarding the specifications of the RTX 40 series graphics cards, the following latest news is for your reference.
-
RTX 4090: GPU model AD102-300-A1, 16384 FP32 core, GPU frequency up to 2750MHz, video memory 384bit 21Gbps 24G GDDR6X, power consumption 450W.
-
RTX 4080: GPU model AD103-300-A1, 10240 FP32 core, video memory is 256bit 21Gbps 16G GDDR6X, power consumption 420W.
-
RTX 4070: GPU model AD104-275-Kx-A1, 7168 FP32 core, video memory is 160bit 18Gbps 10G GDDR6, power consumption is 300W.
.
[related_posts_by_tax taxonomies=”post_tag”]
The post AIDA64 now supports NVIDIA RTX 4090, AMD 600 series AM5 platforms appeared first on Gamingsym.