Main menu

Pages

Nvidia RTX 40 series can increase power consumption

featured image

A new round of specs for the Nvidia GeForce RTX 40 series reveals an interesting feature: configurable TGP. According to the leak, the next-generation of his Nvidia’s premier graphics card will allow users to increase power consumption for better performance. All of these are tuned based on your needs rather than locking your GPU to a specific his TGP.

Also, there is one more surprise. A popular GPU seems to be missing from the lineup.

It is summarized below for ease of reading. Not news.
-RTX 4090, PG136/139-SKU330, 16384FP32, 21Gbps 24G, 450W+.
-RTX 4080 16G, PG136/139-SKU360, 9728FP32, 23Gbps 16G, 340W+.
-RTX 4080 12G, PG141-SKU340/341, 7680FP32, 21Gbps 12G, 285W+.
I basically finished my work.

— kopite7 you (@kopite7 you) September 14, 2022

Kopite7kimi, a well-known leaker in the PC hardware space, has shared a roundup of Nvidia GPU specifications. No major updates, but some changes can be seen. More importantly, Kopite concluded by saying, “I’m done with my work.” This strongly suggests that these specs could be a real deal for his Nvidia to announce soon.

Based on the AD102-300 GPU, we have the flagship RTX 4090 with 16,384 CUDA cores and a boost clock of up to 2,520MHz. The card comes with 24GB of GDDR6X memory clocked at 21Gbps, delivering up to 1TB/s bandwidth. The base TGP is said to be 450 watts.

Instead of offering the RTX 4080 and RTX 4070, it looks like Nvidia has changed its mind about the lineup as we may get two different versions of the RTX 4080.

The first model with the AD103-300 GPU features 16GB of GDDR6X memory running at 23Gbps. It also features 9,728 CUDA cores and a boost clock (2,505MHz) that is very close to the flagship.

This is followed by an AD104-400 based RTX 4080 12GB model which would feature 7,680 CUDA cores and a higher boost clock of up to 2,610MHz. The first version of the card will come with a TGP of 340 watts, while the weaker model should be a more modest 285 watts.

It’s interesting that the RTX 4070 isn’t listed, at least for the time being. Previous leaks indicated that the AD104-400 GPU would fit inside his RTX 4070 Ti or RTX 4070, but according to Kopite7kimi, Nvidia scrapped it and replaced it with his RTX 4080. Decided to release two different versions. VideoCardz also claims that Nvidia’s board partners don’t think this will change for now.

cubit leaks

Let’s go back to configurable TGP. According to VideoCardz, citing his own anonymous source, the RTX 4090’s default TGP is actually 450 watts, but users can adjust it in her BIOS to reach up to 660 watts. For the RTX 4080 16GB, the base TGP (340 Watts) could be boosted to a maximum of 516 Watts, and for the RTX 4080 12GB, the default 285 Watts could be a maximum of 366 Watts.

This is an interesting feature, but unconfirmed. Users can choose between higher power consumption and improved performance. However, even if released, it doesn’t seem to be on all versions of the card.

Most of this will be confirmed or debunked soon. Nvidia will be making a big announcement on September 20th as part of the GTC keynote.

Editor’s pick