Most up-to-date RTX 4080 spec revision adds a lot quicker memory and a better TBP

Rumor mill: We've been hearing a great deal of rumors about Nvidia's approaching RTX 4000 sequence, even though most of them seem to revolve about the RTX 4070, its Ti variant, and the RTX 4090. The most recent assert to arrive from a common leaker includes the substantial-end-but-not-fairly-flagship RTX 4080, which sounds like it will be an complete beast of a card.

As normal, the new Ada Lovelace rumor arrives from the Twitter account of prolific and precise (when it arrived to Ampere) leaker Kopite7kimi (by using VideoCardz). They posted what are claimed to be current specs for the RTX 4080, with the important adjustments relating to the memory pace and the card's TDP.

It experienced been envisioned that the RTX 4070, 4080, and 4090 would use 21 Gbps GDDR6X modules—they ended up described in this month's update on Micron's website—but the RTX 4080 is now considered to function 23 Gbps speeds. Assuming the rumor is accurate, the RTX 4090 will most likely aspect the same 23Gbps GDDR6X.

Let us make some updates.
RTX 4080
PG136/139-SKU360
Advertisement103-300-A1
9728FP32
16GB 23Gbps GDDR6X
complete card ability 340W

— kopite7kimi (@kopite7kimi) August 23, 2022

The up coming revised spec is the RTX 4080's Complete Board Electricity (TBP). This is a person location that leakers appear to transform each and every other 7 days. It was in the beginning reported to eat a monstrous 450W, but that was not too long ago altered to 320W. According to Kopite7kimi, Nvidia has elevated the RTX 4080's TBP yet again, albeit by just 20W, leaving the new amount at 340W. For comparison, the RTX 3080 Ti Founders Edition has a TBP of 350 watts.

The rest of the RTX 4080 specs keep on being unchanged: an Advert103 GPU, 9,728 CUDA cores, and 16GB of memory across a 256-little bit bus.

An additional aspect of the August update on Micron's website was a listing for 24 Gbps GDDR6X. It could stop up in what is imagined to be a new Titan card that was formerly rumored to pack 48GB of VRAM, 18,176 CUDA cores, and an 800 TDP.

Nvidia experiments with the specs of its forthcoming cards as the start dates draw nearer, which is why they feel to transform so generally we may not see the last, locked-in specs until the launch is on the doorstep. The excellent information is that day is now just a pair of months absent.