By using this site, you agree to our Privacy Policy and our Terms of Use. Close
Captain_Yuri said:
hinch said:

Yeah if I had to guess the info on these are rough guidelines given to aib's given from Nvidia and to make sure they can provide enough juice and headroom for power limits for their respective lines of GPU's.

With how AMD are keeping high efficiency with RDNA 3 with their press and maybe internal leaks.. maybe Nvidia got wind of performance of the whole stack, and decided to dial down. And try other configurations with to get similar results without completely throwing away the efficiency/performance gains. It'll be interesting either way. I'd hope if Nvidia can get the 4090 and 4080 TDP's down to sane levels.. for all of our sakes. And if the 3080 does indeed come out with only 320W that makes it much more appealing GPU to me. 

Yea idk how Nvidia is able to get info from AMD before they even launch but Ampere stack against RDNA 2 was pretty solid so I suspect that they have something to go on this time around too. Now that we know Nvidia is also uping the clock speeds similar to AMD, I think we get to see a better picture of where the wattage requirements are coming from. 4090 with 16k cuda cores running at 2.8Ghz will be nuts. And that's before the Kingpin versions.

We will see though. If a 4080 can be 40-50% faster than 3080 while keeping the 320 watts and $700 price point, it should be a very killer GPU. Especially with Nvidia's lead in feature sets. And this time, Nvidia has much better Vram throughout the product stack instead of with Ampere were a 3060 had more Vram than 70 and 80 class loll.

Imagine, just imagine if...they were to make another 1080ti lightening in a bottle with the 4080...

ngl, I would die. 



Mankind, in its arrogance and self-delusion, must believe they are the mirrors to God in both their image and their power. If something shatters that mirror, then it must be totally destroyed.