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

I actually posted the original tweet before wccftech made an article about it a bit ago and I added what red tech gaming said about the IO die and 160 CUs from the tweet thread. Although personally I don't believe in what Red Tech gaming says as he was the one that also stated that RT performance of RDNA 2 will be pretty close to Ampere but we all know how that turned out. But it certainly wouldn't surprise me if it's 3x if the rumors are true.

https://gamrconnect.vgchartz.com/post.php?id=9291448

I don't know how reliable at RedtechGaming are, but we're using Videocardz and WccfTech so...

Anyway, the parts that caught my interest are the ones where they state that it will be thre times more powerful (we'll see if it's true or not but with twice the CUs and shaders it should be close to that), that the RT performance will be around Ampere levels (so Nvidia's next gen should beat AMD again), and that it won't have dedicated cores to handle their Super Resolution tech.

Yea it will be interesting to say the least. I do wonder if the cores will be similar to how Amperes setup is where Ampere has a crazy amount of cores but they are potato cores compared to Turing and RDNA 2. But because there's so many of them, they end up being like a super potato and having really good performance. I remember when people were coming up with crazy performance increases for Ampere after seeing the amount of cores in the rumour mill and while Ampere ended up being a pretty good leap compared to Turing, it's not the crazy leap that some of the others suggested.



                  

PC Specs: CPU: 7800X3D || GPU: Strix 4090 || RAM: 32GB DDR5 6000 || Main SSD: WD 2TB SN850