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

Final prediction is a compromise that is close to the 3080 but not beating it on average. Weaker 4k performance, maybe better lower resolution performance. But ultimately irrelevant because no DLSS and worse RT.

Beating 3080 price by at least $100 for the XT.

Captain_Yuri said:
My prediction is going to be this.

At 1440p, RDNA 2 will match or beat the 3080 in Raster since the 3080 is only 15% faster than 2080 Ti on average at 1440p
At 4k, RDNA 2 will match or be slower than the 3080 outside of AMD's cherry picked games. I doubt they will show it being slower in the actual presentation though.
RT for RDNA 2 will be slower than Ampere by at least 15%, sometimes 20-25%. I don't see the 3080 Raster competitor being as slow as 2080Ti in RT because that would make AMD's first RT implementation worse than Turing. 3080 can be up to 40-45% faster than 2080 Ti when RT is enabled depending on the game.
They will mention Direct ML support but I doubt they will show any games.

Trumpstyle said:
Nothing has changed for me, I'm still 45% over Geforce 2080 ti at 275 Total watt, but worried they might be pushing clocks to beat Geforce 3090.

I'm still 2.1x above Geforce 2080 ti for RTX performance despite redgamingtech source and igor leak.

I win

I'll wait until final reviews are out before saying that definitively as they showed stuff like this with Vega 7. Also if they had the RT performance you were claiming, they would have shown Ray Tracing games or given us a hint for their RT performance.

For pure Raster though... That you do!



                  

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