[SOLVED] Fury X still best gpu of its length?

mihen

Honorable
Oct 11, 2017
466
54
10,890
I have a Fury X and it perfectly fits into an mitx case, but just barely. It's notably shorter than other gpus of its performance level. Just under a GTX 1070. I was wondering if there are any gpus that have superior performance at 7.5"?
 
I compared the cards. The 2060 would only be a 10ish% improvement over the Fury X, and that difference may be due to a newer platform for benchmarking than an actual performance difference. The 2070 is about a 30% improvement, and it's possible to get a custom block for that 2070 on eBay. But... I already had the Fury X for a couple years and that's not worth spending near $1000 on a 2070 with a custom block. The 1660 Super would be a downgrade.

Thanks for digging into it for me.
 
I compared the cards. The 2060 would only be a 10ish% improvement over the Fury X, and that difference may be due to a newer platform for benchmarking than an actual performance difference. The 2070 is about a 30% improvement, and it's possible to get a custom block for that 2070 on eBay. But... I already had the Fury X for a couple years and that's not worth spending near $1000 on a 2070 with a custom block. The 1660 Super would be a downgrade.

Thanks for digging into it for me.
Fury X vs 2060 FE
https://www.anandtech.com/bench/product/2396?vs=2372

Fury X vs 1660 Ti (Super would be almost identical)
https://www.anandtech.com/bench/product/2396?vs=2429

Both tend to be faster than the Fury X by 10% or more. Newer games are even more noticeable than older games as it seems that the Fury X is running out of VRAM in Wolfenstein.
 
Benchmarks are more of a reference with comparable products. Since nVidia puts in the effort to have their gpus work with a wide variety of games, most benchmarks will favor nVidia architecture. There is no way to do an Apples to Apples comparison of dissimilar architectures as they will perform in certain tasks differently.
With any game if you are using AMD. It's best to have an fps counter on, then test the impact of each setting as the developer probably did not optimize the settings for AMD. For instance running at 4k with 16x AA. Completely unnecessary and will impact AMD more.