Retouch4me benchmark
The results of the benchmark conducted using the official testing procedure of Retouch4me. I automated it with action to avoid mistakes and made it available for everyone, so anyone who has Photoshop and R4m DodgeBurn can check their computer – instructions are at the bottom.
Measurement error
Within one computer, the results will be very repeatable with an error below half a second. The license verification of R4m usually takes 0.3 seconds. The differences between versions of a given GPU model and other components may affect the result to some extent, but definitely the most important is the GPU model.
Some of the results come from only one test, so we need to do more tests to see real average timings. For example there is no possibility for 4080 to be faster than 4090, but more results are needed to have a good enough average to correctly indicate such small differences.
Benchmark does not test speed in long batch editing,
so differences between devices may increase in batch processing. For example The MacBook Air can slows down significantly after 1.5m of full load, while the MacBook Pro with fan cooling will continue to work fast. This also applies to laptops with RTX GPUs etc. (they may also experience a decrease in performance when disconnected from the power supply).
Scoreboard
The table shows details – averaged and official results, as well as times of various users on different computers with specific GPUs (U1, U2 etc.).
Legend: | Nvidia | AMD | Mac with Apple Silicon (current Macs) | Mac with Intel CPU (old models) |
Benchmark your computer
Download the test file and check if upgrading your graphics card will make a big difference. The following video explains everything in 2 minutes. Leave a comment with your result on YT.