I tested the new Benchmark option today and would like to draw your attention onto how data on screen are displayed today and 35 years ago when I was programming for the ATARI ST…
today:
35 years ago:
35 years ago I was able to print all sorts of data nicely sorted in clean rows and collumns without any hassle by using the PRINT @ command. I am not sure what you are using today but it looks awful. Sorry. 
This benchmark was done while I was running an other Video (Proteus manual, 2x, h265 NVIDIA, mp4), which slowed down while benchmarking from 5.7 fps to 4.4 fps. That means below values represent my machine when TVAI 3.2 precesses 2 videos in parallel.
This Benchmark was done Topaz Video AI Alpha v3.2.0.1.a
System Information
OS: Windows v11.2009
CPU: AMD Ryzen 7 5800X 8-Core Processor 31.916 GB
GPU: NVIDIA GeForce RTX 3060 11.845 GB
Settings: device: 0 vram: 1 instances: 0
Benchmark Results
Artemis 1X: 4.05 fps 2X: 2.97 fps 4X: 0.96 fps
Proteus 1X: 4.02 fps 2X: 2.99 fps 4X: 0.94 fps
Gaia 1X: 1.32 fps 2X: 0.89 fps 4X: 0.61 fps
4X Slowmo Apollo: 4.75 fps Chronos: 2.84 fps Chronos Fast: 4.76 fps
when running the benchmark solo I got
Topaz Video AI Alpha v3.2.0.1.a
System Information
OS: Windows v11.2009
CPU: AMD Ryzen 7 5800X 8-Core Processor 31.916 GB
GPU: NVIDIA GeForce RTX 3060 11.845 GB
Settings: device: 0 vram: 1 instances: 0
Benchmark Results
Artemis 1X: 9.45 fps 2X: 6.22 fps 4X: 1.96 fps
Proteus 1X: 9.16 fps 2X: 5.89 fps 4X: 1.9 fps
Gaia 1X: 3.29 fps 2X: 2.23 fps 4X: 1.44 fps
4X Slowmo Apollo: 9.67 fps Chronos: 7.25 fps Chronos Fast: 11.81 fps
Near the same results with a 900€ Ultrabook, 21Watt power from the wall while benching.
Topaz Video AI v3.1.9
System Information
OS: Mac v13.02
CPU: Apple M1 8 GB
GPU: Apple M1 5.3333 GB
Processing Settings: device: -2 vram: 0.7 instances: 1
Benchmark Results
Artemis 1X: 8.42 fps 2X: 5.47 fps 4X: 2.3 fps
Proteus 1X: 8.43 fps 2X: 5.15 fps 4X: 1.88 fps
Gaia 1X: 2.78 fps 2X: 2.05 fps 4X: 1.62 fps
4X Slowmo Apollo: 9.76 fps Chronos: 3.18 fps Chronos Fast: 4.56 fps
