Made the switch to Intel FS is slower

Without wishing to causing offence I am not sure why there is a need to change processors at the moment because MSFS is being changed a lot and it my be a different issue in the near or slightly distant future.

So, based on this thread, I switched back to the Ryzen. Graphics performance is back where it was , but with poorer CPU performance, probably reflecting the smaller core count on the 5600x

What puzzles me though is running the same 3DMark benchmarks gave very different GPU performance, suggesting better MB support from the B550 than the Z490.

More testing later

What do you mean exactly by ‘poorer CPU performance’?

Seems that the author of the thread doesn’t have sufficient knowledge of overclocking the system. Its easy to degrade performance with overclocking if not done properly.

Well, i have, but MSFS runs much better without.
Locked at a steady 30fps, vsync on, trackir and all on 1440p, very smooth
 ( high settings) latest AMD drivers, MB and GPU
Ryzen 5 5600x, RX5600XT, 32GB 3200

No CTD’s anymore and no stutters or slowing downs after one hour

I think he means lesser results in benchmarks such as Cinebench multicore. For who does not know how to build computers, it is hard to understand single core performance in poorly optimized games still coded for DX11.

I suggested the 5800X, which has two more cores than the 5600X.

To be exact , running the DX12 benchmark in 3dMark with the AMD platform gave a score of 20k for the GPU and 8k for the CPU. It was the other way round with the Intel. A GPU of 17k and 11k for the CPU. I don’t understand how the the GPU score could differ so much.

Back in Flight Sim though I noted a 10fps improvement with the AMD based board over the Intel. That was using Ryzen master performing some gentle OC. That’s 4k with mostly Ultra settings. Still main thread constrained as others seem to find.

That said though my kit does seem to not perform as well as others I’ve read about. The only other difference is my whole kit is water cooled, and I wonder if something is being constrained. GPU never exceeds 62c and the CPU 75c

Again, I wouldn’t focus on synthetic benchmarks like 3Dmark too much. Since they only test one very specific scenario, they’re very limited in what they can tell you (as you’ve noticed).
The higher core count of the 10900k will of course do better in a synthetic CPU benchmark. Just like a 5900X or 5800X will outperform a 5600X in those types of benchmarks.

Agreed, however it doesn’t quite explain the difference between the GPU performance on the two platforms. Looking a little closer the Intel platform had the GPU maxed at around 80%, while the AMD platform allowed 100% of the GPUs capacity, the Intel MB just didn’t use the GPU to its full potential