Re: [hatari-devel] WinUAE CPU core CPU/FPU/DSP performance according to Centurbo benchmark

[Douglas Little <doug694@xxxxxxxxxxxxxx>]

Hi,

We would need some test where falcon is in a mode where video resolution doesn't steal bandwidth from the cpu (compatible STF 320x200 for example)..

Yes that was the purpose behind shutting off the Videl during tests - to stop the stealing of bus cycles.

ST modes do still steal some cycles I think, but not very much. The impact is quite low.

(It seems there are still other mechanisms stealing some cycles or causing bus slots to get missed because the resulting timings are cycle-fractional - but I guess those causes are more likely to be constants. The timings always yield integers with Hatari)

 

But having some tests later with more bitplanes or TC mode could also help to emulate when videl take some cycles from the CPU.

BTW, is nimbench the program for which you posted results on 19.9.2014 "Hatari WinUAE speed comparisons with Falcon030 / 16MHz" ?

Yes. It's not really a finished program but I have been adding stuff gradually, and it has been helpful when checking out timing compatibility with Hatari sometimes.

D.