What's new

CPS1 Real Arcade vs Fightcade vs MiSTer - 3 Way Showdown...

I did the same test again, both on the Ryu and Blanka stages using the same camera and 60i frame rate, but this time with a 10 MHz crystal instead of the 12 MHz crystal I used before. The results:

12 MHz (from previous test)
Ryu Stage: 54.232 seconds
Blanka Stage: 55.199 seconds

10 MHz
Ryu Stage: 57.366 seconds
Blanka Stage: 57.966 seconds

1.png
2.png
 
It looks like the speed varies by quite a bit even on the same stage. Here are two fights that went the distance on the same arcade machine at the same tournament on the same stage in the same video:

Clipboard01.png
Clipboard02.png


55.422 seconds vs. 53.620 seconds, which is almost a 2-second difference. Ryu vs. Ryu in one of them and Sagat vs. Ryu in the other.
 
A question not a statement. I build some old snes games using SA-1 chip and faster clock oscillators, I noticed when I order clocks they have a PPM +\- tolerance and a ppm +\- that they can loose or gain year to year from nominal. The higher quality oscillators obviously had tighter tolerances and less yearly drift. I found a TI article that also discuses ‘load capacitance’ as having a great deal of effect on oscillator accuracy. Time, temperature, humidity ..etc all contributing to the varying rate of decay.

https://www.ti.com/lit/an/snla290/snla290.pdf?ts=1768802268279

Wouldn’t it be expected to have constant frame/field variation if variation is built into the components?
 
You must log in or register to reply here.
Back
Top