Recent discussion about DK barrels and different play between various versions of mame and the pcb's brings this to mind...
Depending on how the "random" is generated, I would imagine the pcb vs the emulated version could have drastic differences in "random" events, making gameplay on a competative level vastly different.
Another case in point is Hatris. Gatinha plays this alot (and is pending varification of her world record score)... she gets quite a few blue flames with the current version of mame, but has seen only one blue flame among probably fifty or more plays in the version required for score verification (which is based on mame.106). Makes a HUGE difference in game play. I have a hatris pcb... and once I get around to setting up a jamma getup, we'll be able to compare gameplay on the actual pcb as well. Maybe I could do a statistical analysis of blue flame frequency between the two versions of mame and the pcb for comparison purposes.
But getting around to the question at some point... is there any thing that could be done in emulation to approximate the original pcb's "random" frequencies... or is this just taking emulation too far given the huge variety of processors and speeds out there?
|