search

mamedev / mame

MAME
https://www.mamedev.org/
Other
8.32k stars 2.03k forks source link

Minor CPS-1 inaccuracy (Daimakaimura loading times) #5057

Closed oranki closed 2 years ago

oranki commented 5 years ago

In Daimakaimura (Ghouls'n Ghosts), each loading screen between stages/cutscenes is roughly 8 frames shorter than on PCB.

Disclaimer: I'm not expecting this to be taken very seriously, since it affects absolutely nobody except for a handful of speedrunners. Just wanted to let you know, in case it might be helpful for further improving the CPS-1 emulation accuracy.

oranki commented 5 years ago

I apologise for earlier misinformed bug report. I checked a few other recordings and it seems on original PCB, the loading time difference between board and emu is only 4~5 frames, my scrub conversion board just has additional loading time.

rb6502 commented 5 years ago

This is the same issue as SSF2 being too fast or whatever. Lack of wait states. Nobody emulates this correctly yet, it's an annoying problem.

angelosa commented 3 years ago

Or sprite chip delays (cfr. Strider with stage 4 dinosaurs is supposed to run like this ) Known as MAMEtesters 00408 and 01989 and 07443 at very least. I'm not sure about keeping this open since we already have enough proof and reports. ETA: and this, figures.

zaphod77 commented 11 months ago

A

This is the same issue as SSF2 being too fast or whatever. Lack of wait states. Nobody emulates this correctly yet, it's an annoying problem.

Actually, Callus seems to get this correct somehow in every game it supports, according to the internet and many hyper fighting fans, but no one knows how Sardu did it and he's not telling.

Note that Callus supports this game, so please test if you can. :)