pokemon-speedrunning / gambatte-speedrun

Fork of https://github.com/sinamas/gambatte with Pokemon speedrunning-related changes.
GNU General Public License v2.0
94 stars 27 forks source link

Questionable features to consider removing #84

Open entrpntr opened 3 years ago

entrpntr commented 3 years ago

Small list of features I don't see a point in keeping right now (subject to change, if there are compelling reasons to keep any of them)

CasualPokePlayer commented 3 years ago
  1. I do know there are a handful of people who use filters for some reason, could annoy a few users if they're removed, and fadeout being broken is just visual and the timing is still intact (which ultimately is what matters).
  2. I've occasionally seen people use full screen for some reason but it has been quite rare, I don't think anyone would really miss it.
  3. That exists? The fuck? Please nuke it who the fuck would even need that when Alt+F4 exists anyways.
Grogir commented 2 years ago
  1. the timing is still intact (which ultimately is what matters)

Just discovered that the filter ive always been using loses about 0.53s at hard reset compared to no filter. It seems to affect "Bicubic Catmull-Rom spline 3x" and "MaxSt's hq3x". The others hard reset at the same speed as no filter.

CasualPokePlayer commented 2 years ago
  1. the timing is still intact (which ultimately is what matters)

Just discovered that the filter ive always been using loses about 0.53s at hard reset compared to no filter. It seems to affect "Bicubic Catmull-Rom spline 3x" and "MaxSt's hq3x". The others hard reset at the same speed as no filter.

That would be very weird if so, as the reset timing is controlled purely with the emulation timing. How are you measuring this?

Grogir commented 2 years ago

By video. Binding the reset button to a timer works too. I mean its half a second its pretty obvious. Though i'm be curious if the issue is only on my end - see discord

CasualPokePlayer commented 2 years ago

By video. Binding the reset button to a timer works too. I mean its half a second its pretty obvious. Though i'm be curious if the issue is only on my end - see discord

Doing tests on my side showed no significant difference. I could presume your end might have some weird input lag differences between filter/no filter and that caused what you saw (which would be very questionable in itself, but is the best explanation I got).

EDIT: Looked at video closer and audio definitely shows it being off by half a second. Not sure what would be going on other than it somehow messing up the frame pacing.