I was playing around with some 1-M stuff and was looking at ways to skip the 1-6 credits cutscene. Then I found that you could hard reset during the cutscene while still skipping your progress. This also works on the well cutscene after every stage which seems to save around 4 seconds. See the video I made for comparison

Wanted to put this up for discussion before using it in runs since reset times are a bit variable between platforms. What would be your opinion about using this in runs? (It would also seem to affect the first 4 stages run)

Hello!

Nice find! This is something Petracide and I had noticed recently as well and were discussing.

The hard reset saves time on all tested hardware; however, Super Game Boy 2 and Game Boy Player all have longer boot times than an emulator. This would give an immediate (And on 'all levels 1 quests', a significant) advantage over hardware runners. Given, I am the only hardware runner at the moment.

If the game had it, a soft reset would put us all on the same level, but a hard reset offers an edge to Emu players over any hardware (Except maybe DMG/GBC/GBA; but its rare for people to record them). The 2 common hardware standards for GB spreedrunning would be at a disadvantage over Emu if we allowed for hard resets.

These were the things making me hesitant to incorporate it into our runs. With enough interest, we could add a (hard reset) variant for the categories. But in the interest of a fair playfield, I believe we should add 'No hard resetting' to our category rules.

Just my two cents!

out of interest, would gambatte r614 resolve the issue? This is the version where they added a fadeout upon hard reset similar to gameboy player which would make it a more even playing field between emu and at least GBP (I don't know about SGB2). I don't know how the hard resets work on the common hardware standards so I can't form my opinion on if it's comparable enough. I did make a new comparison video with the gambatte r614 hard reset fadeout to show the difference

I will download those tonight and pair them against the SGB2 reset times, out of curiousity! Thanks for the work assembling these; but I still have a few questions and reservations.

"would make it a more even playing field between emu and at least GBP" Do you know how much closer to accurate this is? How many ms it is off between GBP and r614 ?

And are you testing with GBP or GBi(Which is more commonly used in GB speedrunning)

My main concern is that I know GBP and SGB2 have different boot times; so this would still leave a disparity between platforms. As well as limiting people 2 a single emu (I know some people have to use bgb because of issues with gambatte reading all of the controller inputs of some controllers)

I have no option to test with GBP or gbi (don't have the equipment for it unfortunately). The hard reset for gambatte r614 is gbp accurate as far as I know. This is the reason why the pokemon community have allowed it as the only emulator for runs.

I am not opposed to banning the cutscene skips by hard resetting if there is a difference between sgb2 and gbp. I can't test this out myself so i don't know how it behaves there.

Thanks in advance for even considering this!

Oh that's the Pokemon build of Gambatte! I've been meaning to check that out; that just might work even if it's comparable to the DMG I would be happy with that. If we find that we can put hardware and the emu on the same page then I'd be fine with a misc category.

I was only uncomfortable with the idea of an emu being able to use it's tools or quirks for a time advantage. I'll update my with my tests later tonight but I don't see why we couldn't have a misc category limited to that emu either way. :)