Nice run, yes vote. But i have question about very ending.
May be is possible somehow manipulate to die exactly at 99999m, just before overflow to get "maximum score" category?
I think it would be a way better than unremarkable .0027m final score.
Is it possible for the game to even end at 27m normally?
We have decided that because Kooper skip does not work on Japanese, the TAS will be done on English. Though some time will be lost due to text, time will be saved overall.
That video is now obsolete since I've gained greater understanding for how to optimize in this game, so I'll upload my improvements later. Thanks for the support :D
How about the PAL version? Or is that also in English?
For the purpose of encoding or lua botting, would disabling graphics make it run faster? It still emulates it, but just doesn't show on screen (not sure if it makes sense).
From what I've heard, SNES9X 1.52 and 1.53 emulate lag better than 1.51, but unfortunately there is no lua support for them, so they can't be used for TASing if you need to use lua scripts.
If I ever get around to improving ActRaiser, I may end up having to go backwards a version (1.52->1.51) so that I can use lua.
If pacifist ends up losing time, then it's not vaultable. :/
I'm not trolling either.
Edit: And please don't expect me to know all about this game given the informative submission text. >.>
As for the desyncs, I haven't used Dolphin since I finished the TAS with Bass for MM10 so I don't know how far it's capabilities have improved, but I always found excessive rerecording to desync quite often.
Well, for most games, it seems quite fine now, but a select few seem to desync still. If it ever happens, please report it on that thread.
Is it because the game registered input at a later frame without the game lagging? This happens to me as well for Over the Hedge, where input for moving only registers once every other frame despite not lagging at all.
Or make a playaround using every single glitch known including ACE. I'm thinking of something like what Masterjun did back in that joke SMW TAS, but rather than move the player's position, have like a hotkey that spawns more enemies or showcase something normally not able to do.
That'll take a lot of time and effort however.
I'm voting meh. It's well done and all, but something always bothered me.
I mean, I like ACE, but I always wanted that the old warping to final level run wasn't obsoleted. I know that it is too much to ask for, but can we unobsolete the warp run? Like, by a new "warp glitch" branch or something. Just to separate both warp and ACE runs by default.
I like that one better anyways.
I agree with you on the idea that warp glitch runs should be separate from ACE runs, and that the YI warp glitch run should be "reactivated," as it were.
I found the old warp runs kinda.....lackluster after several versions. They stood out when the first version was made, but then it became really repetitive after the route becomes the same and there's not really much room for different playarounds.
Anyways I prefer the first screenshot since it's less "spoiler".
http://tasvideos.org/userfiles/info/19504398734556519
I'm posting the first set of stages so far since it appears the abacus stage is sub-optimal in quite a large number of frames during the first OoB.
Edit: Frustratingly enough the RNG for the safe combination is the only thing making this hard to edit back; every other stage can easily be spliced in with very minor lag differences.
Edit2: Saved 126 frames on abacus stage alone. Also discovered with Hammy and RJ (but not Verne), you can preserve the walking speed in midair by letting go of forward input. I believe it may be possible to save more frames, but using a different behaviour pattern for the dog caused me to drop the abacus too far into the room when I go OoB.
