Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
Most of the stuff being discussed here is in future tense, but it's not a bad idea to start the discussion now.
http://emulation.gametechwiki.com/index.php/Game_%26_Watch#Simulators
Long story short: Libretro is adding simulator cores for G&W games, and MAME is implementing the ability to actually emulate G&W roms.
It seems very likely that TASing those simulators via Libretrohawk might become possible(if ever) before TASing via MAME-core on Bizhawk does. If this turns out to be the case, will movies made with libretro G&W simulators be eligible for publication? Those cores aren't actual emulators, but does the simplistic nature of most G&W games make high-quality simulators good enough to represent the actual game?
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
jlun2 wrote:
Would it be better to get max score on all 3 games there? Also is there any way to view the credits or anything here (GaW 4 for instance) or nah?
Unlike the 4 G&WG games on GB/GBC/GBA, there are no collectibles or credit scenes in the 2 G&WC games on DS. The only information stored in save file is high score information on each game, which seems to max out at 9999.
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
http://tasvideos.org/userfiles/info/43316369425187948
So I extended the run to the 8th loop. The speed of barrels seem to stay the same at the beginning of each loop after this point. Please replace the input file. I'm uploading a new temp encode.
I don't plan to make a new submission because I suppose the feedback won't be much different. :P
P.S.:Although the on screen score counter overflows at 999, which actually happens in the updated movie, the actual high score record of the game can go all the way to 9999.
Edit:
Link to video
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
EZGames69 wrote:
I think a TAS for this would be faster on GaW for GBA, this one just seems too slow and just looks like a human playing it.
First, the classic version of Donkey Kong included in G&W Gallery 2 and 4 wouldn't really be faster than this one.
Second, this one is much more faithful to the original G&W game thanks to the fact that a DS has two screens.
Third, all G&W Gallery games have unlockables. Beating one single game in those games would seem like an arbitrary goal.
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
For some reasons, Pokemon Green actually runs pretty well.
In WL-SML3, the way sprites flicker in a place with many blocks seem to be different than how it looks with Gambatte core. I'm not sure which one is more accurate tho.
Super Connard refuses to run. :P
I know GBHawk only supports DMG right not, but is there a way to force GBC games to run under DMG mode with GBHawk instead of automatically switching to Gambatte?
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
Controls seem to be messed up in all games. I guess it's a controller initialization thing?
Metroid 2 has glitchy graphics.
BTW, are there any chances for a new SGB core based on GBhawk in the future?
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
A question suddenly came to my mind. Does memory corruption affect the other two save files?
Once the corruption is done, I have to reload the save by dying. If I were to reset the game, the save data would be reloaded from external memory and copy over the memory corruption.
Would returning to the title screen make the game reload external memory?
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
I tried to recreate the glitch on 0.9.9, and it didn't work. The accuracy of Desmume is probably not enough to perform this kind of precise save corruption yet.
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
It would be great if the libretro MAME core becomes fully functional.
Coleco Adam would also be a nice addition. I will TAS Coleco Supergames - Donkey Kong Junior when that happens.
Editor, Experienced Forum User, Published Author, Experienced player
(950)
Joined: 10/19/2013
Posts: 1127
Alyosha wrote:
What if instead, we were able to accept runs without the need to publish them? They can exist in movie file / submission page form alone and still be present and searchable. We can save publication for those rus that people find entertaining enough for the effort.
All rejected submissions are archived. The notably good ones are even listed in Gruefood Delight.
I think the most unclear part of the rules is the definition of being "poorly modified". A PAL version that is "not modified" should obviously not be used because the resulted run would be slower than an optimized run done on NTSC version for sure, but what is "poorly modified", exactly? And why is it especially important for PAL games? It's not uncommon for J and U versions of the same game to have different glitches, and the one with more beneficial glitches that result in a faster run is usually chosen (for any%, at least). I don't think anyone would argue that [2565] GBC Pokémon: Silver Version "game end glitch" by MrWint in 30:39.49(there are faster methods now, I know, but this was the fastest run back then) should not obsolete its predecessor for using a glitch resulted from localization, nor do I think anyone was against the use of J version of TTYD for being more broken.
IMHO, the fastest version should be chosen for any% in all cases unless the only differences are texts(using English versions for better understanding is a good idea, unless J is the only NTSC version). If a slower version is considered more authentic or popular, a run of that version should be the one that is judged as a special category run.
TL;DR: The fastest version should be used for Vaultable categories. Choosing English versions with longer texts should be fine unless it's PAL. I fall into the third school mentioned by Nach, I guess.
