Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Is it possible to replace it without rejudging? (see bottom of submission notes for more details)
(I got an error https://postimg.cc/9zntmL5t when trying to post)
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
That link doesn't work, but the userfile ID (and the improvement itself, probably) is valid: https://tasvideos.org/UserFiles/Info/638221025467132301
I think it's quite important now that publication is underway and the file wasn't replaced for some reason.
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Well, ig Bigbass has already made an encode with that console verification, but maybe I'll make my own from the emulator
EDIT: I've made a 4K60 encode, but I set it to go public on Thu, 15 Jun 2023 20:15 +0200 -- since Bigbass already showed the console verification, and also I kinda stopped caring about this encoding stuff because I'm watching my submission being processed, and oops I got a cold too, oh well
Link to video
Oh wait, I forgot about the old/new PPU thing that is buggy a thing since 2.1.2 - before the Qt interface was made. I guess that defaults to the new PPU? Or some other version? 2.2.2 definitely mentioned something about improvements from BizHawk...
So I guess you don't need to change anything in the input? I think we can go home then.
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
The "less faithful" part is true in general, but I think most (if not all) FCEUX 2.2.3 movies sync on newer versions too, and on the Qt frontend as well. If only that didn't stop working for me... which is one of the reasons why I prefer BizHawk too.
I think this ( https://tasvideos.org/UserFiles/Info/638212840144056338 ) is a "resynced" version of this ( https://tasvideos.org/UserFiles/Info/638212840920688175 ), but it actually loses more time due to bad luck or something. Well, it's probably always better to resync to newest stable (Sub)NEShawk anyway, but who knows.
It would probably take a lot for FCEUX to get into the "accepted but not preferred" section, right? asking for a friend +- -+
Do that many people have 32-bit only PCs, though? It's probably on the decline too, especially since Macrohard is killing the ecosystem by turning 60% of the current Windows 10 PCs unupgradable, I mean they're doing Vista 2, I mean Windows 11 released.
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Well, since the "Challenge" TAS ends on the last input required to reach the killscreen, I think this "Freeskate" TAS can end on the last input required to complete the loop (what this updated movie does).
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Well, it looks like it desyncs in Stage 1, Area 1 for some reason, and the previous movie still works just fine. It also looks like the inputs have been corrupted at some point, so it looks like something broke during the BK2 to DTM conversion?
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
I made a demonstration TAS of that: https://tasvideos.org/UserFiles/Info/638212403790869316 (based off this submission)
49 frames slower and it is not a valid IGT strategy, but it looks funny anyway!
I think it is console verifiable as well (Alyosha hey hi hello)
EDIT: oh bruh I forgot about RAM addresses that I used:
0254 seconds
0268 minutes
I did not find one for frame count (maybe I should look for it harder), but that is enough to determine whenever the IGT is actually at zero, and this movie does complete without changing any of these values, although completing the game does not stop the timer.
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Although, isn't this actually from Creative Sparks? This ROM from the TOSEC DAT pack (Commodore C64 - Games - Arcade - [D64]) matches the movie hash and syncs with the movie as well. Is TOSEC and me wrong then? (and yes, I checked the latest DAT pack.)
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Well, ViGadeomes, you kinda screwed that up too: It's not 636177476 cycles, but actually 636177472 cycles. It's definitely enough for BizHawk to complain, so I think the movie file should be updated.
EDIT: I should have not said it that way, I want to apologise.
EDIT 2 because I didn't disable "minor edit", also I wanted to split the message: since ViGadeomes wanted to keep using the version of the original submission and Samsara (to humiliate myself in front of an admin, damn it...) supports it, the cycle count info is completely irrelevant, since BizHawk 2.8 doesn't care anyway,
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
It got accepted! Although, ViGadeomes (I hope you'll still read this), why I'm added in the "additional authors" section? On the other GB submission, I can see that this section is "Only [for] authors not registered for TASVideos".
EDIT: and why add the "Challenge" branch? Can the "Freeskate" mode have an actual ending that TASvideos usually requires?
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Well, I finally finished it: User movie #638177078672416870
I think track four might be slightly unoptimized at the cone parts, but I think it's good enough for a Judge to decide whenever three and/or nine loops are acceptable. I still wonder if I can save at least three frames on track four for sub-38, and/or (much less likely) at least ninety frames total for sub-3.
IGT IL: 46.72, 53.98, 42.75, 38.03 - IGT total: 3:01.48
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
So I think that's as optimized as it can be, unless a bot finds something (maybe not). It's just the case of whenever we want to end input early or aim for IGT, since it's a difference between 13.62 and 13.04 seconds on the last track. I'm aiming for IGT, but I'm also taking time to make it a 9-lap run instead.
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
It turns out that both ways give an IGT of 13.40 seconds for the fourth track, so I think it's fine to end input early. I'd finish at the last frame the internal frame counter stops updating, but that's just me.
Also, you can press R+D+S at the same time to change the difficulty to Hard and start faster. Or you can press L+D+S, it looks like the arrows don't matter in the menu, for some reason...
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
While trying to make my own version, I found additional RAM addresses:
0016 lap number
0017 laps
001B frames
001C seconds
001D tens of seconds
001E hundreds of seconds
Which means you can complete three laps of the first track on Hard mode in 16.32 seconds. Except you can select six or nine laps, should the maximum be chosen here as well?
Experienced Forum User, Published Author, Player
(21)
Joined: 1/13/2023
Posts: 55
Location: Poland
Improved by 20 frames by using SameBoy in AGB + BIOS mode and properly trimming the movie: User movie #638169176801506152
This category is not acceptable for TASvideos afaik, so you have to beat all levels normally ("no level skip"). No weed this time, so hope for leet.
Also, please for god sake use a newer version of BizHawk, I have a feeling 1.x is mostly deprecated by now, AND it desyncs on the latest versions too.