Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
Checked the last thousand frames and saved about 70 frames again, also ended input earlier which resulted in total 137 frames on that final segment.
User movie #638336909747562432
I tried applying this common tech to jump cancellation, and interestingly, if you press Space to jump and then Ctrl on the next frame, the jump continues. But if you press Ctrl to jump and Space on the next frame, the jump cancels instantly, allowing you to jump more rapidly in that section where you have to jump 50 times. Somehow I also didn't have to wait on the yoku block before jumping from it.
I don't think this submission is acceptable as is.
Damn, well, at least I didn't miss any known tech at the very least. Gonna have to cancel this too, I guess, since that makes a total redo necessary- or at least, an intensive enough overhaul that leaving this submission in the queue would take too long
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
Which skip?
23 seconds in, you were able to skip an unnecessary jump; that should be the only mistake like that I made, in which case basically all of my inputs should still be good with some resyncing to account for different timing.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
Alright, in that case that makes fixing this a lot easier- although there was no way I could have known that that existed when I was making this, as a note. No idea if it will be an edit or a full redo, depends on if there's actual movement optimizations at large or if it's just that one timesave at the beginning.
EDIT: also, I was trying earliest frame everywhere too, that's where all the rerecords come from- I even took framedata on earlier possible jumping, and it's always one frame before scroll starts as far as I can tell.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
....how were you able to jump that early? I tried to start input as early as possible, what you see in my recording was as fast as I was able to get it. Plus, besides that, I can't even tell what form your optimization is supposed to be taking just in general from watching the video; I would appreciate a hint, because I can't tell if you somehow managed to cut your jumps earlier or something.
Plus, while patching other recordings is easy enough in BizHawk or libTAS, Hourglass is going to be a serious problem as far as I can tell, I would probably need to redo the entire movie. Making a 1-minute demo is easy enough, especially when the bulk of the timesave is immediate upon starting the game, but this recording took me many many hours to make. I'm not sure that "Movie Must Be Technically Sound" requires me to do that all over again, especially with the crashing, but I'll give it a try like everything else.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
I can record my libtas workflow if you want, because there are useful things that may not be obvious to newcomers. To be efficient you need to know what you want and how to get it without too much overwhelming routine.
Nah, I'm good- what happened here is that I just totally discounted acceleration as a factor in this game, because it mostly isn't a factor along the horizontal, and even isn't a factor in short single upward jumps for the most part, so long as there's enough room ahead of the target platform. Faster acceleration from double-jumping for single jumps or mini-hopping off of platforms just never occurred to me, and even if it did, I would think it would be in the "shaving frames" category rather than saving whole seconds, especially since I had already incorporated recoil-boosted falling from shooting upwards into the run everywhere. Thankfully, this run is teeny tiny, so I should be able to fix this along with the Ball recordings without much issue.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
Okay, damn. The issues with Ball are obvious in retrospect, but I had zero idea about this since it's totally irrelevant to RTA speedrunning- I'm just going to get this over with and cancel it until I incorporate what you're mentioning. Thanks for the input
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
roblox8192 wrote:
I have uploaded the recording file for my TAS if you want to take a look. There's two of them because my TAS in the video you linked does some setup before the run actually begins (since it was meant to show off what the fastest run would look like with RTA rules), so I had an alternate TAS which still has the pauseless and glitchless restrictions while not doing the pre-run setup. You can find the files here:
User movie #https://tasvideos.org/UserFiles/Info/638327585670750437User movie #https://tasvideos.org/UserFiles/Info/638327586812135207
Ah, thanks- I'll see if there's any timesave I missed soon, although possibly not tonight since it's getting late.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
ikuyo wrote:
So, first of all, I want to apologize for the long time it's taken me to get to this judgement. I'll confess I was not too keen on touching this game, but that's exactly why I need to talk to you all about it, and what this means, or could mean, for TASVideos.
In case you haven't played the game before or aren't aware of it otherwise, the game is, very directly, the story of a fictional school shooting. Your character very directly commits one as they go around, and as the author points out, there's also some rather puerile graphics shown. In most cases, I would just object to all this and reject outright, but Pico's school is, as the author points out in the submission, a significant game in the history of Flash as a medium and of Newgrounds as a platform. Heck, they put Pico in Friday Night Funkin' for that very reason.
TASVideos does not punish the authors of movies for questionable content included within the movies, and we've published movies with rapid flashing lights in the past, for example. But I cannot, in good conscience, let this movie be published without a system that allows us to warn viewers of what the game is about. As such, my proposal on how to handle this movie is delaying until such a system exists.
A content warning system would require some significant retooling on the publication side of things, and I cannot guarantee it will be implemented anytime soon, but I think this is the best way to allow this movie to be published, and also a significant improvement to the site overall.
Are you, as the author, ok with a delay on this? What do our users think of this as a feature?
I'm fine with any delay necessary, although I do want to make a significant correction- your player isn't the one committing the school shooting, that's the antagonists. It's possible to shoot a couple of innocents for shits and giggles of course (Nene and the guy in the men's bathroom), and it's possible to kill Alucard after he surrenders, but I don't do any of that since it wastes time, and in general that's in the same category as choosing to dump Yoshi in a pit for the hell of it- player choice and all that.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
CasualPokePlayer asked for feedback on this over on the discord, so I'm just gonna mirror what I posted there:
IMO, there really isn't that much to say in terms of feedback.
The movie is a technical masterclass in TAS tooling and creation and whatnot, and it is a superset of ordinary undertale branches, so that means it qualifies for inclusion into Alternative just as a matter of course, IMO. Naturally it's really boring since it's a whole lot of repeat gameplay with minor variations, but it's way above the bar just in comparison to Desert Bus, and in general I find it really difficult to imagine a recording of this technical caliber getting rejected on entertainment grounds. Strong yes vote.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
Alright, bad news: The table is defective, because none of the lights activate in the right order. So I tried a newer, better-looking table, but that one was also defective, because the drop targets wouldn't come back up after being hit. And after that, I've basically exhausted all of my options- the next newest table version is for VP9, and that works, but I can't get VP9 to run under the PCem hardware configuration, it doesn't even launch to the table editor.
HOWEVER, the Visual Pinball 8 setup does work overall, even if there isn't a functioning Baby Pac-Man table for me to use. I plan on TASing the Atari 7800 version of Baby Pac-Man as a consolation prize, but I'm not done with actual pinball/video arcade hybrids just yet. There are three more pinball games that I'm looking at, CavemanMr. & Mrs. Pac-Man, and Granny and the Gators. Caveman is the actual first hybrid pinball game to release and features what could be considered a full game in the video portion, although it is entire optional and the main mode of play is the pinball portion (as opposed to Baby Pac-Man, where the video portion is the main mode of play). Mr. & Mrs. Pac-Man is really just a pinball game with a small bonus minigame included, although it's definitely way more than nothing at least, and it has Pac-Man's star power going for it. Finally, the most promising is Granny and the Gators; it features the video game as the primary mode of play, which makes "Hardest cycle" viable as a TASing goal, yet the pinball segment is required in order to gain extra ammo for your gun in order to get rid of the gators. Testing all of these will be a monumental effort, so I plan on tackling the 7800 Baby Pac-Man first and getting a break from all the Visual Pinball frustrations I've been having the past couple of days.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
Unfortunate what happened, considering that Bowman is a VERY historically notable Flash game. While it obviously wasn't the first Flash game (that label is disputed and depends on what you consider a "game" vs an "interactive toy", it IS the first ever Flash game to use actual code in the form of ActionScript. Prior to Bowman, Flash games were really just complicated Flash Animations/Movies from a technical standpoint.
I think I might tackle this game myself at some point, although I'll have to start from scratch since I don't want to have to track down a banned user to ask for permission to work off of their recording. We'll see.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
Still, one 3rd party video without a ruffle splash screen is not enough to define a movie rule for all Flash submissions. An FP test on the other hand would be quite conclusive (and verifiable).
I just recorded a video doing just that, hold on a minute while I upload it to youtube
EDIT: here it is https://youtu.be/cwhSNRWZhvw
It's not super visible because of Snipping Tool's FPS and whatnot, but I definitely killed alucard way faster than should be possible with only 12Hz clicking (the max possible at 24FPS)
EDIT EDIT: Also, for reference, skilled spam-clickers can click faster than 12CPS- and while Pico's School isn't a super competitive speedgame, I can definitely imagine some non-autoclickee videos server as proof of the game registering inputs at faster than its stated framerate
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
Technoturnovers wrote:
Can a TAS of a simulated pinball table be considered an actual TAS of the game itself for cataloging purposes, or is it only a TAS of that particular simulated table? If a TAS of a simulated table is considered a "real" TAS of the actual game, should different versions of the reproduced table be considered different game versions?
Need more info. Wasn't the whole Baby Pac-Man machine some kind of a fixed thing? A TAS of the whole thing sounds legitimate, and versions rarely co-exist as separate branches, because they are usually not too different.
The reason why the version of the recreation is significant is because a recreation of a physical pinball table is ultimately subjective, because there's no way to create a 100% accurate simulation, so the person doing the recreation has to decide on the physics parameters based on what feels right. Basically, it all comes down to the fact that the tables just aren't interchangeable, you can't swap in a different version of the table into the TAS and expect it to work at all.
feos wrote:
While the setup looks crazy, it would make sense to have a movie that installs everything but the game, and then the game could be different for each actual TAS. That way one would be able to install the emulator (simulator) once and keep that disk image, using it for every new TAS, just like we're meant to keep the image with xp just installed.
I'm not gonna do that, because it just isn't really practical. Every pinball table has different prerequisites and a different optimal configuration, and the ROM VPinMAME will use has to be on the disk, it can't be run off of the CD drive. It's even entirely possible that future Visual Pinball submissions might use VP9 if that's compatible, which would require completely different files to work.
feos wrote:
Technoturnovers wrote:
"100%"/All bonuses: There are three types of bonuses in Baby Pac-Man; up to 4 Energizers per stage, the Banana as the top-tier fruit (requiring advancing the fruit 7 times), and a maximum tunnel speed of 8 (also requiring advancing the speed 7 times). As such, it is possible to construct a goal of reaching the hardest stage while maxing tunnel speed, acquiring all energizers, maxing out the fruit type, and finally collecting at least one fruit per stage, thus offering a balanced combination of pinball play and maze play. This is the option I'm currently leaning towards, but it also could potentially be considered arbitrary, hence asking for judge input.
Only things you can collect 100% of would be needed for such a goal. "At least one fruit per stage" sounds like there are more, but they are not important?
I revised my goal not too long ago- I said "at least one" fruit because I didn't know how many actually spawn in this game, but now I know that 2 fruits spawn in each maze like normal Pac-Man, so collect 2 fruits it is.
Also, I revised the verification movie, so the second site isn't necessary now thanks to the new and improved table I'm using; I'll revise my thread once I get this all fully settled, I might still need to make some tweaks for performance reasons.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
Verification Movie Details
The verification movie isn't particularly complex, doing the following things in no particular order:
Copy the files from the CD to the Desktop
Install PinMAME, registering the COM library and creating default configurations.
Change Visual Pinball's video settings, disabling cached table rendering and setting the resolution to 1024x768x32 fullscreen so that the entire table is visible, and not obstructed by the VPinMAME DMD.
Launches Baby Pac-Man in order to clear a nag screen, then again to make sure it stuck, deleting any written files afterwards that could affect sync such as NVRAM.
Creates a shortcut to launch Baby Pac-Man and places it in the Startup folder.
Delete Flash Player, because we won't need it.
Finally, shuts down the computer.
One thing that it does NOT do, due to me forgetting, is to disable/hide the mouse cursor in Visual Pinball. This won't affect the game at all, but could be a very minor visual distraction for viewers; however, I rank that as a much lower nuisance than the general ugliness of the table chosen just at base, so it's not a terribly high priority considering how much effort it would take to alter the verification movie. Additionally, due to the way that libTAS handles resolution changes, this video is effectively self-splitting at the moment that Visual Pinball finishes launching. This could be a good or a bad thing, depending on if you want to include the Windows XP bootup and logon in the movie for veracity and completeness purposes.
TAS Goal
After some discussion and advice, I've decided on a set of goals for this game that should complete all unique content, while offering a balanced combination of Pinball and Maze gameplay; I'm tentatively going to call this branch either "100%" or "All bonuses". They are as follows:
Complete the hardest and last unique stage; in this case this would mean ending at Maze #12, as there are 3 unique mazes that cycle, and all inner maze walls turn invisible starting at Maze #10.
Collect all four energizer pellets in each stage. This requires hitting all four of the yellow drop targets multiple times to spawn them all, saving time by using the second-from-left/right drop lanes to even out the P-A-C-M-A-N light progress.
Advance the Fruit type to the most valuable one, the Banana; this requires fully lighting up the leftmost lane multiple times.
Upgrade Tunnel Speed to the maximum value; this requires fully lighting up the rightmost lane multiple times.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
Currently the NG game entry uses ruffle too, but I can't check what it was using back then, because even archived copies from 2009 somehow launch it with ruffle too.
That's just because the Wayback Machine has Ruffle built in. Pico's School is a Flash movie made in 1999, it used flash player because flash player is the player that existed back then
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
It's running in ruffle in that video, and running it in FP at a rate higher than the game's reported value would show that it can indeed react faster (or not). Are there records using both FP and autoclicker?
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
Technoturnovers wrote:
Natively this game really DOES run at thousands of FPS natively
How do you know this?
Through simple observation- Pico's School autoclicker speedrunners achieve uber-fast times in boss fights that would simply be impossible if the game actually ran at only 24FPS. Because this isn't a proper ActionScript based Flash game with an actual input loop based off any kind of tickrate, gameplay functionality isn't determined by the FPS claimed by the SWF file, but by the running speed of the Flash Player itself.
EDIT: for reference, the current Autoclicker RTA WR: https://www.youtube.com/watch?v=AbeJyRXEBLI
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
feos wrote:
If you run it in Adobe Flash Player in PCem, set game scale to 100% while running it on some high resolution monitor, and set monitor refresh rate to 100, does it become possible to click objects every frame and affect them? If yes, that sounds like a reason to allow arbitrary framerate shenanigans for Flash, much like we allowed off-screen inputs (which this run is using I presume).
This run isn't using offscreen inputs, or anything that could actually be considered an "unintended environment" TASing wise. Natively this game really DOES run at thousands of FPS natively, it's just that recording like that with libTAS is a logistical and technical nightmare, you have to hammer frame advance so many times. Using variable framerate effectively wouldn't actually be allowing anything new for this game from a technical standpoint, it's just a convenience of being able to run 5000FPS or whatever on demand rather than for the entire duration of the movie. Again, though, that's dependent on Variable Framerate actually working right on record, which isn't a guarantee.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
For me, I think the current most pressing concern is one of preservation, plain and simple. TAS recordings using specific engines, custom mods, or anything else that are presently impossible to publish could become accepted in the future as these tools either gain broader acceptance, gain necessary features like AVI dumping, or anything else, but that's dependent on the recording files actually surviving to that day, which is anything BUT guaranteed. Many of these demofiles and tools are hosted in very obscure locations, unreliable filehosts, or etc that could cause them to just become lost before they ever get their fair shot. I think a good stopgap solution, one way less strenuous than overhauling the submissions system, would be expanding Userfiles so that it can accept additional movie file types that aren't supported by the site now, but could be in the future; my understanding is that adding new accepted extensions for Userfiles is as simple as editing the relevant System page, but it might be good to add a third system page besides just 'SupportedMovieTypes' and 'SupplementalUserFileExtensions', possibly named 'ExtraMovieTypes' or something, for better organization.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
Alright, so a while back, I learned that libTAS supports variable framerate, which would be perfect for getting the time on this TAS down even lower by completely trashing the bosses without needing to record the entire rest of the movie at 5000FPS where that isn't needed. I'm not quite sure of if variable framerate is actually stable as an option, mind, and it would still require re-recording the entire movie, it would just make doing so SIGNIFICANTLY less strenuous. I'm not sure if doing this is worth it, so I'm gonna need some expert advice on if this submission really needs the extra timesave or not.
Experienced Forum User, Published Author, Player
(61)
Joined: 9/15/2023
Posts: 71
Fair enough; I was still learning libTAS when I made this and Ball 2, so I was kind of sloppy just in general. I'll definitely go back and improve this and the Ball 2 recording, so feel free to set this to Delayed.
