Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
I only watched the first half, but it was an enjoyable watch. The captions are a nice addition and offer additional context and personality while watching. Avoiding the lock screen battles is a good strategy. I can't speak to all the movement intricacies, but it generally seemed to make sense within the constraints of avoiding battles and I imagine the submission notes provide additional details. Overall, a fast-paced TAS and smooth watch.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Short and sweet. It's definitely cool that something like this could exist for the Atari 2600. Initially, I was surprised that level 8 was the easiest difficulty and level 7 was the hardest, but it seems like level 8 being the easiest difficulty was probably an oversight by the developers.
Not much else to say here. I do also think it'd be cool to see the level 7 difficulty TASed; with such long move lengths, there would possibly be an additional optimization aspect of finding a winning game where the AI doesn't have to think too hard.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
My understanding from reading the submission notes is that this run is aiming for fastest completion and that fastest completion happens to include avoiding damage:
Therefore, this probably doesn't need to be branched as "no damage", although the publication will be tagged as 'Forgoes intentional damage' (I think?).
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
The author of the above video posted their TAS to user files here: User movie #638288381497727930 (15:10.80 in TAS time). I'm hoping to try and improve this in the near-ish future. I will probably aim to be a coauthor on this movie, since having this existing movie to build off of will probably be much easier than routing every level entirely from scratch.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Yes vote. This was fun for what it is. This sort of gameplay might get tedious after a few minutes, but in a 40-second TAS, there's a nice buildup of different point-scoring strategies that are clearly superhuman and did surprise me a bit.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
I have my own thoughts on Field of Flowers: I was able to save 3 frames, but it looks like it's a pretty static speed of being able to move every 9 frames, so I don't think any further time save is possible there.
Anyway, after I made some improvements, PinkyNoice is going to finish up the movie and probably upload an improved version in the next few days. We had some trouble with the pegasus minigame desyncing, but it seems to be a 2-frame frame rule. I did some further testing, and it seems a bit more complicated than this. I'm not exactly sure why alterations earlier in the run will desync this section. Fortunately, I've simplified the inputs so it's easier to iterate the minigame further and saved over 30 frames in this minigame alone.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
I'm actually not sure how to access the game's credits, but it's not from advancing the screen forward at the end; after the ending cutscenes, it just brings you back to the gameplay section.
The credits do appear to exist: https://www.youtube.com/watch?v=f_OYpqVB7HA
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Looks like the header became somewhat malformed somehow in the most recent file.
I uploaded this file that re-adds the rerecord count, adds you in the authors field as 'OtakuTAS', adds the cycle count, and generally fixes some other issues. I also played the movie back in 2.8, since it still seems to sync, and that's what the original movie used (easy enough to update to 2.9 if wanted -- but it's really just metadata in this case, since it hasn't affected sync so far).
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
I mention this only because I think it's important improvements are accurately measured, but I think you must have miscalculated how long the boot ROM takes or made some other error in your calculation; I calculate the GBC boot ROM as taking 186 frames. I played back both my movie and yours and they end at the same point. Your movie as uploaded had a final time of 27682 frames, while mine was 27810 frames; 27810 - 27682 = 128 (it's the same if you use cycle counts). The original movie had only one blank frame at the end, so that doesn't explain the discrepancy. Maybe you meant to upload a different movie?
I guess you're also using BizHawk 2.9, instead of 2.8, which the original movie used. It still seems to playback the same on 2.8, but Gambatte did have some changes between 2.8 and 2.9. Probably fine, but worth noting.
Anyway, I look forward to more progress soon, but also no stress. :-)
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
You probably missed my own resync where I cleanly added a few frames to the beginning to account for the boot ROM and trimmed the blank inputs at the end. Comparing to that, your movie is only a 128-frame improvement. (Note: this resync was purely mechanical and took only a few minutes, so I do not deserve or want coauthorship at this point in time).
I might have a look at this. There were a few moments I was curious about, but I might not be able to find anything here. I certainly lack the minigame expertise you have from making a prior TAS, so I appreciate that you've taken the time to give your feedback here.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Hi, I read this post and it's cool you're making an updated TAS with the GBA version of ALTTP.
There was just one thing I wanted to note. You might already be aware, but from the rules for submission standards (Wiki: MovieRules#WhenTasing):
So any text-speed related differences from using the Japanese version would not be taken into account for judgement. Sounds like you're already ahead in other aspects though (it's a bit difficult to tell from the comparison video alone, since it doesn't compare screen-by-screen on the actual gameplay).
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Galaga is a personal favorite of mine. This movie looks pretty strong, and it seems like the botting approach benefited the optimization. Good job!
I didn't even realize that there was RNG, or that 'L' and 'R' presses affected it. A few years back, I was thinking about approaching this game with a shot-sequencing algorithm that would explore various different attack orders to find the optimal one, but I guess that approach would probably have been infeasible with RNG-affected enemy ship movement and the fleeing mechanic.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
I forgot to mention this in my submission notes, but I ended last input at the last dialogue box of the last level. The game does have one last story screen sequence after this, but going through it or skipping it just leads back to level select, so unfortunately the game does not show credits at the end (you can only access them from the title screen, and you can access them even with a fresh save file).
I thought this approach was sensible for this case, as it follows the internal logic of the speedrun, and matches the current realtime speedrun rules.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Near the end of my work on the Shrek 2 GBA TAS, I decided to spend a bit of time looking at Shrek the Third GBA. Shrek the Third has fairly similar basic mechanics to the original, with multi-character levels and locating keys to progress in levels. However, there are also a lot of differences: you have a different cast of characters (Gingy is replaced by Arthur), the basic physics are very different (holding 'A' to jump jumps higher), and the basic progression requires you to collect a certain amount of fairies to get to the next world. The game also has some of the new innovations from Beg for Mercy, like more abilities for Puss and automatic key pickups.
I have to say, I really don't understand this game very well. I haven't looked into the fairy threshold recently, but I know you don't need to collect them all for any%. 100% is probably easier to route because you just collect everything instead of being more discerning. There are also optional special abilities each character gets after collecting 4 specific objects. There used to be fairly detailed speedrunning resources two years ago, but all of the links have died since then. Almost all of the speedruns were dead links as well, thanks to Twitch not automatically saving VODs. Fortunately, someone did a decent speedrun a few weeks ago:
Link to video
I also used rewind to get started on a TAS, but it's kind of a mess. The early levels are particularly bad, but my entire playthrough has routing issues. Theoretically, I was aiming for 100%, but I missed some required objects and fairies. Most likely, I'll try and update this with a file that finishes the game, and then take a break for a while. I might just take a break before finishing a complete movie, though.
My movies: 001 (19 July 2023, 25:42.04, through 3-3)
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Having just finished a TAS of the original Shrek 2 GBA, I thought I'd make a topic for one of the sequels.
This game is generally pretty similar mechanics-wise to the original. Puss in Boots has a new down-attack and can double jump. Unlike the original, there are no multi-character levels. Instead of needing to spend time pressing select to pick up items, items like keys are picked up automatically.
Someone else already uploaded a video-only TAS of the game that completes the game in 14:57.83:
Link to video
Having done my own TAS, I think there are a few seconds of time save from improved movement (for example: sword slash landings and less bonking on corners), but it's generally a good TAS and a solid foundation.
Edit: I uploaded a bk2 of a real-time speedrun from a few months ago; it's almost twice as long as the TAS: BfM_001 (1 December 2022, 28:59.97)
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
This doesn't use a boot ROM, but it probably should. BizHawk will warn you about this every time you open the movie file.
Fortunately, it's pretty easy to add a boot ROM and resync the movie: User movie #638248589572532942
I also created a standard speed encode:
Link to video
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
I concur. I didn't think this TAS was quite right for our traditional movie tiers at the time (although the rules have changed since then), but I do think this run is a good fit for playground.
I think this is also in compliance with the Playgrounds rule "Playground will never house all runs of any given game.", since the DOS version exists and could be TASed.
Editor, Experienced Forum User, Reviewer, Player
(166)
Joined: 8/8/2019
Posts: 131
Even if LibTAS worked well with web browsers, The Password Game wouldn't work anyway, because it requires networked resources for at least two of the sections (the GeoGuessr one and the chess puzzle one). I guess it probably also requires them for validating the YouTube video length one as well.