Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
EZGames69 because his C64 TAS of Snoopy was one of the few TASes that actually made me laugh, and when it comes to TASes on obscure or less popular consoles, that's about as much as you could ask for in a TAS.
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
Hi everybody. I just made a TAS of Bobby is Going Home for the Atari 2600, which can be found here: http://tasvideos.org/userfiles/info/60426653132727859
This movie is 3 frames faster than the current TAS in the first level, and also completes the second level (which the current TAS doesn't). I'm still trying to determine if there's a faster way to do the second level of the game. As such, if anybody can find any strategies to save time in this TAS such that the final time for the TAS is lowered, then they can be listed as coauthors with me.
This is a pretty simple game, so if nobody has found anything new by Sunday, then I'm going to submit this TAS for publication. Good luck!
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
feos wrote:
Here's the rule we use:
http://tasvideos.org/MovieRules.html#GamesWithoutClearEnding
We require that difficulty no longer increases and all unique gameplay has been beaten. BG color doesn't seem to affect gameplay, but are you sure the difficulty stops at level 2?
Hmmm... I'm not sure. I just TASed the first level, and am 3 frames ahead of the current publication right now. After I finish the second level, i'll look for a difference on the third level (though I'm not 100% sure what I would be looking for. I guess I could find the position of the enemies in memory and see if their speed is increasing in each level?)
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
I recently noticed that the submission for "Bobby is Going Home" for the Atari 2600 stops input before all new content has been seen in the game.
More specifically, the TAS beats only the first level. However, from level 2 onwards, the first screen of each level has a disappearing bridge of variable length, which the first screen of level 1 doesn't have.
Beyond that, level 2 appears to be the same as every level after that (up to the highest level in the game, which is 32 according to the instruction manual), except that the color of the background changes on each loop.
Based on this, what would be the best stopping point for a TAS of the game? Would it be the current place that the published movie ends, after the first screen of the second level (when all new content has been seen), after finishing the second level, or after finishing all 32 levels (which are the same as level 2 except for having different colors).
I'm assuming that this TAS should be required to beat the second level of the game based on my experiences TASing Atari 2600 games and seeing how they are judged for "full completion", but I wanted to get a definitive answer from a judge.
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
This is a great demonstration of the power of TASing to completely bend a games' logic to work in favor of the TASer. A definite yes vote for me. Additionally, this movie should definitely be accepted to stars.
Thanks for taking the time to make this. It looks like a good start towards getting BizHawk to work on MacOS. From testing out this version of BizHawk on my Mac on Catalina, I did find the following errors that I wanted to bring up, in case you hadn't found them yet.
1. Atari 2600 games won't load on this version. More specifically, they cause a NullReferenceException to be thrown. The exact error message begins "System.NullReferenceException: Object reference not set to an instance of an object at BizHawk.Common.SettingsUtil.CreateSetter(System.Type t) ...". I tested this for the games Pitfall!, RoadRunner, and Adventure and all 3 couldn't load.
2. Trying to load the trace logger while playing an NES game causes BizHawk to violently crash with an exception.
3. Trying to open up TASStudio to create a movie for an NES game also causes BizHawk to violently crash. However, it is possible to play back movies which have already been recorded without the emulator crashing, so that's a plus.
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
feos wrote:
We didn't just silently reject his previous runs, we provided tons of feedback describing how to get it done right. And yes, this game's nature means it's non-trivial to check how optimal a movie is, so judging it took time for a reason.
I figured that it probably took a while to judge the previous submission. Even if it weren't for the hard-to-check optimality of the game, being over 2 hours long puts it well above the 50th percentile for movie lengths on TASVideos, and I would assume probably over the 90th percentile for SNES games.
feos wrote:
And apparently that's not true :D Here's my recent judgment that showcases how this works for Vault.
With regards to acceptable timeloss in vault movies compared to published runs, how much timeloss is considered acceptable? More specifically, how does that relate to the following rule: "If you found a shortcut that saves 30 seconds, your movie should be faster by at least 30 seconds. Losing time elsewhere for no good reason is unacceptable."
Basically, what I mean is, in what cases can a movie lose time to a published run and be faster overall and get published, and in what cases would this cause a movie to be rejected? Is the rule above only referring to cases where a runner loses a grotesque amount of time (where the gameplay looks noticeably sloppy)? Or is the rule itself obsolete?
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
Also, to Technickle, I have something I would like to say about this new submission.
Since there is already a published movie for this category on the website, the standard that new submissions for this category are held to is that they should include all of the time save/speed tricks that the previous TAS included, plus whatever new optimizations were found. To this end, TASVideos rules specify that if a new movie is submitted that obsoletes an old movie, it can't lose time over the original TAS for no reason in certain areas and then just be faster overall due to a few route or trick differences and get published. Rather, it should be as fast or faster throughout all the sections of the game (provided that the new route/strategies don't affect the other portions of the game in a way that slows them down), and then it should incorporate the new timesave. Otherwise, it is likely to be rejected.
I know it's tempting when you find a large timesave in a 2+ hour long TAS to just submit a new TAS with a few improvements that save a couple of minutes, and ignore the TAS being slower than the published TAS in other parts. However, if you want to be able to obsolete a published movie, you need to be able to complete the TAS faster through all sections, even if it does make the TAS more tedious to make.
In the long run, remaking TASes like this is good for you as a whole, as it improves your abilities as a TASer to TAS future games as well, so I would few it more as an opportunity than an inconvenience if I were you.
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
To play devil's advocate to the comments above, I'll just say this.
While Technickle hasn't said why he waited so long to submit this improved TAS, I'm assuming the story goes like this: he initially created his first Top Gear TAS in mid-2018, which was rejected. Then, he submitted a few more TASes for the game in different categories, which were also rejected. Then, after 5 failed submissions, including one which was for a 2 hour TAS, he finally got his TAS accepted. However, due to concerns over his Top Gear 2 TAS that was published last week being suboptimal, and due to the length of the TAS, it took almost 3 months for it to actually get published from the time it was originally submitted.
Technickle probably noticed while checking for sub-optimality after receiving feedback for his TAS that he had about 4 minutes worth of time save that he could achieve if he re-did the TAS. However, since he had never had a TAS published on the site before, and there were already concerns about his current TAS not being up to standards, he decided to wait until his first TAS was published, and then a little while later he would create a new TAS to obsolete it once he was sure that he would get a movie published on the site. By the time months later when his TAS had actually been published, he got impatient waiting to make his next submission, and decided to just submit the improved TAS later on in the week.
Of course, this is all just speculation on my part. But since Technickle hasn't acknowledged what his motivation was for his actions, this seems like the most logical reason to me.
Having said that, in the comments you're making to Technickle, you're looking at the situation from the perspectives of an experienced publisher for the website and a senior judge for the website, who each have over 500 player points and lots of published movies on the site. You're not looking at this from the perspective of somebody with 0 accepted movies whose been trying for a year to get a single movie accepted.
Again, don't get me wrong, I'm not defending Technickle's actions, and I understand that you're annoyed with him for wasting your time both in terms of the time it took to judge his movie and the time it took to make and upload an encode for his TAS. I'm just saying that there's 2 sides to every story, and it's important to keep that in mind whenever judging someone's actions.
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
To be honest, I almost feel like there should be a rule that no one can submit a TAS of Super Mario Bros for the NES to be published on TASVideos unless either
A) They've already had a SMB1 TAS published on the site at some point
or...
B) They have at least 100 player points.
I could see some people saying that it's completely arbitrary and pointless to automatically delete submissions for one particular game for inexperienced players, but considering the massive number of submitted SMB1 runs by new TASers and the fact that virtually all of these TASes have been suboptimal and rejected, it wouldn't be the craziest idea in the world...
...Anyways, back to the discussion of this movie in particular. If I watched it, I probably would have voted no on entertainment, because this is intrinsically a pretty boring goal choice which is essentially a less interesting and less optimized version of the any% TAS of the game. However, I'm too lazy to download an SMB1 ROM and to play the movie file back in it, so I'm abstaining from voting until someone puts up a temporary encode.
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
Nice TAS.
I found it entertaining due to the impressive technical feat of corrupting the game's linked list of officers to point to a spot in memory you want to beat the game quickly. This gets a yes vote for me.
I do have 2 questions for you though:
1. Without this save glitch, roughly how long does the game take to complete?
2. How did you discover that the game's officers were being stored in a linked list, and how did you find where the linked list was in memory?
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
El TAS más largo que he visto es "Ocarina of Time No Doors All Dungeons." Enlace: http://tasvideos.org/3537M.html
Si quiere ver una lista de los videos TAS más largos, puede hacer clic en esto: http://tasvideos.org/MovieStatistics.html
Si tienes el tiempo, el video TAS más largo en TASVideos es "Toca Touring Car Championship Max Points," un video que es diez horas de duración. Un enlace para esto es aquí: http://tasvideos.org/3386M.html ¡Disfrute!
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
Spikestuff wrote:
Lobsterzelda wrote:
The gameplay doesn't really stand out significantly from non tool-assisted gameplay. The main reason why this is so much faster than the RTA record is that the game is a memory game, so the TAS can always select the right option without mistakes.
There's like there's a term for this. Ah, that's right it's "Superhuman".
We have other TASes accepted that are basically Superhuman TASes.
I'm not sure I understand what you're trying to say.
Experienced Forum User, Published Author, Skilled player
(1222)
Joined: 3/17/2019
Posts: 280
I don't believe this TAS should be published for a couple of reasons.
The gameplay doesn't really stand out significantly from non tool-assisted gameplay. The main reason why this is so much faster than the RTA record is that the game is a memory game, so the TAS can always select the right option without mistakes. Additionally, the TAS can immediately click on the next mii, but this isn't a very significant difference over an RTA run. It reminds me of a TAS of a bowling game where the main impressive thing about the run is that every throw is a strike.
In and of itself, this minigame also isn't particularly entertaining to watch, which would eliminate the possibility that this minigame could be accepted to moons tier.
I also wonder if there's any way to get more matches per section of the map then this TAS does. When the minigame loads, are all of the colors of each mii fixed after that point?
