Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Mothrayas wrote:
"It's faster than unassisted play" is not a sufficient explanation.
There you go:
I still remember the first time I've watched a TAS that uses the Wall Surfing technique. At that time I was still willing to help nahoc with the making of his first TAS. As one of the first things to do, I suggested him to look for existing TASes on YouTube... In a matter of seconds, I got him giving an "oh my god" follow by the video url of Matt Schepcar's unfinished TAS. I was completely clueless of what he was about, until I opened the video and I instantly understood, as my reaction was the same. It was so impressive that at first I thought that some kind of cheats were used, also because there was no mention about TASing in that video. With some Googling, we both found Matt Schepcar's blog, where there was a post about how to write a bot... for TASing purposes. Still unbelieving, I've taken the laziest route possible: I've tried replicating the TAS of that first level by visually analyzing the video frame-by-frame. And I succeeded. But there was no point in copying without understanding what was going on... so I've tried to replicate the Wall Surfing trick from scratch. And I failed. All I could do was to note that the trick couldn't be executed without a bot. Seeing there was some serious work to do in order to achieve actual results, I gave up with this game and left the job to nahoc and E-Sh4rk. It has not been hard for them to figure how to make the appropriate bot, as they're both programmers. The only job left was to write a Lua script and explain them the Movie Rules thoroughly. On their own, nahoc and E-Sh4rk, did great with the rest. Not only they mastered botting of this game, but they even managed to beat Matt Schepcar's times! But that's not all. They also introduced the Out of Bounds techs, that allow to take advantage of something that we could describe as parallel universes...
The final product turned out even better than what I imagined. The character moves so fast that you can barely follow what's going on... actually, the first time anyone witnesses it, has a hard time just at realizing that the level has been already beaten. Everything happens so fast that the voice saying "start!" and "goal!" fills more time than actual gameplay itself. It's also nice to hear the continuous noises of the walls getting hit, as anomalous and erratic sounds are often a characteristic of advanced TASing. And it gets better. Later levels will end before the goal zone could even get in sight, thanks to sophisticated OoB trickery. It's not just faster than RTA, it's MUCH faster. To the point that even people used to watch TASes could not expect it. Other TASes that achieve similar results exist, but I didn't seem them appearing during this year. For these reasons, I deem [3933] GBA Kuru Kuru Kururin by mohoc & E-Sh4rk in 06:11.59 to be a movie applicable for this nomination.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Magician Lord is just too much obscure, on multiple levels. It's not just the game itself being little known, it's also the gameplay visually appearing too much simplified by the high levels of optimization, so it can't serve as a good showcase for newcomers. For example, the bosses die just too quickly, before anything could be observed.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I'm also against PAL usage unless it features interesting game mechanics compared to NTSC. Stuff like cutscenes and dialogue length can't be considered as gameplay differences at all, as they are completely unaffected by the player.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
TASVideoAgent wrote:
Game version: 1.0 (Note: The database incorrectly labels this version as "Battle Kid - Fortress of Peril Demo1 by Sivik (PD)")
What evidences are there that point out that this isn't a demo version?
Edit: nevermind, I solved it myself. From the relative Wikipedia page:
Wikipedia wrote:
There have been three U.S. releases of Battle Kid: Fortress of Peril. The first one was a limited edition run of 33 copies for early pre-orders, game play testers and production staff, with each copy being uniquely numbered. February 2010's version 1.000 was the second, main, and currently most common release. A third release of version 1.100 was released on July 2, 2010.
The main differences between the 1.000 and 1.100 releases of Battle Kid was the fixing of a glitch discovered in the 1.000 version of the game. This allowed the player to become invincible, therefore being able to beat the game with ease. Also fixed was a sound effect and the letters "D" and "V" of the game's font, which were sometimes confused with "O" and "U" respectively. Exclusive to the 1.100 release as well is an extra level originally showcased at the 2010 ScrewAttack Gaming Convention which can be accessed by typing "SGCLEVEL" into the password screen.
Edit 2: Additionally, the game version used for this submission features the developer logo right after booting the game, just like like for version 1.100, which highly suggests that it was used as a release version as well. Differently, any other ROM found are either non-release versions or unofficial hacks.
That's enough evidence for me.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I wonder if the publications for this game should get a different labelling. As of now, the fastest completion has blank label, while the other branch is labelled as "deathless". I think that the current situation has started due to a misunderstanding, as the fastest completion movie was initially thought to be using a built-in warp mechanic, while instead it ultimately turned out to be exploiting a programming overlook. From the judgment note of that submission:
feos wrote:
This run uses a technique that looked like a debug combination at the first glance - being able to advance the level number after Game Over - but it turns out to be an actual game bug. The manual says that after Game Over you can revisit levels that you've beaten. This is how it works.
After the Game Over screen, you go to the first room of the level you've died in. You press Up or Down, and then B to confirm your choice and enter the room you've chosen. If you press Up after you've reached the last room you're supposed to enter, you get back to 1-1, and vice versa. The glitch here is that if you press Up at the same frame as B, you get access to second room of the level you've died in. And if you press Down in that case, you go back to the title screen.
So that's an example of an actual game bug, erroneous assumption made by the developer, or simply an oversight, doesn't matter how you call it. This run is legitimate and it also got great feedback, accepting to Moons.
The policy for branch labelling is to leave it blank for goals that feature the most common way to play a game, while instead giving an accurate label for more exotic goals. So in my opinion, the fastest completion movie should be labelled as "warp glitch", while making the deathless movie without label.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
By the way, FilipeTales agreed to make a new submission that includes the Extra levels, so the current submission should be cancelled soon. If you know some tricks for other levels, please write them in a PM to me or to FilipeTales.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I didn't know about that glitch, that's interesting. Your solution saved 6 frames in 1-10, well done! I will look into 5-10 later.
Edit: ok 5-10 is 54 frames faster now... wow.
Even if you didn't edit the movie file yourself, I think in this case you should be added, because you basically provided the solutions themselves. Do you agree to be added as an author for this movie?
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
FilipeTales wrote:
FilipeTales wrote:
every failed attempts i reverted-saved as different file.
I used this expression because no one would believe me i was able to make a TAS without any rerecords, so to make everyone believe me i used this expression to delete the failed attempt which rerecorded and start over again from beginning without rerecords.
Even if the Puzzle mode of Tetris Attack is very simple, it still requires to make some precise inputs in order to optimize the outcome, and your movie even featured some frame-perfect tricks, which doesn't have anything to do with casual play, so it doesn't really matter how many years you did spend playing this game until now, you still needed to, at the very least, performing extensive trial and error with frame advance, which basically and practically means to go back and improve your previous attempts. That's because TASing itself consists in improving the previous attempts. If a movie really required 1 attempt in order to be made, that would mean that the game chosen was too much simple for a meaningful TAS, which is not the case for this game, as I confirmed firsthand while judging this submission.
Now, you may wonder how did some people make TASes with so few rerecords. The answer is very simple: those rerecord counts are wrong. For a reason or another, something have gone wrong with their files, so that the resulting rerecords value ended up being unfaithful. For example, take a look the submission that seems to feature the lowest rerecord count ever: #5470: Winslinator's INTV Advanced Dungeons & Dragons: Cloudy Mountain in 00:11.62. From its submission text, we can read this:
Submission #5470 wrote:
Speaking of which, the number of re-records should be 4117, as the movie I originally made for Bizhawk 1.12.0 had that many re-records (it had to be redone as it refused to play in v1.12.1).
As I already said, TASing itself consists in making as many attempts as possible, trying every possibility, fixing mistakes, and improving your previous strategies. The rerecord count exists in order to keep track of how many efforts were put into the making of each movie, ideally speaking.
However, rerecord count is in no way an accurate measure of how optimized a movie actually is, not to mention that the rerecords value could be faked altogether.
And if you think about it, the rerecord count is just a number, which doesn't tell much about a movie. Instead, if you want to know if a movie is good, you should try to improve it yourself, that's the only way to know for sure how much good it is.
So, reassuming:
Rerecords count should be high.
Rerecords count can't be trusted.
Rerecords count is not very important.
By the way, it seems to me that you're doing this 0-rerecords-thing just because you want to place yourself at the top of this page: FewestRerecords
You see, the fact is that this page it's just a system page that was created for statistic reasons, in order to keep track of some very minor aspects of the site data. It's not a podium pedestal, it's not an awards room, it's not a hall of fame. There is no merit or honor into getting a movie into that list.
With that said, if you still hope to place your future TASes into a honorable place, you should instead try to do something like winning a award, or making a Stars movie, or winning a contest, or improving a very old movie or a very optimized movie.
And if you think that you can't do any of that, don't worry: submitting acceptable TASes is already enough for being considered a worthy community member. Just stop it with wanting to make a low rerecord movie. Please.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
GJTASer2018 wrote:
Vynneve is only 2nd on NMG though, and the one run ahead of him on the board is faster than this submission (about 1:49:12). This is not TAS timing because Speedrun.com starts their clock when you first gain control of the prince, however it's hard to tell whether that run would actually be faster adjusting for that because they don't start from power-on.
Actually, it seems to me that it's faster mostly because of the lack of cinematics. The audio commentary for this TAS mentioned something about deleting the video folder for speedrunning, didn't they? 🙄
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Now that I think about, what about figuring out a possible full-completion definition for this game? That could be a better way to showcase more that the game has to offer. Something like "maximum score" or "all parts unlocked". Note that full-completion can also be accepted for Vault.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Truncated wrote:
An alternative category might be best ending (no commandos or hostages injured).
Nice idea. Seeing the final message telling about injuries left me with a dissatisfaction aftertaste. Could it be considered as full completion for this game? And what about maximum score?
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Also note that the game features long unskippable cinematics.
In any case, please focus more on what can be seen from watching the movie. (btw I still didn't fully watch it myself, it's long)
Edit: ok, I finished watching. I had the go in three rounds, but I didn't get bored at any point of the movie. It was good enough for me, despite the deliberate abstention from glitching too hard.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Zinfidel wrote:
What does everyone (anyone?) think about what would constitute an entertaining run of this game?
I think that it would be a good idea to follow what the speedrunning community does: "no fail missions/no OoB". However I'm not completely sure about the "no OoB" part. Why do they forbid OoBs? Are these too much overpowered?
Regarding the playaround idea, I think it would be harder to make and less likely to be entertaining, so be careful.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
FilipeTales wrote:
The only optimal 0 rerecords TAS is Tetris Attack which contains stages that helps TAS to be optimal and without rerecords, since i played this game every day since 2012, i'm pretty sure it's going to be optimal as possible!
Why willing to make a 0 rerecords TAS? Look, there is really no merit in doing such a thing. Rerecord count is just a triviality, so it doesn't affect the judging process.
On the other hand, we recommend you to try to improve your TAS as many times as possible, before submitting it. The art of TASing is about trying all possibilities, and improving your movie repeatedly, until you can't find more improvements.
However... If you're still insisting to make a 0 rerecord TAS, please upload it to userfiles first. Then, a judge will watch it and tell you if you should make a submission or not.
