Posts for ThunderAxe31


Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
The movie desyncs to me. I get this: Not sure why this is happening though.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Thanks for making a new run, your efforts are highly appreciated! Also, you've much improved since your first run. I'm looking forward to watch your next movie.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
MUGG wrote:
In level 1, you neglected to keep going over the edge at the end?
Because that route caused way more lag.
MUGG wrote:
In level 3, isn't it beneficial to walk on the carts so they start moving, giving you a boost? I don't remember if it works like that.
Indeed... Except that after these carts start moving, the game will lag like hell.
MUGG wrote:
You bumped into a wall after bouncing off a penguin. Can't that be prevented?
No. It can't be prevented. And a different route would cause much more lag.
MUGG wrote:
I think some lag can be reduced if you opt to take the time to kill enemies.
I thought that, too... but no. From my experience, I can say that common sense doesn't work out with bootlegs.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
[1639] SGB Pocket Monsters: Green "warp glitch" by p4wn3r in 04:24.37 to Vault. Rating is 6.6 and the obsoleted movie, which is only 10 frames different, is already in Vault.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Here are the suggestion that currently come to mind: [1639] SGB Pocket Monsters: Green "warp glitch" by p4wn3r in 04:24.37 [941] SGB Pocket Monsters: Green "warp glitch" by primorial_soup in 04:24.53 (obsoleted) These need the label "warp glitch", for the same reason I noted above. Fixed.
Since all Pokémon runs have been covered at this point, let's go to Mario now. [1895] NES Super Mario Bros. "warpless, walkathon" by Mars608 in 25:30.05 Should be labelled just "walkathon"; the warpless part is already omitted in the "maximum conins" and "all items" runs. Didn't need to be fixed. [3370] DS Super Mario 64 DS by Sharkey91, Really_Tall & ALAKTORN in 08:12.93 (also 4 obsoleted movies) These perform major skipping, however I'm not sure how it should be labelled. Maybe something like "infinite stairs skip"? [2210] FDS Super Mario Bros. 2 "warps, Luigi" by HappyLee & KFCMARIO in 08:13.77 (also 4 obsoleted movies) Should be "warps, Luigi" or "Luigi, warps". Didn't need to be fixed. [3348] FDS Super Mario Bros. 2 "warps, Mario" by HappyLee in 08:04.83 (also 12 obsoleted movies) Should be "warps". Didn't need to be fixed. [3622] DS Newer Super Mario Bros. DS "warps" by Soig in 26:04.37 Should be "warps". Didn't need to be fixed. [2810] GBC Super Mario Bros. Deluxe "warps" by negative seven in 04:55.99 Should be "warps". Didn't need to be fixed. [3643] Wii New Super Mario Bros. Wii "warps" by Monster in 25:02.10 Should be "warps". Didn't need to be fixed. Also, there are many Super Mario World movies, including hacks, that lack an appropriate label, but we also have to decide if going for the "warps"-"warpless" naming or "low exits"-"max exits" naming
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I apologize if I was too much pedantic. I consider this movie category very important and in need to be carefully managed with the appropriate rules in order to shine to the fullest. I thank everyone for the contribute to the discussion, because it helped me in being able to appreciate the ruleset as is, to the point to tempting me to join the challenge and making a faster TAS. I just want to point out one last thing: while movies done on Blue and Red can obsolete each other in any case due to their similarity, Yellow Version instead features many little differences, and thus I think this branch should be kept separated from them. Also, future submissions should also use the English version, unless there is the introduction of major technical and entertaining meritis. However, this can't be said for sure unless we get an actual new submission to see and judge, so don't take my words as a rule.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Sorry for the late reply, feos. Indeed, while the ruleset for this movie fits nicely in the "no memory corruption" description, it must be noted that it wouldn't point out to anything in particular about the uniqueness of this movie. Also, the goal for this movie reflects the way the game is normally played. So in the end I agree that labelless is the best solution.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Zowayix wrote:
Brock Through Walls is memory corruption (game reads a ton of unrelated values as cutscene walk data). The precursor, Brock skip, is not.
However Brock skip in unnecessary for Glitchless route.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
feos wrote:
What I'm getting at, this submission would be a blank branch by our standards. Note that blank branch not equal to our definition of any%. We do not explicitly highlight any% in any way, because it doesn't help with organizing the branches. So when labeling runs, we rely on statistics instead. If a run sets unique goals that other runs don't, we put them in a label if such goals are actually rare overall. "ACE" is a rare category, hence we call it out. "No ACE" is not a rare category, it is overwhelmingly common through the whole speedrunning community. Hence in the case of this submission, we don't need to say "no ACE" in the branch: it's so common that it's implied. Same about SRAM or memory corruption. They are rare, avoiding them is common, no branch label needed.
While there are no doubts that "SRAM glitch" is a very rare category, there isn't much difference in spread if we compare the RTA branches "Any% Glitchless" and "Any% No Save Corruption". Take a look at the current amount of RTA runs on speedrun.com: Pokémon Red/Blue ( https://www.speedrun.com/pkmnredblue ) (ENG) "Any% Glitchless" has 288 runs; (ENG) "Any% No Save Corruption" has 197 runs; all other branches have less than 30 runs for each. Pokémon Yellow ( https://www.speedrun.com/pkmnyellow ) (ENG) "Any% Glitchless" has 64 runs; (ENG) "Any% No Save Corruption" has 40 runs; all other branches have 4-6 runs for each. I'm not sure if this difference is sufficent to require a blank label. Also, I'm taking in account the fact that there are numerous possible goals for Pokémon games, even some not present on speedrun.com nor TASVideos, so an appropriate label would also help to avoid confusion for this reason.
feos wrote:
However, this doesn't mean that the actual goals of this movie are arbitrary. In no way they are. They are quite common, and they were what the overwhelming majority of the runs were for games where no memory or SRAM corruption techniques were known. But then they were discovered. Does that mean the whole branch that avoids them is effectively dead? Of course not. As long as people keep competing in it, and as long as it features unique TAS content, it's alive.
I agree that the goals for this movie are not arbitrary, differently from what I initially thought. However, I think that not only because of how common are the runs made under the relative ruleset, but also because the ruleset itself happens to follow a clear logic: no memory corruption (thanks again to Zowayix). In my opinion, both reasons are necessary in order to warrant the new branch, and thankfully both are met. Additionally, I find there are also entertaining and technical merits for this run. So this is my idea for the label names: TASes that use the equivalent of the RTA ruleset for "Any% No Save Corruption" should be published with label "no SRAM glitch" or "no save/reset"; TASes that use the equivalent of the RTA ruleset for "Any% Glitchless" should be published with label "no memory corruption".
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I'll elaborate about my current thoughts. My major concern was finding a clear and absolute ruleset that would result in making a movie that features all mandatory game events. For this reason, I've brought the argument that an exploit should be considered as a glitch if it allows to break event sequence, i.e. to save time by skipping a game event intended to be mandatory. However, by following this logic we could argue that using Dig from inside a building would be an uninteded use, and then we could also argue that getting red bar in order to cut sound effects would be an unintended use, and so on. Not only this would be very arbitrary, but it would also remove much of the challenge in optimization and would make the TASes much less interesting. On the other hand, banning only glitches that can lead to ACE possibilities is a much clearer definition. In fact, as Zowayix pointed out, this simply results in avoiding memory corruption, which is also already recognized on the site as a movie category. Lastly, this makes a clear cut from the No Save Corruption branch, which is also important. By the way, while getting instant text via the Bike Shop could be considered as a form of memory corruption, it must be noted that menuing glitches are pretty common in Pokémon Gen I games and are often unavoidable, like the 1/256 misses.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
TiKevin83 wrote:
I disagree with the assumption of difficulty in defining glitch. The "glitchless" category used in the RTA community is based on a very straightforward definition of "no arbitrary code execution."
Yes, this definition is more clear.
TiKevin83 wrote:
On the other hand, trying to follow the intended sequence of the game would also necessitate beating Erika before Koga before Sabrina before Blaine, which has never been suggested as a requirement for glitchless Pokemon runs before by anyone, classic or no classic.
There is nothing in the game design that is attempting to force the player to beat these gym leaders in this order, but rather a couple of hints: the order of the medals in the player status screen and the different level of the gym leader's Pokémon. So I don't think this can be considered as sequence breaking, with exceptions like beating Blaine before Koga, due to HMs.
TiKevin83 wrote:
I do understand the desire to not publish competing classic and non-classic runs that have very similar category definitions if they both existed, but for the moment there is no classic category TAS.
Yes, but we need to carefully evaluate all aspects beforehand. We can't publish a movie just because there is free space for new branches. At this point I'm beginning to be in favor of accepting this movie. However, I still need to evaluate more data and hear more opinions.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Memory wrote:
Not specific enough tbh. In some situations it might be a glitch and other situations it might not. Depends on the exact kind of movement required to pass through.
The point is that I'm supposing a situation in which there is an obstacle with the same purpose as the gap you have hypothesized, that can be bypassed as easily as doing a tight jump as you hypothesized, but that it would look impressive to perform from the point of view of a person not familiar with the game.
Memory wrote:
Can you really draw an absolute line through intended sequence vs unintended sequence? For plenty of games, the intended sequence itself is unclear. Sure, this particular instance seems pretty clear but there are many games where it is not, and developer intent is largely ignored anyways.
Indeed, with most games this is impossible, due to the complexity of game design (amount of possible routes, amount of level objects, etc.) and even if we can draw the line, in most cases the resulting run doesn't result enough different due to relatively small amount of exploit used in the fastest run. However, with Pokémon games we can, because the game is both very linear and has many known exploits that can be put in use. This allows to beat the game by using different approaches and not having redaundant branches.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Memory wrote:
Again, I would describe this as a sequence break rather than a glitch. I'll give you an example in another genre. If a developer intends you to not be able to make it across a gap, but you can still do it with a tight jump, is that a glitch? I wouldn't think so.
But what if instead of a gap the developer decided to put a wall, and you could go beyond this wall by using some frame-perfect or subpixel-perfect movement? The obstacle is different and the way to bypass it is different, but the purpose of the obstacle and the result of the action performed by the player would be the same, but would you also consider this as not being a glitch? I see that drawing the line between "glitch" and "non-glitch" is impossible; but on the other hand we can instead draw an absolute line between "intended sequence" and "unintended sequence", and that's why I insist that we should use this yardstick instead. Then we can name the branch whatever is more appropriate, glitchless or not.
Memory wrote:
We have the co-op diploma Red/Blue branch that maximizes game content (including going through the Rocket Hideout), this should not be an issue.
The game(s) and the goal for that run are different, so we can't really make a comparison.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Memory wrote:
I don't see what intended methods of game progression has to do with whether or not something is a glitch.
Well, the fact itself that something intended to happen did not happen, can in itself be described as a glitch. In this case, the player is supposed to go in a certain place and clear some fights in order to obtain the intended item required for progressing beyond Marowak's ghost.
Memory wrote:
The fact that this is a sequence break does not seem relevant to the branch goal.
The purpose of a branch that forgoe major skips is to show off more game content, or different gameplay content, compared to other branches. Skipping part of the intended events sequence simply defeates such purpose.
Memory wrote:
Would using a pokedoll on Marowak after using silph scope be allowed? According to the glitchless classic rules it would not be allowed, but it has nothing to do with event sequence in that case, which was your primary argument.
You can't encounter Marowak's ghost again after having used either Silph Scope or Poké Doll on it. The player necessarily have to decide which method to use in order to advance.
Zowayix wrote:
My interpretation of RBY "glitchless" has always been "no memory corruption", since that conveniently handles unavoidable glitches such as 1/256 miss and Viridian Forest no-encounters oversight. Then using the Poke Doll on Marowak and the (fixed in Yellow) Bike Shop keep-instant-text oversight don't count as memory corruption.
We really need to avoid branches that are too much similar to each other, and thus we can't publish both Glitchless and Glitchless Classic. And the RTA ruleset of the latter branch looks much more appropriate in my opinion, because it's more inclusive of game content.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
While Mewtwo is certainly part of the main story, it's not required in order to advance towards the Hall of Fame and the credits roll. Otherwise, we should also argue that the Legendary Birds are also major events, and so on adding extra goals. The point is that clearing the Rocket Hideout is supposed to be mandatory in order to advance from the point of view of a casual gameplay. Also, since you've mentioned Pokémon Origins: in the second episode, Silph Scope is used for Marowak's ghost, and not a Poké Doll. :P So I still think that the ruleset of Glitchless Classic is more coherent. About the red bar, I see that not abusing red bar would make it easy to beat this TAS, but I don't see why we should compare the timing of TASes that make use of it with TASes that don't.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Post subject: Re: look, i was trying my best..
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
ItsameMario203 wrote:
i know that my tas isnt good enough. its just my first tas.
Please don't submit a TAS if you know that isn't good enough. Instead, upload your movie to userfiles space: http://tasvideos.org/userfiles/my#uploadfile
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I think that using Poké Doll to skip Marowak's ghost goes aganist the purpose of the branch of beating all mandatory opponents since it arbitrarily skips the whole Rocket Hideout, not to mention that it can be considered as an actual glitch since this usage is clearly possible due to a development oversight. In fact, this is not allowed for the Any% Glitchless (Classic) RTA branch on speedrun.com. About the entertaining side, I have to say that I disliked the choice of exploiting the red bar due to its annoying noise. Also, I don't like the idea of getting hit just for the sake of saving time, because it breaks the flow. While I appreciate all the efforts put into optimizing everything optimally, included abuse of red bar, I'm not much entertained and I voted Meh. By the way, why did you set Earphone 3 option? There is no mention of it in the submission text.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Indeed, technically speaking, this run is not beating any existing record, nor showing off any different content from the "all levels" runs, whose go for the full completion. However, since the restriction of avoiding redaundant content is only applied for non-Vaultable goals, I'm assuming that this submission could be accepted.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Yes vote! The tricks were cool, and I especially appreciated the luck manipulation at 1:50!
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
You need to go on GB > Settings > Enable BIOS > True Then reboot the core and play the .bk2 file.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Post subject: Back again
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Sorry for having paused the project again... So, continuing from when I left: I searched thoroughly for a way to trigger in stage 10 that glitch. Despite having now a better understanding of how the glitch works, I was unable to find any way to trigger it there... I'm sorry. But let's go in detail. When you land on a box, that box will break and Crash will rebounce on it. But if before landing on it you broke one or more boxes with a spin during that jump, Crash won't rebounce on the box, and the box will not break. At this point Crash will also be able to walk on that box without breaking it. I've come to the conclusion that this was a lazy workaround aganist the glitch. In fact, if there are at least 3 boxes stacked and you break the middle one(s), the bottom box will make Crash rebounce even if Crash is still spinning. But since the game does not expect Crash to be able to rebounce on *anything* while spinning, then the game freaks out a resets Crash's maximum falling speed to 0, causing him to get stuck in air almost instantly. It looks like for some reason, while a box is falling over another one, the bottom one temporarily loses the feature of being able to prevent Crash from rebouncing during a spin. So triggering this glitch is going to be impossible with less than 3 boxes stacked... Unless we find a different way to force Crash to rebounce while spinning. Note: when you land on an enemy while spinning, that enemy will be killed by spin instead. So, I had to simply re-TAS the whole Stage 10 from scratch. In the process, I managed to save 95 frames. The new WIP file desyncs in level 14. Edit: new WIP file: resynced up to level 16 complete on BizHawk 2.2.1, 30 frames saved in stage 15. Movie #45784678238323512
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
ViGadeomes wrote:
Hello, Is it normal a submitter can vote for his own movie? One vote can change everything in the mind of judges to say if a movie go on Vault or on Moons x). I don't log in automatically if it's a bug.
No, one single vote can't change much, especially if it's unexplained. Also, submission votes are not the only factor for a judge decision.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Post subject: Just perfect.
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
My favorite kind of TAS goals are the ones that allow to show off as much game content and TASing skills as possible, and I don't think there could be a better goal for doing it in SMB. Thank you so much for this piece of art.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Post subject: speed display script!
Experienced Forum User, Judge, Published Author, Skilled player (1278)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Today I wanted to write a script for showing the speed values on screen. Have a screenshot: Click here to download. Edit: new version below. It basically shows position and speed value for the current and last 20 frames. The lines scroll automatically each frame, from top to bottom. It automatically resets when you load a savestate. The speed colors change in order to reflect the known target speed amounts. Red is for when standing still, white is average, green is good (maximum known speed), and blue is used if by any change you manage to break the known speed limit by whatever means. Note that the speed value can randomly vary slightly due to a bug in the game engine, not because of this script. For example, when you jump, the speed for the first frame is 1.8, then gets to 1.84, then 1.81, and then it finally starts lowering. This is why I don't trust speed values found in the game memory. Edit: turns out it was actually my script's fault: the initial speed for jumping is 1.87. I've fixed the bug and you can download the v1.1 script here. Still, it's the game's fault if the horizontal speed does occasionally drop by 0.01 when walking from left to right. Don't trust speed values found in game memory. And I've also created a script for calculating the average speed. You can download it here. Edit 2: new version below. Aesthetically speaking it's the same thing, even for the colors. The only difference is that the speed amounts are calculated as average amount of all the speeds from script launch onwards. If you want to start a new measurement, you must restart script execution. Starting the script execution in the right frame is vital in order to get the exact average. Also, in order to be able to do proper comparisons, make sure to keep track of the amount of frames; for this reason I added a frame counter that starts from script launch. Edit 2: fixed in the average speed script a nasty bug caused due to bad handling of exponential notation. You can download v1.2 here.
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"