Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
ThunderAxe31 wrote:
However, I think that it should allow the usage of this feature for Vault, since I think that there is a sensible way to draw a borderline. Usually, a password system is a secondary feature of a game, for which its purpose is to give an alternative way to play the game (may it an advantage, a disadvantage, or a completely cosmetic option); instead, for Monster Racher 1/2 it's the main feature proposed by the game, as it's suggested at the beginning of the gameplay and it's strongly encouraged on the back of the official game cover, as the concept of spawning monsters from disks is the main theme of the franchise.
Thinking it again, I've concluded that this logic doesn't work. There is another game which features a password-like system as a main game mechanic: Ketai Denjuu Telefang. However, we have a precedent of a rejected submission for this game, due to having used a secret password for unlocking a strong monster: #5882: jlun2's GBA Keitai Denjuu Telefang 2: Speed Version "secret Denjuu" in 1:20:32.53
With Monster Rancher 1/2, the situation is slightly different due to the fact that its monster-generating system is mainly based on an algorithm, with few exceptions.
Also, Monster Rancher 1/2 is much more balanced when it comes to potential unfair advantages, as it requires the player to complete certain portion of the game before being allowed to generate special monsters from CD:
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
brunovalads wrote:
ThunderAxe31 wrote:
I've just found a way to pass through closed doors.
Oh nice, so that will make the run "fruit-less"?
Yep. Edit: No, there are two doors that I couldn't glitch through. Also, in the first level I get a banana just because it's on the way. Edit 2: The second door clip turned out to be 2 frames slower.......
I was hoping that this glitch could have saved a lot of time, but instead is gonna save few seconds... But on the other hand, I've found another glitch. I'm now looking for useful applications... Edit: Great, the new glitch as well saved some time... 1 frame. You can see it at frame 24670.
Edit 2: Now I'm done with the improvements, I apologize for the inconvenient.
By the way, have here a cam hack script. Note that it causes desync if used before approaching room exits.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I've just found a way to pass through closed doors. This is going to save some seconds of gameplay at least, in the best case scenario it could save even a minute overall. So please, set this submission to delayed until I include the new improvements.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Hello there! I and mohoc are attempting to make a new TAS of this game, as there are a ton of possible improvements thanks to the OoB discovered by mohoc and the wall collisions abuse discovered by another person on YouTube:
Link to video
I've managed to reproduce manually the inputs for the first level: User movie #49439186419057236
However, I wonder if we could figure a way to beat these records, if possible... In any case, I'm trying to contact the author for proposing a collaboration.
Edit: Oh, almost forgot to mention. I've written some lua scripts for this game:
OoB viewer v1.1 is a script that displays a graphical representation of map while you're outside of the screen. It's a bit cpu-hungry, so emulation is gonna slow down a lot on older computers. There is also a VBA-rr version of this script, in case you have troubles using BizHawk.
Speed hud v1.2 shows the player position and current speed in the lower left of the screen. Digits before the decimal point are the exact pixel amount, while digits after the decimal point are a representation of subpixels. The script does also show an arrow in the center of the screen, which will point the direction that the player is going to during the last frame, and will change length and colour according to the speed.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I agree that using the CD feature of Monster Rancher 1/2 is like using secret passwords, as it's based on external informations (where "informations" may both refer to the data in the CD or to the knowledge itself that the specific CD gives the desired monster).
However, I think that it should allow the usage of this feature for Vault, since I think that there is a sensible way to draw a borderline. Usually, a password system is a secondary feature of a game, for which its purpose is to give an alternative way to play the game (may it an advantage, a disadvantage, or a completely cosmetic option); instead, for Monster Racher 1/2 it's the main feature proposed by the game, as it's suggested at the beginning of the gameplay and it's strongly encouraged on the back of the official game cover, as the concept of spawning monsters from disks is the main theme of the franchise.
I also have to note that the feature is able to work with any CD in existence, as the monster and its stats are decided through an algorithm, with few exceptions (as mentioned by Omnigamer).
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Ok, so now the text in the full-completion rules has been made clearer: arbitrary definitions of full-completion are applicable with a community agreement.
I think there is a downside in this. When full-completion requirements definied by the games are considered boring and thus superseded by the full-completion requirements defined by the community, this would mean that the former would not be acceptable for Vault, while the latter would be acceptable anyway thanks of the entertaining (otherwise why would the community pick different requirements than the ones proposed by the game?)
Let's make an example. The speedrunning community for Metroid Prime Hunters has three RTA branches:
"any%" : the usual fastest completion branch,
"100%" : the full-completion defined in-game,
"all items" : the full-completion defined by the community.
The "100%" goal requires the player to get all upgrades and scan every type of item drop at least once. Since scanning is considered annoying, the community did come with the "all items" branch, which consists in just getting all upgrades.
Now, since our rules allow for full-completion requirements defined by the community to supersede the requirements defined by the game, this means that a "100%" TAS would not be applicable for Vault, and thus would have no chance of getting published since it would be likely to be boring; instead an "all items" TAS would be acceptable even if it wasn't considered as full-completion, due to the fact that it could still be acceptable for Moons anyway thanks to its high entertaining value.
Vault is often the only hope for many TASes to be accepted, and allowing community-based definitions of full-completion does effectively result in replacing Vaultable movies with movies that would be acceptable anyway as a separate branch, thanks to their higher entertaining and differing contents from the in-game definitions that has been disregarded because are considered boring.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I think that "100%" should be only used for games that do clearly display a 100% value, because it refers only to a specific form of full completion, while for many games it's instead "best ending" or "maximum points". Using "100%" outside of its specific context is just like labelling fastest completion as "any%": a label that indicates a vague and moot concept that is forced regardless of how much fitting it is for that specific game.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
feos wrote:
To finally answer to your question, since avoiding things like a "game end glitch" for games where it's possible is not vaultable, we can't use such a flag for them. And that is our borderline that is inherent to your definition of any%, which is in turn inherent to our definition of the Vault tier (which I'd still love to be renamed to Coins).
This submission was unacceplable due to the triviality of its execution.
The problem is that we already have a published movie that beats the game without making use of the glitch from the above mentioned submission. Should the current publication be flagged as fastest completion or not?
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Mothrayas wrote:
This post just reaffirms my point, which is that you cannot know for sure if a set of completion criteria is full and has nothing left (unless the game explicitly says so), and therefore it makes no sense to judge full completion criteria by the game not telling you if anything is left hidden.
I didn't talked about cases in which the game isn't telling it, but rather cases where the game is telling and lying, which is a different matter.
The yardstick is based upon what the game proposes. Even if the actual full % could be wrong, the request of the game is still clear. A game may lie when it says that there are only 100% items to be found, but it's not the problem: the important fact is that it's telling you to get the items, regardless of how many additional ones are hidden around in secret easter eggs.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Mothrayas wrote:
ThunderAxe31 wrote:
Memory wrote:
This is not Pokemon. Pokemon only visibly tracks the Pokedex. This game VISIBLY tracks far more than defeating death. Full completion is to be determined on a per game basis rather than applying nonsensical precedents to cases that it does not match.
Spectral Interlude does indeed show how many upgrades you currently have, but it doesn't tell you in any way how many are left to be found, nor does it tell if you already found them all. This means that it's not a goal proposed by the game, but arbitrarily self-imposed by the player.
Many games actually do not tell you if you already found every item. Even Super Metroid does not give any special indicator for 100% completion. (And "100%" by itself is not a reliable indicator - see also examples of other games where percentage indicators go beyond "100").
I see you're referring to the fact that Super Metroid does display the percentage counter at the end of the credits, but it doesn't tell if that is corresponding to actual full completion, since we have many cases of games in which full completion is displayed in-game with a percentage above 100%.
The fact is that 100% does implicitly refer to the concept of full completion; it's basically a synonymous (though sometimes it could be less accurate than other wordings, like "all X levels" or "all victims saved"). In the case of Super Metroid, the game is simply happening to be minimalistic on their awarding message, by using the same X% function even if the player didn't reach full completion. And we can't disassemble every single game that shows a bare X% just to make sure that it's or isn't the maximum percentage intended by the developers, since we can't really discern with certain about their original intentions.
The only thing that we can do is trusting what the game is displaying, even if it may be a lie. If there could ever happen to stumble into a "100%" run which is not indicative of the true full completion progress, and assuming the remote chance that no one in the world has yet discovered about the existence of the remaining %, how could we ever avoid that? Disassembling every bit of the game? Contacting the original developers for asking them (and assuming they'll not lie as well)? What if someone at some point discovered a way to reach 106% completion that no one ever discovered before in Crash Bandicoot Warped, through intended game mechanics and without glitching the progress display? In the end there will always remain a chance that it could happen. The only thing we can do is going with the highest % equal or superior of 100 known to be the achievable, and obsolete a run in the case a higher legit % is disovered.
By the way, all this only applies to games that do display a bare X% value, without giving any additional awarding message or alternative cutscene, which is uncommon.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Memory wrote:
This is not Pokemon. Pokemon only visibly tracks the Pokedex. This game VISIBLY tracks far more than defeating death. Full completion is to be determined on a per game basis rather than applying nonsensical precedents to cases that it does not match.
Spectral Interlude does indeed show how many upgrades you currently have, but it doesn't tell you in any way how many are left to be found, nor does it tell if you already found them all. This means that it's not a goal proposed by the game, but arbitrarily self-imposed by the player.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Memory wrote:
A clear consensus is required on what constitutes full-completion.
Some games reward the player for something internally defined as full completion.
Sometime full completion requirements are explicitly mentioned in the game manual.
Conditions that are imposed unofficially by players and do not originate from fundamental game-play features are considered arbitrary and are not eligible
I'd quite argue that these conditions do originate from fundamental game-play features. The definition given basically includes every permanent upgrade to the character and to the game world. It does not explicitly state that requirements must be explicitly listed in the game itself, nor should it.
Then this movie would not count as full completion: [3687] GB Pokémon: Blue Version "Gotta Catch 'Em All!" by luckytyphlosion in 37:55.33
In the mentioned movie, the player catches all Pokémon species, which is considered as the sole full completion requirement, since the game keeps track and awards the player with an unique congratulation message for having filled all Pokédex entries.
However, if we disregard goals that are proposed by the game itself, we end up with a lot of other things that could be considered as permanent upgrades: The Key Items can be obtained only once, and can't be removed from the player's inventory in any way. Same can be said for the trainer fights: once a trainer is beaten you can't fight them again, which is indeed a permanent game progress. In my opinion these are examples of self-imposed arbitrary goals, since the game does not keep track or award the player in any way for having achieved them.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
brunovalads wrote:
ThunderAxe31 wrote:
Regarding Lua, do you have any ideas in particular?
Not quite, I'm not familiar with all the particularities of this game. When I play a platformer I usually think on player info (position, speed, timers), sprite/enemy info (position, speed, hitboxes, timers) and some general info (RNG, timers, level info), but it seems that your script does the job for this TAS. And yea, a map feature would be nice, after some deep research.
Sorry, I don't feel like adding much details right now, and the fact that I'm busy with other stuff doesn't help either. I've added my address list to the submission text, though.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
brunovalads wrote:
Do you intend to make a "100%" TAS of it someday?
Yes.
brunovalads wrote:
And did you make any documentation or Lua script of this game?
Not sure what you mean by documentation... Subtitles would be a bad idea, since you can't really watch this movie and read at the same time. But I guess I should list the zips in the submission text. Edit: added.
Regarding Lua, do you have any ideas in particular? The only thing that comes to mind would be to make a map display, but that would require much reverse-engeneering probably not be that useful in the end...
In any case, you can get the script I used for TASing: http://tasvideos.org/userfiles/info/49345614448199520
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
DrD2k9, in this post I brought up why I consider obtaining monsters with this feature to result in a non-Vaultable movie; I'd like to hear opinions and thoughts about it.
Since it's in the previous thread page, allow me to copy-paste:
ThunderAxe31 wrote:
The usage of arbitrary images introduces the same problem of secret passwords, codes, and input sequences: these rely on a knowledge external to the game and its official manual, and thus giving an unfair advantage to the player. So a movie that makes use of an arbitrary image should be allowed only if it introduces gameplay that results enough different from the Vaultable branches, as well as meeting the entertainment requirements of Moons tier.
Below I show the relative text in the Movie Rules page (bold mine):
We allow playing unlockable content using in-game passwords
Using in-game codes[1] or passwords at the start of a game is allowed if it makes the game harder or if it makes cosmetic changes to the game, as long as parts of the game are not skipped.
Using in-game codes or passwords at the start of a game to unlock a special game mode, character, level sets, or otherwise play the game in some unusual way is allowed. However movies of this nature are not considered to be one of the primary branches for the game.
Please tell me if you agree with my interpretation.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
feos wrote:
What if no game gives optimal outcome here? Having to brute-force this problem by trying them all doesn't feel like anything we want to approve.
There is no such thing as optimal outcome anyway. We had a lot of movies in the past that looked unbeatable, and despite that they got bested at some point, again and again. The only thing we can do is assess if a movie looks reasonably optimized for our standards.
This is especially true for games like Monster Rancher 1/2: obtaining a strong monster is not the only factor that determines the optimality of a run, since the game requires the TASer to do an immense amount of planning of all the possible choices and strategies, which also differ in relation to the chosen monster.
feos wrote:
When we design the perfect image by hand, it removes all the possible problems in my eyes.
I suppose you're referring to the optimality and legality problems.
For what concernes optimality: designing an image by hand requires reverse-engeneering of the game code, and I don't think that we should require the TASers to work to such extent for making an acceptable TAS. There are some games that provide an insane amount of possible routes, making it pratically impossible to know beforehand how much a movie is close (or far) to perfection. Even if we struggled to design a seemingly perfect image by hand, it could still be beaten by using a game image.
For what concernes legality, I already said that there shouldn't be problems if the image used is a PSX game as well. What's wrong with my point?
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I've tried replacing the secondary image used for this submission with a blank image... While I succeeded in unlocking the monster, I immediately encountered an irreversible desync. If anyone wants to try to resync the movie, you can get the hand-crafted image from here: https://www.dropbox.com/s/5zdf5jd98568d7q/MR2%20Pixie.zip?dl=1
From what I could see, the size of the track files determine the monster species, while the contents of the track files affect sync. I've tried padding the track files with 0x00 and then 0xFF bytes, but the result was the same. It's some specific file structure in the first track that makes the movie different... But no idea what.
feos wrote:
this particular game is known to have content that can only be unlocked if irrelevant images that don't belong to the game are used.
What do you mean with "irrelevant"? Since it's the game itself that supports and encourges it, any CD in existence is potentially able to be used for the purpose of the intended game mechanic. I understand that this looks ridiculous, but in the end Monster Rancher 1/2 is just making use of the hardware possibilities in a much more deep way than most PSX games do, and it's reproducible. It's not like it's asking the player to kick the console while playing or put the Memory Card in an oven.
feos wrote:
I believe that using arbitrary images even for cases like this one should not be allowed under any circumstances. Arbitrary means literally arbitrary: it can be an unreleased Demo disk of some Gore Grind band from Antarctica, and the movie may completely refuse to sync on anything else. Allowing arbitrary images means we invite such scenarios for all such cases. We can't limit this to "obtainable" disks either, because this definition is even weaker, and introduces actual legal problems.
I agree that we should forbid any copyrighted material, except for official PSX game images, because they are necessary for making movies in first place. We already allow that, so it would be contradictory to forbid usage of the specific image for being from an official PSX CD.
On the other hand, using non-copyrighted, hand-crafted, blank images would mean that accessing to them is easy and legal. This could be encouraged, but not required, since we can't forbid usage of PSX images without making movie creation impossible.
So this specific point shouldn't be a problem for this submission, since it's making use of good dumps of official PSX games only.
feos wrote:
In that sense, due to all the complexity of this case-by-case exception, I do not think that we should allow use of unintended images for Vault. And the definition of intended image is the same: either the game explicitly asks for some particular image, or the publisher encourages using it, like it happens with Sonic & Knuckles and Sonic 3 (or modern DLCs maybe).
I totally agree with this point. The usage of arbitrary images introduces the same problem of secret passwords, codes, and input sequences: these rely on a knowledge external to the game and its official manual, and thus giving an unfair advantage to the player. So a movie that makes use of an arbitrary image should be allowed only if it introduces gameplay that results enough different from the Vaultable branches, as well as meeting the entertainment requirements of Moons tier.
So, given my thoughts expressed above, I think that we could allow usage of both hand-crafted images and arbitrary PSX images, just for Monster Rancher 1/2, and just when the game explicitly prompts the player to.
I also have an alternative idea. Since some games are known to require emulation of additional hardware functionalities, like for example Kirby Tilt 'n' Tumble, it would make sense to implement in BizHawk the ability to generate a blank image on the fly, just for Monster Rancher 1/2, since these are the only games that make official use of this hardware possibility.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Since lately the discussion in this thread has derailed from the specific submitted movie to the game research in general, I wanted the last 18 posts moved into the relative game thread.
About this submission, I'm postponing the verdict until I'm sure that the game can be reasonably beaten faster by making use of the level warp glitch, as well as making sure that such hypothetical run would not break any rule.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Uhm, about 4 mounths since last post.
So, the current publication was made before that GB BIOS emulation was introduced, I wanted to try to improve it... And I failed, no matter what. I wish that someone else could improve it in my place...
I resynced the movie on BizHawk 2.3.2, download it from here: Movie #56195387747101377
Use it as a base if you want to improve the currently published movie, no need to give me credit.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Challenger wrote:
OK, the glitch is really good. I played the remaining of the game with the ThunderAxe's glitch strategy and you can still beat this game and see the credits without problems aside some glitched sprites.
I suppose this submission can be cancelled now.
Challenger wrote:
Great! However there is much hidden lag at the beginning of level 8. I hope to be able to adjust the glitch a little.
Challenger wrote:
EDIT: I also tested your lua script and for some reason, the lag counter resets if you load a state.
That's the intended behaviour I wanted for my script. The alternative was to let the count keeping growing even after loading savestates, which would have been confusing. Instead, by resetting it when loading a state, you're enabled to do state comparisons better. Tell me if you have a different preference.
Challenger wrote:
EDIT 2: We also need address for the timer overall: Minutes, seconds, and frames.
Right now I don't have time, I'll do it later today.
Note: I'm feel its speed have little problem, sometimes boost up, sometimes slow down
I think the speed is always normal, except in the level 8 warp movie I've uploaded: for some reason it did generate a lot of "hidden" lag while in the first room of level 8. Probably because of the way it was corrupted (prince is made of blocks); in fact this doesn't happen if corruption is less severe.
I'm working on making another movie file anyway, for triggering the warp to level 8 optimally... Then the new TAS can be started.
Experienced Forum User, Judge, Published Author, Skilled player
(1289)
Joined: 9/12/2016
Posts: 1645
Location: Italy
xxNKxx wrote:
Wow, ThunderAxe31 had done done successfully warp to lv 8.
Actually, MESHUGGAH was the first who managed to pull off :)
xxNKxx wrote:
That's what I got before, but I can't make it after that
Is that because the corrupted player graphics are hard to follow? I'm making a lua script that will help for that. Are there other problems?
xxNKxx wrote:
So, should I cancel this?
Not yet. We first need to comfirm that the game can be beaten. The level 8 warp could freeze later levels, who knows?
