Glitch City Laboratories Archives

Glitch City Laboratories closed on 1 September 2020 (announcement). This is an archived copy of a thread from Glitch City Laboratories Forums.

You can join Glitch City Research Institute to ask questions or discuss current developments.

You may also download the archive of this forum in .tar.gz, .sql.gz, or .sqlite.gz formats.

Pokémon Glitch Discussion

Glitching in 2019: Approaches and Common Goals - Page 1

Glitching in 2019: Approaches and Common Goals

Posted by: Sherkel
Date: 2019-03-19 22:52:46
This is a quick exchange I had with luckytyphlosion a few days ago about the state of glitchology today as compared to when this site began, in addition to how other communities have approached the field and why they may have had more success. I don't have anything more to add, but it brings up some good points and I think it's worth a read for anyone wondering why this site is the way it is.

Sherkel Last Saturday at 8:00 PM
@luckytyphlosion Raven has been documenting stats/sprites/dex entries/etc. of Pokemon with invalid index numbers
So like, "classic" glitch research so to speak

luckytyphlosion Last Saturday at 8:00 PM
empirical
is he dumping them from the ROM

Sherkel Last Saturday at 8:00 PM
I don't know

Ravenfreak Last Saturday at 8:59 PM
Im actually just documenting them manually using gameshark codes to encounter each glitch Pokémon. :V Im taking the names, types and glitch moves from my save file. Its not the quickest way to get the information but it works. :stuck_out_tongue:

ShinySkitty Last Saturday at 9:00 PM
in which game?

bbbbbbbbba Last Saturday at 9:01 PM
How could you know they are not coming from the RAM?

luckytyphlosion Last Saturday at 9:10 PM
sad
:morphoff:

Sherkel Last Saturday at 9:40 PM
It's just how it was done in the olden days

luckytyphlosion Last Saturday at 9:41 PM
OLDEN
yeah, back when glitching was actually fun :morphon3d:

Sherkel Last Saturday at 9:42 PM
Now we know that keeping the current enemy mon byte set throughout the battle/dex appearing can have other effects as those read from RAM, but keep in mind when this whole thing started and how the views of relevance of changed

luckytyphlosion Last Saturday at 9:42 PM
hey maybe I can convince @bbbbbbbbba to do what I'm too bothered to do and actually dump glitch mon info by reading the ROM

Sherkel Last Saturday at 9:42 PM
If we ever had some sort of "GlitchDex 3.0" it would be pulled straight from the ROM
Actually, @Melodic Evie the Bird Mother ❤ 🦉 may have done that when revising the English RBY entries, but idk for sure

luckytyphlosion Last Saturday at 9:44 PM
you mean the literal glitchdex entries?
wait

Sherkel Last Saturday at 9:44 PM
No, what the site calls the GlitchDex
It dates back to before GCL, I think

luckytyphlosion Last Saturday at 9:44 PM
how does missingno's pokedex entry have that long-ass cry
wait
is that missingno's cry
or are those sounds generated from the pokedex entry
because that shouldn't be possible
for the sounds to be generated from the pokedex entry

Sherkel Last Saturday at 9:46 PM
I don't mean to dismiss what @Ravenfreak is doing, in fact my point is that doing this sort of thing without obsessive precision is fine for this purpose
Uhh, lemme see that page

luckytyphlosion Last Saturday at 9:46 PM
what even is the purpose of the glitchdex

Sherkel Last Saturday at 9:46 PM
I doubt it has the 30 minute Yellow VC cry with the 20khz beep

luckytyphlosion Last Saturday at 9:46 PM
kek
yellow missingno cry is superficial at best

Sherkel Last Saturday at 9:46 PM
> what even is the purpose of the glitchdex
As far as "glitching" today goes, nothing

luckytyphlosion Last Saturday at 9:46 PM
it's the result of random corruption
based on the corrupted audio bank
when its sprite is decompressed

Sherkel Last Saturday at 9:47 PM
Any that execute arbitrary code are referenced on the pages with the exploits that use that call
That's basically all glitch cries

luckytyphlosion Last Saturday at 9:47 PM
the reason it's so long on yellow VC is because the cry somehow ends up pointing to OAM (on both console and VC)

Sherkel Last Saturday at 9:47 PM
I mean, some are consistent, but consistently reading from somewhere that isn't meant to be a cry nonetheless

luckytyphlosion Last Saturday at 9:48 PM
because of inaccessible OAM the cry eventually ends when it reads an $ff
but VC doesn't have inaccessible OAM
so it actually reads a lot more non-zero entries
with that adding 8 more minutes to its so-called cry

Sherkel Last Saturday at 9:50 PM
What were you referring to earlier, by the way? I think the links to cries in the GlitchDex for the most part got broken
And I see none in MissingNo.'s
Anyway, the pret/PSR perspective is very different from what's in the GCL databases

luckytyphlosion Last Saturday at 9:52 PM
uh this is neither of those perspectives
this is my perspective

Sherkel Last Saturday at 9:52 PM
I mean the pragmatic one

luckytyphlosion Last Saturday at 9:52 PM
pret and PSR care even less about documenting glitches/behaviour than I do
ah

Sherkel Last Saturday at 9:52 PM
GCL wants to incorporate that but keep in mind but also dates back to before 2005
When the perspective was "huh, what does this do? Oh, cool."
And I have a lot of sympathy for that one

luckytyphlosion Last Saturday at 9:53 PM
"glitches r kewl"

Sherkel Last Saturday at 9:55 PM
Yes, you do care more about documenting from what I can tell. I don't know exactly what PSR thinks of that, but based on how hard it is to find their actual  findings I can guess they don't care that much
And pret is just focused on the purpose that's in the name

luckytyphlosion Last Saturday at 9:56 PM
well the research PSR does is mainly focused around speedrunning
you can really see the appliedness of it with manips
extracting all the relevant info of how the game ticks to a cycle towards manips

Sherkel Last Saturday at 9:57 PM
Of course
I'd say they've done more for glitchology than any other group

luckytyphlosion Last Saturday at 9:57 PM
pokecrystal's maintainers are more focused around making it a usable base for hacking
while still keeping it matching

Sherkel Last Saturday at 9:57 PM
But all they seem to publish are dumbed-down descriptions of routes

luckytyphlosion Last Saturday at 9:58 PM
yes, because the people who use the routes don't know how to glitch
they just execute
it's not necessary to know how the glitches work in most cases (although 151 is an exception, you do need to at least know the techniques behind some of the glitches)

Sherkel Last Saturday at 9:58 PM
Hence why I want GCL to do that

luckytyphlosion Last Saturday at 9:58 PM
GCL to do what

Sherkel Last Saturday at 9:59 PM
Document stuff properly

luckytyphlosion Last Saturday at 9:59 PM
well I think a good way to motivate research is to tie that research in with a meaningful purpose that can unite people together
speedrunning is one

Sherkel Last Saturday at 9:59 PM
The problem is that with the obvious exceptions, the people involved/trusted/active/etc are very out of touch
Myself included

luckytyphlosion Last Saturday at 10:00 PM
for pret the common cause is improving the disassemblies/decompilations
also, a button challenge for sm64
so if you can find some common cause that can unite people together then you can get your research done
because right now it's not really defined what the research is for

Sherkel Last Saturday at 10:00 PM
That's a good point

luckytyphlosion Last Saturday at 10:01 PM
it doesn't seem like anyone wants to document for the intrinsic value of doing so

Sherkel Last Saturday at 10:01 PM
And full disclosure, GCL's still is (as far as I know) "glitches r kewl"
So that could be why we've fallen behind

Re: Glitching in 2019: Approaches and Common Goals

Posted by: coloradohugge
Date: 2019-03-20 02:12:40
well. I too get it done by manually using gameshark codes to encounter each glitch Pokémon, cause i don't know how to do it in any other way.

If that is the case, then ALL the pages for the international glitchdex's also have roots in being generated thay way. infact, all the name strings for all the pages are directly from me encountering them all and slowly write down each name of them all. and then slowly adding each to a new page entry.

So if there is a "proper" way to do things, then the reason why i haven't done it personally is because i don't know how to do it. and i can only do as much as i can do. :0

Re: Glitching in 2019: Approaches and Common Goals

Posted by: Parzival
Date: 2019-03-20 08:21:03
honestly glitches aren't fun anymore because everyone has a huge, raging hate boner for empirical testing.

Re: Glitching in 2019: Approaches and Common Goals

Posted by: Sherkel
Date: 2019-03-20 14:03:29

well. I too get it done by manually using gameshark codes to encounter each glitch Pokémon, cause i don't know how to do it in any other way.

If that is the case, then ALL the pages for the international glitchdex's also have roots in being generated thay way. infact, all the name strings for all the pages are directly from me encountering them all and slowly write down each name of them all. and then slowly adding each to a new page entry.

So if there is a "proper" way to do things, then the reason why i haven't done it personally is because i don't know how to do it. and i can only do as much as i can do. :0
You would take the data from the ROM, as explained here. To find where the starting address for base stats is in other language versions, just search the bytes for Bulbasaur's.

But, like I said, I have no problem with doing them "the old way". I gave you your current rank for it. The data is still there and most likely correct. Is that a polite way of saying I don't especially care? Maybe, but it doesn't mean others can't and that it isn't still work done for the site. The whole Decamark thing made me realize how needless it was to go too far with documenting glitch Pokémon species (aside from ????? (000) and MissingNo./'M, their functionality is fairly nonexistent), but others are free to disagree.

Basically, I have trouble seeing the point in documenting glitch species just for its own sake, but also respect what kicked this site off originally, the idea of tearing into these games with whatever means are available and reporting what happens.

Re: Glitching in 2019: Approaches and Common Goals

Posted by: Raven Freak
Date: 2019-03-20 19:00:08
I suppose I should start documenting the glitches by doing the methods luckytyphlosion mentioned. I'm all for getting correct information, and honestly I didn't think that the gameshark codes would affect information such as the base stats, as the only codes I used were the instant battle and level codes. (But they do affect RAM…) So in theory, those glitches are still getting their base stats from elsewhere of course. I haven't even documented them yet, but I will get back to doing so. Edit: The base stats are located in the same offset in German Yellow as the English equivalent btw for those who are curious.

Re: Glitching in 2019: Approaches and Common Goals

Posted by: Sherkel
Date: 2019-03-21 12:58:08
To be honest, the whole matter of which parts of RAM are modified is a trifle compared to how I was making it sound. I think for the researcher it's more a matter of tediousness to check which TMs they learn, when they evolve and such, provided they're not used to doing it another way. What was going through my mind was more ideological, as in "why modify memory externally to research things that are accessible without that?". Anyway, I never originally meant for this to be all about empirical testing, or hate thereof. Instead I'd rather just direct anyone interested in glitching to the rest of the chat.

Re: Glitching in 2019: Approaches and Common Goals

Posted by: ISSOtm
Date: 2019-03-21 13:51:19

So if there is a "proper" way to do things, then the reason why i haven't done it personally is because i don't know how to do it. and i can only do as much as i can do. :0

The "proper" way is to examine what the game is doing, so basically analysis of the ROM. (Can be done at runtime, not just by looking at code)
I wouldn't blame you for not knowing, though.



honestly glitches aren't fun anymore because everyone has a huge, raging hate boner for empirical testing.

Empirical testing has brought a lot of errors into people's minds. It's also very susceptible to missing / hiding key facts.
It is always the basis for discoveries (hardly anything's discovered by looking at source code), but not a good basis for documentation, except maybe through strict testing.



But, like I said, I have no problem with doing them "the old way". I gave you your current rank for it. The data is still there and most likely correct. Is that a polite way of saying I don't especially care? Maybe, but it doesn't mean others can't and that it isn't still work done for the site. The whole Decamark thing made me realize how needless it was to go too far with documenting glitch Pokémon species (aside from ????? (000) and MissingNo./'M, their functionality is fairly nonexistent), but others are free to disagree.

Documenting glitch species has interests beyond preservation; for example, reliable documentation against false rumours, myths, and fabrications (see the recent thread about just that), or being a reference for stats and so on.
Functionality is still interesting, for example consider that 8F used to be obtained through Pokémon flags corrupting item IDs; this is overshadowed by more advanced techniques (including ACE), but it's still good to have them documented somewhere if someone accidentally locked themselves out of these better methods, or is doing a challenge run that somehow bans them.


Basically, I have trouble seeing the point in documenting glitch species just for its own sake, but also respect what kicked this site off originally, the idea of tearing into these games with whatever means are available and reporting what happens.

Mixed opinion. I understand people wanting to glitch just for fun - ACE did take the charm away, in a way - but I think site documentation should be done accurately, because I do want to keep the mood of "that era", but certainly not the dis-/mis-information.

Re: Glitching in 2019: Approaches and Common Goals

Posted by: Parzival
Date: 2019-03-22 20:27:10
Let's just write a script or smth that does the testing and documentation for us, then? If we're gonna be systematic about it, let's just do it with some helper program for a (hacked) console or emulator and take human error out entirely?


Seriously.

Re: Glitching in 2019: Approaches and Common Goals

Posted by: Sherkel
Date: 2019-03-23 16:24:44

Let's just write a script or smth that does the testing and documentation for us, then? If we're gonna be systematic about it, let's just do it with some helper program for a (hacked) console or emulator and take human error out entirely?


Seriously.
That's exactly what happened for the Decamarks. Not sure where the data is, but I'm pretty sure it's floating around somewhere. Granted, it didn't grab sprites, screenshots, or witness other potentially interesting effects. I don't see the need to have an opinion on it myself, as none of them can be obtained within the bounds of normal gameplay.

I didn't know about the older 8F thing, Isso! That is a good point in favor of testing possible glitches "in the field" so to speak.