Thoughts on reworking TTYD item drop information?

[jdaster64]

Hi all, I've made a handful of contributions directly or indirectly to MarioWiki over the years, and appreciate a lot of the in-depth list articles, bestiary tables, etc. on the site.

I've also been an active member of a few communities that discuss a lot of challenge running and game mechanical stuff in various Mario RPGs, and some of the Wiki's articles are a bit unclear or inconsistent in how some of the data is presented. We've gotten by relying on secondary resources when necessary, but I thought I'd reach out here and see if there's interest in systematically changing up the way certain data is presented.

In particular, calculating item drop rates in The Thousand-Year Door is something that I think can't really be completely described with numbers in a table; the actual drop rate for an item varies based on the total weight of the items in an enemy's table, whether or not the battle takes place in the Pit, whether or not the player has an Item Hog badge (or multiple!) equipped, and in the case of the Switch release, exactly how many and which enemy types appear in a battle, since all enemies' drop tables are checked in sequence, rather than just the front.

I'm wondering if there's precedent for having an article that goes into item drop calculations in detail (similar to Bulbapedia's page on damage), which could be linked in a tooltip or in the header of an particular enemy's item table, with the table just containing the raw weights?

If that's not preferred, then it would probably be best to remove the numerical information entirely so it's not misleading (as the current values are incorrect in any case), and use a more abstract rarity scale instead. I know there's a Japanese strategy guide for the original game that presenting the information in a similar fashion (with circles and triangles for a scale of very common to very rare), but I can't find a screenshot.

I'm curious as to what other folks think would be best!

 

[jdaster64]

(I'd also be interested in helping clean up any other statistical information in which I might have some expertise for PM64, TTYD, SPM, or some of the M&L games, if anyone has any suggestions for improvements!)

 

[Waluigi Time]

This certainly sounds like something we could (and should) do. I'm not familiar with TTYD's internal workings in the slightest, so bear with me. Are the Pit of 100 Trials and/or Item Hog "flat" modifiers (i.e. do those variables always increase/decrease the chance of drops by a specific percentage) or is it different for each enemy? That would go a long way towards deciding what info should go with enemy statistics and what should be covered on a general page for how item drops work, which seems like a good idea. (I'm not sure if there's direct precedent for this specific situation, but we already have articles covering the nitty-gritty of Mario Kart stats. That seems close enough.)

 

[jdaster64]

How things work in general are that there two sets of "weights" per item type, plus a weight for no item, used for choosing held items and as part of choosing random drops, e.g.:

Hyper Goomba
Nothing - 200 H, 300 D
Life Shroom - 5 H, 5 D
Tasty Tonic - 0 H, 5 D
Ice Storm - 5, 5
Boo's Sheet - 20, 20
Repel Cape - 10, 10
Last Stand - 2, 1
Last Stand P - 2, 1
Charge - 0, 2

[These are the GameCube weights; the Switch ones are different but relatively equivalent for most enemies.]

For enemies holding an item, these weights are used directly to determine whether they hold something / what it is they hold (and the same weights are used for Mowz's Kiss Thief), but for dropping items, the game first does a meta-check for whether to try to drop a held item, drop a random item, or drop nothing. The relative chances of those cases are 25/25/50 (1/1/2) for most fights but 20/20/60 (1/1/3) for the Pit, and wearing N Item Hogs multiplies the "held drop" and "random drop" cases' chances by N + 1.

If neither the "held drop" or "random drop" case is hit, or one of them is hit and doesn't succeed (either no held item exists or no random item is chosen), if Item Hog is equipped, there is a 50% chance of one of a special pool of items dropping. In the event a "held drop" is attempted, if one or more enemies were holding items at the start of the fight, it picks one of them at random to drop, guaranteed. In the event a "random drop" is attempted, it uses the "drop" weights above to determine whether / what to drop.

Notably, the GameCube and Switch versions treat the random drop case differently; the former only checks the front enemy's table for random drops, but the latter checks each enemy's drop table in order (dropping the first one to come up non-None, if any), meaning the exact odds of an item dropping are dependent on how many of a particular enemy exist and where they are in the loadout (as other enemies might have been checked first).

I suppose we could present the weight tables as the "base chance" of an item getting chosen in the event a random drop check occurs for a single enemy, but at that point I think it's kind of needlessly obscuring the actual underlying weights. (I realize this is somewhat hypocritical, having come up with my own preferred scales for Mario Kart stats back in the day, lol)

 

[Waluigi Time]

So if I'm understanding this right, each item has its own base drop chance, but the actual drop chance is affected by the rest of those factors? If that's the case, I would suggest this:

• If it would be useful information, include the chance each item has to be dropped without any other modifiers on their pages.
• Keep loot tables in the enemy statistics, but remove the percentages.
• Create a page that goes more in-depth into the mechanics of item drops.

That would be how I would tackle it at least, assuming I'm understanding everything correctly!

 

[jdaster64]

Sorry, I've been busy the last week or so and will probably remain as such for the rest of the month, lol.

Each item has a base weight, which could be phrased as just a relative number (e.g. 5 held, 5 drop for a Life Shroom in Hyper Goomba's table above) or a "base chance" out of the total weights, including the weight for nothing (e.g. 5/244 held, 5/349 drop for a Life Shroom in Hyper Goomba's table). I personally think just dropping the denominators looks cleaner, but could go either way.

Either way, I'd be down to help make those changes.

 

[jdaster64]

Sorry to drop the ball on this, but if the rates are currently only on badges' individual pages, then here's an example of what a new version might look like:

Using fractions instead of internal weights makes more sense without the rest of the drop table as context.

My proposal for new roll-over text:
- Hold rate: "The rate of an enemy holding an item or badge in battle, as well as the chance of that item being stolen using Ms. Mowz's Kiss Thief if an enemy is not holding anything. In 25% of battles (20% in the Pit of 100 Trials), one of the held items, if any exist, will be guaranteed to drop. Wearing Item Hog badges will increase the chance of this occurring."
- Base random-drop rate: "The base chance of an enemy randomly dropping an item or badge at the end of a battle. In 25% of battles (20% in the Pit of 100 Trials), the front enemy (in the GameCube release) / all enemies in turn (in the Nintendo Switch release) will perform a check for a random drop. Wearing Item Hog badges will increase the chance of random-drop checks being performed."

Let me know if that copy sounds reasonable, or if you'd prefer I use different terminology for the different releases of the game.

If we wanted a page to go over drop mechanics in more detail and list all of the drop tables, that could be added to the template and linked from the bestiary entries at a later time.

 

[Lee Chaolan]

Hey jdaster (I contacted you a while back on the Glitz Pit about information on badges), this is a late response and I'm not sure if you're going to see it, but for what it's worth I feel that you're the authority on TTYD info on the English side on the internet and I personally feel like whatever you decide will be best. I don't think anyone else on here or the wiki has anywhere near enough in-depth game knowledge to make any legitimate complaints if you decide to change things to be more accurate, so if you ever decide to get back to this, I'd say just go for it.

Perhaps the wiki mods might have some stronger opinions on how to format things as I'm not one myself, but that can always be addressed later, imo it's better to work towards more accurate data than keeping things as they are for no reason. What you've shown here looks fine to me and a separate "list of ttyd badge drop rates" page could be useful as well, possibly even preferred.

 

[jdaster64]

Been pretty busy for a while, sorry for the radio silence.

I couldn't find a template I particularly liked for specific badge drops, but I drafted a new template for a bestiary-like page of item/badge drop tables:

I'm imagining there'd be either one page with separate tables or two entirely separate pages for the GameCube and Nintendo Switch releases, and we can get rid of the "drop rates" sections on TTYD badges' pages. I'm neutral as to whether percentage roll-over rates need to be included, but it wouldn't be that hard to add to the template.

Would anyone have a good resource for where to start when adding new templates to the wiki?

 

[OmegaRuby]

The go-to resource is Help:Templates. That'll get the ball rolling on this, and I'd imagine what you're going for is a Content template. Hope this helps!

 

[jdaster64]

That looks helpful; thanks! I'll see if I can find time to work on this some over the weekend.

 

[jdaster64]

Paper Mario: The Thousand-Year Door enemy item drop tables

In the game Paper Mario: The Thousand-Year Door, enemies may sometimes drop an item or badge on the field after being defeated, or hold items and badges while in battle. This page lists all possible items and badges that a particular enemy may hold...

www.mariowiki.com

Sorry for the long wait; had a bunch of other side projects and forgot about this for a while. Here's my attempt at the page; let me know if any formatting changes or further information needs to be added. I can link it in the TTYD category box and link it to/from other relevant pages as needed if everything looks good.

 

[Lee Chaolan]

Looks great! Appreciate the effort to make this information easily accessible and add it to the wiki.

A few notes:

- The page title will probably need to be changed, generally wiki articles won't include the "table" in the title. List of Paper Mario: The Thousand Year Door item drops would probably be my suggestion.

- Likewise the subheadings should drop the "table" out of the title, as we're primarily focused on presenting the game information and not how we're showing it. Replacing it with "item drops" or even "item drop values" should be fine.

In 25% of battles (20% in the Pit of 100 Trials), one of the items held by an enemy at the start of the battle (if any exist) will be guaranteed to drop.

- The phrasing here doesn't quite make sense imo, it's the equivalent of saying "25% of the time, something will drop 100% of the time". The way it's phrased in the Drop section reads better so I would change this to be closer to that.

Rolling over the fractional hold and drop rates will show them as percentages.

- I'm a little confused on what this sentence means, are you able to explain a little more? Maybe this is just me not understanding maths lol

Notes: Each table's title lists the enemy it is most commonly associated with. Generally enemies of a particular type use a single table, but exceptions (such as instances of the table being used by a different enemy, or instances of that enemy not being able to drop items) will be listed here.

- If I understand correctly, the game itself doesn't use tables to determine the drop rates, so we should try and reflect how the game itself handles that information. Changing this to something like "enemies of a particular type share item drop rates" would be better.

- The footnotes and references section can be deleted if there's nothing in them.

Anyway I always hate saying that I appreciate something and then nitpicking at it because I feel like it makes me look like I'm being ungrateful so just want to reiterate that I appreciate your work here haha

 

[jdaster64]

Will take a look at changing some stuff early next week; I just got back from a long vacation before putting that page together and still have a lot of stuff to catch up this weekend.

- The page title will probably need to be changed, generally wiki articles won't include the "table" in the title. List of Paper Mario: The Thousand Year Door item drops would probably be my suggestion.

- Likewise the subheadings should drop the "table" out of the title, as we're primarily focused on presenting the game information and not how we're showing it. Replacing it with "item drops" or even "item drop values" should be fine.

To clarify, "drop tables" (or "loot tables", etc.) exist to me as a concept regardless of how they're presented in code or on reference sites; I feel like that's a pretty common way to describe the different item drops and their probabilities for enemies in RPGs or whatnot. I'm fine changing it to just "item drops" or whatever else, though.

If I understand correctly, the game itself doesn't use tables to determine the drop rates, so we should try and reflect how the game itself handles that information. Changing this to something like "enemies of a particular type share item drop rates" would be better.

I'm not sure exactly what you mean, but they kind of do? Each individual enemy in each individual loadout has its own lookup pointer / string that points to a lookup table of possible drops and probabilities. The names of the data structures and their use in practice imply that they are meant to correspond 1:1, but the fact that each enemy has to do a separate lookup is why there's the possibility for some enemies not to have an item table at all, or to have the wrong one, like Clefts using the Dark Puff lookup table in the Pit of 100 Trials in the GameCube version (they have a similar internal name so a dev probably made a copy-paste error that propagated across all the Pit loadouts), or one of the Glitz Pit fights using the lookup tables for Ruff, Ice, and Poison Puffs even though those enemies don't appear anywhere in the Glitz Pit (weirdly, which fight does that is different between the GameCube and Switch releases!)

Again, if there's a way to describe it that'd be less technical but still get across the nuance that there are exceptions, I'd be fine not using the word "table".

The phrasing here doesn't quite make sense imo, it's the equivalent of saying "25% of the time, something will drop 100% of the time". The way it's phrased in the Drop section reads better so I would change this to be closer to that.

That's fair, lol; I'll rework the phrasing on that to be more consistent.

Rolling over the fractional hold and drop rates will show them as percentages.

That's just referring to being able to hover over them to see the percentage. I can reword that, or just delete it and leave it up to the reader to discover from the formatting.

The footnotes and references section can be deleted if there's nothing in them.

Can do.

 

[Lee Chaolan]

To clarify, "drop tables" (or "loot tables", etc.) exist to me as a concept regardless of how they're presented in code or on reference sites; I feel like that's a pretty common way to describe the different item drops and their probabilities for enemies in RPGs or whatnot. I'm fine changing it to just "item drops" or whatever else, though.

I'm not sure exactly what you mean, but they kind of do? Each individual enemy in each individual loadout has its own lookup pointer / string that points to a lookup table of possible drops and probabilities. The names of the data structures and their use in practice imply that they are meant to correspond 1:1, but the fact that each enemy has to do a separate lookup is why there's the possibility for some enemies not to have an item table at all, or to have the wrong one, like Clefts using the Dark Puff lookup table in the Pit of 100 Trials in the GameCube version (they have a similar internal name so a dev probably made a copy-paste error that propagated across all the Pit loadouts), or one of the Glitz Pit fights using the lookup tables for Ruff, Ice, and Poison Puffs even though those enemies don't appear anywhere in the Glitz Pit (weirdly, which fight does that is different between the GameCube and Switch releases!)

Again, if there's a way to describe it that'd be less technical but still get across the nuance that there are exceptions, I'd be fine not using the word "table".

I admittedly know basically nothing about how games are programmed so I'll leave it up to your discretion on how to word things then. The mariowiki in general doesn't tend to have this type of technical information all too often and will usually generalise things to be a bit more basic and less video game guide-y, but I'm not familiar enough with drop tables to know what needs to stay or not. Maybe it's fine as is, I'll try and get some additional input on this.

That's just referring to being able to hover over them to see the percentage. I can reword that, or just delete it and leave it up to the reader to discover from the formatting.

Oh right the word "rolling" was confusing me haha, I think it's fine to include but maybe reword it slightly since this can also be viewed on mobile and that type of text can also be tapped on instead of hovered over. I feel like there's a word for it but it's escaping me rn lol