The bulk update request #27 is pending approval.

Reason: Implications for size differences. Considered "extremely_much_larger_monster" / "extremely_much_smaller_monster"

We should probably discuss how to classify size difference scale. So far I haven't seen much differentiation except for size_difference and macro / micro on other boorus.
Can you please give some guidelines on how you'd like to tag size differences? A good reference would be using Pokemon heights whether they're within the same species, interspecies or compared to humans as we'll typically do. So to begin with in the modern games how does an average sized Pokemon species compare to its largest and smallest sizes (+-20%) or if comparing the smallest and the largest directly to each other?
 

Edit: Oops I didn't see the discussion in topic #13 already. I'll bring that over here so it has its own place.

anonymous said:
Size Differences
It's probably sufficient to tag larger_monster / smaller_monster and not both larger_monster / smaller_human or just use aliasing.

In regards to size differences, I'm not sure what should constitute a "larger" or "much larger" monster. I'd guess a monster more than ~1.5x the size of the human would be a "larger monster", a monster ~3x the size of the human would be a "much larger" monster, and anything that's "how would you even hug that thing" is an "extremely much larger monster".

For smaller monsters, I think a rule of thumb could be "if it's lower than the chest, it's a smaller monster, if lower than the knees, it's a much smaller monster", and if smaller than a foot or a hand it's an "extremely much smaller monster".

dronedartdrain-tc said:
i had the same thought but i didnt really comment on it because the host seemed to like it. maybe i should ask after all.

anonymous said:
I'd just alias smaller_human / larger_human to their corresponding larger_monster / smaller_monster equivalents.

Aliasing is probably smarter than relying on users to tag both the monster and human's size. Feel free to add that to the BUR!

I think monster scaling definitely should be strictly compared to the human in the picture as I'd imagine that's what someone using them would be looking for, and judging by other things or canon forms gets really complicated really fast.

Oh whoops I think I was explaining how to compare size differences the wrong way in my post. What I was trying to do was use Pokemon sizes as a guide to determine what scale the size difference tag we would use. Like you were doing in forum #109.

anonymous said:
In regards to size differences, I'm not sure what should constitute a "larger" or "much larger" monster. I'd guess a monster more than ~1.5x the size of the human would be a "larger monster", a monster ~3x the size of the human would be a "much larger" monster, and anything that's "how would you even hug that thing" is an "extremely much larger monster".

We would still only use size difference tags for comparing human and monster heights. You're right that trying to use them to compare canon forms or monsters to each other would be too complicated.

Let's use the heights of a single Pokemon like Gengar to start mapping this out. His canon heights are:

• Average (M): 5ft (rounded) / 1,5m
• Smallest (XXXS): 4ft / 1,2m
• Largest (XXXL): 6ft / 1,8m
• Alpha: 10ft / 3m

Since we're trying to find out what scale size difference tag to use we need 2 characters (a monster and a human) with their own heights. We'll use Gengar's heights above to compare.

We can use other Pokemon sizes for more comparisons. These use rounded averages.

• Eevee: 1ft / 0,3m
• Riolu: 2,5ft / 0,75m
• Arceus: 10ft / 3m
• Giratina (altered form): 15ft / 4,5m
• Milotic: 20ft / 6m

Finally if you wanted to not limit yourself to the more typical range of human sizes that are roughly between ~4,5ft to 7ft or compare larger differences you could also compare Pokemon heights directly to each other as well. For these you might say that the human has taken the size of a Pokemon but this is more for determining varied values for the unit and relative difference columns as well as asking what tag to apply.

At the end of the day we're trying to find what to put in the Size Difference Tag column. Is this a good way of discussing this topic or am I overthinking it?

Anyway I think this is a good idea since we probably will need more size difference tags than size_difference and macro / micro because of posts like these. I'm sure some will greatly appreciate us having these extra details! XP

I think any size difference should be based on something that is feasible to measure on a large amount of images firsthand, and in your measurements it seems like numbers 2x, 4x, 8x/10x tend to come up. I think we might be able to use similar measures to those I suggested for when human is the larger, that is (assuming a bipedal-ish monster with roughly human proportions) belly (~50% of height), knees ( ~25% of height) or feet/hands (~10%). This would also allow some rough guidelines for when a monsters full stature isn't visible:

- Can the human reach about to the monsters belly? -> larger_monster
- (If the monster is quadrupedal) Can the human reach about halfway to the monsters "main body"? -> larger_monster
- Is the human about half the size of the monsters leg? -> much_larger_monster
- Does it seem like the human could roughly be held in the monsters hand? -> extremely_much_larger_monster

Of course, monsters come in all shapes and sizes and these wouldn't always be feasible either for monsters with weird proportions or oversized hands or whatever, but it at least offers some alternative

I guess you could also have a tag for astronomical size differences where the human or monster is about the size of an insect in comparison to the other or smaller than their nails.

Here's a new suggestion vaguely based on the above: https://files.catbox.moe/eqxbud.png

anonymous said:
I think any size difference should be based on something that is feasible to measure on a large amount of images firsthand, and in your measurements it seems like numbers 2x, 4x, 8x/10x tend to come up.

I intentionally chose comparisons that would make larger size difference multipliers to try and give enough examples to find where the boundary of each classification tag would be. The differences are more likely to be closer to the first table than the others I think. So about the size of Riolu or bigger if they're smaller than the yume though the monster could be tinier if the artists is going for the cute factor, or smaller than a Seviper (9ft / 3m).

anonymous said:
Here's a new suggestion vaguely based on the above: https://files.catbox.moe/eqxbud.png

That said I really like the image you linked! It not only provides a good visual basis for size differences but also gives really nice descriptive words we can use too! I was almost going to create a chart myself with my examples.
One question I have though is do we need to determine more definitive size different measurements when applying these tags, kind of like I was implying in my tables above, or will we just let the tagger decide? Does that matter alot when searching?

wawanya said:
I intentionally chose comparisons that would make larger size difference multipliers to try and give enough examples to find where the boundary of each classification tag would be. The differences are more likely to be closer to the first table than the others I think. So about the size of Riolu or bigger if they're smaller than the yume though the monster could be tinier if the artists is going for the cute factor, or smaller than a Seviper (9ft / 3m).

That said I really like the image you linked! It not only provides a good visual basis for size differences but also gives really nice descriptive words we can use too! I was almost going to create a chart myself with my examples.
One question I have though is do we need to determine more definitive size different measurements when applying these tags, kind of like I was implying in my tables above, or will we just let the tagger decide? Does that matter alot when searching?

Some cutoff points could possibly go elsewhere, but I don't think we'd find them using a table. Figuring out the sizes of subjects in an image is really hard and would probably always be based on some easy to spot gauge like "seems to reach up to their belly", "could fit in your hand" and so on and rarely be accurate. Even with my suggestions, I assume you'd end up with large amounts of images "misclassified" because of enlarged hands, non-humanoid monsters and monsters whose size isn't primarily height, but could probably still serve some "basic" use for people looking for those types of gauges ("someone I feel like I couldn't reach for a kiss even on my toes", "someone that could hold me in their hand" and so on).

I think the biggest thing lacking right now is a basic basis for how to measure quadrupeds. One way might be to imagine both subjects lying down next to each others with their legs stretched out rather than standing.

anonymous said:
I think the biggest thing lacking right now is a basic basis for how to measure quadrupeds. One way might be to imagine both subjects lying down next to each others with their legs stretched out rather than standing.

Measuring quadrupeds by comparing their size or length laying down is a good idea. It's actually kind of funny since even if they're the same size they would appear very big lol. And there's a massive difference between a quadruped who's the size of a human lying down vs one who's as tall as a human when the monster is standing on 4 legs.
In any case we'll take a lenient approach to more specific size tagging. We'll keep an eye on it but I don't think mistagging will get out of control before anyone spots an issue.
Are you able to modify your BUR in the OP with these ideas in mind? If not I can tweak it. I'll also update the Tagging Checklist with this additional guidance.

should we really measure based on standing up cause i did so far which was logical to me but then theres things like this and i mean..
is it really correct to tag larger human in this image just because she would be taller when standing.. kinda doesnt make sense in this one.
maybe we should just tag it based on how it looks in the image. this would also solve the quadruped problem a little but yeah.. difficult.

post #349

edit: so basically right now we are only taking height into account but not really width at all or poses which might lead to strange tagging..

edit: ok so heres an idea..
maybe just measure from a rolled up/fetal position or a sitting position instead for quadros/animals?
not thinking about a height line anymore but rather something like hitboxes. this would account for overall mass and not just length.

https://files.catbox.moe/k0i6pu.png

edit: rolled up is probably preferable because of things like snakes? the basic idea is to compare hitboxes instead of any sort of height line. basically abandoning those for at least animal images...
might even make sense to apply this in general.

edit: ok after thinking maybe we do just use it for animals and strange monsters like blobs? whats your thoughts though.

dronedartdrain-tc said:
should we really measure based on standing up cause i did so far which was logical to me but then theres things like this and i mean..
is it really correct to tag larger human in this image just because she would be taller when standing.. kinda doesnt make sense in this one.
maybe we should just tag it based on how it looks in the image. this would also solve the quadruped problem a little but yeah.. difficult.

post #349

edit: so basically right now we are only taking height into account but not really width at all or poses which might lead to strange tagging..

I think I might have focused too much on a single dimension like height which gives the wrong impression. The tables above are supposed to ask at what point do we consider using a larger scale size_difference tag, using Pokemon as a reference and multipliers to simplify that question. In the end the best answer was the image in forum #125.
It's a bit different applying this in practice. A 5ft Gengar would still be larger than a 5ft yume since he's roughly square shaped. For example this pic shows a yume who's taller than her Gengar but he still looks larger because of his body shape though it's up to users to decide whether it needs a size_difference tag (it doesn't on e6).
 

dronedartdrain-tc said:
edit: ok so heres an idea..
maybe just measure from a rolled up/fetal position or a sitting position instead for quadros/animals?
not thinking about a height line anymore but rather something like hitboxes. this would account for overall mass and not just length.

https://files.catbox.moe/k0i6pu.png

What I meant for comparing quadrupeds lying down is if both the yume and monster were lying down together. This can help determine the difference like how your top right illustration shows the yume is still much smaller. Adding rough bounding boxes to their bodies is a great way of determining their size! (Also those mons are super cute too! XD)
 

dronedartdrain-tc said:
edit: rolled up is probably preferable because of things like snakes? the basic idea is to compare hitboxes instead of any sort of height line. basically abandoning those for at least animal images...
might even make sense to apply this in general.

edit: ok after thinking maybe we do just use it for animals and strange monsters like blobs? whats your thoughts though.

I think those curled up examples are good guides that can help people out when tagging. They're useful if someone isn't sure if whether to tag their images or not but I think we should keep a relaxed approach for tagging size differences since it's hard to eyeball each character alot of the time.

just saying im gonna alias the human sizes and delete that part from the rules.

wawanya said:
I think I might have focused too much on a single dimension like height which gives the wrong impression.

i meant that more in a general booru way (that we were kinda using length to tag right now) but also that you guys were focusing more on length me thinks.

wawanya said:
I think those curled up examples are good guides that can help people out when tagging. They're useful if someone isn't sure if whether to tag their images or not but I think we should keep a relaxed approach for tagging size differences since it's hard to eyeball each character alot of the time.

Adding rough bounding boxes to their bodies is a great way of determining their size! (Also those mons are super cute too! XD)

yeah i just thought it makes sense that way. just some sort of guideline for those cases. also i think more in visuals so maybe its just more normal to my art brain.
(also thanks!)

wawanya said:
For example this pic shows a yume who's taller than her Gengar but he still looks larger because of his body shape though it's up to users to decide whether it needs a size_difference tag (it doesn't on e6).

e6 doesnt really have the issue of comparing humans to monsters so for them this is kinda irrelevant. here i would say it does need the larger tag but thats me. just strange to see a massively wide mon not be considered larger lolol.

wawanya said:
What I meant for comparing quadrupeds lying down is if both the yume and monster were lying down together.

yeah that i understood. just that one was focusing more on length.

also i didnt even see this
>create implication larger_monster (53) -> size_difference (70)
ahh i was thinking larger and smaller just means any difference while size difference refers to a larger difference than that..
would be good to know the minimum difference in size you guys wanna settle on in the end to be able to use the tags.

dronedartdrain-tc said:
here i would say it does need the larger tag but thats me. just strange to see a massively wide mon not be considered larger lolol.

I'm happy for examples like that to have a larger_monster tag. We can use size_difference tags for similar images.
 

dronedartdrain-tc said:
would be good to know the minimum difference in size you guys wanna settle on in the end to be able to use the tags.

Some communities I've seen usually accept that a 1ft difference should use a size_difference tag. What kind of cases with smaller differences might need tagging if any?
Of course that's just for length. How should we determine when to apply a tag if the size varies by shape in another way like in the bounding box method? Should we use rough percentages? Maybe if the characters have something that looks like 15 % overall difference?

im gonna drop another thought.
we should have a smiliar size tag cause im coming across images where im like "well they are the same size" so im not tagging anything and it looks strange to me.
its as if im forgetting to tag a size.

I'd say you're right if we were designing a data structure for a programming project and decided that relative size was a mandatory attribute for each post. In that case it'd be best to provide an option for all differences, including the lack of significant variation.
As for now, mandating a similar_size tag could be useful since size_difference is more important to us than e6. However it should still be possible to find comparably sized partners by instead excluding size_difference results. Is that enough or does our tighter focus give more value for having a tag for this too?

wawanya said:
I'd say you're right if we were designing a data structure for a programming project and decided that relative size was a mandatory attribute for each post. In that case it'd be best to provide an option for all differences, including the lack of significant variation.
As for now, mandating a similar_size tag could be useful since size_difference is more important to us than e6. However it should still be possible to find comparably sized partners by instead excluding size_difference results. Is that enough or does our tighter focus give more value for having a tag for this too?

yeahh it might not be necessary but i always think ah hm isnt it good to tag it cause directly searching for it gives you more accurate results in case larger/smaller is missing somewhere.
i dont really super duper need the tag though.

dronedartdrain-tc said:
yeahh it might not be necessary but i always think ah hm isnt it good to tag it cause directly searching for it gives you more accurate results in case larger/smaller is missing somewhere.

That's fair. Other boorus can have ambiguity problems for certain tags where you could describe the possibilities as true (tag for an attribute in the post), false (antonym tag to specifically highlight an attribute is not in the post) or undefined (unknown whether a tag applies). This matters alot if particular attributes or tag groups are mandatory but when they're optional there's space for confusion.