ROM Hack "Lightweight" Pokémon SUMO/USUM? (models)

VeniaSilente

Well-Known Member
OP
Newcomer
Joined
Sep 6, 2014
Messages
94
Trophies
0
XP
488
Country
Cote d'Ivoire
Hey.

Basically one question that I guess it fits here than anywhere else, regarding the Gen7 Pokémon 3DS games.

One thing that makes the SUMO / USUM games "USM" -"Ultra Slow Mo"- is the Pokémon models stored inside. I've heard a number of reasonings: that the models are "overdone" and designed for the Switch or the PS4 instead of for a console with ~240p resolution; that GF is just incompetent and the models are completely unoptimized. Whatever the case, the games just run painfully slow in particular whenever there's a battle (kind of the point of a Pokémon mainline game) and even more when in a o3DS where this compounds with a number of other issues such as having to reboot in and out of the game.

But, since we already do have things like the Pokémon retextures hack for OR/AS, I was wondering, would it improve the performance of USM to have a romhack that replaces the models for more lightweight / more optimized ones? Has there been research in this direction? I haven't really been able to find anything. Perhaps there's something already done even but googling for "USUM lightweight models" or similar leads me to, at most, news articles about the latest Legendaries.
 

Sliter

Well-Known Member
Member
Joined
Dec 7, 2013
Messages
3,259
Trophies
0
Location
ᕕ( ᐛ )ᕗ
XP
1,688
Country
Brazil
making less poly models for 666+ monster would be a pain project and probably not finished anytime soon ... at least until gen2 they could be get from pokemon stadium GS :v so it like 251 less to make but to adapt to US/M xD
 

VeniaSilente

Well-Known Member
OP
Newcomer
Joined
Sep 6, 2014
Messages
94
Trophies
0
XP
488
Country
Cote d'Ivoire
making less poly models for 666+ monster would be a pain project and probably not finished anytime soon ... at least until gen2 they could be get from pokemon stadium GS :v so it like 251 less to make but to adapt to US/M xD
Are the models in X/Y the same as the ones in G7? If they weren't that would save enough time. The X/Y games feel randomly laggy at times, but not to the "get used to it or switch to a PS4" level that Ultra Slow Mo does.
 
  • Like
Reactions: Sliter

DeathChaos

Unmei wo koeru!
Member
Joined
Oct 21, 2015
Messages
1,364
Trophies
0
Age
27
XP
1,705
Country
Puerto Rico
I was wanting to do some tests, but I cant seem to figure out how to import a modified model with ohana3ds.
If we ever manage to get animations working, I'd love to get back the pre-Gen 4 Starmie animations, the new one is just so lifeless.
hHOx52g.gif
 

VeniaSilente

Well-Known Member
OP
Newcomer
Joined
Sep 6, 2014
Messages
94
Trophies
0
XP
488
Country
Cote d'Ivoire
Old Starmie animations? Now that would be nice.

Regarding the load of work that could be reducing the models, I do hold a hope that this can be done in an easier way, ideally iteratively.

Best case scenario the ~700 models in X/Y are reduced ones already and those can be used, leaving only the new mons in G7 to reduce. Even in the worst case scenario (no reduced models previously available at all), it is not necessary to reduce all the models before producing a working product, for the most part it suffices with reducing only the G7 starters and a number of com mons, and even then it suffices with partial reduction (let's say from 10000 polygons to 9000) for a start. A first release could simply cover a selection of starters plus the top [X amount of]%ile heaviest model mons for testing.

All this thanks to the fact that save data information would not be affected by this kind of hack, so eg.: releasing a first version and continue working on an upgrade does not necessitate clients (players) to restart their saves like what happens in the Pokémon GBA hackscene.

Now, testing this kind of romhack I am not sure how would it go. Any references here? At early stages of progress with minimal reduction in model weight, I would assume (?) that gains in performance would only mostly be observable via something like an onscreen FPS counter / NTR plugin, comparing battery life across sessions, or via comparing pre-programmes sequences of battles recorded with vs without the patch applied. While I an not that knowledgeable in the subject I would also presume that any gains would only mostly be appreciable in the o3DS but then again as a user of that is the audience / target I expect for this kind of hack (kinda like how there is a SNES emulator intended specifically for the original 3DS).
 
D

Deleted User

Guest
With such good plan, why dont ya start doing something yourself? I will be happy to see your final product!
 

Rayzark

giooig
Member
Joined
Nov 11, 2015
Messages
150
Trophies
0
Location
Roma
XP
140
Country
Italy
I was wanting to do some tests, but I cant seem to figure out how to import a modified model with ohana3ds.
you have to export the .obj part first, edit them and reimport them. Unfortunately you can't edit the model adding polygons or vertexes, it crashes in ohana 3ds if you do that
 

VeniaSilente

Well-Known Member
OP
Newcomer
Joined
Sep 6, 2014
Messages
94
Trophies
0
XP
488
Country
Cote d'Ivoire
With such good plan, why dont ya start doing something yourself? I will be happy to see your final product!
Alas, it's not even the plan I have. Just a problem and a clear perception of what a solution would look like. My area of expertise is worldbuilding, not actually working with the technical resources, so it's not something I would approach by myself if I expect to see success. But that's one of the reasons why I decided to ask - I can't be the only one with this problem ( = "the only person in the world with a o3DS and Pokémon Sun/Moon") so I thought it'd be worth checking if there was already any existing work on this directions, or plans of.

Now, that I can assist a lot with testing developments, I can.
 

You may also like...

General chit-chat
Help Users
    AncientBoi @ AncientBoi: lol