/co/ - Comics & Cartoons

Where cartoons and comics collide!

Index Catalog Archive Bottom Refresh
+
-
Name
Options
Subject
Message

Max message length: 12000

files

Max file size: 32.00 MB

Total max file size: 50.00 MB

Max files: 5

Supported file types: GIF, JPG, PNG, WebM, OGG, and more

E-mail
Password

(used to delete files and posts)

Misc

Remember to follow the Rules

The backup domains are located at 8chan.se and 8chan.cc. TOR access can be found here, or you can access the TOR portal from the clearnet at Redchannit 3.0 (Temporarily Dead).

Ghost Screen
Celebrating its fifth anniversary all September


8chan.moe is a hobby project with no affiliation whatsoever to the administration of any other "8chan" site, past or present.

Affiliated boards /ac/ Looks like our rules page got nuked, will have to repost it again later on.

(1.98 MB 1000x1472 GGG.png)

Anonymous 07/12/2025 (Sat) 00:40:20 No. 53528
Is this the best Cartoon of all time about "the power of science"??? Any other one that can compare?? Dexter is trash btw.
futurama
Jimmy Neutron was fun. Still a shame the final season was never made, if a reboot were to happen I feel the last season plans would be ignored completely. The janky CGI was also a part of the charm, a reboot wouldn't do it justice.
>>53613 >janky CGI they used the movie models which were made with movie"s budget. The movie was an excuse for them to kickstart a franchise with models that would last a decade. This whole "jimmy had bad cgi" thing is a meme. >b-but this tiny animation mistake
>>53616 Even so I feel like using updated modern models wouldn't feel the same if they rebooted it today.
(390.51 KB 640x480 20250713-182936.png)

>>53613 I have spent way too much time with frame advance on Jimmy Neutron as part of making >>>/v/1539653 I think the blatant clipping problems are mainly something that occurs in season 1 rather than the series as a whole. It absolutely looks way better than most TV CGI shows that came out before it, and even stuff from years after JN ended looks awful in comparison (pic related came out 2005 and manages to combine the worst of a PS2 pre-rendered cutscene with the worst of an in-engine cutscene of the same era).
>>53672 >>>/v/1539653 how many pictures do you need to train a model like this?
(257.44 KB 454x537 022.png)

>>53818 The thresholds for images needed for a style is a subject I'm not entirely sure on myself, with dataset quantity thresholds being poorly documented (even numbers needed for a cartoon character were all over the place when I started. Now that I've done enough character models with low data I can safely say for those). As a sometimes style maker, frequent style user, and frequent guide reader I can say the following with confidence >base models vary in dataset quantity needed for training (Flux apparently needs a lot less than Illustrious) >image size and quality matters (and don't upscale, you'll teach upscale artifacts) >at least for certain styles (pixel art likes low noise) your training settings plays a big part >training settings are easily the least understood part of the process even for experienced trainers (most people leave everything on default or only changes a few settings to their default preference) >a model that only wants to replicate one specific type/framing of image (e.g., video game portraits or promo character art on a blank background) needs a lot less than a style able to do "anything", but it won't be any good outside of that specific thing (only ~30+ of that style with flip augmentation is good enough?) >~50 works fine for a general purpose style when it works, but some things are just frustratingly completely undoable by the resulting model <for example my LucasArts cutscene style model at 58 really sucks at adult women because there's only a small number of those in the dataset, and the default anime bias in the base model will overshine it on them >the 398 images for the Jimmy Neutron style and nearly 1k images I used for my Gundam X style was almost certainly overkill >do not try to handtag 1k images in a single go, you will burn yourself out >making a bunch of smaller LoRAs for characters with screenshots, then combining them and some generalist data, is a much healthier way to get a big handtagged dataset >if a model is going to make characters (instead of just objects), it needs subject variety to avoid turning every character into the most seen characters (there was a guy who made series styles with mass automated screenshots and autolabel, which worked but they a habit of making everyone's shape and clothes match the main character due to screentime bias) I'm thinking ~100 is probably enough for a general style. I can't tell you how big an impact hand-tagging (with or without suggestions from auto-tagger) vs. auto-tag only vs. cleaned up auto-tags (e.g., correcting 1boy vs. 1girl, replacing hand on hips with the actual booru tag) is in practice for styles, though consistent (a "blouse" is a "blouse" every time it appears, never a "shirt") is essential to getting the most out of low datasets for characters. For a character or something like a pose instead of a style, the answer is a lot easier and more definite: Results get way stiffer and less accurate with under 20 images and improve as you get to 30. After 30 images there seems be enough diminishing returns that excluding the lower quality images is as big an improvement as including more images. Extra costumes need extra images, but less than a whole new character. ~10 properly tagged seems to do it, but it seems to vary by outfit complexity and if you don't have enough the worst that will happen is they won't work or gens won't be a perfect match (e.g., if you only have 3 pics of a character in a particular bikini and add them to 30 of the character in normal clothes, at worst you'll just get bikinis with a vague resemblance to that official one when prompting for bikini).
(152.13 KB 900x1366 STK637980.jpeg)

>>53819 thanks for the answer. Seems like 30 per character, 100 for a style tends to be the sweet spot. Would you say it get's bad if you keep feeding it pictures after a while? <for example my LucasArts cutscene style model at 58 really sucks at adult women because there's only a small number of those in the dataset, and the default anime bias in the base model will overshine it on them if you haven't you could check on the artists other stuff to feed the model with. This is by Gary Winnick the guy that did the art for maniac mansion for example, the thing with artists is that they always have worked on other stuff
>Dexter is trash btw Midwit take
>>53938 >Would you say it get's bad if you keep feeding it pictures after a while? For characters? If they're all high quality and properly tagged, the only problem is the model will not only pick up the character's traits, but begin to make everything adopt the artstyle of the source more strictly (assuming you're feeding screencaps to get that many images rather than dozens of fanartists with slightly different styles). Most of the time for characters this is fine, it merely makes gens resemble the character's source material very closely , but it makes them less compatible with style LoRAs and makes them more expensive in compute time to train. For styles, no, as long as there aren't near duplicates (watch out, shows love to reuse cels or the digital equivalent of them across episodes/scenes and you start to notice when doing this) or a strong bias towards certain characters (e.g., half your pictures of females are of the main girl and another quarter are of the second girl you'll have all girls resemble those two very strongly) the only downsides to more and more images (at least up to 1k, which is the max Civitai's onsite generator can handle) is that you have to actually acquire and tag them and that it costs more in compute time to train the thing. I actually did look up Anson Jew's other stuff (he was the common name with the source material, and his website confirmed at least one of the scenes used from Full Throttle was drawn by him) but his style doesn't match enough, plus I wanted to capture the technical artifacts of the medium (beyond just being pixel art). Even the other LucasArts games from the time period with his name on them, Outlaws and Herc's Adventures, had their own distinct style.
>>53946 Savino took a huge shit on Dexter's legacy, while Jimmy died with some dignity.
>>54036 Planet Sheen sucked but the FOP crossovers were fun
>>54036 >doesn't post the car commerical >>54071 Yes. >>53938 Just made a style with 104 images of a cartoon. Working better but still struggles with inherent bias of anime characters. Might be 125, or it might be the weaknesses in the training data.
>>54071 i liked the one with the galactus thing, but i don't appreciate timmy trying to steal cindy from jimmy.
>>54081 >car commerical That commercial was a lot better than Planet Sheen as a whole.


Forms
Delete
Report
Quick Reply