@TheShovel

i gave the interpolation model some further adjustments and tweaks and stuff to get stuff as good as i possibly could. theres surprisingly a lot of stuff you can tweak internally!

Hello! since my first ship, i have gotten a lot of little suggestions and bug reports, so i have worked on implementing most of them, along with other stuff that i felt would be worth adding along the way:

• replaced the ugly svg doodles on the home page with cool epic krita drawings made by yours truly. rip to the drawing effect though…
• i have re written the docs to be in markdown, and then parsed through a script to automatically build the docs from those, making it a lot easier for me to update them
• fixed a bunch of random ui issues and inconsistencies
• added filter effects to layers in the drawing tool

aaaand last but not least, themes! You can now choose a theme from a couple of pre-made ones in the settings.

I have also went out of my way to update the github readme to say how much stuff was made by me and how much was made by ai. Hopefully that way, i dont get falsely accused of using ai again. i am still salty about that >:(

the shipwrights keep marking my readme as ai, even after removing the little ai i used in the first place. SO HERE IS AN ENTIRE 16 MINUTE VIDEO OF ME REWRITING THE ENTIRE README, AGAIN. What a waste of my nerves bro. 3 times. 3 times my own hand written text was called ai. actually mental

And so, one more project comes to an end this summer, and probably my last one before school starts!

Since last time, i just did a lot of testing, gave it to people to try out, implemented some QOL features and thats it. Just polish stuff :D

heyyyyy. long time no see but its almost over!!!! i have started to port the krita drawing tools to the app, so you can draw your frames directly in the app!!!! ooooo crazy i know. it was actually a lot of work, and its still not done, but its getting there and im quite proud of it!!!
I also have added camera controls and effects! theres a separate special layer for camera keyframes, and you can animate the camera in all kinds of ways! we also have layer blending modes and soooo much more in the app itself.

I have also worked on making the website less “i vibe coded this in 20 minutes” type stuff. its getting along pretty well as well! Even put animated gifs on the front page :D

Great news:

Forza Horizon 2 finally gets to the title screen!

hiiii. been a while… but ive been hard at work! i have made a website, docs, tools for automating the docs, sound, camera and sooooo much more stuff. I also started work on a krita compatible built in image editor! So that you can edit your images directly in the software easily! Yep. well id love to write more but ill be honest i dont feel like it. Look at the github commits i guess lmao

oh also its now like, super fast. so fast, it can run on your phone, hell, ive been even working on a possible mobile version, but youll only see that if it actually happens hehe. ok byeeeeee!!!!

ok yeah im done for real now. This it. Last one. I just worked on improving its coding capabilities… a little. its still not great, but its still something. Pretty much making this devlog just to bank it the last few hours on this project.

hey yall!!! i finally have more stuff to show!
I did quite a lot of fairly difficult stuff, that im glad is now over. First of all, the ui is a lot better (still not perfect but better), now you can do interpolation on mutiple layers, and you can interpolate TRANSPARENT IMAGES. Thats right, i just made the first ever (that i heard of) actually good video interpolator that supports transparency. How i did it? Its pretty silly. I take every pixel with opacity in the current frame of that layer, and then i assign a opaque color to it. I make sure that the color that gets chosen is not a color that is in the image, to avoid breaking it… for the 2nd step! After i replace every pixel with opacity with an opaque pixel, i interpolate the frame. The interpolation model does a great job at full images. After that, i take all of the colors i assigned to the pixels i replaced, and do the reverse operation to turn them back into transparent pixels. And ta da! Its not the fastest approach, but other stuff that i tried, like masks and so on, dont seem as good.

Along with that, I have also worked on some dynamic coloring tools! Right now this is still pretty basic, but the idea is: you have a layer, you place these dots on it, you choose a color, and it will fill that shape with the color. Its pretty basic, quite slow and still buggy right now but stuff is happening!!!!

Hiii!!! The shovel, back at it again, with another BANGER. You know how it be. always on the grind.
So i was working on cutscenes for my passion project. Its a game blah blah doesnt matter. I was animating cutscenes frame by frame and then went like “man, this sure does look choppy, and not like in a cool stylized way. I wish i could easily make more inbetweens…” and then i realized, im a software developer and i can do it!!! So i thought of all those crappy “anime in 60 fps” videos on youtube. Those are ugly, and i thought… why? and realized that its because the artists didnt intend for it to be interpolated! So what if i make an animation editor, where you can drop your frames, you can set their timing and length and all that, and then you can interpolate them with full control!
So this is what this project is. So far you can drag frames in, interpolate them with a small onnx interpolation model that runs locally in your browser, you can then download the frames that look bad, fix them up, drop them back in and improve the interpolation! This sounds like it wouldnt do much but it would save someone like me soooooooo much time!

i think i am some kind of masochist cuz for some reason i keep coming back to this project. for some reason i still think this could be something bigger, even though i doubt ill reach that point. well, anyway, i started working on its coding capabilities! still pretty basic and rough, but it does work and it does seem to scale decently! you can ask it to make edits, add logging and explain the code and so on. pretty cooool ok back to data engineering hell i go

i was playing LSD dream emulator for the ps1, as you do, and then went “how hard would it be to port this game to computers”

so now im porting LSD dream emulator to computers.
Im using the efforts of the WIP decomp and the fan remaster, but trying to make it as faithful to the original as i can. Right now, i got some maps to load (kinda), some shaders to work (kinda) and sound (kinda).

Very early, but still something! Also its incredibly laggy sometimes. Decoding that old poopy ps1 format to modern formats is not the fastest in a browser

ok, im gonna be so deadass, im so done with this project. This is it, this is the state im leaving it in. It was fun at the start but now its just very frustrating and boring to look at walls of data go on my screen and me going “yep that looks good enough”.

I want to work on cooler things that interest me more, so this project, is done.

Since last time i implemented:

• text editing
• coding capabilities
• typo detection and fixing
• more templates and so on.

Thats it. It was a interesting research project, but i think i came to a conclusion: The reason why no one did this, is because it is stupidly tedious. Training a classic LLM is so much easier even if it destroys the environment and costs a lot of money. Its a lot less human work, which i guess is more valuable than anything. If we should learn a lesson from this, is that humans cant be replaced, and our efforts are worth more than we think.

Also i made a website

First of all, i again didnt do devlogs every day because i wasnt sure where this project was going. I couldnt tell if it as going to work or not or if it was even viable. But, now i think i can start making some actual consistent devlogs, because i think its working!

This is a no-inference chatbot, its meant to function like chatgpt and all the other slop generators out there, BUT, without an LLM AT ALL! Thats right. Its all just hand made agorithms, web search and data association. This project has been a pain in the butt to work on due to how tedious it is, but its finally starting to shape up.

As you can see, it can respond to basic greetings, write essays, iterate on them and so on. It fails quite often due to its lack of information but thats nothing that cant be fixed with more templates and data. Ill be going back to the grind and try to improve this further, because of how much time and effort building this takes idk how much further i can get it, but ill try my best. When i reach a point where i go “ok yeah this is all i feel like doing”, im going to build an api server people can host and document it and call it done.

This really is more of an experiment and research project rather than a thing i know i can give people to use in a “production” environment if you know what i mean.

Ok yeah anyway see yah mwah

Hi! so, i am making this devlog with 15 hours on record because i was honestly not sure if this would work. In my last stardance project, i tried making an agentic chatbot utilizing LLMs, but trying to make it as light and as compatible as possible so it can run even on my poopy 7gb of ram laptop.

That whole thing got me thinking. Do we even need LLMs? Do we even need generative AI for this?

So this project started. This is an attempt at making agentic AI without LLMs, or weights or anything similar. The entire idea is that this can be fast and easily modified without having to use an entire datacenter to retrain the model, because there is no model.

Right now, i have worked on a conversational “engine”, that basically puts words together algorithmically by detecting certain properties from the users prompt. Its not very impressive right now, but i think im onto something that could be quite a big deal if i get right. I dont even know if this project will be shippable but ill still document its development because i think its interesting. Alright smooches! See you later! <3