The glass shader will need a closer look soon (prototype) image - Star Wars Jedi Academy Remastered mod for Star Wars: Jedi Academy

Ive found 4 bugs in the game. Z fighting in 3 places, and shadows rendering through structures from character models outside of them. One is the chair (ending), and the shadows in this mission (by two doors with characters above waiting in ambush on roof), the droid recovery mission (outside high ground), and the crime lord mission (elevator).

Most people would never see them but because im being extremely thorough I notice them.

Also considering a pact with .JPG 4:2:0 sub 100% more than original philosophy allowed for. But honestly things like sand textures dont need super radically varied beyond human appreciation detail complexity. A sort of sameness version in these noise fests equals what compression artifact do you speak of? Looks the same in this case really to humans, and saves hundreds and hundreds and hundreds of MBs. With fine tuning at god level resolutions, using this slider in appropriate scenarios is a way to get more texture resolution bang for buck since its by design supposed to throw out data that makes it overall look the same really to a human observer. Resolution is king. There comes a point otherwise undesirable compression technology like JPG passes a threshold where its a Genius option despite how primitive it is nowindayz.

I mean ignoring the stupid shader nonsense ill have to isolate the file for and figure out (fix), that doors a solid 16k 4:2:0 100% .JPG one lossy pass. Just like the walls that are different textures on both sides of it. Its very detailed as you can see, and its reasonably so when some assets have really stretched out textures that flat out require the maximum texture rez the engine will put up with no matter how small it is in memory consumption to not look awful. When you dont have detail texture tech to the extent of Aliens vs Predator 2 (to its inferior DX7 credit) you can still just brute force resolution your way into detail texturing everywhere, and JPG actually makes God resolutions very mundane and tolerable. Its not arbitrary where I am ending either in getting to such a large size texture mod. Im truly going to MAX out id tech 3, which has wildly superior potential to upgrade compared to id tech 4. And I am not just blindly upscaling anything. Everything I do has to be visually visible in 3rd person even getting the camera up close, which it might sometimes do just because. To make 1st person detail texturing work in terms of what the engine can actually render, youll need proper detail textures, which would be the way to optimize the mod size if someone wanted to program that layer into the game. If im understanding that correctly. Im not as sophisticated as someone like that is, but ill happily brute force the groundwork for a newer version of the game.

The switch texture modding it to be superior would take a monkey. 2x and go. Its 720P trash rez. But a PS5 release with full blown path tracing and in and out but extensive visual eye treats. Thats nerd crack. Or maybe make it a timed exclusive for the newest X Box. Paint a bunch of green in the marketing and blow everyone's mind without necessarily a total engine rewrite or crazy millions of dollars. Keep the budget reasonable and reheat this classic and open the microwave oven for the good of mankind.

+ its kind of cool. This is literally an engine from 1999 that with the tiniest little bit of help from where its going to be left off on my end of ability, that starts looking like a legitimately modern game until you see the face geometry or some of the lower polygon level areas. Which gives id tech 3 its final chance to historically shine as the superhuman **** that it is with cybernetic enhancements in place. Gives very real vibes its incredibly advanced for its time. And this is bootstrapping it under the x86 maximum size law here at Remastered Retro PC Games. Too bad moddb name size limitations are too oppressive.

Its almost like taking a RX7 FD and dropping in a LS9 because: yes we can.