Blogs - skurtyyskirts

As each release of Remix comes out, I feel like my knowledge of how to use it grows with it. With 0.3 I learned the install process, the compatibility process and API wrappers, with 0.4 I had completed learning all of the upscalers and implementing those workflows into my projects. And now with 0.5 I'm learning more of the anti-culling techniques and remedy's when fixing a scene, along with scene scale. Culling is such a complex issue because you can't tell if it's Remix that's causing the culling issues, or the game itself. Remix has built in anti-culling options that correlate to the scene's scale size. The default anti-culling settings default to a large scale scene I believe, lowering the settings has completely fixed my culling issues in Tom Clancy's Splinter Cell (an experimental WIP).

Along with the release of Remix 0.5, a HUGE workflow automation improvement was shared amongst the Remix discord by Mechantie. They asked ChatGPT to create a script that would automate the process of replacing the albedo textures, normal maps, etc. by simply running the script. Before this, I had to replace each and every texture by hand. Replacing 1 single texture was a process of selecting 3 textures (normal maps, roughness, and albedo). I assume Remix will implement this auto-replace process natively soon as more updates come out closer to release.

I don't have a particular game I'm zoned in on at the moment. I keep cycling between Garrysmod, Tomb Raider, Battlefront 2 (2005), the COD titles. Each of them have their own hurdles to mount over when remixing, but they're keeping me on my toes. I'm hoping to get into Hitman more the following weeks. The hashes are very stable in these titles, I'm wanting to get better practice with lighting and finding a balance with exposure, intensity, etc. Lots I want to do, so many ideas with so many titles.

My first few RTX mods are out, and I'm pretty satisfied with how they're turning out. I'm getting used to the workflow of Remix and how to use the toolkit. I think the process is pretty straight forward once you've gotten the hang of it. I've been testing out a few games to mod with, and it turns out Source games are the easiest to work with in the toolkit. The trickiest part working with the toolkit is finding stable hashes. A lot of hashes are unstable, and because of that making changes that can be seen within a mod becomes challenging. It's trial and error, but I'm starting to find stable hashes easier and know what to look for.

For a program that's only in a beta release and only on it's second update, this software is incredible. Adding pathtracing to older games (the games I played growing up) is incredible to me and what's keeping me going. The community has a great Discord that's very active, with Nvidia employees that are so helpful and responsive - it's been really cool to see them take the time to answer questions and help us out. They're very open to feedback on their software and I've been able to pass down many ideas on how to better it.

Amongst all my mods, I'm hoping to put the most time into a Tomb Raider 2 RTX mod. An entire remaster of that game is my goal for Remix. I've gotten a few levels remixed, but the project is currently on hold. The project is only playable using a branch of OpenLara - but OpenLara is not currently programmed to work with Tomb Raider 2 just yet - at least not the desktop release. Once Tomb Raider 2's compatible, I'll be right back on the project. I actually prefer that it's worked out this way - this gives me time to practice my skills with Remix so when the time comes to remaster TR2, I'll be ready.