I will at some point. When I'm sure it's ready. The goal has always been to make a modder's resource. Long term, I don't want to become the person everyone asks to shinify mods - eventually I want it to be so easy to do that there is just always a shinified version available. But I also don't want to spend a ton of energy supporting it if it's half-baked.
I can tell you HOW I do it - XEdit has a seldom documented script that converts .nif files to a readable .json format ('Nif - Batch JSON Converter'). So I do that. Then I use a set of powershell scripts I built to regex my way through them all, replacing relevant texture sets and adjusting shader settings to match SRShiny as closely as possible. This was a LOT of trial and error to get right and really pushed my regex knowledge beyond its limits, so I had to brute force a LOT of tutorials. But in the end, it works. Then run the XEdit script again to convert back.
What I'm trying to do currently when I work on the suite is make it more user friendly/readable, and also build in some filtering methods to make it even more time efficient by default. This process involves a lot of ForEach loops, which means running through all the files in the project, in order. In its 'worst,' most inefficient version, with the filters turned off, a big mod can be done in a few minutes - keeping in mind that it used to take me hours by hand, I'd call that pretty great! With smarter filtering to ensure it only loops over the files it needs to (say, by only running the 'Iron' script on things named 'Iron' or in a folder named 'Iron'), I've been able to cut that down to a few seconds for some mods, which is phenomenal. So I'm working on making the fastest mode as effective as possible.
That's for vanilla style mods, understand. Anything involving mashups of vanilla texture sets, or mods that use unconventional naming schemes, demands the unfiltered mode ('minutes'). There's not a lot I'll be able to do about that except recommend that users go make a cup of tea while it runs.
Anyway yes the script will be released, I just want to make absolutely sure I'm not throwing gasoline into a dumpster fire first.
These looks pretty dang nice n smooth. Any plans to touch up some modded armours, like the ones in Immersive Armors? Obviously after you're satisfied with the vanilla and DLC armour sets. Thanks for making this mod, will try it out in a future playthrough.
We'll see. I'll probably start with meshes I'm actually still using. Eventually I'll publish the resource so other people can do converts this way too. That's the only real 'plan,' everything else is just doing whatever strikes my fancy.
My hope was always that I'd put out a tutorial and my scripts and then the envmapping would spread everywhere. So far it hasn't happened, so it's time for phase two - get more public about it.
For now, I'm going to keep up with TBOS's progress on Talos Blessed and do my best to get shiny versions covered for that project. Besides that it just depends what grabs me.
9 comments
This is helping me revise my scripts for shinification very quickly. Plusses all around.
I can tell you HOW I do it - XEdit has a seldom documented script that converts .nif files to a readable .json format ('Nif - Batch JSON Converter'). So I do that. Then I use a set of powershell scripts I built to regex my way through them all, replacing relevant texture sets and adjusting shader settings to match SRShiny as closely as possible. This was a LOT of trial and error to get right and really pushed my regex knowledge beyond its limits, so I had to brute force a LOT of tutorials. But in the end, it works. Then run the XEdit script again to convert back.
What I'm trying to do currently when I work on the suite is make it more user friendly/readable, and also build in some filtering methods to make it even more time efficient by default. This process involves a lot of ForEach loops, which means running through all the files in the project, in order. In its 'worst,' most inefficient version, with the filters turned off, a big mod can be done in a few minutes - keeping in mind that it used to take me hours by hand, I'd call that pretty great! With smarter filtering to ensure it only loops over the files it needs to (say, by only running the 'Iron' script on things named 'Iron' or in a folder named 'Iron'), I've been able to cut that down to a few seconds for some mods, which is phenomenal. So I'm working on making the fastest mode as effective as possible.
That's for vanilla style mods, understand. Anything involving mashups of vanilla texture sets, or mods that use unconventional naming schemes, demands the unfiltered mode ('minutes'). There's not a lot I'll be able to do about that except recommend that users go make a cup of tea while it runs.
Anyway yes the script will be released, I just want to make absolutely sure I'm not throwing gasoline into a dumpster fire first.
My hope was always that I'd put out a tutorial and my scripts and then the envmapping would spread everywhere. So far it hasn't happened, so it's time for phase two - get more public about it.
For now, I'm going to keep up with TBOS's progress on Talos Blessed and do my best to get shiny versions covered for that project. Besides that it just depends what grabs me.