Holy shit! I didn’t realize it’s been so long since I posted some news. In fact, I’ve been so busy with process improvements it’s giving poor Lupita a headache…poor girl.
March has been a crazy month! Aside from the real job encroaching upon my passions, we also have been crazy with spring break (Florida’s kind of a destination for that crap if you didn’t know).
So Roman, what have you been up to? Well, I’m glad you asked!
Well we were doing good with the streams but kind of went on hiatus while I figured some shit out. The good news is, I figured the shit out and now we can get back on track. What was the shit you asked?
During production of the Cassie chapter we hit a point in post production where we got seriously bogged down with verifying we had all the images. Making sure they were properly named. And then tracking down lost or missing images at the 11th hour to try and get everything together for release. It was a huge pain in the ass and a huge time suck. Now that I’m on the verge of releasing playtests for Lupita, I realized I didn’t want to spend hours prior to a release scanning images and cranking out last minute missed images.
By now I’m sure you know my philosophy, why do something I can program a computer to do? So I did. I spent some serious time upgrading the python script that turns my Scrivener text file into a functional Renpy script. Here’s what we added…
A script that looks at the output file and scans for all the images required in the script. It then compares that list to all the images currently in my directory of rendered images and if any are labeled incorrectly our missing, it flags and notifies me immediately. This way we fix errors as we go and don’t have to spend any time during post production tracking down or re-rendering missing images.
I’ve always wanted to structure it so that patrons can download the latest content without downloading the whole game. Now that v2.0.0 is up and running and is stable, we can do this. However, sometimes there are bug fixes, corrections, typos, and all kinds of other stuff that gets fixed between releases. Fortunately I use git, so I can just compare two versions and see what files changed.
But that still leaves me with having to manually go in and copy those files into a separate folder that I can zip and offer up to patrons. That’s when I said, fuck it! Let the computer do it!
So now I have a module that will read my git diff output file for all the files that have changed. Then it scans the game directory for those files and copies them into a new directory with an identical structure. Once done, it zips it and moves it to Mega. Once it’s uploaded all I have to do is go in and create a link.
This works for standard content as well as alternate game modes, etc. So no more hunting and pecking for files. Now we can upload the entire game for those who want to download it that way or just the new content for those who want to grab a smaller download and dump it into their existing game.
Additional Image Variation:
Those who watched some of the Lupita live streams know that we ended up breaking with tradition a bit and mixing up the image variety. We had one scene in particular where it made sense to have Roger visible if you were on the red path but not the blue. At the time, the converter wasn’t set up to recognize this combination. It was kind of an all or nothing. Roger is either there for both paths or he’s not.
Well, I took some time and tweaked the image-making part of the module and included some new notations I can use to define more dynamic compositions. So now we can expand our storytelling capabilities even more by having a wider range of image combinations available.
This took some time to program because I didn’t want to just write a notation for every possible combination because that would be too cumbersome. But I managed to develop a short hand that converter can use to then direct it’s own iterations over the image and build the different combinations itself.
So for those wanting more differentiation based on paths, kinks and choices, good news. We can!
Now that all this is complete, I can get back to rendering and streaming. It was important to get this down now because I want to use and test these tools during the production of Lupita’s chapter and see if we can’t use them to crank out the play tests quicker.
Provided everything works, then when we are completely done with Lupita’s renders we should be ready to launch final play test within hours and the update within a day or two. Unlike Cassie’s chapter where we finished rendering and then spent almost two weeks putting all the final pieces into place.
It’s all about hitting a faster turn around on the content.
Speaking of which, I’m in the process of putting music and final touches on the first of Lupita’s playtests and should have that out before the end of the week.
If that goes smooth, then I’d like to spend Friday and Saturday doing some streams and pushing forward with the chapter.
As always, thank you for your patience while I work to build us a faster, better process!
We’ll talk soon!