Direct from Patreon - Late Aug News Update
You must be registered to see the links
- This is a long-ass post. I separated it into parts so you can skip to the part you want to read
- The First part is Technical stuff.
- The second is me just sharing stuff.
- The third is some ideas I have
Technical stuff
Animation Layering:
In the last post, I described my attempts at this goal.
In my last attempt, I described my idea of storing the partial animations on a map and using the frame number as the key and the animations pixels as the value.
So I tried doing that, and once I stored it as files on my pc, they were too big.
I'm talking around 14Mb for 1 frame, which is not acceptable.
To reduce the file size, I compressed the files using Zlib, which is built-in in NodeJs, which you can utilize when using Electron.
Once I compressed the files, their size reduced significantly. (56kb)
and the total animation size was around 14Mb.
But this was not it...
Once I tried putting everything together, another issue occurred. In order to replace the pixels of the main animation with the partial one, I needed to loop over all the pixels while the animation was playing, which caused major FPS drops and dropped the animation from 24FPS (The FPS I'm using to render) down to around 16-20.
Attempt #5:
My new goal was to paint the partial animation on top of the main one, checking for pixels.
But the problem was that the canvas I'm painting on requires a height, width, and pixels array, and when I'm storing the pixels into an array I don't know the height and width it creates.
And yes, it's possible to calculate that, but I felt like I'm creating Skynet instead of some minor helper function that should help me improve the quality of the game.
Attempt #6 (And final):
I went back to the start and tried a completely different approach.
Now I'm at a point where I can fetch the frame's number while the animation is playing.
I realized I could just store the partial images as normal images (Webp) and paint them on the canvas.
Surprisingly, it was even smaller in size compared to the JSON files.
And since they are images, I can crop just the needed parts!
So now, I only need to preload them into the memory before the animation starts (which is really fast) and then paint them by frame number.
Here is a POC:
You must be registered to see the links
#wdSC0SNP6A_eMDs7XcpT33EbGcwk1brOtxX75rLS8eQ
You can see that I fucked up with the cropping in the beginning, so you can see a bit of Ann's black curly hair on the right.
I didn't fix it because that's how you can see better how it works!
TTrick's Sharing
Heart-to-heart talk:
This part might seems like it comes from nowhere. But I just felt I needed to write this, so I am.
I wanted to share with you why the upcoming version is taking so long.
As you know, the last update was about Ann's love path.
I've already mentioned how hard it was to write since it's not my strong size. But I always wanted to improve that route, and I feel like I did.
I started with that route because I got a lot of criticism about it, saying I don't give it as much attention and love as the others.
Even though I tend to ignore most criticism since people tend to complain about everything. That one was true, and I knew it, and I wanted to improve it for a long time, and I felt like it was the right time.
So the day came, and I finally released the version and was really happy with the result. I was also so motivated and hyped that I didn't even take a single day off! I immediately started working on the next version. (Even before I published the poll asking for you to vote for it - I worked on scenes that I knew would be in all routes).
But then a couple more days passed, and nothing... The version was by far the most "quiet" version I've ever released.
I barely gained followers and supporters. No one talked or cared about that version on various sites.
At one point, I actually checked that I clicked the publish button because I thought there was some kind of a mistake. But there was none.
Not only that. I even got a few messages from people who have been my supporters for a long time saying they are dropping tiers or canceling for a few months.
Needless to say, that wasn't the most fun experience.
Devs work on the game for a few good months to release 20-30 minutes of content at best. And what makes them keep pushing is seeing the people get excited about their work.
And I know some people will think to themself now: "He should make the game because he likes it and not to please other people."
To those people - Well, I AM making the game I want, but if it was only for me, I wouldn't need to put that story into writing, I wouldn't need to fix poke-through holes in renders, and so on.
Anyway! It made me realize that I tried pleasing some vocal people that aren't my supporters and apparently wouldn't be in the future anyways.
Apologies for those here reading this, and this is their preferred route - I'm just sharing my own experience and how I saw things.
It also doesn't mean I'm abandoning this route or won't give it more love in the future!
So even though I worked on the game, I also feel like I didn't dedicate enough time to it as I should've, so I'm going to pause the payment for this month.
I will keep working on it, but it will make me feel better doing that.
I'm not going to make any changes or stuff like this. I'm happy with the last version and am always excited to work on LSS.
If I had a PR team, they would probably delete this part, but since I love to be transparent for good and the worst... Either way, you get walls of texts that most would probably skip.
Ideas:
So while working on the animations layering project, it brought up some ideas that might be cool ( or not ), depending on you.
The best part about that project was that I had a stage to share my progress with you.
I was thinking of doing this as a regular thing. But this time, I could list all the projects (or other stuff) I need to work on, and you could vote on what you want to see first.
It could be anything!
- From LSS girls in a bikini contest (Suggested by Prowly on discord!) In this idea, I will upload the girls in swimsuits and show you my thought process and how I chose each girl her swimsuit.
- Or it can be even DAZ tutorials. I could post DAZ tips that I learned over the years.
- It can be some coding stuff for the engine - for example, me coding in a gallery or a reply gallery (finally) and showing you how I'm progressing.
If you worry that I'll just invest my time in that instead of working on the story, you shouldn't worry. Because I will do it regardless, the only added time will be me posting about it.
Let me know what you think!
If some parts of this post felt like I'm depressed, so far from that! I'm happy with the game, the community, and my life in general!
Also, I have many private renders that I shared on discord that some of you that aren't there might have missed!
I'm going to share those in a separate message tomorrow!
Cheers ,
TTrick.
If your still reading this.. maybe show him some support guys.. not always has to be financially.. maybe a little verbal written support from time to time.. he deserves it.. (Crowangel)
