Today I had to present my final year project called segmentation and detection of glioma using deep learning.

It took 15 minutes for the evaluators to understand what an mri image dataset (BRaTs) dataset looks like (they are voxels and not pixels). My god, these peasants...

And I was there expecting them to understand down sampling convolutions and up scaling convolutions of U Net model 😂. Life is so convoluted right now!

Finally getting off my proverbial ass and doing something about the lack of games I like. Going to focus on making an engine for the kind of games I want to play.

No, I am not starting from scratch. Going to base my engine on Godot and use it for my own titles. I am not insane. Making it from scratch is too much work these days. But the indies are shifting from Unity to other engines right now. So a lot of wanted attention will be placed on better alternatives. This means more content and plugin choices will be available to Godot devs.

I kept making excuses as to how hard it will be or it will take forever. It only ended up taking me further away from what I wanted. I have my wishlist of features and I will focus on modularizing them so they can be used as needed. If it makes sense I will make these modules available to the community at Godot. This will help get feedback on what can be improved and generalized further. It will also reduce development costs in the long run. I want to take the approach that No Man's Sky has taken for content and generate as much as I can. I am fascinated by generating objects using algorithms. This seems to be a trend in games.

The struggle I have with games: I want to build things like structures in game (aka Minecraft), I want to build characters in game (aka RPGs), I also want to deform terrain (aka organic voxels), and I want a mixed genre (guns and dragons). Nothing like this exists in a form I want to pay for. I also want to be able to mod the game and for other people to be able to mod the game. That really narrows the list of games down to nothing. Sure there are few games that hit these bullet points, but not all in the same game.

I am finding I struggle to be engaged intellectually at work. I do what I have to for a paycheck. I think having a side project will help with this. One that is radically different than what I do at work is going to be helpful. I need to be realistic about expectations. I probably shouldn't expect any real progress for at least 2 to 3 years and probably more likely 5 years. I have some experience with the tool chains from other engines I have worked with. I also want something that I own and is mine. Even if it sucks.

If you guys remember, i was teasing from time to time, that i'm working on some Rust Project in my free time.

Well here it is, i put up a whole bunch of Editor Windows in it, to showcase it a little bit. (It also reminded me, that i need to update the Version to 2024.01).

It's essentially a toolkit, with which i can create all the content, that is later used as a data basis, that is being fed into the Client + Server Combo of the actual Game. My Plan for this year is to go beyond the Editor and create a first version of the Client + Server to be able to playtest the stuff.

And sorry if it kinda sounds like an ad, but i'm more posting it here to show, how nice it actually is to build stuff with rust.

Let me know what you think ^^

I started down my career path to make games. I have never really made anything good. I think I am finally fed up with not doing this. So I am going to be working on selecting a game engine. I want 3D, I want to experiment with voxels, I want a permissive license, I don't want something huge, I want to contribute back in a meaningful way, it needs to support 3D. So after looking around I found Godot. Another programmer who lives near me uses this as well.

Does anyone else have some good positive experience with game engines for smaller projects? I have played with UE4 and Torque3D/2D. I don't like the bloated feel of UE4 even though its a very cool engine. I didn't like their install system at all. T3D is old and not up to date.

Algorithm suggestions for 2d and 3d volumetric combination of voxels?

I built an image to voxel converter for exporting to the game Avorion over the weekend. I am using a naive approach and treating each pixel as a box shape. I need to learn how to write an algorithm for combining voxels with the same color into rectangular cuboids. This will reduce the imported shapes. As is the game has issues with 64x64x1 images on import. It would be good to reduce the object count by creating cuboids with the pixels that can be combined. I would like to learn to write the algorithm for both 2d and 3d.