I spent some time messing around with WebGL trying to render some better looking 3d mazes, which I have no doubt you can do. There are plenty of examples of WebGL being used to make some captivating 3d content and that challenge is enticing. However, it wasn’t long until I came across WebGPU, the Next Generation™ of 3d rendering technology for the web. Digging in a bit, it felt very familiar. The API is very ‘Vuklanesque’, inspired by it, as well as other modern graphics APIs (Metal, Direct3D 12 etc.). ...

Not a faithful recreation. I believe the Microsoft one was following the left-hand rule solver (and right when you flipped upside down), which means the mazes must have been ‘simply connected’ or ‘perfect mazes’. There isn’t the topsy-turvy geodesic shape, psychedelic fractals, OpenGL logos, Smiley Face end-point or… rat either. This one is using the Recursive Backtracker to create passages and then Dijkstra’s to find the longest connected path through the maze, and then it follows that. ...

Probably a surprise to no one, but while going through the maze generation exercise I thought it would be nice to display a maze in 3d. There is (though I haven’t gotten to it yet) an entire section of the Mazes for Programmers book dedicated to generating mazes that wrap around 3d shapes, like mazes on cubes, spheres, even a möbius strip! Displaying a maze in 3d isn’t in the scope of the book, but it’s not hard to take the concepts and translate them to 3d. ...

Apparently my last devblog entry was just over 3 years ago today, and… well, a lot has happened in that time. Professionally, I moved from supporting and writing tools for animators in Maya, of which I had years of experience, to supporting and writing tools for FX artists in Houdini, of which I had precisely zip. Today, 3 years later, I can say with confidence that I have only just started to scratch the surface. ...

took a break from the renderer as the next step involves getting render passes setup so I can render shadows and do fun stuff like post processing etc. This poses two problems for me.. Once I get started, I won’t have a lot to show for a while because I’m sure it’ll involve a lot of refactoring work I actually have no real clue what I’m doing, or what order makes the most sense to do things in Thankfully on the latter point, I found a YouTube channel called “The Cherno” which has an incredible amount of content, one of his series is an educational series about building a Game Engine from scratch which is very pertinent to my interests, his Hazel engine is definitely a much more ambitious and fully featured engine than I was ever planning, but it’s been very helpful to see how a professional approaches engine development. ...

You know that thing you do when you’re learning something new, so you make every mistake under the sun and / or massively over complicate / engineer things? Yeah, I did that. As I mentioned in my last post, I did get textures working – but there was a large caveat. I only had a single shared descriptor set, which meant all meshes shared the same textures. So I did a bit of refactoring ...

Now we’re all caught up, I reached a bit of a milestone yesterday. I managed to get textures loading and I’m fairly happy with the solution for now. This one is a milestone for me because I really wasn’t sure how to organise my resources, so the “final” design was a bit of trial and a lot of error. Right now, if you request to load the same texture twice, it’ll oblige no questions asked, but the design allows for future optimisations in this area. ...

The following is progress made over a few weeks, a few weeks ago. I was still wrapping my head around how Vulkan works and really just trying to get stuff to work. It was one step forward, two steps back for a time, that’s just how it is sometimes. Jumping off from where I was at the end of the last post, I implemented a camera controller which at this time handled all user input. I spent a fair whack of time coming up with a way to handle what happened when the mouse goes outside the bounds of the window, I was quite proud of that. I later found out that GLFW (the cross-platform window framework I am using) has an option to do this for you, you just have to turn it on (and know it exists). ...

Static Rectangle Engine lives! I started this project a few weeks ago, the goal? make my own game engine. Why? purely educational reasons honestly. In my humble (and uneducated) opinion, there are a number of reasons someone may want to build their own game engine today, even in the land of Unity / Unreal and Godot etc. Full unbridled control and comprehension would be first and foremost. For me this is rather important, over my many years I have seen projects of great potential buckle and fold when those appointed to maintain them weren’t given the time to fully grasp them. ...

It has been many years since my last attempt to set up a blog (according to my web host 10 years to be precise) and you would think that in that time it would have been made easier. Well, dear reader, it has not. Anyway, I have set this devblog up with the hope of keeping my motivations up and to share progress on the seemingly endless array of projects I toil away at, perhaps I may even finish one of them. ...