Yes, I use it to get the depth value to the scene which reduces how many times I have to ray march later.

Not necessarily, if the object is really far away then yeah it will lose the detail since the number of pixels the object will occupy on the screen will be very small, but if the object is very close to the camera and is occupying a lot of the screen, then it doesn't matter. I guess to answer your question: its relative to how many pixels the object is occupying the screen. Oh and thank you!

I'm glad the video helped. For your question though, I am assuming you are taking about how the ray marching plane is being calculated from the camera. The idea is to keep the plane on top of the near plane of the camera right, so calculating that position of the near plane globally is what we are trying to find. the near plane is always in front of the camera so we first get the global position of the camera, then move in the direction that the camera is looking at by how far the near plane is. Now when I look back at the video and the article, you are right. I should have explained the details of that and probably some more illustrations to showcase that in action. I am new to the tutorial making scene so I'll try to be more meticulous next time. I hope my explanation here helped though

Okay I came up with a quick illustration to explain what I mean. I don't use excaldraw for sharing so I hope this link isn't broken after a while excalidraw.com/#json=QI6S4Ify4eEwRAQ0GZE_q,NtoAV0HPR4RSaF9Yc4zerA

Thanks!

Glad you liked it!

The style is perfect for developers.

Do you have more stuff on three js? An just another question. How did you learn raymarching?

@@richardramirez5746 Glad to hear that!

@@richardramirez5746 I will soon. I will be making one more video about the ray marching optimization and it will also be in three.js. As a side note, I am trying to transition my content into three.js so I'll prob make more in the future. As to ray marching, I essentially started with watching inigo quilez's live streams. I then learned shader programming and then tried to make a whole game with it. U learn by actually coding. U know what, that actually might be a good video idea 🤔

@@nabilnymansour Great, so first viewer here! I am subscribed.

I also think that making the tools for it is not that easy or straightforward since you want artists to be the ones using it and I can't find a simple way of making a nice and seamless way of doing that. I know that womp.com/index have tried to do that (I do think they end up polygonizing the SDF, but still it's pretty good).

Well that is one way of doing things, but you can also make it as a post effect shader which is what I am doing. You could also make separate computerations to see how far your player is from the terrain (which can be used for collisions). That is what I do in the game. You should check out my rendering engine since I have some demos about that github.com/NabilNYMansour/Unity-FractiX

Yeah no problem, I am using this SDF: float sphereFold(float4 z, float minR, float maxR, float bloatFactor) { // bloat = 1 will not change size. float r2 = dot(z.xyz, z.xyz); return max(maxR / max(minR, r2), bloatFactor); } void boxFold(inout float4 z, float3 r) { z.xyz = clamp(z.xyz, -r, r) * 2.0 - z.xyz; } float apollonianHills(float3 pos) { float4 p = float4(pos, 1.); float3 col = float3Init(1e20); for (int i = 0; i < 11; ++i) { boxFold(p, float3(1, 1.5, .5)); p *= sphereFold(p, 0., 2, 0.02); } return p.y * 0.2 / p.w; } idk if sending code here is good, but Ill make a couple of shadertoys later that will have all the maps. I've just been super busy with other things and haven't been able to focus on youtube recently. I hope this helps though!

@@nabilnymansour Hmm, this seems like more of a variation on a mandelbox fractal than a sphereflake. Thank you for sharing the code though.

Haha, I've never really gotten a job with shaders, you can email me if you have more information. As to templeOS, I haven't really used it before so Idk much about, but I think it's doable. I am busy these days, but around the end of this month, I should be done with a cone marching tutorial.

Thank you!

Glad I can be of help!

That is very true, I suppose the only thing about this is I don't really know how to make that into a game mechanic...

@@nabilnymansour how about as a replacement for the standard collision mechanics? if (distance-speed < threshold): scale

Hmm 🤔Thanks for the suggestion.

Oh yeah, the final game actually has a different sound all together!

Thank u!

would be interesting to see if it would be possible to texture the fractal procedurally and maybe spawn regular assets that way too. e.g. having water under a certain height, and have the fractals look like sand or stone caves, and higher up could be mountains and plant like fractal colouring plus some plant assets maybe

@@user-eh5wo8re3d I do think thats all possible. My idea with this game is to try to control it in some fashion. I also made the rendering engine open source so that others could also experiment with this. I still need to document it though: github.com/NabilNYMansour/Unity-FractiX

Tha could be an option outside of the game itself, but if its in game (while ur flying) you get really dizzy and nauseous. But not a bad idea!

It took me 3 years to learn shaders and how to effectively use them (though I wasn't giving 100% of my time for it) and it took 3 months of work to make this unity project. The rendering engine took 2 months, and the game itself was 1 month.

Soon 😭I've just been busy with other stuff recently. However, you can actually check out the code right now. I made the rendering engine open source: github.com/NabilNYMansour/Unity-FractiX/

I did want to have some dithering effect applied, but I would have to go back to my custom rendering engine and add that functionality which seems a bit more complicated .

Though now that I think about it, it might be not as complicated as I thought it would be. Hmm, I'll experiment with this later. Thanks 👍

u read my mind, i had a lot more ideas that i wanted to add, but I really didn't want to push the scope of the game further from what it is. I was really scared of feature creep so I kept on saying to myself "just stick to the original plan". But yes, I do agree that at some point, the game does become repetitive. I might fix some bugs that I've been informed of, but idk if ill end up adding really new big features and script different logic. Its not like I'm 100 certain I wont work on this game from now on, but at least for now, I don't think I'll be adding stuff.

​@@nabilnymansour Sure! They were just recommendations, but I was saying that because I see potential for an extremely good incremental/Roguelite game, like the new game HASTE from landfall. But I struggle to keep my scope small in my own games and I see you are doing well in that sense so it's a good choice.

@@o1497 Oh yeah for sure. The focus was really to explore making the fractals and turning it into a game, but I do think taking that aside, the gliding mechanic has a lot of potential and there are many things that u can expand on it.

You are very kind thank u so much 🙏

haha i suppose. In that sense this is a 3 years project. But in terms of the actual hours I've spent on this exact unity project, I spent 2 months before the fall semester doing the rendering, and I spent this Jan finishing it up with boids and other levels.

Thats a good point, but i am a bit afraid of adding another light source since I actually have to ray march to every light source when calculating the light for the scene. Though u r right, this is something I overlooked...

After thinking about it, I might be able to change the color of the terrain based on how close the player is to it, that might serve as a way to tackle the issue you mentioned without really making another light source. I'll test that out when I get some time.

@@nabilnymansour sounds cool ! i do feel like that'd enhance the feeling quite a bit and make the game more enjoyable

I hope u enjoy it! The demo has the first 2 levels unlocked, the full game will have 3 more

I think I had it turned off maybe when I recorded this since I have added AO to my rendering engine. Also, this is just a proof-of-concept game so I haven't thought much in terms of a story 😅. You can check it out on Steam right now. I have a demo also if u wanna try it out. The game is called FRACTAL GLIDE.

@@nabilnymansour Awesome. I totally will

🙌

Thank you!

Hmm, I think its not impossible, but the logic might be a bit expensive. I am trying to think of different ways of computing the closest distance, but I would imagine that if you can query the environment in a way, then you can make some type of navigation logic. I wanted to do that really with the mengerites, but saw that it would be too expensive. TLDR, I 'believe' its doable.

Honestly, same 😴 It has been a bit tiring working on this. There was so much more I wanted to add, but feature creep is something I wanted to avoid as much as I can.

Thank you!

Okay and?

Thanks

haha maybe, the new version has a different sound though

Hmm, I did put a slider for the volume in the final build, maybe I can do a slider for each component of the game. We'll see.

I will make a tutorial about raymarching and conemarching in the future and I do intend to have my rendering engine be open source. so stay tuned 👍

I had an idea to incorporate the construction of the fractal as a game mechanic but its not very easy to visualize and idk how I would make that fun. But u r right, the concept of fractals can be expanded on in many ways. I will make my rendering engine open source after release but for now I think just to test if fractals is even something that would attract people is enough. also for making the fractal change, that is doable but as you fly in the game, having the terrain change ended up making me nauseous when working on it. I do think its doable, but experimenting on that just gave me a headache lol