🎉 Celebrating 25 Years of GameDev.net! 🎉

Not many can claim 25 years on the Internet! Join us in celebrating this milestone. Learn more about our history, and thank you for being a part of our community!

Why the sudden boom in marching cubes? [Possible target]

Started by
20 comments, last by Turbo14 6 years, 2 months ago

So what is up? What did I miss?

I finally get my PC working and am bombarded with emails asking about marching cubes. Get on Gamedev.net and it's even here people are researching marching cubes. All of the gaming forums is a buzz with players theories on how marching cubes work, developers looking for teams to build a marching cube games.

So why the spike? Reminds me of when Minecraft released.

Advertisement

Just random coincidence?

Although i consider it as well at the moment as a intermediate format to convert inefficient poly soup to something more efficient.

But there is growing interest in SDF, voxels as well - anything different from triangles. This happens all the time since games like Comanche.

We want easier editing (user generated content as in Minecraft), more details, non static worlds... stuff like that.

 

Personally i'll probably prefer Elastic Surface Nets - see here for a comparison: https://0fps.net/2012/07/12/smooth-voxel-terrain-part-2/

5 minutes ago, JoeJ said:

ust random coincidence?

Baader-Meinhof Phenomenon? Although I doubt it because marching cubes is something I explored right after Minecraft. Google doesn't show a rise, yet it feels like something.

It's strange because marching cubes is used in games already so why the interest now, it's not like a new game using marching cubes has been released?

Maybe it's just the start or maybe it is just coincidence, even so there could be some real progress to be made.

Found the possible culprit: https://wildmagegame.com/ I missed the Imgur post by a day I think https://imgur.com/gallery/1JfQa

Asked the players on the other forum what go them interested. Google has some correlation between this game and marching cubes. Looks like a 1:4 ratio, not much but still a possible target.

The main reason this could be the one is how the game is posted. The developer constantly talks about the game as if it was made by one person and the game looks like it could be using Unreal from it's environments.

Never heard of this game, but looks awesome :) 

There is also new Atomontage video with lots of media coverage (voxels but volumetric).

... and related upcoming games like Dual Universe, Dreams, Claybook

 

9 minutes ago, JoeJ said:

Atomontage video with lots of media coverage (voxels but volumetric).

... and related upcoming games like Dual Universe, Dreams, Claybook

This really could be just a coincidence. I am getting different answers from different places. Almost like everyone saw a marching cube game and all decided at once that it's the type of game to make.

Personally I think this could be good, environments in games always feel dead, would be nice if more of them interacted with the players.

11 minutes ago, Scouting Ninja said:

Personally I think this could be good, environments in games always feel dead, would be nice if more of them interacted with the players.

Yeah, i'm not convinced of Atomontage or such things, but Voxelfarm for example seems really nice tech for user generated content and destruction. I always wondered why no larger game came up using it since Everquest has been canceled. IIRC It has plugins for UE and Unity and is not that expensive.

Well I just posted a MC tutorial on GameDev just yesterday. I had been working on that tutorial for a long time and picked Sunday due to personal timing, so I expect it's just co-incidence.

5 minutes ago, Boris The Brave said:

I had been working on that tutorial for a long time and picked Sunday due to personal timing, so I expect it's just co-incidence.

I could pass this as co-incidence because I know from personal experience that making a tutorial takes a lot of time and planing.

However it's not only your topic, there is two other on gamedev.net and a few similar post cropping up on other game development forums. Players are talking about it and so are other developers.

One more thing it could be is that maybe the anti-hype around No Man's Sky settled down so everyone who wanted to clone it's mechanics is now trying.

 

I don't know what it is but now seams like a good time to invest into marching cubes. Could be that it is what the players around the world is looking for.

37 minutes ago, JoeJ said:

Yeah, i'm not convinced of Atomontage or such things, but Voxelfarm for example seems really nice tech for user generated content and destruction.

It's the same problem as Euclideon in that it looks like a good idea but polygons are so far in the lead that even with huge amounts of money there is no way to close the gap: https://www.youtube.com/watch?v=gM6QkPsA2ds

The main thing most voxel engines forget is that polygons do in fact use the same data types as voxels. A vertices is a voxel. So investors can keep investing in polygons and get a quick return and this advances both voxel studies and polygon studies. That is why we have global illumination based on a voxel like grid and use lots of volume tricks.

 

The main problem and the thing I think most relative to this topic is that voxels are static by nature. Tracking every data point  consumes too much resources and is why things like Marching Cubes was mostly used in sculpting software, where nothing has to move much.

Maybe we have finally reach the point where things like this could be considered to be used at real time.

 

Eh...

It's partly just that marching cubes has this weird aura of complexity around it, and it comes around whenever people want to make "minecraft, but better", or "infinite universe via procedural generation". And then most learning resources badly overcomplicating the process, so people run around looking for guidance. In practice, going from edge intersection -> marching squares -> marching cubes is dead simple, and things only get complicated when you go to make a whole world out of voxels.

Most of these efforts also founder, because of the aforementioned complexity of making whole worlds out of voxels (or "whole worlds" all by itself). No Man's Sky is pretty, but it cheats like crazy on the voxel planet front, and it's one of a very few large-scale voxel games to even make it to market.

Voxel Farm looks legitimately interesting, albeit I haven't had a chance to actually use it. With any luck high-quality voxel middleware will open up the field a bit and we get to see what games can actually be made more interesting by the addition of Voxels :)

Tristam MacDonald. Ex-BigTech Software Engineer. Future farmer. [https://trist.am]

This topic is closed to new replies.

Advertisement