Update
Added inventory, with mouse wheel selection, Workbench, Furnace + mockup GUI.
Added axe movement (not show in video)
Refactoring of code with comments.
emMQi-YiaAo
Update
Added inventory, with mouse wheel selection, Workbench, Furnace + mockup GUI.
Added axe movement (not show in video)
Refactoring of code with comments.
emMQi-YiaAo
I think you should use a shader to make the whole world look like the Furnace GUI does right now 
TBH, I’ve never seen a non-random-gen voxel engine. I think it would be a really cool direction to take, as you could do a lot with it. And in my opinion, built worlds always look much nicer than otherwise.
I think you should use a shader to make the whole world look like the Furnace GUI does right now
TBH, I’ve never seen a non-random-gen voxel engine. I think it would be a really cool direction to take, as you could do a lot with it. And in my opinion, built worlds always look much nicer than otherwise.
Random generation can sometimes be just as nice as custom-built. Take a look at Delver for example*.
(*If you look around, you may be able to find the old free version. If not, just watch the videos and look at the screenshots.)
I was more referring to large, outdoor spaces rather than enclosed spaces like that of Delver’s.
That said, that does look like a very nice game.
Ah, yes.
Minecraft especially has some fixing to do in procedural generation…
But even then, custom design never beats Simplex/Perlin noise for natural landscape generation.
Well if I was doing by-hand terrain, in my opinion, I would start with noise, and touch up upon that. Shaping a world entirely from scratch would be a completely tedious task, while not yielding the result one would want when making a game, of, as you say, using procedural generation.
Random generation can sometimes be just as nice as custom-built. Take a look at Delver for example*.
(*If you look around, you may be able to find the old free version. If not, just watch the videos and look at the screenshots.)
I was more referring to large, outdoor spaces rather than enclosed spaces like that of Delver’s.
That said, that does look like a very nice game.
Ah, yes.
Minecraft especially has some fixing to do in procedural generation…
But even then, custom design never beats Simplex/Perlin noise for natural landscape generation.
Well if I was doing by-hand terrain, in my opinion, I would start with noise, and touch up upon that. Shaping a world entirely from scratch would be a completely tedious task, while not yielding the result one would want when making a game, of, as you say, using procedural generation.
That’s a good idea, I would like it to look different. How I would do that I’m not sure, ill take a look at cell shaders, then maybe some vertex noise?
The landscape is random, I use a diamond start algorithm. But your suggestion is one I’d not considered, and it would Lend itself to small missions?
Delver is very interesting, and has a lot in common with my other game. The old free one uses prefabs joined together I believe. Like say-animal crossings!
Making a whole landscape by hand would be a challenge! I’d need to either make a tool to do that, or add tools from within the game, or utilise some minecraft world editing tools and read the format and convert it.
All very interesting. I currently changing the picking to pick faces rather than cubes.
Has anyone ever used the openGL name stack? (Though I thing I will solve the problem geometrically)
That’s a good idea, I would like it to look different. How I would do that I’m not sure, ill take a look at cell shaders, then maybe some vertex noise?
The landscape is random, I use a diamond start algorithm. But your suggestion is one I’d not considered, and it would Lend itself to small missions?
Delver is very interesting, and has a lot in common with my other game. The old free one uses prefabs joined together I believe. Like say-animal crossings!
Making a whole landscape by hand would be a challenge! I’d need to either make a tool to do that, or add tools from within the game, or utilise some minecraft world editing tools and read the format and convert it.
All very interesting. I currently changing the picking to pick faces rather than cubes.
Has anyone ever used the openGL name stack? (Though I thing I will solve the problem geometrically)
Update
I have been working on the picking, rather than tracing a in ray increments along the ray and testing block intersection, I have finally decided to ditch this
and do 3 separate axis tests at 1 block intervals, using parametric equations. This is not only 100% accurate, it eliminates problems with corners, and assuming you can pick a block 10
units away on each axis, that is 30 calculations for worst case. Rather than the 100 or so I was doing before for 0.1 unit accuracy.
Now, I’m testing for collision with a face, not a cube!
I can give details of how this is done if anyone should require a mathematical explanation with code.
HtjWJiodjhs
Despite thinking of that method and posting about it a few months ago, I never manged to implement it properly myself. 
I would like to see the code (and the mathematical explanation, so I can see what I missed).
Thanks
Update
I have been working on the picking, rather than tracing a in ray increments along the ray and testing block intersection, I have finally decided to ditch this
and do 3 separate axis tests at 1 block intervals, using parametric equations. This is not only 100% accurate, it eliminates problems with corners, and assuming you can pick a block 10
units away on each axis, that is 30 calculations for worst case. Rather than the 100 or so I was doing before for 0.1 unit accuracy.
Now, I’m testing for collision with a face, not a cube!
I can give details of how this is done if anyone should require a mathematical explanation with code.
HtjWJiodjhs
Despite thinking of that method and posting about it a few months ago, I never manged to implement it properly myself.
I would like to see the code (and the mathematical explanation, so I can see what I missed).
Thanks