Видео

Suggestion for testing (if you want to go this route) : Set up an MC server, and load in chunks from that, while filtering through or converting only blocks you are interested in, and ofc set vertical block limits, to build up your world. This would give you market-established world generation to test/code your traversal/gameplay/invisible blocks. Then, once the fundamentals are set, replace the MC server with your own generation :)

Hi! I'm a business dev based in South Africa (with a keen interest in gaming dev), and had a dream about a voxel-based top-down game two nights ago. I searched RUclips for elements I saw in my dream and came across your project. Well-freakin' done, it looks great! I'll be keeping a close eye, and feel free to reach out...as maybe you would like some of my idea's, especially for the multiplayer aspect. Keep up the great work!

This is more or less my life, although I have no dog, and no pretty lake around. But I also work as a consultant and work on my game around that. I just also have two kids that require attention. Progress is slow, but it is there. Nice video.

Very cool, want to make something similar myself!

Can you make setup tour?

quite rem;inds me of dtronghold crusader

Hi there, keep it up

Game looks rather nice visually. You know, while I like the approach you went with with the walls and ceilings in the end, and I think you made the right choice, it didn't have to be that complicated if you didn't want it to be. Just wanted to put that out there for any potential novice gamedevs that might run into the same problem and also consider dropping their project because they lack the skill to come up with a system like you did. No need to give up! There is simpler alternatives. Alot of the isometric games from the late 90s simply had the player manually control camera height by level to hide and show stuff as and when required (and those games didn't have the advantage of the ability to rotate the camera seems as they were sprite based rather than voxel based, which would make this approach even more practical). It was definitely a bit clunky, but it did work. You could take a look at any of the classic X-Com titles as an example of this, off the top of my head, but other games took this approach too.

0:30 yeah that is so relatable

What mic you do use?

good content man, subbing to keep you motivated and to see more comming

Love to see that apex pro keyboard, amazing imo.

Just saw this recommended on my feed and wow man we have similar game projects! My project is basically mix of Minecraft Dungeons and Terraria and I have a dilemma on what to do with the building system and oh my god I have never thought of that way, this video is the perfect solution! About your walls problem... I figure it out by basically use the scroll wheel to manually adjust how you see the height level of the blocks. (As you scroll down, the blocks on the top Y-Level disappears downwards until it reaches the players feet) And then I quickly realize that manually adjusting the height view back and forth is annoying and your solution feels better lol. Edit: I just watched your latest video and I think the scroll wheel solution is useful for navigating tall mountains!!

Creating a shader that turns blocks alpha down sounds a lot more performant then reloading chunks. Also easier to animate or add any other pretty visuals. Because these jumps when blocks disappear don't look good

There is a rogue-lite dragon survival game for PSP, where player roams the map, eating knights and stealing princesses. It is really dated now and requires emulator, so yours could have been a massive success on Steam and phones.

Your shadows have two issues: their are noisy making impression of ripples on the surface of deep dark water; they don't fade with the square of a distance, so when the sun hits horizon, a shadow from a tiny tree scales entire world! I suggest to just once calculate the occlusion map, and then update it as the cells get modified. Otherwise good solutions to non-trivial problems.

Your video is an interesting glimpse into what my life could be. Currently working in IT operations with the goal of moving into a DevOps role, pursuing game dev as a hobby. Thanks for sharing.

Hey! I love your project, can you play Elden Lands map for yourself?

really cool project dude! hey, which method you're using to render voxels? are you using greedy mesh, marching cubes or something like that?

I LOVE that water shader, how did you make it and I think that would be a great tutorial. lots of games need water and I haven't seen tonnes of good tutorials.

I can see how changing your game would be a game-changer

looking great

Any update of game is game changer (literally)

For the issue when voxels are blocking the camera, you should just use a see-through shader which draws a big radius around the player and doesn't render blocking that radius. You can also scale that radius how you want or for different scenarios. There's quite a few tutorials on shaders like that

This is interesting , I had a slightly similar idea for a concept of my own dream game which I have in my head. Which was if rim world and the sims combined basically. Still learning stuff , so I can't implement much yet. But I honestly thought it wasn't possible. BTW just an idea that is probably not useful. But I noticed you have small objects that are not the same size as regular voxels in your game . In a mincraft mod and games like vintage story there is the chisel which allows you to create objects that go beyond the usual appearance. What if the player could create these objects giving them functions . Like a chair for example , pretty simple, make a custom chair appearance and have it inherit the properties of the base chair... (kind of like platos concept of the world of forms only real because its in a video game) A table which I saw you have in game could be customized as well. Might go beyond the scope of what your doing or be practically impossible as I havn't But thanks for doing this I will be following your progress as I imagine if I try to make my dream game the struggles will be similar

+++

Very cool project! You probably already considered it, but adding an over the shoulder camera mode or even a first person mode (toggleable) could be a fun way to solve the terrain hiding issue. I think it would be cool to walk into a mountain valley/chasm, then pop into over-the-shoulder mode to get a real sense of the scale (and explore it properly).

9:23 cant you just make a occlusion culling cone that starts from the player that hides the voxels? Why hide lots?

Wonderful. Absolutely.

My biggest issue with the project so far, is the jarring aestethic difference between the cubic terrain and the varied lool of the doodads and character. I realise non of the art is anywhere close to final, but I'm just wondering if the game would be better served with smoothing your terrain? Voxels behind the scene, but "realistic" terrain visually? This would also get rid of the "jump simulator" issue and could even improve building aestethics by adding tiny variance to placed blocks of the same type.

May I suggest you raycast a cylinder from the camera to the player and hide (or make transluscent) any blocks that collides with said cylinder that are at player Z+1 or above (presuming Z = height). With that, you would have a field of view just and only around the player and you can control the cylinder's width making bigger (for the player's build) or smaller (say he's in a cave without a light source) Instead of sectioning the entire world.

Instead of having to flatten out the terrain. What about generating an invisible copy which is more smoothed out? That way visually you get the blocky look but what the player is walking on is actually a smoother transition.

do you have a discord server? i saw one guy ask about seeing the code and i'd be interested in giving my opinion aswell

it looks brilliant, i've been struggling a bit with my game since i moved to unreal i no longer trust unity so unreal engine it is for me ... glad you are making a tone of progress!

I think an auto jump for 1 tile is a must. Minecraft can be annoying enough in first person constantly jumping that they added an auto jump feature. With a game with this perspective it would be difficult to play without

such a cool idea, loving the progress! keep it up!

Looks awesome! Are you doing anything to optimize rendering so many objects at once?

At 6:09 it's suddenly beautiful. Looks like a world worth exploring.

This looks fantastic! Very pleasing - you could generate small environments and have them stand as little art pieces on their own.

Awesome devlog, and great seeing you back in action! :-D And I have to say, as soon as you introduced your hills together with more flatlands, it all seemed way more natural and better looking. Nice job. Looking forward to the next one!

I think you should definitely add an auto jump feature for 1-block high jumps, as well as maybe a climb for larger ledges.

Amazing solution to the terrain!! Love it. A few things I noticed: - no shallow waters in the newest generator. though the deep ones are also cool... what are your future plans? what about rivers? - hiding the whole hill you are behind. how about clipping the overlapping section relative to the camera? I think other games use it too, some for clipping the player rather than the terrain though (clipping when to switch to a silhouette) - foliage sprintes. I think your game could actually benefit from shelling (due to fixed camera angle). Acerola relently made a great video on this subject named "How Are Games Rendering Fur?" where he showcases its broad useability. Keep up the great work, always looking forward to the next vid 💪

To solve the jumping issue, lift your character's collision box up by 1 voxel so it doesn't cover his feet. This will let you move into 1-voxel-tall ledges. Then slowly push up the player when they are inside a block.

That icy terrain at 2:50 looks siiiick