Saplings and Seeds

[tillpp]

output.mp4

Be able to to add saplings, and other structures growable structures.

.{
	..
	.onTick = .{
		.type = .sapling,
		.structures = .{
			.{
				.id = "cubyz:sbb",
				.structure = "cubyz:tree/oak/white", // what structure should be grown?
				.placeMode = .degradable,
				.chance = 0.03, // chance, that tick will cause this to grow
			},
			// other alternative structures, this sapling can grow into
		},		
	},
	..
}

[IntegratedQuantum]

Why do some of them decay instantly?

[tillpp]

Why do some of them decay instantly?

I think it's because default tree generation creates too many leafs, and they then instantly decay because their are too far away from a branch/log

[IntegratedQuantum]

Are you sure that's the case?
These things could also happen if it does the updates during your placement.

[tillpp]

Are you sure that's the case? These things could also happen if it does the updates during your placement.

oh you mean, sometimes leafs are generated before stems are generated?

[IntegratedQuantum]

yes

[tillpp]

I think this is a general problem with the Blueprint.paste() function, so I think It's better to fix that in a separate PR.

[IntegratedQuantum]

There are some flaws with your approach. Instead of unnecessarily trying to bend your code around the terrain generation system, I'd suggest to directly reference the underlying SBB.

[tillpp]

One problem i have, is that It feels too laggy if you have too many trees growing at once,
what do you think?

Maybe the way we paste Blueprint has to be optimized.

(Btw does someone know why the check failed)

[Wunka]

(Btw does someone know why the check failed)

You get a recursion of imports through pub const SimpleStructures = @import("simple_structures/_list.zig"); in biomes.zig
At least that is what I think because the test passes If I remove that and instead do
const sbbGen = @import("../../../server/terrain/simple_structures/SbbGen.zig"); inside sapling.zig but that is pretty cursed so I hope for you that you find some other way

[IntegratedQuantum]

One problem i have, is that It feels too laggy if you have too many trees growing at once,
what do you think?

Well, if you have a performance problem: Measure it.

If the problem is mesh light sampling then #2405 might help improve it a bit. Other than that, there is a good chance that it's just the light updates.

[tillpp]

I think the lagg spikes are soo big, that i dont really need to meassure something:
(each line is a tree that grows)
I think optimizing Blueprint pasting is something we need to work on

also thanks @Wunka for the tip

[IntegratedQuantum]

I think the lagg spikes are soo big, that i dont really need to meassure something:

The measuring is to figure out what exactly is responsible for this. And I doubt that it's the blueprint pasting.

[IntegratedQuantum]

How many leaf blocks are there in a tree?

[tillpp]

probably less then 1000

[IntegratedQuantum]

It would be nice to have those numbers, I think for all saplings the number of saplings per tree should be consistent.

[tillpp]

feel free to count them

[tillpp]

also the amount of leaves varies because of the generation

[careeoki]

How close are we to having saplings with growth stages that have different models?

[Argmaster]

Nowhere close

[tillpp]

requires: #3043

[tillpp]

I think my PR broke the formatter:

(it still keeps printing)

but the game compiles

[tillpp]

good news: no problem on client side
bad news: server gets overwhelmed if you spam saplings

[tillpp]

@careeoki

How close are we to having saplings with growth stages that have different models?

it depends.
In this PR one block and summon a SBB structure.
of course the resulting SBB structure can itself summon a SBB structure.
You can use this to chain SBB together to simulate something growing.
(like sugar cane or something)

but more complex stuff isn't possible with this yet. (like checking if the sugar cane isn't too big already (well technically it is possible, but it wouldn't be nice to implement), or removing a bigger SBB with a smaller SBB afterwards)

[Crepestrom]

do you have any clue why the server is dieing?

[IntegratedQuantum]

server gets overwhelmed if you spam saplings

even in release?

[tillpp]

even in release?

its better in release.
But my computer is also unrepresentative of the average gamer. (humble flex)
So I hope someone with a bad PC can test this by doing:

1. select a medium sized region with the selection wand (but z height = 1)
2. /set cubyz:sapling/oak
3. while seeing the saplings grow, try to drop item
4. can you see the items drop and move?

[IntegratedQuantum]

There was an issue in chunk saving, it always saved it immediately instead of waiting 5 seconds in case other changes happened.
With bccdbda it should be more balanced now.

[tillpp]

i did merge master and my computer still struggles
nothing grows, no items drop when pressing Q. but when rejoining i can see the dropped items and the trees growing. So maybe it is the network who is overwhelmed:

[IntegratedQuantum]

yes, this is #3250, it should at least be a little better though.

[IntegratedQuantum]

When I was testing this, I also noticed that the roots are replacing all ground blocks, instead of checking for decayable.

[tillpp]

decayable

I assume you meant degradeable?

[tillpp]

it should at least be a little better though.

yeah a little bit:

[IntegratedQuantum]

Should be configurable via self.placeMode, like when placed in generation.

[IntegratedQuantum]

It makes no sense to make this public

[IntegratedQuantum]

Please use an enum to improve reasability at the callsite

[Argmaster]

I think I will look into taking over this PR soon, after I'm done with #3446