Render and growth radius not equal?

[Gavin Jenkins]

After testing, it seems that things only grow out to 2 chunks in front of you (previous direction of movement), although 4 in the direction you walked from. You can still see 4 all around though. So I can look forward and watch some plants not growing and others growing. This seems like a mistake, especially with it being asymmetric.

I'm guessing it's a config parameter, not RB, since it affects vines as well as other crops.

[Matthías Valdimarsson]

Anything loaded by the server will grow or nothing will grow at all, if you are not seeing growth from some specific crop you should probably include what crop that is.

[Gavin Jenkins]

No, it's just anything that isn't persistent. I think even chicken egg laying. Definitely vines and cocoa and melons and cane too.

I can see the chunks visually and they are within 4 chunks of me, but they are acting dead beyond 2 chunks. Only in one direction, though. Wherever I walked from, it grows, etc. out to 4 chunks. I can reverse this by walking in the other direction then back to the center point, now things grow 4 chunks in that direction and 2 in the other.

[Gavin Jenkins]

Correction: SOME things still work, such as redstone, furnaces burning, baby animals growing up in the "Semi dead" chunks. So they are definitely loaded by the server, just weirdly, asymmetrically, not growing things.

[programmerdan]

I've experienced this before when AFK'ing render distance fields, although I can't give any additional specifics as once I discovered this I simply put in a larger circle minecart to AFK in, and the problem was resolved. This related to pumpkin growth, so would tie into the random block updates that MC leverages to cause things to "grow", if indeed there is a bug of some sort here

[Gavin Jenkins]

A minecart loop solves the asymmetry, but still results in everything growing at 1/2 or 1/4 speed (portion of time your minecart is within render distance of that corner farm), when there's no real reason for it not to be at 100%, since the server is using the resources to load the chunk anyway already.

Also, constantly loading and unloading edge chunks in a looping cart might be really bad for lag, as the server constantly has to look up the same data from disk over and over. Unless it's better than I think it is about caching.

[Daniel Boston]

Overstating the problem a lot, unless your farm is so immense that to load each corner is to fully unload all other corners. I've only experienced it as an edge case, so perhaps I've not seen exactly what you've seen. However, this might be worth looking in to. I'm guessing this is Realistic Biomes (still getting acquainted with the code behind the mod soup we've got going on here)?

Can you also fully describe a reproducible setup that reliably illustrates the issue, including plant type, orientation, size in chunks, location of player, etc. so others can reproduce and confirm?

[Gavin Jenkins]

I will build a mockup on civtest when I get the chance.

[Rourke750]

I don't really know what's going but you all know rb handles growth right?  The crops will not grow unless its their time in the config.  When they are allowed to grow they will update in batches with a block update.

[programmerdan]

Yeah, we're aware. Gavin's proposition is that growth isn't occurring even in loaded chunks, under certain conditions. My challenge to him is to prove these conditions are reproducible, so one of the devs can actually intelligently investigate. So far I haven't seen anything I could actually look at the code for. 

[Gavin Jenkins]

I'm not 100% sure it IS realistic biomes, which is why I posted it here instead of github. It could be something like one of the spigot or bukkit parameters filtering out the relevant events or registrations before RB has a chance to interpret them.

[Sean Chambers]

So then it would probably be useful to have a way to reproduce to help pinpoint the problem and isolate where it actually is. Right?

Sorry, don't mean to be harsh but I work with a product team and it drives me up the wall when someone is talking about a problem without reproducible steps because without evidence it's just wasted effort and spinning wheels. 

Sean

Sent from my iPhone

[Daniel Boston]

Hence my request for exact definition of parameters. I feel the same way, and get exactly as angered as you're describing. I hate chasing bugs without some measure of specificity. 

[Sean Chambers]

Imho, it is useful to quickly discuss the issue but if no one immediately comes forward and says they know what it is, there's little use discussing it further without something to focus on. If someone has further information then of course that's useful to add but you get the point. 

Usually there's just way too many moving parts in most software to try to apply a fix without a failing test case that transitions to passing after a fix has been applied. My devs know to come to me if they don't write a unit test to resolve a bug as a good test bed for resolved issues is a must to ensure a bug doesn't pop up again. I digress however. 

Sean

Sent from my iPhone