What I expected:
Iron golems spawning with more then 50 villagers and more then 100 doors,
The doors were counting towards a house, and the villagers were first bred, later added with spawners when I had around 30
Now I expected the golems to spawn when I waited for 5 hours, however I never found one, or even heard one underground like I had to when it was still 1.2.13.5.
Even so, they no longer seem to spawn, and the villagers were spawning in a world I created since I got the beta update, meaning I didn't use an older world, and yet there are no golems spawning naturally.
Please fix this as this might also effect iron farms more then the previous bug we faced.
Hope to hear from you soon, or see the fix standing in the next update,
StrategyMamber
Linked issues
is cloned by 1
Attachments
Comments 27
Unsure if I should be adding my "me too" here, or creating a new issue. My problem is basically the same, but I'm willing to provide details of testing (see discussion in MCPE-33751 for example). I'm working in a new world created in 1.4.2 on Xbox One.
In short: iron golem spawns are at best very inconsistent and at worst they are broken in a way that I haven't been able to narrow down yet. I have a snail farm (24 doors, at least 14 villagers) that has been confirmed to produce golems, but which goes an hour (and several hours, cumulatively across play sessions) without spawning a single golem. This is with me in the vicinity (no more than 50 blocks away from a village edge) doing other stuff or just waiting for a golem. I'm happy to copy/paste the discussion from the other issue here so devs have what they need.
My farm design is attached. It's flowing water to a central drop hole (corner blocks of pool are water source blocks); pretty standard design. Numbers indicate height above ground level (doors are placed on 0-level blocks).
[media]
Since the original reporter doesn't seem to be responding, I suppose we may as well look at your problem instead.
In the other ticket, I don't think I thought too much about the 50 block distance, but now it strikes me as potentially important. If you're 50 blocks from the village edge, you're at least 64 blocks (Euclidean distance) from the nominal village center. For the farm to work, I think the village center has to be within an active area ("ticking area"). Your player, of course, creates an active area that moves with him, but it's aligned to chunk boundaries, so (1) it moves in discrete jumps of 16 blocks at a time, which means small movements of the player can have large effects on the active area, and (2) it isn't round, which means the Euclidean distance to its boundary depends on what direction you measure it. In fact, along the diagonals the size in blocks of your active area is as little as 32 blocks, if the player happens to be near the closest corner of a chunk. To determine whether this is contributing to your problem, we need to look at the following:
The coordinates of the drop hole, which we're using as a proxy for the village center, and
The coordinates, roughly, of the farthest block in the area you were idling in.
What your Simulation Distance is set to.
I'm also pretty sure that the golem can only spawn at a point that's within an active area, but the mechanics are a little different in that case. If the village center is in range, the spawn algorithm will run, but if it picks a random point that's outside the active area, the spawn will be initiated but remain suspended until the area becomes active. In that case, to the game's way of thinking there's a golem within the village, so the next time it tries to spawn it will abort for that reason. (Unless there are enough villagers to qualify for 2 golems, and in that case you might get one or you might spawn another suspended one and stop everything that way.) An interesting effect of this analysis is that you would not see any golems spawn while you're far away, but later they will suddenly appear to spawn as soon as you come into range. I have seen this effect in my own farms but never understood it until just now.
If it turns out that the idling distance is affecting your results, the next step is to see whether staying a bit closer improves the rate.
Okay good idea. I have spent a half hour or more just sitting on the roof of the village (...uh...20 or so taxicab blocks from the hole I guess?), but I have also spent time nearby breaking up the old village, or spelunking the Hellmouth that is right below that. I have had golems spawn while I was out bulldozing the old village (Golems 3 and 4 from my previous discussion). My processing range simulation distance is set to 10 chunks.
I will map out the area and provide coordinates for the drop hole, the doors, and the farthest that I can remember nerd-poling to wait. For what it's worth, the villagers are always active when I'm watching from a distance. Dunno if that implies their area is a ticking area.
ETA: I wonder if this relates at all to MCPE-29902 ("Village data does not load")?
It took me perhaps 30 minutes to collect this information, during which zero golems spawned.
Seed: 884635330
15 Villagers
Ground Level is 64 (Level 0 in the diagram)
NW water source block: 206, 65, -608
NW block of drop hole in middle of pool: 213, 64, -601
Farthest ground-level area I was working in while waiting for golems: 171,...,-603
Farthest nerd-pole used while waiting: 187,95,-633
Doors (all at height 64):
225,-608
225,-604
225,-601
225,-600
225,-597
225,-593
221,-589
217,-589
214,-589
213,-589
210,-589
206,-589
202,-608
202,-604
202,-601
202,-600
202,-597
202,-593
221,-612
217,-612
214,-612
213,-612
210,-612
206,-612
(Doors in this screenshot are slightly different; this is before I made them symmetrical on all 4 sides.)
[media]Water may the source of the problem - more like quirky behavior. Semi-transparent facing the door isn't ideal to build within roof space limits. It will be either transparent or opaque, depending how the water being setup, build in order, or block update affected by sky light level. Here are some behaviors that happened during the test sessions. MD's build was used for the test.
-Placed doors before water placed with lower sky light (SL), doors will be detected. It would produce 24 villagers in daylight. However, it won't produce more than 23 from 2 repopulation sessions, strange enough to spawn a golem
-Placed water before doors with lower SL, some door(s) detected ( When place water say on a light level 0 at night, surface block beneath water will be stuck with lower SL during the day. Remove water to get the surface block updated with higher SL, replace the water, the door will be detected.
-Sometime if build in order, you could have up to 24 villagers but golem spawning won't be on top of rate all the time due to vary causes by semi transparent.
-Semi transparent causes a lot of interfence, lowers spawn rate
-Water flowing near or by source, door wouldn't be detected. Water flowing farther from source, door would be detected
With these quirky behaviors, golem spawn may work, stops working soon or later, may comes back working again itself, quits working altogether, etc
Replacing water with opaque blocks same amount roof space as behind door do definitely halt golem from spawning. 24 doors were detected because they were detected soon after I replaced them.
I think i have been experiencing more quirky behaviors other than above. I have decided to quit farther experimenting.
As a rule of thumb, build anything that is an opaque as a roof. Front and behind door has different X of roof space. If water sit within roof space, increase or decrease roof space on the opposite side.
Here's another attachment showing red concrete blocks replace water. It halts golem from spawning altogether by the same roof space on both sides.
[media]Yes, I believe you found the root of my problem. This bug report is not mine (I just kind of hijacked it as a generic "iron golem spawning problem"), but as far as I'm concerned I don't have any issue any more. Mods can close this bug if they want.
After dropping the spawn zone down a block, the farm produces golems as expected. I realized that the 8 centered doors (4 pairs of 2 doors) would still have been valid, and probably also the other 4 doors on one entire side of my particular village, because I did build an add-on base on the other side of the back wall of the villager apartments. So for those apartments, anyway, they had the full 5 roof blocks.
I'm probably going to next try reversing the village, so "outside" points away from the pool, allowing me to build two spawn platforms.
As for your question about them spawning up to 24, I didn't do rigorous testing, and with your findings regarding order of placement and how I was, for a time, playing with removing and adding water, who knows if my observations are all that accurate. But from what I recall, when they got to 24 villagers, they would breed another right away if an excess Farmer accidentally got shoved into the pool and down the kill hole.
So this was just user error - sorry for wasting everybody's time! And thanks for the help.
Oh no, as Auldrick stated, "...we may as well look at...". I was gladly to help tackle down the problem you were having. Your issue was fit by the title on this report.
It's good to hear that the modified spawn worked very well on your end.
Actually, I thought it was 8 doors (these 4 pairs) were detected from the early test sessions. From last few sessions, there were another 8 doors detected, where they are facing water flowing that is farther from the source. You probably had 16 detected doors in the first place. Golem spawn conditions may be stricter than villager breed when it comes down to roof check. When detecting a door occurs, it adds to village data permanently- not confirmed. May be the reason we manage to breed new villager instantly soon after one dies. Golem requires detected doors and roof check the second the player make a change. I could have written a long comment earlier. But, now that we know not to wrestle with semi transparent as wiki stated, "..it could be opaque."
I believe you can have 2 platforms with the proper setup. I have thought about that while testing. You could decrease fail attempts spawn by adding more valid blocks.
Best of luck
Cleaning up old tickets: This ticket has not been updated recently (~1 year+) so is being closed as Cannot Reproduce. If you feel this is still a valid issue then please comment, or create a new ticket following the Issue Guidelines.
Quick Links:
📓 Issue Guidelines – 💬 Mojang Support – 📧 Suggestions – 📖 Minecraft Wiki
Can you confirm if you have mob spawning switched on or off in the settings?