There was a few updates back a bug that prevented us from building Tango's Iron Foundry, the last update, addressed that bug and I'm happy to say it is buildable now.
Last night I got around to testing it on a test world, and am happy to say that it was working, and furthermore saved, and loaded back with no problems. There are a few things that i should note however.
First and foremost whenever you create a new village during the chaining process, give it at least 1 full minute if not more for villagers to detect the newly created village. Second when you are in the process of breaking doors to force them to bounce over to its satellite or home, also give it a full minute or more for this process. It seems to take the Xbox 1, and 360 quite awhile to detect changes to villages, while on the pc counterpart takes only a few seconds.
The advantage to this system is that you can use only 16 villagers, and get a whopping amount of iron. I built the 28 version, and I know I didn't wait long enough for some of the village to form correctly (I estimate that I probably have 12 to 14 villages, which in 15 minutes created almost 2 full stacks of iron ingots, and around 32 roses.
I key sign that you can watch for to see if you waited long enough during the actual village relocation part, (part where you begin placing all the doors for each village), is to keep an eye on your villagers, if you use 10 in each spot for a total of 20 villagers (supporting 2 iron golems in active area at a time), if they begin to breed during door placement that you made a mistake, or went too fast.
I'm going to go through the process again only much slower and see if i can get the full 44 stacked villages running.
The main advantages to this system over JL2579's Iron farm (made famous by DocM77) is that 1) his only produces around 1 iron golem every 6 minutes, and uses 16 villagers per farm, with our current villager cap, which I believe is 50, means that you can have a max of 3 farms, which will use 48 villagers, leaving you only 2 villagers for other projects, and will produce an iron golem on average about every 2 minutes.
The foundry on the other hand, can use as little as 16 villagers (I recommend at least 20), as 20 villagers will allow at least 2 iron golems to spawn at a time, It also supports up to i believe 44 villages, and thus you should be spawning an iron golem roughly every 8 seconds. Which is ALOT of iron, but more so leaves you with 30 villagers to use in other projects, such as a merchant hall, or using them in mob sorters, ect.
Kinda impressed that this works through save considering that was always an issue with xbox and a lack of spawn chunks doubt i will ever get round to building one but lovely to know i could
Actually it don't, I did alot more testing on it just now, and 'slowly' rebuilt a new one. There is no problem with it working through a save, but what seems to be happening is that there is a 'village' cap of 4. I'm going to try and get some info from 4J on this, if there is no cap then there is still a merger bug. This is the one thing I hate about the console version, getting information on the internal workings is like pulling teeth, ie painfully hard.
Kinda impressed that this works through save considering that was always an issue with xbox and a lack of spawn chunks doubt i will ever get round to building one but lovely to know i could
There is spawn chunks on xbox. There are two ways you can test this, first and formost is where you first spawn, attempt to place lava, and you will get a message that its not allowed in 'player spawn area', or something similar. Second is to goto the end, kill the dragon and throw items through the portal. Provided that you havn't changed your spawn, should you go back through the portal, items will appear within the spawn chunks. Final test you can use to confirm the exists of spawn chunks is to travel to the spawn area, drop some items, and quickly fly a good distance away, then save and exit, reload the world, wait 5 to 8 minutes. Now if there was no spawn chunks when you return to the spawn area there should be items on the ground, as the area was 'unloaded', which is not the case, when you load up your world, those chunks are loaded, and thus the 5 to 8 minutes you waited when reloading the world was enough time for those items to despawn.
I made the full 44 village version and it seems to work great just sometimes there is a pause between the golems spawning always 30 seconds or less though and i guess the problem is i either didn't make it in the spawn chunks or they don't stay loaded because when i go away and come back it stops
except I have done the spawn chunk before, at the spawn area, and the items have stayed loaded, so that plus discussions on this forum and minecraft reddits also confirming we dont have spawn chunks (we have a spawn thats not the same) is what made me say it, plus all the various videos of items made in spawn chunks on PC where the youtuber has commented on sorry it wont work on consoles because of this..
I made the full 44 village version and it seems to work great just sometimes there is a pause between the golems spawning always 30 seconds or less though and i guess the problem is i either didn't make it in the spawn chunks or they don't stay loaded because when i go away and come back it stops
It doesn't work, I've tried doing it very very very very slowly, you can see that it don't work by using just 16 villagers, (8 in each chamber), once you start actually adding the internal doors to each village, they will begin to breed, however they don't start breeding until after the 4th village is in place, around the fifth village they will start breeding, and this is a sure indication that 'additional' villages don't work. Otherwize, we could place it anywhere and pretty much automate chaining and village insertion, very much like the iron trench did. I've been playing around with the chaining process so that I could automate it via redstone, and managed to get it working, and it really wasn't that hard, and not nearly as complex as the trench was, reason was a) I only used 16 villages, The iron foundry was quite a bit easier to work with then the trench layout was.
Tonight I'm going to play more with it only using the PC (so that I can use the village marker mod), and see how I can compact and debug the process.
except I have done the spawn chunk before, at the spawn area, and the items have stayed loaded, so that plus discussions on this forum and minecraft reddits also confirming we dont have spawn chunks (we have a spawn thats not the same) is what made me say it, plus all the various videos of items made in spawn chunks on PC where the youtuber has commented on sorry it wont work on consoles because of this..
Yeah I done some further testing and found out that the spawn area does become 'inert'. Odd thing is that I thought I tested this on the xbox 360 and that the chunks do constantly remain active and loaded. I'm going to test it on the 360, and see if they do have spawn chunks. I'm really curious as to if 4J ever plans to implement spawn chunks at some point in time. Its too bad they wouldn't entertain a Q/A period with us.
OK, been doing a lot of testing and work, and pretty much got the 'chaining' process complete. Anyone who watched tangos tutorial on the Iron trench and shuttered in fear because of all the redstone will be glad to know that we greatly simplified the process, to which i'll explain in detail in a minute. When we sat down and begin hammering out ideas, we pretty much knew from the get go that we would have to have a farm that could rebuild itself, since obviously spawn chunks were out of the question. What we opted on was based on 2 of tangos farms, the iron trench, and the iron Foundry. The trench because it could rebuild itself, and the foundry because it was quite simple to build, and was scalable.
What we so far have come up with is a design that supports either a 19 village limit, and another which supports 43 village limit. We then wanted to greatly simplify the redstone involved in the rebuild, and I think we accomplished it, at least so far.
Heres our breakdown so far (sorry no pictures or videos yet as its still a work in progress)
Similar to tangos trench we first wanted a system that could detect when iron stops flowing and if so trigger a reset or rebuild. To accomplish this we used a timer that if it cycles 2x and no iron is collected during this time, it will trigger a rebuild. The timer is currently set at 3 minutes, thus if no iron is collected in 6 minutes a rebuild will trigger automatically.
The rebuild begins by simply sending a 1 tick pulse to a dropper that is loaded with items, which will place 1 item into a dropper chain (which we refer to as a sequencer), The first dropper is the prestart sequence, a comparator detects an item in the dropper and initialized the prestart sequence, which more or less involes, closing all sky access to doors, and raising any 'detector' villagers. Once its done this is begins a 1 minute timer.
The one minute timer simply will send a 1 tick pulse to the sequencer chain, which will move the item to the next dropper, it then disables itself (the timer),
Again a comparator reads the signal out of the dropper and preforms the correct set of actions on either the satellite side, or the main spawn platform side, which is to first raise the villager, wait 60 seconds (this is the same as breaking all doors), then set correct sky access for the doors in question, and finally lower the villager, which in turn detects a new village, and 'links' up to other villages. It finally restarts the 1 minute timer, and the process continues till the sequence chain is complete.
Once the chain is complete, the item is pulled out of the dropper and placed back into the start sequence dropper, which keeps things neat and tidy. and triggers a stage 2 process, which should be quite a bit easier to do, but its late now and we are tired and headed to bed.
I hope I explained this well enough, and will gladly entertain questions, comments, and advice. Hopefully by tomorrow or next day we'll finish it up, completely debug it, and somehow showcase it.
It doesn't work, I've tried doing it very very very very slowly, you can see that it don't work by using just 16 villagers, (8 in each chamber), once you start actually adding the internal doors to each village, they will begin to breed, however they don't start breeding until after the 4th village is in place, around the fifth village they will start breeding, and this is a sure indication that 'additional' villages don't work. Otherwize, we could place it anywhere and pretty much automate chaining and village insertion, very much like the iron trench did. I've been playing around with the chaining process so that I could automate it via redstone, and managed to get it working, and it really wasn't that hard, and not nearly as complex as the trench was, reason was a) I only used 16 villages, The iron foundry was quite a bit easier to work with then the trench layout was.
Tonight I'm going to play more with it only using the PC (so that I can use the village marker mod), and see how I can compact and debug the process.
Mine does not start to breed and works great until i go away and comeback or enter another dimension
Kinda, we keep optomizing areas that don't need to be looked at. Well at least not at this point in time, for example, we came up with a brand new spawn cage, similar to that of what ZipCrowd uses on thier 1.8 golem farm, which really helps improve getting the golems out of the spawn area. They arn't in the area anymore then 1 to 2 seconds at most. Main reason for this is when we manually fire the farm with lots of villages (32+), using a 10 villager limit, using the old floor system, it was taking 5 to 10 seconds for a golem to get out of the village sphere.
Second issue we ran across that we were not aware of till we did some testing on the PC was that, 'linking' was actually pretty damn slow, way slower then a villager takes to actually 'create' the village. The process more or less involves 2 platforms as you are likely aware of, the area or 'home' where the golems are actually farmed which I will refer to as P1, and a 2nd platform that's 64 blocks away that is used in creating, and stacking villages, which I will refer to as P2.
The process begins with P1, lowering its villager, and creating the 1st village on Row 1, then P2 lowers its villager, and creates a new village on Row 2, after 30 seconds (to be safe), Row 1 door on P2, needs to link itself to Row 1 on P2 (effecitivly giving village 1, 2 doors, but this is only temporary, then the villager in P1 is raised 1 block, which in turn places him too far away from the door to the village he created, after 60 seconds, that door is no longer part of Village 1, and Village 1 now belongs to P2. So now we have 2 villages on P2, each with 1 door.
So, P1 now closes access to all doors, and the villager is lowered, This time Row 3's door is exposed, which in turn creates a new village. Next the 2 other villages needs to be linked, so Row 2's door needs to be exposed after 30 seconds, then Row 1's door needs exposed after 30 seconds. Finally The villager in P2 needs to be raised, and at least 60 seconds wait time, When this is done P1 will now have 3 villages, each with 1 door.
This process needs to be repeated over and over until the desired number of villages is reached. To which the second phase is extremely easy, its just a matter of sliding the villages from P2, into P1 from the newest created village to the oldest village (village 1). linking each village as it comes in with its own set of doors, this part is quite simple.
The problem is that linking requires more or less a incremental (and can be reset counter), that can be used in iterations, this would make walking through the linking phase much much much simpler then how we currently are doing it, with a bucketload of nor latches.
The last thing we want to be able to do, is create 2 villages and link at the same time, much faster reset times, but this shouldn't be too hard once we actually get back to the project.
So yes we have done it, yes it works, but currently the redstone involved in the creating and linking phase is wayyy to much, with too many areas that could break.
FYI, we also experimented, with the idea of creating, and linking to doors right away then semi positing the villages to a somewhat center point, but this required, 128/128 along x/z/y axis, but it worked well, the advantage to this is that villages started right away creating iron golems, and the process was greatly simplified, and fast, 32 villages took slightly longer then 32 minutes to work fully, and started right away within 1 minute with 1 village functional. Around each minute to 2 minutes another village became active. But as I said, 128x128x128 seemed a bit extreme. It didn't need all this area, just some platforms each 64 blocks away from center.
Few things I've learned while playing around, with automation in respects to village stacking.
a) Villagers will detect doors upto 16 blocks away across the x,z plane, and +-4 blocks on the Y axis.
Where the door is located on the Y access, should the villager that is detecting it be raised 6 blocks, he will 'detach' himself from the door, if its the only door for the village, the village will be deleted. By simply placing a 2 roof blocks over a door, a sticky piston facing up on top of the roof block, and a block on top of the sticky piston allows one to easily create a villager that can create or link to existing villages. Place a villager on top of the block that the piston controls, and encase around it so the villager can't escape. By raising the piston, and waiting around 60 seconds the villager will loose track of the lower door. Lowering the piston will cause the villager to detect the door again after a short period of time (usually 2 to 10 seconds). If there is no villages around he will create a new village. If there is another village that its center point is less then 64 blocks away, then this door will link up with that village, which in turn will 'move' the village center, and possibly increase its radius.
c) Most automatically building iron golem farms out there don't really show you that you need to keep your 'villager' stack, which consists of around 10 to 20 villagers usually packed into a small basket so to speak at least 16 blocks away from any doors in the farm, or at least we figured out that keeping them at or near the center point of the village, and at least 16 blocks away from any of the doors in the village, is a smart move. The main reason is that eventually those villages will merge, for one reason or another, when this happens if your villagers are within range of the doors they will begin breeding, using up precious villager numbers that console users really can't afford to loose. The second reason is that should these villagers detect any doors they could and often do mess up linking and creating of villages.
d) When linking, doors are ALWAYS added to the oldest village, no matter if a newer village is closer this is really a big hindrance to < 1.8 versions. In 1.8 (PC), its the opposite, in that doors are ALWAYS added to the newest village, As a result one MUST construct all villages first before attempting to link them with thier main doors, the main reason is because once a village links up with its 21 doors, its takes many many many doors linking to that village to move its center point away from the area where your main doors are added.
e) A misconception is that you need 16 villagers to spawn iron golems, this is false, you really on need 10 villagers. For every 10 villagers the village can support 1 additional iron golem, so if the village had 30 villagers, the area could support 3 iron golems. However additional villagers will not increase the spawn speed. Simply put for iron golems to spawn requires 21 doors, and at least 10 villagers, once the village meets this criteria, there is a 1 in 7000 chance per tick that an iron golem will spawn. They have a 16x16x6 spawn area, centered on the village center point. Iron golems can spawn with transparent blocks in their head, chest, feet. The only checks is that there is solid blocks for them to spawn on and non solid blocks above the solid blocks.
f) Another misconception is that villagers will influence the village center, and thus where golems will spawn. Again this is false. Only the village doors have an impact on the village center point. Its radius is defined by the distance from the center point of the village to the furthest door. If this this value is less then 32 then it defaults to a radius of 32, and thus villages never have a radius of less then 32.
g) Villagers do NOT need to 'see' a door for them to detect it. It just needs to be a valid house, which most of us already know what defines a house to a villager. The villager just needs to be within range for them to detect it.
Hope this helps, and was useful info to some, heaven knows it took us hours of testing, and research to find this all out.
After messing around with some early designs i've discovered that during the village chaining process, that sometimes, the linkining process will allow you to after creating the village, expose all doors that need to be linked instantly and they will all link perfectly while other times it requires that each door that needs to be linked be linked in order, waiting for each door to link before attempting to link another. Failure to wait for the door link will result in it linking to another village which can really cause the village chaining process to mess up.
After several different designs behaved so oddly we more or less duplicated the world on the PC, and installed 'Village Marker', which is a mod by KaboPC, which more or less shows villages, the doors they are linked to, golem spawn areas, ect for each village, each using a different color.
What we believe is that where the village chaining begins, is key to allowing one to use the 'easy link' system. In essence the direction influences the link system in that if its either positive or negative along each axis.
I'm about to do some testing and see if I can figure it out. Will post results as soon as i know them.
SWEET! Figured it out, and VERY Happy, this will make the actual redstone MUCH easier!
This is what I found. If your dealing with the north, south directions, in your linking. (Basically your either creating a village, and then linking to it which is to the south or north. Here is how you ensure that linking does not require you to wait for each door you need to link. Taking the first village you create, If you link to the north along the Negative Z axis, it will work, If your linking to the south, along the positive Z axis it will work.
If your working with the east/west axis, and the village your creating requires linking to the east, you should be moving along the positive axis. If the village requires linking to the west, then you should be moving along the negative x axis.
Taking the above descriptions, I'll give an example, we create a brand new village, at (x:0,y:0), if we want to link to it to the north, we must ensure that as 'we travel' north x goes negative, or if we are linking to the south, then x must go positive as we travel south. I hope this makes since.
When I'm referring to work, i mean that you do NOT need to wait for each door that you need to link to to link before attempting to link to the next one. Lets say you just created village number 10, and thus have 9 villages to link to, you can expose all 9 doors at the same time to skylight and they will all link properly within 5 or 6 seconds.
If you attempt to link opposite directions then what I listed, it still can work, however after you create the newest village, for each village you need to link to you will need to wait probably 10-15 seconds (to ensure) that it links up with its matching pair, before attempting to link to the next. Thus the process is, create new village (wait 15 seconds), (expose 1st door link skylight) (Wait 15 seconds), (expose 2nd door to link to skylight) (wait 15 seconds), repeat as necessary until all doors are linked. Move to next platform, raise villager (Wait 60 seconds for linked doors to detach), lower villager and repeat.
If your chaining just 10 villages, the amount of overheat time (Based on 15 second wait times) is. 675 seconds, or 11 minutes and 15 seconds, this is JUST overhead for linking. Thus we clearly see that its clearly advantageous to link as I mentioned above.
If you link as i described above correctly, then your process is Create new village, (Wait 15 seconds), Expose all doors that require linking to skylight. (Wait 15 seconds), Move to next platform, raise villager (Wait 60 seconds for linked doors to detach), lower villager and repeat process.
I really hope the information i've posted in the last few days helps people, one way or another. Now that we have all the information we need, our team has some work to do, and hopefully in a few days I'll get back to you with a easy to construct self-rebuilding iron farm.
For anyone interested, I will be streaming at http://twitch.tv/cire360 I also have opened the testworld to everyone, however no permissions will be given to anyone who joins. I will be constructing a self-repairing-36 Stackable village Iron golem farm. Don't expect it to be finished however as I'm somewhat of a perfectionist. My goals will be to get it functional, and then compact and simplify the redstone involved. The 2nd goals, are to make it as easy as possible for anyone to build (redstone wize).
Nope villager mechanics are still broken sadly. It looks like the game does NOT do a census within the village area, or is not taking into account any villagers that are not within 16 blocks of a door.
Figured it out, there are parts of console that take along time to 'work', Census population is the key one, however I did get everything functional for phase one, in other words I got the redstone functional that can create, chain and stack villages, and it really wasn't that hard. Now I just need to do the redstone for the other platform as I built duel platforms in my test world, and then, do the redstone that actually pulls the villages in, which I figure should be pretty easy.
I've finished all the redstone required on both platforms to create 36 villages, and align them on outside platforms, in essence all the redstone involved for creating, and linking the villages is done. Still need to do a 'reset' button, which isn't a big deal, and i think adding a control room where you can watch it step through the process as it goes through phase 1 (creating villages), would be kinda cool, but not sure the amount of work to run 36 different lines would be worth the effort.
Right now I'm manually pulling in the villages once phase one is complete, gona start doing some rough designs to see what i can come up with to complete phase 2 over the next couple of days.
Success, I managed to create, link, and pull to a center point 12 villages tonight (I messed up 2 of them so rather then restart the process, i just skipped those 2). Right now I have to manually pull them in, but soon I'll automate this. At which point I have built a heavily modified titan, that self repairs itself.
There is only 1 issue left to solve, and that is how to deal with additional villagers once villages merge to become a super village. I'm currently running it using a max of around 14 villagers.
Pictures, soon, and i may even do a tutorail for it. The full version supports, 16/20/36 villages.
Last night I got around to testing it on a test world, and am happy to say that it was working, and furthermore saved, and loaded back with no problems. There are a few things that i should note however.
First and foremost whenever you create a new village during the chaining process, give it at least 1 full minute if not more for villagers to detect the newly created village. Second when you are in the process of breaking doors to force them to bounce over to its satellite or home, also give it a full minute or more for this process. It seems to take the Xbox 1, and 360 quite awhile to detect changes to villages, while on the pc counterpart takes only a few seconds.
The advantage to this system is that you can use only 16 villagers, and get a whopping amount of iron. I built the 28 version, and I know I didn't wait long enough for some of the village to form correctly (I estimate that I probably have 12 to 14 villages, which in 15 minutes created almost 2 full stacks of iron ingots, and around 32 roses.
I key sign that you can watch for to see if you waited long enough during the actual village relocation part, (part where you begin placing all the doors for each village), is to keep an eye on your villagers, if you use 10 in each spot for a total of 20 villagers (supporting 2 iron golems in active area at a time), if they begin to breed during door placement that you made a mistake, or went too fast.
I'm going to go through the process again only much slower and see if i can get the full 44 stacked villages running.
The main advantages to this system over JL2579's Iron farm (made famous by DocM77) is that 1) his only produces around 1 iron golem every 6 minutes, and uses 16 villagers per farm, with our current villager cap, which I believe is 50, means that you can have a max of 3 farms, which will use 48 villagers, leaving you only 2 villagers for other projects, and will produce an iron golem on average about every 2 minutes.
The foundry on the other hand, can use as little as 16 villagers (I recommend at least 20), as 20 villagers will allow at least 2 iron golems to spawn at a time, It also supports up to i believe 44 villages, and thus you should be spawning an iron golem roughly every 8 seconds. Which is ALOT of iron, but more so leaves you with 30 villagers to use in other projects, such as a merchant hall, or using them in mob sorters, ect.
There is spawn chunks on xbox. There are two ways you can test this, first and formost is where you first spawn, attempt to place lava, and you will get a message that its not allowed in 'player spawn area', or something similar. Second is to goto the end, kill the dragon and throw items through the portal. Provided that you havn't changed your spawn, should you go back through the portal, items will appear within the spawn chunks. Final test you can use to confirm the exists of spawn chunks is to travel to the spawn area, drop some items, and quickly fly a good distance away, then save and exit, reload the world, wait 5 to 8 minutes. Now if there was no spawn chunks when you return to the spawn area there should be items on the ground, as the area was 'unloaded', which is not the case, when you load up your world, those chunks are loaded, and thus the 5 to 8 minutes you waited when reloading the world was enough time for those items to despawn.
It doesn't work, I've tried doing it very very very very slowly, you can see that it don't work by using just 16 villagers, (8 in each chamber), once you start actually adding the internal doors to each village, they will begin to breed, however they don't start breeding until after the 4th village is in place, around the fifth village they will start breeding, and this is a sure indication that 'additional' villages don't work. Otherwize, we could place it anywhere and pretty much automate chaining and village insertion, very much like the iron trench did. I've been playing around with the chaining process so that I could automate it via redstone, and managed to get it working, and it really wasn't that hard, and not nearly as complex as the trench was, reason was a) I only used 16 villages, The iron foundry was quite a bit easier to work with then the trench layout was.
Tonight I'm going to play more with it only using the PC (so that I can use the village marker mod), and see how I can compact and debug the process.
Yeah I done some further testing and found out that the spawn area does become 'inert'. Odd thing is that I thought I tested this on the xbox 360 and that the chunks do constantly remain active and loaded. I'm going to test it on the 360, and see if they do have spawn chunks. I'm really curious as to if 4J ever plans to implement spawn chunks at some point in time. Its too bad they wouldn't entertain a Q/A period with us.
What we so far have come up with is a design that supports either a 19 village limit, and another which supports 43 village limit. We then wanted to greatly simplify the redstone involved in the rebuild, and I think we accomplished it, at least so far.
Heres our breakdown so far (sorry no pictures or videos yet as its still a work in progress)
Similar to tangos trench we first wanted a system that could detect when iron stops flowing and if so trigger a reset or rebuild. To accomplish this we used a timer that if it cycles 2x and no iron is collected during this time, it will trigger a rebuild. The timer is currently set at 3 minutes, thus if no iron is collected in 6 minutes a rebuild will trigger automatically.
The rebuild begins by simply sending a 1 tick pulse to a dropper that is loaded with items, which will place 1 item into a dropper chain (which we refer to as a sequencer), The first dropper is the prestart sequence, a comparator detects an item in the dropper and initialized the prestart sequence, which more or less involes, closing all sky access to doors, and raising any 'detector' villagers. Once its done this is begins a 1 minute timer.
The one minute timer simply will send a 1 tick pulse to the sequencer chain, which will move the item to the next dropper, it then disables itself (the timer),
Again a comparator reads the signal out of the dropper and preforms the correct set of actions on either the satellite side, or the main spawn platform side, which is to first raise the villager, wait 60 seconds (this is the same as breaking all doors), then set correct sky access for the doors in question, and finally lower the villager, which in turn detects a new village, and 'links' up to other villages. It finally restarts the 1 minute timer, and the process continues till the sequence chain is complete.
Once the chain is complete, the item is pulled out of the dropper and placed back into the start sequence dropper, which keeps things neat and tidy. and triggers a stage 2 process, which should be quite a bit easier to do, but its late now and we are tired and headed to bed.
I hope I explained this well enough, and will gladly entertain questions, comments, and advice. Hopefully by tomorrow or next day we'll finish it up, completely debug it, and somehow showcase it.
Mine does not start to breed and works great until i go away and comeback or enter another dimension
Kinda, we keep optomizing areas that don't need to be looked at. Well at least not at this point in time, for example, we came up with a brand new spawn cage, similar to that of what ZipCrowd uses on thier 1.8 golem farm, which really helps improve getting the golems out of the spawn area. They arn't in the area anymore then 1 to 2 seconds at most. Main reason for this is when we manually fire the farm with lots of villages (32+), using a 10 villager limit, using the old floor system, it was taking 5 to 10 seconds for a golem to get out of the village sphere.
Second issue we ran across that we were not aware of till we did some testing on the PC was that, 'linking' was actually pretty damn slow, way slower then a villager takes to actually 'create' the village. The process more or less involves 2 platforms as you are likely aware of, the area or 'home' where the golems are actually farmed which I will refer to as P1, and a 2nd platform that's 64 blocks away that is used in creating, and stacking villages, which I will refer to as P2.
The process begins with P1, lowering its villager, and creating the 1st village on Row 1, then P2 lowers its villager, and creates a new village on Row 2, after 30 seconds (to be safe), Row 1 door on P2, needs to link itself to Row 1 on P2 (effecitivly giving village 1, 2 doors, but this is only temporary, then the villager in P1 is raised 1 block, which in turn places him too far away from the door to the village he created, after 60 seconds, that door is no longer part of Village 1, and Village 1 now belongs to P2. So now we have 2 villages on P2, each with 1 door.
So, P1 now closes access to all doors, and the villager is lowered, This time Row 3's door is exposed, which in turn creates a new village. Next the 2 other villages needs to be linked, so Row 2's door needs to be exposed after 30 seconds, then Row 1's door needs exposed after 30 seconds. Finally The villager in P2 needs to be raised, and at least 60 seconds wait time, When this is done P1 will now have 3 villages, each with 1 door.
This process needs to be repeated over and over until the desired number of villages is reached. To which the second phase is extremely easy, its just a matter of sliding the villages from P2, into P1 from the newest created village to the oldest village (village 1). linking each village as it comes in with its own set of doors, this part is quite simple.
The problem is that linking requires more or less a incremental (and can be reset counter), that can be used in iterations, this would make walking through the linking phase much much much simpler then how we currently are doing it, with a bucketload of nor latches.
The last thing we want to be able to do, is create 2 villages and link at the same time, much faster reset times, but this shouldn't be too hard once we actually get back to the project.
So yes we have done it, yes it works, but currently the redstone involved in the creating and linking phase is wayyy to much, with too many areas that could break.
FYI, we also experimented, with the idea of creating, and linking to doors right away then semi positing the villages to a somewhat center point, but this required, 128/128 along x/z/y axis, but it worked well, the advantage to this is that villages started right away creating iron golems, and the process was greatly simplified, and fast, 32 villages took slightly longer then 32 minutes to work fully, and started right away within 1 minute with 1 village functional. Around each minute to 2 minutes another village became active. But as I said, 128x128x128 seemed a bit extreme. It didn't need all this area, just some platforms each 64 blocks away from center.
a) Villagers will detect doors upto 16 blocks away across the x,z plane, and +-4 blocks on the Y axis.
Where the door is located on the Y access, should the villager that is detecting it be raised 6 blocks, he will 'detach' himself from the door, if its the only door for the village, the village will be deleted. By simply placing a 2 roof blocks over a door, a sticky piston facing up on top of the roof block, and a block on top of the sticky piston allows one to easily create a villager that can create or link to existing villages. Place a villager on top of the block that the piston controls, and encase around it so the villager can't escape. By raising the piston, and waiting around 60 seconds the villager will loose track of the lower door. Lowering the piston will cause the villager to detect the door again after a short period of time (usually 2 to 10 seconds). If there is no villages around he will create a new village. If there is another village that its center point is less then 64 blocks away, then this door will link up with that village, which in turn will 'move' the village center, and possibly increase its radius.
c) Most automatically building iron golem farms out there don't really show you that you need to keep your 'villager' stack, which consists of around 10 to 20 villagers usually packed into a small basket so to speak at least 16 blocks away from any doors in the farm, or at least we figured out that keeping them at or near the center point of the village, and at least 16 blocks away from any of the doors in the village, is a smart move. The main reason is that eventually those villages will merge, for one reason or another, when this happens if your villagers are within range of the doors they will begin breeding, using up precious villager numbers that console users really can't afford to loose. The second reason is that should these villagers detect any doors they could and often do mess up linking and creating of villages.
d) When linking, doors are ALWAYS added to the oldest village, no matter if a newer village is closer this is really a big hindrance to < 1.8 versions. In 1.8 (PC), its the opposite, in that doors are ALWAYS added to the newest village, As a result one MUST construct all villages first before attempting to link them with thier main doors, the main reason is because once a village links up with its 21 doors, its takes many many many doors linking to that village to move its center point away from the area where your main doors are added.
e) A misconception is that you need 16 villagers to spawn iron golems, this is false, you really on need 10 villagers. For every 10 villagers the village can support 1 additional iron golem, so if the village had 30 villagers, the area could support 3 iron golems. However additional villagers will not increase the spawn speed. Simply put for iron golems to spawn requires 21 doors, and at least 10 villagers, once the village meets this criteria, there is a 1 in 7000 chance per tick that an iron golem will spawn. They have a 16x16x6 spawn area, centered on the village center point. Iron golems can spawn with transparent blocks in their head, chest, feet. The only checks is that there is solid blocks for them to spawn on and non solid blocks above the solid blocks.
f) Another misconception is that villagers will influence the village center, and thus where golems will spawn. Again this is false. Only the village doors have an impact on the village center point. Its radius is defined by the distance from the center point of the village to the furthest door. If this this value is less then 32 then it defaults to a radius of 32, and thus villages never have a radius of less then 32.
g) Villagers do NOT need to 'see' a door for them to detect it. It just needs to be a valid house, which most of us already know what defines a house to a villager. The villager just needs to be within range for them to detect it.
Hope this helps, and was useful info to some, heaven knows it took us hours of testing, and research to find this all out.
After several different designs behaved so oddly we more or less duplicated the world on the PC, and installed 'Village Marker', which is a mod by KaboPC, which more or less shows villages, the doors they are linked to, golem spawn areas, ect for each village, each using a different color.
What we believe is that where the village chaining begins, is key to allowing one to use the 'easy link' system. In essence the direction influences the link system in that if its either positive or negative along each axis.
I'm about to do some testing and see if I can figure it out. Will post results as soon as i know them.
This is what I found. If your dealing with the north, south directions, in your linking. (Basically your either creating a village, and then linking to it which is to the south or north. Here is how you ensure that linking does not require you to wait for each door you need to link. Taking the first village you create, If you link to the north along the Negative Z axis, it will work, If your linking to the south, along the positive Z axis it will work.
If your working with the east/west axis, and the village your creating requires linking to the east, you should be moving along the positive axis. If the village requires linking to the west, then you should be moving along the negative x axis.
Taking the above descriptions, I'll give an example, we create a brand new village, at (x:0,y:0), if we want to link to it to the north, we must ensure that as 'we travel' north x goes negative, or if we are linking to the south, then x must go positive as we travel south. I hope this makes since.
When I'm referring to work, i mean that you do NOT need to wait for each door that you need to link to to link before attempting to link to the next one. Lets say you just created village number 10, and thus have 9 villages to link to, you can expose all 9 doors at the same time to skylight and they will all link properly within 5 or 6 seconds.
If you attempt to link opposite directions then what I listed, it still can work, however after you create the newest village, for each village you need to link to you will need to wait probably 10-15 seconds (to ensure) that it links up with its matching pair, before attempting to link to the next. Thus the process is, create new village (wait 15 seconds), (expose 1st door link skylight) (Wait 15 seconds), (expose 2nd door to link to skylight) (wait 15 seconds), repeat as necessary until all doors are linked. Move to next platform, raise villager (Wait 60 seconds for linked doors to detach), lower villager and repeat.
If your chaining just 10 villages, the amount of overheat time (Based on 15 second wait times) is. 675 seconds, or 11 minutes and 15 seconds, this is JUST overhead for linking. Thus we clearly see that its clearly advantageous to link as I mentioned above.
If you link as i described above correctly, then your process is Create new village, (Wait 15 seconds), Expose all doors that require linking to skylight. (Wait 15 seconds), Move to next platform, raise villager (Wait 60 seconds for linked doors to detach), lower villager and repeat process.
I really hope the information i've posted in the last few days helps people, one way or another. Now that we have all the information we need, our team has some work to do, and hopefully in a few days I'll get back to you with a easy to construct self-rebuilding iron farm.
Streaming again twitch.tv/cire360
Right now I'm manually pulling in the villages once phase one is complete, gona start doing some rough designs to see what i can come up with to complete phase 2 over the next couple of days.
There is only 1 issue left to solve, and that is how to deal with additional villagers once villages merge to become a super village. I'm currently running it using a max of around 14 villagers.
Pictures, soon, and i may even do a tutorail for it. The full version supports, 16/20/36 villages.