Major update. See the original post for details, updated schematics, and downloads.
As the new title says, I managed to increase the output by 73% to 11510 items/hr thanks to a couple of suggestions from people in this post and a few minor modifications. Yes, that means I am getting over 3 items a second (3.2 to be more precise) and will get a full inventory in 12 minutes.
Done. Left it at the bottom too since the diagrams take up a lot of vertical space. That way you can scroll up or down depending on preference and/or which one you are near.
That's amazing. You must have a hard time getting wood fast enough to build chests for all this!
Haha, well I placed 8 double chests using MCEdit for storing items after I ran tests. I didn't think I would have any issue with space and thought having 8 of them was overkill. However, after running three tests (15 minutes at the old rate of 6654 items/hr and 10 minutes + 30 minutes at the new rate of 11510 items/hr) I've already filled 6 of the 8 double chests in the trap room. I ended up placing 8 more double chests in the collection room to dump items to so I could run a longer test (30 minutes) as my inventory was filling in just over 10 minutes.
That's amazing. You must have a hard time getting wood fast enough to build chests for all this!
Haha, well I placed 8 double chests using MCEdit for storing items after I ran tests. I didn't think I would have any issue with space and thought having 8 of them was overkill. However, after running three tests (15 minutes at the old rate of 6654 items/hr and 10 minutes + 30 minutes at the new rate of 11510 items/hr) I've already filled 6 of the 8 double chests in the trap room. I ended up placing 8 more double chests in the collection room to dump items to so I could run a longer test (30 minutes) as my inventory was filling in just over 10 minutes.
I'm guessing the biggest problem is with bacon that won't stack? I can see how that would fill an inventory VERY fast. I wouldn't be surprised if you dropped that stuff right away. :tongue.gif: I wish Notch would make a "fridge" block that allows for stacking of food.
Do mobs really need allot of space to spawn ? If not you could try to build something like this
<snip>
or this
<snip>
With appropiate height. I would really see someone try :smile.gif:
The point is to make the mobs fall faster, enabling quicker spawns. Since I think we are already hitting the 200 limit fast.
You bring up a good point on the 200 mob limit and I decided to do some quick math to see how close I am to the theoretical max with my current trap. The following is just for calculating in the 0 chunk space:
The tower has 28 floors of 16x16 spawning space for a total of 7168 blocks that mobs can spawn on.
The entirety of the chunk is composed of 32768 blocks (16x16x128).
Dividing the area mobs can spawn on by the total volume of the chunk gives us: 7168/32768 = 0.21875
This means that since the algorithm picks a random position in 3D space in a chunk to try and spawn a mob that there is a 21.875% chance that any random (x,y,z) position will qualify as a valid spawning surface.
The spawning algorithm is called 20 times every second and since it picks a random position in 3D space that means that 21.785% of the calls have a chance of succeeding and spawning a monster.
20 calls/second * 21.785% success rate yields a theoretical average of 4.375 mob spawns/sec.
I am not sure of how mob drop rates average out over time, but lets assume that, on average, every mob drops one item. Since my trap is currently yielding 3.2 items/second and the theoretical average is 4.375 items/sec, then I am at 3.2/4.375 = 73.14% efficiency (of the max possible of the average rate).
Now that math isn't likely to be entirely accurate, but I think it's a close enough approximation. So that would say that there is room for improvement, but by decreasing the available surface for mobs to spawn on you push down the theoretical maximum average. So as long as such a decrease can be offset by an increase in efficiency, then such changes to the tower will be positive.
For example: Removing 25 blocks from the floors in the tower reduces the theoretical maximum average item drop rate down to 3.95 items/second. That requires an efficiency of 81% just to maintain the same drop rate of 3.2 items/second.
That's amazing. You must have a hard time getting wood fast enough to build chests for all this!
Haha, well I placed 8 double chests using MCEdit for storing items after I ran tests. I didn't think I would have any issue with space and thought having 8 of them was overkill. However, after running three tests (15 minutes at the old rate of 6654 items/hr and 10 minutes + 30 minutes at the new rate of 11510 items/hr) I've already filled 6 of the 8 double chests in the trap room. I ended up placing 8 more double chests in the collection room to dump items to so I could run a longer test (30 minutes) as my inventory was filling in just over 10 minutes.
I'm guessing the biggest problem is with bacon that won't stack? I can see how that would fill an inventory VERY fast. I wouldn't be surprised if you dropped that stuff right away. :tongue.gif: I wish Notch would make a "fridge" block that allows for stacking of food.
Well, luckily, no pigs or other such annoying farm animals find there way into the trap. While I have no opposition to bacon, I certainly appreciate its absence from my trap :tongue.gif:
I don't understand how it works. The pics arent helpful. Can someone explain
I plan on making a video tonight, but what do you need explained? There's also a save file that you can download in the OP so you can run around and look at the trap in person all you want.
Maybe you want to take a long at the spider problem. Spiders seem to get stuck there for some time in the corners.
Hmm, just checked that issue again and it seems they are still getting stuck (in just two corners?). Seems like the eventually get unstuck though. I'll have to think about it and see what I can come up with. Damn spiders always being so difficult.
Screenshot of items coming down the collection canal from the trap room:
<snip>
.....
... Holy ****!
Haha, that's what I said when I first saw that when I ran my first test after the changes. I quickly tried to log onto the forums only to find they were down so I typed the update in notepad while I waited :tongue.gif:
Seriously, though, this is amazing. How do you find chunk 0?
The easiest way I know is to use NBT Forge and find the area from coordinates [0,0] to [15,15]. That 16x16 square is chunk 0 (from the bottom of bedrock to the sky limit). Another way is to make a copy of you save and delete c.0.0.dat and then look for the missing chunk (you can use cartographer to locate it).
As the new title says, I managed to increase the output by 73% to 11510 items/hr thanks to a couple of suggestions from people in this post and a few minor modifications. Yes, that means I am getting over 3 items a second (3.2 to be more precise) and will get a full inventory in 12 minutes.
Done. Left it at the bottom too since the diagrams take up a lot of vertical space. That way you can scroll up or down depending on preference and/or which one you are near.
You heard that, green and red.
The link to the save is in the OP, but here it is again.
Haha, well I placed 8 double chests using MCEdit for storing items after I ran tests. I didn't think I would have any issue with space and thought having 8 of them was overkill. However, after running three tests (15 minutes at the old rate of 6654 items/hr and 10 minutes + 30 minutes at the new rate of 11510 items/hr) I've already filled 6 of the 8 double chests in the trap room. I ended up placing 8 more double chests in the collection room to dump items to so I could run a longer test (30 minutes) as my inventory was filling in just over 10 minutes.
I'm guessing the biggest problem is with bacon that won't stack? I can see how that would fill an inventory VERY fast. I wouldn't be surprised if you dropped that stuff right away. :tongue.gif: I wish Notch would make a "fridge" block that allows for stacking of food.
You heard that, green and red.
... Holy ****!
You bring up a good point on the 200 mob limit and I decided to do some quick math to see how close I am to the theoretical max with my current trap. The following is just for calculating in the 0 chunk space:
The tower has 28 floors of 16x16 spawning space for a total of 7168 blocks that mobs can spawn on.
The entirety of the chunk is composed of 32768 blocks (16x16x128).
Dividing the area mobs can spawn on by the total volume of the chunk gives us: 7168/32768 = 0.21875
This means that since the algorithm picks a random position in 3D space in a chunk to try and spawn a mob that there is a 21.875% chance that any random (x,y,z) position will qualify as a valid spawning surface.
The spawning algorithm is called 20 times every second and since it picks a random position in 3D space that means that 21.785% of the calls have a chance of succeeding and spawning a monster.
20 calls/second * 21.785% success rate yields a theoretical average of 4.375 mob spawns/sec.
I am not sure of how mob drop rates average out over time, but lets assume that, on average, every mob drops one item. Since my trap is currently yielding 3.2 items/second and the theoretical average is 4.375 items/sec, then I am at 3.2/4.375 = 73.14% efficiency (of the max possible of the average rate).
Now that math isn't likely to be entirely accurate, but I think it's a close enough approximation. So that would say that there is room for improvement, but by decreasing the available surface for mobs to spawn on you push down the theoretical maximum average. So as long as such a decrease can be offset by an increase in efficiency, then such changes to the tower will be positive.
For example: Removing 25 blocks from the floors in the tower reduces the theoretical maximum average item drop rate down to 3.95 items/second. That requires an efficiency of 81% just to maintain the same drop rate of 3.2 items/second.
Well, luckily, no pigs or other such annoying farm animals find there way into the trap. While I have no opposition to bacon, I certainly appreciate its absence from my trap :tongue.gif:
I plan on making a video tonight, but what do you need explained? There's also a save file that you can download in the OP so you can run around and look at the trap in person all you want.
Hmm, just checked that issue again and it seems they are still getting stuck (in just two corners?). Seems like the eventually get unstuck though. I'll have to think about it and see what I can come up with. Damn spiders always being so difficult.
Haha, that's what I said when I first saw that when I ran my first test after the changes. I quickly tried to log onto the forums only to find they were down so I typed the update in notepad while I waited :tongue.gif:
Seriously, though, this is amazing. How do you find chunk 0?
The easiest way I know is to use NBT Forge and find the area from coordinates [0,0] to [15,15]. That 16x16 square is chunk 0 (from the bottom of bedrock to the sky limit). Another way is to make a copy of you save and delete c.0.0.dat and then look for the missing chunk (you can use cartographer to locate it).