Mob trap science: how big should I go?

A place to talk to other users about the mod.
Post Reply
User avatar
agentwiggles
Posts: 188
Joined: Fri Apr 06, 2012 2:31 pm

Mob trap science: how big should I go?

Post by agentwiggles »

I've just started work on a mob trap in a server that my friends and I play on. The location I've chosen is perfect - the middle of a small lake in the center of my base, meaning it will run very often and making the use of water in its design very easy.
Spoiler
Show
Image
That's the spot, with the giant square hole where the mobtrap will eventually be. This is roughly at sea level, so I can go about 60 more blocks upwards if I want. As a result, I'm wondering how worthwhile it is to dig down deeper. Obviously, the amount of space I have affects how many spawning pads I have, but I can't imagine that I need to have this thing go down to like bedrock level. How deep would you dig it if you were building this?
Padfoote
Posts: 156
Joined: Sat Nov 03, 2012 10:09 pm

Re: Mob trap science: how big should I go?

Post by Padfoote »

I just started work on my own mob trap, and I know exactly how you feel trying to design this. The area you have planned out that's below sea level I would personally use for the sorting system. From there, the height I would use, (depending on if you want to see a massive tower or not), would be about 60-80 blocks above sea level, giving about 25-35 floors of spawning pads if you have a 2 high space between each level. Also, it would be fairly easy to add more floors if you decide you want more later on.
Utterbob
Posts: 62
Joined: Thu Jul 26, 2012 6:04 am

Re: Mob trap science: how big should I go?

Post by Utterbob »

As for height, you want to avoid breaking the 128 mark unless you intend to use all of the next 128 too. I forget the specific explanation other than it's something to do with the way the game handles the additional height since Anvil maps (and this includes your roof blocks too). As for depth, its a matter of choice and the factors I'm about to talk about.. the only thing to remember, really, is the 5 min item despawn. An item that arrived just after an elevator left for a 64 block lift could be sitting for up to 3 mins before the elevator cycles and gets it to the top (depending on the circuit to clear and reverse the elevator), leaving only 2 min to sort it. Won't be a problem unless your sorting is very complex but worth noting.

As for footprint, its a matter of choice. You can go pretty small if you light everything up really well but the stats on mob spawning/despawning show that even 5% of the area around you being dark can drop the mob trap output by huge amounts (can't recall the exact amount).

I make mine huge to go for percentage of near-by darkness, saving the need to light everything up (which feels like abusing the mechanic a little to me personally). That way I just explore as I go and maintain the adventure feel of my caves in the process rather than systematically lighting them for the sake of lighting them. No matter how big the trap though, lighting surrounding caves will make the biggest difference.

So long story short it's a matter of play style preference, if you want maximum efficiency you don't need a very big trap, just a lot of lit up areas and a fast spawn-to-kill turnover. If you like playing more in the adventure feel and not being so specific to mechanics then build it bigger to compensate for lower efficiency.

P.S. There is a vid that explains all this really well. I will have another look for it later, can't find it at the moment.
User avatar
Benanov
Posts: 259
Joined: Sat Nov 03, 2012 5:12 pm

Re: Mob trap science: how big should I go?

Post by Benanov »

Honestly, you can build until your server admin finds out the java process is consuming 100% CPU on one core and the game is complaining regularly that it can't keep up. :)

You will get more lag and inability to keep up if your processing is weak. Keep your processing chain short; you may need to coalesce things into stacks by feeding everything through a intermittently powered hopper (and make your XP orbs early) in order to make fewer entities that represent the same amount of items.
There's only one V in my name, thanks.

<TaterBoy> I figured out why there's so much lag. We have too much iron.
EtherealWrath
Posts: 298
Joined: Sun Dec 23, 2012 1:37 pm

Re: Mob trap science: how big should I go?

Post by EtherealWrath »

It depends on how you're doing for materials;

Even a small grinder can be very resource intensive depending on build style/mechanism.
(You'll almost certainly want a processing unless you want a massive cobblestone box right outside your house)

Personally I'd start off small, and gradually build more layers as time and materials allow. Avoid going above y128, or 128 in horizontal radius (I think 128 is the despawn/chunk limit.)

Items/EXP will despawn after 5 min, but hoppers will reset this timer.
Phantom screams echo through the ruined facility
A horrible silence builds an eerie tranquility
The souls of many innocent fill the air
The hope they all died with scattered down there
User avatar
thekyz
Posts: 266
Joined: Wed Jul 20, 2011 5:52 am
Location: Rennes, France

Re: Mob trap science: how big should I go?

Post by thekyz »

I'm actually intrigued, why do you guys think it's a bad idea to go above height 128 ?
The great randomo
Posts: 596
Joined: Fri Sep 30, 2011 2:25 pm
Location: In the ether...

Re: Mob trap science: how big should I go?

Post by The great randomo »

If you have time, this shows how to build an efficient one.
Chris Martin wrote:The morning is for sleeping.
User avatar
DNoved1
Posts: 221
Joined: Wed Jan 23, 2013 5:29 pm

Re: Mob trap science: how big should I go?

Post by DNoved1 »

thekyz wrote:I'm actually intrigued, why do you guys think it's a bad idea to go above height 128 ?
A guy on youtube called JL2579 has a video or two going extensively into the mechanics behind mob traps and how spawning works.

Basically, the game only tries to spawn monsters in the bottom 128 layers of the map unless additional constructs (anything but air) is above that level. If a single block exists at level 256 in each chunk for example, it should halve spawn rates. FC might have changed this though, so it may not be applicable.
The great randomo
Posts: 596
Joined: Fri Sep 30, 2011 2:25 pm
Location: In the ether...

Re: Mob trap science: how big should I go?

Post by The great randomo »

DNoved1 wrote:A guy on youtube called JL2579 has a video or two going extensively into the mechanics behind mob traps and how spawning works.
^
|
Chris Martin wrote:The morning is for sleeping.
User avatar
SterlingRed
Posts: 1466
Joined: Tue Jul 05, 2011 11:02 am

Re: Mob trap science: how big should I go?

Post by SterlingRed »

Keep in mind the concept of "bang for the buck/time." At some point, the increased yield gained from a bigger mob grinder is simply not worth the effort or resources. Especially considering size may not be your largest drop rate limiting factor.

There are 3 main things to consider: mob farm size, lighting, and mob life span.
The shorter mobs live, the more will spawn in a given time period. Thus the faster you can move mobs from spawn pads to death, the better. The bigger the mob farm is, the harder this gets.
To balance size and lighting without wasting time building floors that do little to increase yield, start with one or two floors, cap it off, and measure your output in a given time period. Then add a floor, and do the same. If you see a reasonable increase in yield, continue adding floors one by one until the floors don't have much effect on the output (or aren't worth the effort.) If adding a floor has little impact on yield rates, you need to focus on lighting, or you're hitting the max mob count. In the latter scenario, try to adjust the mob grinder to make mobs lifespans as short as possible.
User avatar
DNoved1
Posts: 221
Joined: Wed Jan 23, 2013 5:29 pm

Re: Mob trap science: how big should I go?

Post by DNoved1 »

@randomo Sorry about being redundant, youtube was blocked when I posted that and so I didn't realize you had just posted that very video.

@wiggles In then end I guess it really comes down to how much xp/nitre/scrolls/whatever it is you want from your trap. For comparison, I have 3 floors of this...
Spoiler
Show
Image
...each with about 220 spawning spaces, for about 660 spawning spaces total. I've done minimal lighting (just as I come across caves whilst mining), and over the weekend (including friday), I've gotten this much nitre, as well as enough xp for several enchants.
Spoiler
Show
Image
This is enough for me (at least until I get an infernal enchanter, but then I'll be wanting an enderman farm), but like I said before, it really depends on how much you need.
Utterbob
Posts: 62
Joined: Thu Jul 26, 2012 6:04 am

Re: Mob trap science: how big should I go?

Post by Utterbob »

The great randomo wrote:
Spoiler
Show
If you have time, this shows how to build an efficient one.
This was the one I was thinking of in my post too, thanks for relinking! Must remember to bookmark stuff like this :)
User avatar
thekyz
Posts: 266
Joined: Wed Jul 20, 2011 5:52 am
Location: Rennes, France

Re: Mob trap science: how big should I go?

Post by thekyz »

DNoved1 wrote:Basically, the game only tries to spawn monsters in the bottom 128 layers of the map unless additional constructs (anything but air) is above that level. If a single block exists at level 256 in each chunk for example, it should halve spawn rates.
Am I misinterpreting this change FC talked about a while ago then ?
FlowerChild wrote:
Battosay wrote: I don't like that, sorry. Means that above a certain level, my mobtrap won't spawn mobs in it, right ?
Well, not cool.
No, that's not what it means. It means that I'd only be testing distance taking x and z into account, but not y. So instead of being a sphere around the player, it becomes a cylinder, giving an even distribution of mobs across the height levels, instead of concentrated at the top.

It appears this is already what the spawning code is doing, but the despawning code is reverting the shape to a sphere, causing the monsters that are beyond 128 distance, including the vertical, to despawn shortly after spawning.
http://sargunster.com/btwforum/viewtopi ... 128#p81290
odemarken
Posts: 17
Joined: Sat Dec 08, 2012 2:46 am

Re: Mob trap science: how big should I go?

Post by odemarken »

Ok, my two cents, don't want to hijack the discussion, but I feel this may be relevant.
I have a mob trap with a... decent output (like, a stack of rotten flash and 25 levels of exp in an hour, maybe?) but I always wonder if I could improve it, mainly for hoarding lots of iron. I've got most caves under it lit, though I'm pretty sure I haven't found all of them within 128 blocks.
Now, I played a bit with F3 and entity counters (hope that's not considered cheating here), and noticed two things:
1) by checking at the rendered entities counter while looking around, I'm guessing only about 10-20 mobs are inside the trap at any given time,
2) more strangely, when I switch to peaceful and back to normal, the total entity count only drops and rises by 30 or so, while the hostile mob cap AFAIK is about 80 for single player, right?

Now, it may be relevant that I have rather large herds of pigs, cows and chickens - about 60 of each. Suspecting that it may affect the mob cap, I culled the herds (not the chickens though, they're too hard to breed), which brought down the total entity count from ~250 to like 120. But this didn't change much: still only about 30 hostile mobs spawn, 10-20 of which seem to be in my mob trap.
Other possible relevant info: the trap is based on the "Mobs-on-demand" design, but I gave up on pistons and water and covered the canals with vines instead, except the bottom level where water always flows. It doesn't spawn spiders or endermen, has 5 levels, giving it, i guess, like 500 spawning spaces?
So, is it possible that my friendly mobs are affecting the hostile mob cap? (I don't experience lag or dropped frames, btw). And if not, would it make sense to add more levels if so few mobs spawn anyway?

TL;DR:
Only about 30 hostile mobs spawn in my world according to F3, 10-20 of which seem to spawn in my mob trap. I also have 120-200 animals in my base. Do you think it's possible that animals limit the hostile mob cap, and if not, why do I get so few spawns, and would it make sense to add levels to the trap? The output rate of the trap is not that bad, but I'd like to improve it.

Thanks!
User avatar
DNoved1
Posts: 221
Joined: Wed Jan 23, 2013 5:29 pm

Re: Mob trap science: how big should I go?

Post by DNoved1 »

Passive mobs most definitely do not affect hostile spawn rates. To improve your spawns I would recommend following the advice given in this thread and either light more caves or increase the amount of spawning spaces you have like you were thinking about.
Post Reply