Yeah they still die in SMP. I've been trying different types and in all of them my chickens die if there are more than one per coop. I've tried higher ceilings, all glass, trapdoor pens and more and I always find dead chickens after a while. Is there any good way to mitigate this? With animals not spawning it's becoming very hard especially since pigs and cows also seem to be able to escape fenced in areas on occasion.
I find that when using a 4x4 square of space with water at the corners (pushing the chickens to the center) and glass for walls the death rate is extremely slow unless I have a massive number of chickens. Once the chicken count gets down to 10-15, deaths nearly stop. In my first version, where I did not use glass for the walls, deaths would take the count down to 4-5 chickens before deaths became rare. I've never tried anything smaller than 4x4, so not sure how much this contributes to the survival.
From what I've seen Mystify post, he knows the **** he's talking about. I don't mean in this thread, I mean overall. Don't argue with a man that knows when you, in fact, take guesses.
Keep in mind that this is coded in java, a language not quite as convenient as it could have been made.
I had to laugh. I'm probably in the what, .01%? of minecraft players that has actually browsed the code. I'm not saying that as a means of saying I know what I'm talking about - my earlier statements are more on my general experience as a developer. It's always a chuckle to see people blindly follow someone else because they can't quite decide what's right for themselves.
Mystify and I just differ in the extent we'd think he would have to change the code. He seems to think Notch would need to update the data structures, which definitely would require more extensive changes. I came back with a solution that only updates the pattern, not the breadth of values. This would a much smaller solution that would likely require only localized changes in the code that updates the light levels. It's not as "pretty" a solution as Mystify would have created; there are trade-offs to any solution.
One problem is that light does not add in game. A large glass dome covered in lava is still dark not to far below; while in real life, having many distant light sources can give as much light as a single dim one close by.
Yeah, it'd be interesting if there was an additive effect, and that additive effect would increase the reach. But I'm pretty sure that's where you'd start to incur some pretty significant computational load to figure it all out.
1
I find that when using a 4x4 square of space with water at the corners (pushing the chickens to the center) and glass for walls the death rate is extremely slow unless I have a massive number of chickens. Once the chicken count gets down to 10-15, deaths nearly stop. In my first version, where I did not use glass for the walls, deaths would take the count down to 4-5 chickens before deaths became rare. I've never tried anything smaller than 4x4, so not sure how much this contributes to the survival.
1
I had to laugh. I'm probably in the what, .01%? of minecraft players that has actually browsed the code. I'm not saying that as a means of saying I know what I'm talking about - my earlier statements are more on my general experience as a developer. It's always a chuckle to see people blindly follow someone else because they can't quite decide what's right for themselves.
Mystify and I just differ in the extent we'd think he would have to change the code. He seems to think Notch would need to update the data structures, which definitely would require more extensive changes. I came back with a solution that only updates the pattern, not the breadth of values. This would a much smaller solution that would likely require only localized changes in the code that updates the light levels. It's not as "pretty" a solution as Mystify would have created; there are trade-offs to any solution.
1
Yeah, it'd be interesting if there was an additive effect, and that additive effect would increase the reach. But I'm pretty sure that's where you'd start to incur some pretty significant computational load to figure it all out.