We recently had a pretty animated discussion regarding the presumptive change to MCXBLA terrain generation algorithms coming with the 1.8 update (which is still far enough off that nobody really knows anything about it).
The summation of this conversation is that builders like the 1.8 style and explorers like the 1.7 and prior (big surprise ). However, someone made the point that the limitation of MCXBLA worlds to 1024X1024 implied that 4J could not slavishly implement the 1.8 terrain code simply because the size of biomes in 1.8 would overwhelm the available space. (i.e. one biome might take up the entire MCXBLA world).
We know that 4J has been pretty good so far at (some would say) slavishly re-creating the PC code when they do the ports (down to re-creating known BUGS from the the PC that have already been fixed over there). However, the above comment got me thinking that there is no way 4J can't see the possibly see the biome problem coming and that they are probably smart enough to know that they're going to need to change something in the 1.8 code to make the whole thing work within the confines of 1024X1024.
This realization led me down two related thought paths:
A: The easiest way to address the MCXBLA size issue is to treat the world's edge as a portal to another overland. Since the conceptual code already exists for transferring between "alternate dimensions" in the form of the nether portal, the easiest answer is to accept that there will never be comingled 1024X1024 worlds but there could be cojoined worlds.
B: How deterministic is the world seed? If you generate two new worlds on MCXBLA using the exact same seed, will the worlds be exactly the same or will they only have the same theme. If setting the seed is deterministic then my idea in A will not work because you would only wander from one copy of your world to another carbon copy.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
The summation of this conversation is that builders like the 1.8 style and explorers like the 1.7 and prior (big surprise
We know that 4J has been pretty good so far at (some would say) slavishly re-creating the PC code when they do the ports (down to re-creating known BUGS from the the PC that have already been fixed over there). However, the above comment got me thinking that there is no way 4J can't see the possibly see the biome problem coming and that they are probably smart enough to know that they're going to need to change something in the 1.8 code to make the whole thing work within the confines of 1024X1024.
This realization led me down two related thought paths:
A: The easiest way to address the MCXBLA size issue is to treat the world's edge as a portal to another overland. Since the conceptual code already exists for transferring between "alternate dimensions" in the form of the nether portal, the easiest answer is to accept that there will never be comingled 1024X1024 worlds but there could be cojoined worlds.
B: How deterministic is the world seed? If you generate two new worlds on MCXBLA using the exact same seed, will the worlds be exactly the same or will they only have the same theme. If setting the seed is deterministic then my idea in A will not work because you would only wander from one copy of your world to another carbon copy.