Posts by Player
-
-
nanitozo: This is a mod bug/general architectural problem it's tripping over, I've added a workaround.
A new build is available at http://files.player.to/tmp/fastcraft-1.22-ctest21.jar
-
-
New build: http://files.player.to/tmp/fastcraft-1.22-ctest19.jar
It contains mostly bugfixes and should improve compatibility.
-
-
Sunconure11: Hmm that's just an empty mod shipped with the ctest18 jar to detect if the fastcraft jar is broken, e.g. lack of the META-INF folder. I will look into it. Are you using kcauldron by any chance?
Edit: It seems to be an issue with FML's mod detection in general. It won't load the mod, but still scan for it and take it into account for some operations...Entoarox: I'll have to check and patch it out. Btw. MC 1.8 most likely has this fixed in the vanilla version already.
-
-
I've uploaded a new test build for MC 1.7.10 at http://files.player.to/tmp/fastcraft-1.22-ctest18.jar
It contains some misc. fixes, hopefully the last test build before the 1.22 release. Please let me know if you run into issues.
-
Hey player long time no text ... so I have to ask and please don't be offended ... um when is the invisible entities bug fix going to be released? Can I please have an ETA? Anywhere within 60 Days would be great. I run a server but it is not 1/2 as stable without your mod. Unfortunately Squids and Certain Pokemon go perm invisible. Thank you very much this is just a question so I can stop checking this forum Daily.
It's almost done, but I'm still debugging. It could be a few days to weeks.
-
-
-
ajthemacboy: I think I have this fixed for the next build.
Kittykiller: Most likely not a FC bug, the mention is just an unrelated hook. Unfortunately the stack trace is incompletely recorded, so it's hard to figure out what actually happened.
Fayti1703: CoFHTweaks is incompatible and without changes on their side I can't change that in a reasonable way.
Phishphan420: You can only use it in FTB packs atm (that get mirrored to Curse afaik), FC may be available from there soon.
-
TooGoodForYou999: Getting a solid 1.22 release out for 1.7.10 has priority. After that's done I'll consider it.
Eogen: The file contains a stack trace for every thread 30 seconds into the tick. The one for "Server thread" is most important, anything you see there is usually what's causing the delay. Stall detection is only time based, it doesn't know if the thread is stuck infinitely as in a deadlock or infinite loop.
I've checked a few of the reports and they suggest that the stalls are only large scale chained world generation. This happens if the code generating a chunk accesses another chunk that wasn't generated either, thus starting generation there as well. MC prevents this by delaying decoration (2nd stage of world gen) until 2x2 chunks are generated, but that also implies that only those 2x2 are allowed to be accessed. Many mods access more, in your case tfc, which is a bug.
Just waiting it out works around the problem, especially with pre-generation. The effect of switching to Java 7 is most likely just a coincidence in this case. -
You'll have to copy the level.dat, possibly editing the seed in there. It contains all the block/item <-> id mappings, which are being assigned on a per-save basis. It'd be much harder to adjust the chunks instead.
BC's blueprints may be also worth a thought for the coarse structure, unfortunately they aren't widely supported yet, including IC2 blocks.
-
The conditions for the steam explosions are overheating or being unable to drain all steam. Presumably one of those is met, but I'm not too familiar with the code involved.
-
It'd be easiest if you just decompile IC2 with BytecodeViewer or similar and look at IC2.generate around line 650.
IC2 scales its vertical ore distribution and chunk count proportionally to the world height. The quantities are being randomly varied to reduce the bias from the generation (=chunk) grid. There's also a config option to scale the generation quantities.
The values except for rubber trees with a normal world height are:
copper: avg. 15 chunks, size 10, y 10 to 70, non-uniformly with peak at 40
tin: avg. 25 chunks, size 6, y 0 to 40 uniform
uranium: 20 chunks, size 3, y 0 to 60 uniform
lead: 8 chunks, size 4, y 0 to 60 uniformThe size is the WorldGenMinable parameter, which is not linear.
-
It's been most likely triggered by an ingame-action, unrelated to the build.
The underlying problem is presumably Cauldron's tile entity placement simulation, which leaks some state and doesn't clean up properly by at least calling invalidate on the tile entity.
You could investigate changing the log4j configuration in the cauldron jar (log4j2.xml) to filter the warnings, but IC2 already throttles them heavily. The only "damage" is that the affected block may be unable to send/receive energy and your logfile grows somewhat. It's not CPU intense.
-
I've found the problem causing the squid glitch after a lot of debugging. It may be related to other disappearing entity issues.
A new build will follow soon.
cindylo: This is unlikely to be related to Fastcraft, the mention at the bottom of the stack trace is fairly transparent and does nothing even remotely related to the error. It looks like one of the inner classes in ChatComponentStyle is missing, probably from a corrupted jar or a broken coremod.
-
-