quadraxis

Looking at the FML log (line 61): GL info: ' Vendor: 'Intel' Version: '4.4.0 - Build 20.19.15.4474' Renderer: 'Intel(R) HD Graphics 530' It looks like this instance of Minecraft isn't using the dedicated graphics. Can you compare what the F3 screen says when it is/isn't working?

Sounds a bit like this issue: Do you see any exceptions in the console output? For that issue in particular, the fml-junk-earlystartup.log file would be better than the regular log.

If the regular download isn't working, try the direct link. (Click the "i" icon next to the main link)

I've opened an issue for this crash here: https://github.com/MinecraftForge/MinecraftForge/issues/4285

https://files.minecraftforge.net/maven/net/minecraftforge/forge/index_1.12.1.html

See https://github.com/MinecraftForge/MinecraftForge/issues/3933

This has been fixed in 1.12 versions of Forge, but not 1.11.2.

As a note for the future, using the @Override annotation on all the methods you override from Block will help prevent issues like this - it'd cause a compile time error, as the getAiPathNodeType method doesn't exist in the Block class in 1.10.

Is there even any need to have two methods? I feel like it would be better just to relocate the existing call to onLoad() to a more suitable place.

The issue isn't infinite recursion, it's just loading cascading to a large number of chunks. A TE in chunk A causes chunk B to load, which causes another TE to load chunk C, etc. Forge flags up when something similar happens during chunk population (see https://github.com/MinecraftForge/MinecraftForge/pull/3756) as it has been noted to cause issues there. Here the issue might not be so bad, it depends on what block positions the tile entities check - if they only check individual neighbouring blocks, for example, it's not likely to trigger very many spurious chunk loads.

When a chunk is loaded, onChunkLoad() gets called. If something happens then to cause another chunk to be loaded, onChunkLoad() will get called on that chunk as well. That is the potential issue. update() is only called on chunks that are ticked, so a chunk load there won't necessarily cause a call to update() on the new chunk.

You can overflow into unloaded chunks in update() but those chunks won't get ticked, preventing recursive loads. ChunkEvent.Load is fired from the end of Chunk#onChunkLoad(), and is a 'safe' place to access the world from. onChunkLoad() could be a viable place for calling another TE method, after World#addTileEntities() is called.

Some of those optimisations have been adopted. Of the ones that haven't, I'm not aware of if they were even considered or, if they were, the reasons given for not adopting them. Without that information, it seems somewhat hypocritical to call something 'dubious' on the basis of no real evidence. By all means make the point that this is not an officially endorsed solution, and I would also recommend using 64-bit Java if at all possible.

Well the log says: That's System.getProperty("os.arch") And FoamFix is not intended to be a "performance optimisation" mod, it's intended to reduce memory usage, which it does.

You have a 32-bit OS, which is limiting the amount of memory you can allocate. Try 1280MB, that might work. Also consider adding FoamFix (https://asie.pl/Projects/Minecraft/Mods/FoamFix), which will help save some memory.

Something that might be worth checking: make sure both java.exe and javaw.exe have any needed firewall permissions etc

Code blocks, quotes and text boxes all have really low contrast for text using the dark theme.

This is a known issue. The graphics drivers for HD 2000/3000 integrated graphics don't support Windows 10 properly. You'll need to use a version of Java before 8u60. The version that the Mojang launchers bundle should work.

Thanks for the descriptive report. This is a known issue: https://github.com/MinecraftForge/MinecraftForge/issues/3095

Consider making an issue for this on the Forge GitHub page: https://github.com/MinecraftForge/MinecraftForge/issues

For 1.8 to 1.9, read: https://gist.github.com/williewillus/e37edde85dc78d2e138c Particularly: https://github.com/kashike/migration/wiki/1.8.9-to-1.9 for renames. They also link this https://gist.github.com/bonii-xx/dcb1b6a9a1d13b69a1c9, which you can use with IDEA to do renaming for you somehow? Hope this helps.

Is it the same problem as in this thread?

I've been running into this issue too in 1.9.4, and it's pretty vexing. Does anyone have any ideas as to what causes it? Relevant Block code: @Override public void onEntityCollidedWithBlock(World worldIn, BlockPos pos, IBlockState state, Entity entityIn) { if (!worldIn.isRemote && entityIn instanceof EntityPlayerMP && PortalUtils.checkEntity(entityIn)) { PortalType typeIn = state.getValue(TYPE); int origin = worldIn.provider.getDimension(); BlockArea area = PortalUtils.isInsideActivePortal(origin, pos); if (area == null) { area = PortalUtils.isInsidePortal(origin, pos); IBlockState border = Blocks.QUARTZ_BLOCK.getDefaultState(); if (area == null || !PortalUtils.checkPortal(worldIn, area, border, state)) return; } int destination = PortalUtils.getDestinationDimension(typeIn, origin); WorldServer worldOut = MiscUtils.worldServerForDimension(destination); PortalType typeOut = PortalUtils.getTypeMapping(destination, origin); Teleporter teleporter = new PortalTeleporter(worldOut, area.getSize(), typeOut); EntityPlayerMP player = (EntityPlayerMP) entityIn; worldIn.getMinecraftServer().getPlayerList().transferPlayerToDimension(player, destination, teleporter); } } Teleporter code: private void placeEntity(Entity entity, Vec3d location) { double x = location.xCoord, y = location.yCoord, z = location.zCoord; float yaw = entity.rotationYaw, pitch = 0.0f; entity.motionX = entity.motionY = entity.motionZ = 0.0; if (entity instanceof EntityPlayerMP) { EntityPlayerMP player = (EntityPlayerMP) entity; player.connection.setPlayerLocation(x, y, z, yaw, pitch); } else { entity.setLocationAndAngles(x, y, z, yaw, pitch); } }

I'm trying to set up some blocks with differently textured variants. When I try to register a block with an ItemMultiTexture as its ItemBlock, the game crashes with the stacktrace: java.lang.NoSuchMethodException: net.minecraft.item.ItemMultiTexture.<init>(net.minecraft.block.Block, mod.block.PortalFrameBlock, mod.block.VariantBlock$1) at java.lang.Class.getConstructor0(Class.java:3082) ~[?:1.8.0_92] at java.lang.Class.getConstructor(Class.java:1825) ~[?:1.8.0_92] at net.minecraftforge.fml.common.registry.GameRegistry.registerBlock(GameRegistry.java:272) [GameRegistry.class:?] at net.minecraftforge.fml.common.registry.GameRegistry.registerBlock(GameRegistry.java:239) [GameRegistry.class:?] at mod.block.VariantBlock.<init>(VariantBlock.java:24) [VariantBlock.class:?] at mod.block.PortalFrameBlock.<init>(PortalFrameBlock.java:30) [PortalFrameBlock.class:?] at mod.block.ModBlocks.init(ModBlocks.java:18) [ModBlocks.class:?] at mod.proxy.CommonProxy.preInit(CommonProxy.java:16) [CommonProxy.class:?] at mod.proxy.ClientProxy.preInit(ClientProxy.java:14) [ClientProxy.class:?] at mod.TestMod.preInit(TestMod.java:34) [TestMod.class:?] Context from my code: protected VariantBlock(String name, Material material) { super(material); setUnlocalizedName(name); setRegistryName(name); Function<ItemStack, String> function = new Function<ItemStack, String>() { @Override public String apply(ItemStack input) { return getTypeForMeta(input.getMetadata()).getName(); } }; GameRegistry.registerBlock(this, ItemMultiTexture.class, this, function); } registerBlock() wants me to pass in the constructor arguments for ItemMultiTexture, not the types, but ItemMultiTexture has a (Block, Block, Function<>) constructor, which doesn't match the subtypes produced from the arguments. Is there any way to work around this, or should I just begin porting everything to 1.9.4?