[1.8] Mod Development Help Thread/Updating to 1.8 Thread
- Thread starter Chaka
- Start date
-
The FTB Forum is now read-only, and is here as an archive. To participate in our community discussions, please join our Discord! https://ftb.team/discord
So what happens when CoFHCore, AsieLib, Mantle, etc., etc., all do the same thing because they're not willing to take a dep on DragonAPI? Are all these hooks going to be able to coexist?
Yes, they can coexist. If Forge refuses to include the fix on its own, there is no choice but for us to put it in our own code.
http://www.reddit.com/r/feedthebeast/comments/2ni6av/eli5_the_18_block_model_change/cmdyh4x
Quote from skyboy on the TB thing to confirm it
Quote from skyboy on the TB thing to confirm it
Also @asiekierka
http://www.reddit.com/r/feedthebeast/comments/2ni6av/eli5_the_18_block_model_change/cmdvu02
That basically means We have to reimplement something. For a more in-depth explanation of the whole problem asie provides againasiekierka said:
http://www.reddit.com/r/feedthebeast/comments/2ni6av/eli5_the_18_block_model_change/cmdvu02
I posted that on the previous page of this thread xD
This link may be of use to some people:
http://minecraft.gamepedia.com/Block_models
It's basically a description of the JSON file structure for blockstates, models, and items. Have a look at various assets in the mojang folder to compare and see these things in play. After reading the above link I'm pretty confident in what each of the parts of the JSON's do
http://minecraft.gamepedia.com/Block_models
It's basically a description of the JSON file structure for blockstates, models, and items. Have a look at various assets in the mojang folder to compare and see these things in play. After reading the above link I'm pretty confident in what each of the parts of the JSON's do
holy **** what where they thinking with even using a JSON format...
even minimized that thing is too verbose to even be considered to be used in any decent title. granted I find obj to already be too clunky for anything but experiments and learning.
There should be an option for dynamically choosing to render only some parts of the models. an extension of the variant system should do where if multiple variants match then render all of them.
even minimized that thing is too verbose to even be considered to be used in any decent title. granted I find obj to already be too clunky for anything but experiments and learning.
There should be an option for dynamically choosing to render only some parts of the models. an extension of the variant system should do where if multiple variants match then render all of them.
If it does get into Forge, is there a significant reason to update to 1.8 yet?
Well 1.8 did add some nice mechanics to the game. The new underwater mob (I haven't actually seen that many mods that add something similar to that to be honest) and the totally new villager AI (who can now plant crops and do stuff like that). I sure would like to play with those at some point.
I'm pretty sure the underwater stuff was added to Gany's Surface, and I never care about villagers.
Ok, that's your personal choice (about the villagers) but other players may actually like to work with villagers (like me).
The reason why individually you should update to 1.8 is because:
1. Despite your lack of interest in updating and the difficulties it will bring with modding you must realise you aren't going to be able to stop everyone. It's just not going to happen
2. If you want your mods to be played they need to be played on what everyone else is playing. I guarantee you. People will move to 1.8 eventually and you will be left behind.
I should mention that you obviously don't need to update now, you can wait until everyone else has started moving if you wish since I don't see any good 1.8 packs coming out in the next month or two.
1. Despite your lack of interest in updating and the difficulties it will bring with modding you must realise you aren't going to be able to stop everyone. It's just not going to happen
2. If you want your mods to be played they need to be played on what everyone else is playing. I guarantee you. People will move to 1.8 eventually and you will be left behind.
I should mention that you obviously don't need to update now, you can wait until everyone else has started moving if you wish since I don't see any good 1.8 packs coming out in the next month or two.
I'm kinda worried about all the BlockPos, Integer, IBlockState etc. instances that are created on the heap and burden the gabage collector.
It might look more elegant in code to pass a BlockPos to a method instead of x, y, and z but god knows how much RAM and GC time it costs.
Looks like Block States haven't completely replaced metadata: There are getStateFromMeta and getMetaFromState which have to be overridden. Before 1.8 metadata only goes up to 16 and if that's still the case then you can't store more than 4 bits in a block state. Now compare that to how much RAM all the pointers and allocated objects around a block state take.
I doubt that 4 GB of RAM will be enough for any medium sized modpack in the future.
I updated a small mod of mine to 1.8. Seems like the blocks work functionally now but they all look like black and purple squares yet.
It might look more elegant in code to pass a BlockPos to a method instead of x, y, and z but god knows how much RAM and GC time it costs.
Looks like Block States haven't completely replaced metadata: There are getStateFromMeta and getMetaFromState which have to be overridden. Before 1.8 metadata only goes up to 16 and if that's still the case then you can't store more than 4 bits in a block state. Now compare that to how much RAM all the pointers and allocated objects around a block state take.
I doubt that 4 GB of RAM will be enough for any medium sized modpack in the future.
I updated a small mod of mine to 1.8. Seems like the blocks work functionally now but they all look like black and purple squares yet.