While playing my current survival mode and making houses I realized you can't stack slabs of opposite trees together I suggest this be fixed and we should be able to add say a birch slab and a spruce slab, or oak and stone slab, together to make house designs more aesthetically pleasing to the eye. Eliminating the gap between a slab and a separate block but only for another slab... wooden or stone.
I know this is a super late reply, but I think that would require every single combination of slab mixing, which would take up a TON of block IDs. At least that's how I remember it being, I wouldn't be surprised if someone stepped in and hit me with a "actually no that's not how that works" post.
This would likely be impossible unless it became possible to have two (or more) blocks in the same space. If this change happened, it could also become possible to have flowing water in non-solid blocks like stairs, slabs or fences, lava-logged blocks, redstone on or under slabs, carpets on slabs or stairs, torches on slabs, and more; however, it would also probably break commands like /fill and /testforblock.
This would likely be impossible unless it became possible to have two (or more) blocks in the same space. If this change happened, it could also become possible to have flowing water in non-solid blocks like stairs, slabs or fences, lava-logged blocks, redstone on or under slabs, carpets on slabs or stairs, torches on slabs, and more; however, it would also probably break commands like /fill and /testforblock.
This is not impossible, just add a new block for every slab combination. Of course, that would massively clutter up the Creative inventory and add thousands of new block states, and all their models and stuff, which are extremely inefficiently coded, and you'd need to have the game automatically generate each block (you wouldn't want to have to manually add them, even a single new slab would require two new combinations with every existing slab). Alternatively, they could make a new tile entity which stores two slab variants, which would work in reasonable amounts (a few thousand in loaded chunks wouldn't have an impact, and only tile entities with dynamic models, such as chests, are bad for FPS performance*).
*Case in point - this is a Superflat world with a layer of flower pots, which I turned into a tile entity so they can hold an unlimited number of plants (Mojang did the same in 1.7 before later making them separate blocks after 1.13):
This was before I added more variants, including multiple variants of the same block, e.g. saplings can render as any type of tree, no matter though, as over 2 billion block variants can be stored in a single 32 bit value:
See also:
So, in other words, even on potato machines like mine, tile entities that do nothing don't really contribute to performance loss all that much. And really, does the average person use more than 65k slabs in any build?
Performance impact of various blocks; chests are so bad since they are dynamically rendered in order to have a smooth animation, but functionally identical barrels are just an ordinary block and have no more direct impact than a block like cobblestone (TEs do use around 10x more memory than basic blocks, at least with the data formats used back in 1.6.4, but even 65,536 is only a few MB for a basic TE that only stores a couple numerical "blockstates"):
There did used to be a performance issue when having a lot of tile entities, causing lag spikes when unloading chunks, but that has since been fixed (and the fix was incredibly simple, literally change a "list" into a "hashmap"), and you'd only use a tile entity for merged slabs, not single or same-double slabs (this can be detected when you place a slab item; if the block it is placed on is a half-slab of a different type then create the merged tile entity. Commands could also work in a similar manner; e.g. "stone_double_slab" places a normal double slab block while "stone_oak_double_slab" places a combined stone and oak slab block, no need to mess around with data values or tile entity NBT data unless you are a modder that is adding more such blocks (likewise, Mojang could have easily eliminated the need to use data values in commands before 1.13; "pink_wool" maps to ID 35, DV 6, rather than having to type in "(35 1 6" or "wool 1 6").
While playing my current survival mode and making houses I realized you can't stack slabs of opposite trees together I suggest this be fixed and we should be able to add say a birch slab and a spruce slab, or oak and stone slab, together to make house designs more aesthetically pleasing to the eye. Eliminating the gap between a slab and a separate block but only for another slab... wooden or stone.
I know this is a super late reply, but I think that would require every single combination of slab mixing, which would take up a TON of block IDs. At least that's how I remember it being, I wouldn't be surprised if someone stepped in and hit me with a "actually no that's not how that works" post.
This would likely be impossible unless it became possible to have two (or more) blocks in the same space. If this change happened, it could also become possible to have flowing water in non-solid blocks like stairs, slabs or fences, lava-logged blocks, redstone on or under slabs, carpets on slabs or stairs, torches on slabs, and more; however, it would also probably break commands like /fill and /testforblock.
My suggestions: Enhancements - Throwable Fire Charges - On Phantoms and Elytra. Also check out The Minecraftian Language. This signature is not here to waste your space.
I believe that combining two different slabs is on the list of commonly suggested features that Mojang refuses to add to the game.
This is not impossible, just add a new block for every slab combination. Of course, that would massively clutter up the Creative inventory and add thousands of new block states, and all their models and stuff, which are extremely inefficiently coded, and you'd need to have the game automatically generate each block (you wouldn't want to have to manually add them, even a single new slab would require two new combinations with every existing slab). Alternatively, they could make a new tile entity which stores two slab variants, which would work in reasonable amounts (a few thousand in loaded chunks wouldn't have an impact, and only tile entities with dynamic models, such as chests, are bad for FPS performance*).
*Case in point - this is a Superflat world with a layer of flower pots, which I turned into a tile entity so they can hold an unlimited number of plants (Mojang did the same in 1.7 before later making them separate blocks after 1.13):
This was before I added more variants, including multiple variants of the same block, e.g. saplings can render as any type of tree, no matter though, as over 2 billion block variants can be stored in a single 32 bit value:
See also:
There did used to be a performance issue when having a lot of tile entities, causing lag spikes when unloading chunks, but that has since been fixed (and the fix was incredibly simple, literally change a "list" into a "hashmap"), and you'd only use a tile entity for merged slabs, not single or same-double slabs (this can be detected when you place a slab item; if the block it is placed on is a half-slab of a different type then create the merged tile entity. Commands could also work in a similar manner; e.g. "stone_double_slab" places a normal double slab block while "stone_oak_double_slab" places a combined stone and oak slab block, no need to mess around with data values or tile entity NBT data unless you are a modder that is adding more such blocks (likewise, Mojang could have easily eliminated the need to use data values in commands before 1.13; "pink_wool" maps to ID 35, DV 6, rather than having to type in "(35 1 6" or "wool 1 6").
TheMasterCaver's First World - possibly the most caved-out world in Minecraft history - includes world download.
TheMasterCaver's World - my own version of Minecraft largely based on my views of how the game should have evolved since 1.6.4.
Why do I still play in 1.6.4?