[0.16.27] inconsistent inserter to splitter behaviour

[Dupl3xxx]

When inserting directly onto a splitter, the lane selected is not consistent with rotation. When inserting onto a belt facing away from the inserter, the behaviour is consistent.

[Loewchen]

viewtopic.php?f=23&t=24617

[kitcat]

I’d say this is not a duplicate because 24617 was about inserters placing items before or after a splitter. This one is about inserters placing items on the wrong side of a splitter’s belt. The inserters facing west or south are clearly placing items on the left lane even though inserters placing items on belts facing away from then should always place items on the right lane, which they do correctly in the screenshot when placing on belts, no splitters.

[Dupl3xxx]

The problem here is not that inserting onto splitters is different than belts, but that it's not consistent with rotation. As far as I know, blueprint mirroring is not getting implemented in vanilla because not everything works with mirroring, like rails or oil. However, this problem breaks rotation. And, as far as I'm aware, this is the only thing that breaks rotation. With train-unloading, inserting directly onto a splitter is fairly common, so this isn't some weird, highly unlikely scenario either.

[Loewchen]

The problem here is not that inserting onto splitters is different than belts, but that it's not consistent with rotation. As far as I know, blueprint mirroring is not getting implemented in vanilla because not everything works with mirroring, like rails or oil. However, this problem breaks rotation. And, as far as I'm aware, this is the only thing that breaks rotation. With train-unloading, inserting directly onto a splitter is fairly common, so this isn't some weird, highly unlikely scenario either.

So you mean this one: viewtopic.php?t=43461?

[Dupl3xxx]

Yes. The fact that it doesn't 'split' is a mild annoyance. The fact that it does not follow the 'standard' for which lane to put items on is the real issue here.