To me the Bishop's move doesn't look very natural either. The fact that after an even number of steps each destination can be reached through two paths makes it essentially different from a normal slider, and more like a 'crooked' piece.
Sure, it's more like a kind of crooked piece, but the same thing goes with the rook, it simply looks like a straight path. It is the idea that was suggested, to make the rook and bishop work by the same rule.
But of course you can make the pieces move as you want; there doesn't need to be a justification.
Justification is the key element here, brought by this variant's idea.
There's another way, which may suit the purpose better.
In the ideal condition for sliders, since the triangle has only three sides (and corners) and two of them are equal to each other in terms of trajectory selection, the branching of their directions should cover the whole board. That's what I also suggested with the only remark that the sliding pieces should always stay further from the initial cell and levels the piece already crossed, which can somewhat limit their power and make the game stable.
@H. G. Muller, the corrections were made.
Sure, it's more like a kind of crooked piece, but the same thing goes with the rook, it simply looks like a straight path. It is the idea that was suggested, to make the rook and bishop work by the same rule.
Justification is the key element here, brought by this variant's idea. There's another way, which may suit the purpose better.
In the ideal condition for sliders, since the triangle has only three sides (and corners) and two of them are equal to each other in terms of trajectory selection, the branching of their directions should cover the whole board. That's what I also suggested with the only remark that the sliding pieces should always stay further from the initial cell and levels the piece already crossed, which can somewhat limit their power and make the game stable.