Nukta clashing with below-base subscripts in Telugu and Kannada

[sridatta1]

Based on this tweet and my earlier testing in Kannada
https://twitter.com/virtualvinodh/status/1542533851133755393
Kannada
Some sample clusters, can be tested for more ಕ಼್ತ ಕ್ತ ಜ಼್ಜ಼ ಜ್ಜ ಫ಼್ಫ಼ ಫ಼್ತ ಫ್ತ

Telugu
క఼్త క్త ఫ఼్త ఫ్త

[tiroj]

Hmm. This gets tricky in the Kannada, given that the distance between the nukta and the subscript is going to vary based on the width of the base letter. Lowering the subscripts contextually would be the easier solution. Adjusting their position horizontally is a lot more complicated.

[Increasingly, I find myself wishing that Kannada had retained the convention of vertical arrangement of base+subscript like Telugu. While the offset horizontal arrangement of subscripts seems simpler because it can be mostly resolved with kerning, in fact it introduces more problems as soon as one goes beyond the common combinations of Kannada language.]

The Telugu should be an easier fix: a contextual vertical adjustment.