Unless you’re using a separate MIDI processor to split the voices, the last Shruthi in the chain will have a lag of almost 15 ms. Unless you’re doing six Shruthi XTs, programming is much better on Ambika than on Shruthi (but I wouldn’t describe the process for either as “menu diving”, the page system is nearly as nice as knob-per-function on Ambika, and is some ways is better than a typical knobby implementation). Ambika also has three LFOs and envelopes, plus an additional per-voice LFO, Shruthi only has two each (although if you were doing six Shruthis they’d all be per-voice), and Ambika has a larger mod matrix with twice as many operators. Ambika’s more powerful processor, larger screen, and additional controls opens it up for more interesting firmware mods than Shruthi. You’d need an additional mixer to combine the inputs of six Shruthis and get them to equal levels. Two Ambika can be polychained for twelve voices, infinitely more convenient than a twelve-unit Shruthi chain even if money is no object. Where did you read that Shruthi responds better to CC than Ambika? In what way?