I played the pre-release demo a year ago on the developer’s request (thanks to this thread, incidentally). This is the feedback I gave:
The investigation part did not feel very interesting. It was just a matter of finding every hotspot (which could take quite long even with the hotspot indicators). The core problem was that there was no point in engaging with the findings until you were ready to start solving, and no point in starting solving until you had unlocked all the clues. (And since the clue text is mostly very short and fact based. Reading a clue wasn’t its own reward.)
I liked the last puzzle since there was both an attempt at framing and a way to tell it was a frame. That’s a satisfying kind of deduction from the environment rather than just spoonfed clues.
The multiple choice gameplay worked well enough, and I can see the Obra Dinn influences. It didn’t quite give the same effect, due to the much more constrained scenarios and more explicit hints. Satisfying as a logic puzzle, but it felt a bit mechanical rather than the intuition and leaps of logic from a fictional master detective.
I’m curious about what they changed, and especially if they managed to design their additional scenarios to lean more into the subtle environmental clues than explicit “here’s a word you can enter into the multiple choice question” ones that were the meat and potatoes of the demo. It’s definitely worth a shot. but I think the avclub review is overselling it.