Personally, I found Opus Magnum far too easy. In TIS-100, the constraints are brutal, and I often made multiple attempts before I was successful. In Opus Magnum, I finished every puzzle on my first try. You have almost no constraints. While optimizing is still a challenge, it felt much less rewarding to me. I feel no compulsion to return to OM after finishing the main campaign, but I return to TIS-100 every few months or so to mess around with the harder puzzles.
The 3 saved puzzles is a bit annoying, but you can always backup the puzzle files since they're just text.
That was by design in Opus Magnum. The game is still brutally hard by sticking to the (self-imposed) constraints, but it wants to open up to a much broader audience.
Celeste follows the same designed approach to difficulty.
> In Opus Magnum, I finished every puzzle on my first try.
I never was close to failing to finish a puzzle in OM. It was just a matter of how poor my solution performed whatever metric I chose to optimize first. The puzzles at the end of the campaign had me "give up" and settle for an ugly slow solution though.
> I feel no compulsion to return to OM after finishing the main campaign
If you haven't tried the Journal puzzles in OM, I would recommend you give them a shot. Many of them are in confined spaces rather than an infinite plane. I have yet to find a puzzle as hard as the ones in the TIS-NET directory, but they are a step up from the rest of the OM puzzles in terms of difficulty.
I find the OM approach to be much more fun when you have friends to compete with - "Making a solution is easy" opens it up to a lot more people, but "Making a GOOD solution is hard" makes it great for competitive play.
(Or even without friends, like me, I just refused to proceed to the next level until I was in the global top 1% for the current level :P)
The 3 saved puzzles is a bit annoying, but you can always backup the puzzle files since they're just text.