By varied, I mean that I seem to do better when I have some variety among my tasks. Fix a bug, make a texture, read XYZ article as research, rather than fix bugs all day.
There's a cost to context shifting my brain, but there's also diminishing returns by staying on something too long.
Going back to pomodoro, a TTRPG creator / influencer I follow was recently plugging that as a creativity tool. As I recall it was something like: one idea is easy to brainstorm, but your first idea isn't necessarily your best - push yourself to come up with stuff for X minutes (I think maybe he used 15) before jumping on something.
Here's a link that jumps to that spot - it looks like I mixed it up - he's talking about using it for productivity & focus not necessarily for ideation - so maybe it fits with your query about avoiding burn out better than I previously thought.
Yeah, context switching is generally seen as expensive, but if overall it works best this way for you, that's good! I suppose I could do something like that too.
I also have a "too much time in chair leads to spine pain" factor.
But yeah, it's really about "this is something that works for me" rather than "this is the way". If AM/PM switch off works for you, by all means, lean into it.