Linking this picture below back to the link shared in the post above covering the idea of how the Euclidean steps relate to the ‘underlying’ seq steps, it should help visualize the words …
Think about the specific case discussed here to serve as a example …
It’s a simple case a 16 step pattern and you assign Pulse1=5
Think of the distribution of the Euclidean steps like a stretched ‘window’ into the underlying pattern … it is always stretched unless you have 16 active Euclidean steps lit in purple (Pulse1=16 would do this here with that being the only parameter set)
The ‘gaps’ between Euclidean steps are of zero length on the ‘underlying’ pattern, but of variable length on the Euclidean view as we use the pulse seeds and logic to snap the ‘windowed’ trigs into different positions in time - the trigs on the Euclidean trig do have a time interval between them, but it’s varying depending on settings. But it also explains why you cannot p-lock on the unlit Euclidean steps - they are non-existent on the source pattern. For consistency, this is how it works, think of it as a different way to view the same data, but you selectively stretch and rotate the positions of a (generally) reduced set of the potential trigs on or destined to be on the source pattern. If you print/freeze, you basically start with a new underlying pattern and if those frozen steps had p-locks on them, the next attempt with the Euclidean mode (using the same parameters) will generally be different than the first as you have a new linear sequence of ‘underlying’ trigs to pull in (if you p-lock step one on the first attempt, it will still be on one next try, but all other permutations will likely re-order)
I find visualizing a stretched (re-timed) version of the ‘underlying’ steps to be the easiest way to follow how the step views of the same data ‘connect’ , it makes rotating easier to conceptualise - the operators and seeding just affect the number and placement of trigs to look at in the window we audition
