Of course you are right about the general inadvisability of turning off
the interrupts for a long time in a multitasking environmnent. And although
Thom's test program is not something you would expect to run when there could
be competition, it does include an initial delay of several seconds to give a
chance for some other process that might be inadvertently present to finish.
The point I was trying to make is that perhaps there is yet more evidence
that the Amiga's multitasking design is robust. For example, you can run the
Gronk test from Wack, and Wack does not choke because the interrupts go off.
How much that speaks for the system and how much for Wack is not clear to me.
But maybe it says something.