Hi Art:
My report on V1.15 posted this morning seems to have vaporized. I think my session timed out and I didn't notice. I guess that's what happens when I'm so long winded.

To summarize, it looks like you've killed all the bugs.
The leds showing the wrong states, homing, crashing, and whatever else you've tackled over the last few days are all fixed here and it seems to be doing everything properly.
Great Work !!

I presume the issue with CPU loading has been corrected with M4 V1936??
I haven't had a problem even with the earlier versions and my system runs at about zero % as far as I can see. I ran the roadrunner Gcode at warp speed and still only had the odd quick spike up to maybe 15%.
So what were the conditions that folks were seeing high CPU load??
Seems like a lot of folks are asking what the minimum machine requirement is for M4.
Perhaps the developers should come up with a standard test condition.
1. Gcode file for all to execute with dimensions that fit on a small machine so all can execute it.
2. A standardized tool path display setting.
3. A reasonable machine movement speed that everyone can achieve (even if it's slow or a little fast for some folks)
etc. so that folks can evaluate different machines and report back what they are using and their loading. This way a set of features that are good/bad or indifferent for machine hardware can be determined.
Sage