On Jackpot, the first two stepper (X, Y) are on the first mux I2S chip.
Adding Z and A then uses two mux chips.
Adding B and C then uses all three mux chips.
I bet this has to do with he ESP-32 built in I2S bit clocking acceleration that is THE enabling feature for FluidNC supporting 6 stepper drivers.
The JL1 configuration doesn’t need to clock any bits out to anywhere but the first chip.
Fix for the Jackpot driver initialization problem is out as pre9. I’ll try it later this evening.
I’ll put this on the JL1 this evening as well.
I have it running on a test board right now, All seems well, but this is just a test bench. I can slap it on the LR in a bit and home it a few dozen times.
homing was working ok on the last 2, but you couldn’t drive Y, they were different speeds
I’m still connected and both Z steppers are in sync, still.
I tested out pre9 and everything works as expected again. Drivers initialize properly. Jogging works correctly. Multiple homing attempts of each axis and home all worked every time. Checked probing and ran a short job. All looks good.
So fingers crossed 
Yea, I know pre7 changed a ton of code related to spindles/VFD. That doesn’t affect me personally but something else that probably needs some testing before release.
OK, so I loaded up Pre9 this evening on the JL1.
Drove around moves and trying to get a webUI crash with reloads and stuff, couldn’t.
Busted out my old Windows Laptop and updated to the latest LightBurn.
Threw in a quick test pattern.
I still have the JL1 partially apart from using it as a testbed for my stacking pendant adapter (lets you use a pendant and still have the lower 2 GPIOs pass through so it can e.g. run an isolated RS-485 spindle.)
Didn’t bother to position the scrap plywood I threw in, and pardon the mess… but repeated this several times:
So at least from the standpoint of laser ops on my JL1, this is working fine.
I’ll post this on discord as well.
on 3.8.4 LightBurn did not work correctly, literally at all.
Part of the cut could have been done in another place and if I put 2 passes, then he makes 4 passes.
Now I installed 3.8.9 and have not yet noticed any strangeness. But I haven’t seriously tested it yet.
The latest release is 3.8.4-pre9 as described above. The issues you describe are part of known bugs, with the known stable release prior to this being 3.7.17. Anything between these up to 3.8.4-pr8 have significant issues.
Link to the latest pre-release which seems to be fixed:
I tested it a little and now I’m not sure.
Wrong return home after laser cutting
What exactly does this mean so we can try to duplicate the issue?
Agreed, it seemed flawless to me last night.
I CAN’T EXACTLY IDENTIFY THE PROBLEM
I run tests and get the same problem in 3.7.17.
Most likely something with LightBurn and my settings.
Once I figure it out, I’ll go back to 3.8.9 and try again.
And here’s the problem. that after cutting, the Y axis goes to its original position, and the X axis remains in place, and the Y axis rests against the limit switch and turns off
I’ll figure it out
Well, at least this isn’t introduced with a 3.8 version. Hopefully it’s just some kind of config.
Hooray! Released as 3.9.0
Hopefully lol
We hooray’d for 3.8 also 
