There's one in every group...

I saw the uc-100 cncdrive.com/UC100.html
but isn't that going the other direction?
Still confused.

That is not compatible with Linuxcnc either. It is mostly the same electronics as are on your board, stuffed into a parallel port plug adapter, and is used to connect a simple parallel port machine to a Mach3 computer using a USB port.

Linuxcnc works by sending the actual pulses that control the machine to the parallel port. While Mach3 can work that way. These USB adapters don't. Mach3 uses a driver that bypasses its own motion controller, and sends the movement commands serially through the USB to the adapter. Then the adapter, using its own motion controller, sends the pulses to the parallel port or in the case of your board directly to the step motor driver chips.

It might be possible to solder leads to the optos on that board and connect them in turn to a parallel port, but it seams hardly worth the effort, when stepper drivers of that quality are so inexpensive.

kopcicle wrote: I was afraid of that. Sorry I wasn't clear.

I really want to ditch windows. Linux seems to not see the controller correctly or completely

Product: By BlueSkyT
[1041932.787825] usb 5-2: Manufacturer: USB_HID
[1041932.787827] usb 5-2: SerialNumber: 6D7254715456
[1041932.796467] hid-generic 0003:3698:0107.0169: hiddev0,hidraw1: USB HID v1.10 Device [ USB_HID By BlueSkyT] on usb-0000:00:1d.3-2/input0

I have yet to try the live disc but so far it's been a dead end.
Although I have something that works switching between *Nix and M$ is getting old.

Linuxcnc does not support USB motion control due to latency issues with the USB.
You can most probably ditch that controller and wire drives to a parallel port on the PC. That always works with Linuxcnc, or if you feel adventurous, get a Mesa board like the 7i96 or 7i76E.

> Linuxcnc works by sending the actual pulses that control the machine to the parallel port. While Mach3 can work that way. These USB adapters don't. Mach3 uses a driver that bypasses its own motion controller, and sends the movement commands serially through the USB to the adapter. Then the adapter, using its own motion controller, sends the pulses to the parallel port or in the case of your board directly to the step motor driver chips.

If that is the case, then there is a protocol which must be functionally similar to gcode that is passed down the USB to the MX3 controller. Does anyone know the details of this protocol? Maybe via a USB sniffer? Interesting that you say there is a motion controller in the MX3 board - the folks on the Genmitsu 3018 FB group have all been saying that it is a dumb controller that merely passes on the parallel port data that has been tunnelled over a USB connection - but it does make sense given how people here have commented on the latency issues over USB.