I wanted to be able to go back to the stock ECU without having to re wire anything. with the switch, its as simple as setting it to narrowband, and plugging the stock ECU into the harness, and then plugging the MAF back in.
It allows me the freedom to still reliably use my car until I get the tune working like I want it.

Ok, so I used that tune and it seemed to run better than before, however, it will not idle without throttle input, and the AFR readings started around 14.7 then worked their way up to 16.5 once warm, before I shut it off in worry. The advance is about 17* warm.
The AFR readings between the gauge and tunerstudio were basically the same, and this was while the sensor was grounded to the head. I'm going to splice the ground into the harness at pin 3O on the mazda connector. Diagram says the harness wire color is black/blue.
I'll report back and see if that changes anything, but we will see.

On a side note, the wideband has it's positive power coming from the white/red +12v in the harness, and therefore starts when the ECU does.
The MTXL goes into heater mode for a good 15 seconds before sending signal to the ECU. Should I let it heat up before starting or should I just start it.
It seems like the megasquirt has provisions for waiting on the AFR to start sending signal, as well as the fact the MTXL manual says not to let it heat up before starting because it could shock cool the sensor and break it. Just want to see what you guys say.

Also, I ran the O2 signal wire to the stock harness using the plug from my old O2 sensor.
The wires on the plug were made of stainless steel, which I soldered to about 6 inches down from the plug. Would that cause a problem in terms of resistance and create false readings?

As for the idle, I can't tell why it's dying. If I let off I can feel the idle getting bumped up by the ECU, so I think the IAC valve is working at least. It's set to 385Hz, which I think is correct. This leads me to believe that maybe the AFR is just too lean for it to stay idling.

Ok so after making the changes to the ground I outlined above, my gauge and tunerstudio AFR's seem to be in line with one another.

At around 2000 rpm:

AFR is between 12.8 and 14 (warm)
Timing advance goes from 15 - 35 depending on load.
Dwell stays around 3 ms.
The engine feels very stumby and seems to misfire.

It STILL will not idle. I've tried every setting on the idle valve options. I made sure the output was "idle" on the MS3X board.
I know it's working too because the "CL Idle" light kicks on in tunerstudio, and I can feel it try to work when I lower the RPM, but it just doesn't want to work. I'll post the msq tommorow along with composite data logs.

I'm definitely going to spend the dosh on VEanalyze :vash:

Problem solved. Wired the flyback diode in line towards harness, which because its a negative switched wire, blocked all the current and the IAC valve wouldn't even work.
Fixed that. It now idles rock solid after a few tweaks. Took it on the road for a straight hour and let VEanalyze do it's thing. Car feels so much different.

Thanks for all your help guys.

Congrats, it all looks a bit like dark magic at first, but once you figure it all out, it's not that difficult :)

why are you wiring in flyback diodes?