One thing I noticed and posted on PT.net... he's using DYNCYLAIR_DMA as his airmass PID, but his MAF is not disabled. If his VE table does not match up with his MAF calibration, then it's quite possible that his airmass is being incorrectly reported.
GM.DYNCYLAIR_DMA is incorrectly shown as the PID for spark reference in the tuning tool.
This link here will show a screenshot that proves this. I went back to running a 1 bar MAF for a while but still boosted. As you can see, GM.DYNCYLAIR_DMA is flatlined, yet timing follows the change in DYNCYLAIR and CYLAIR_DMA.