Just saw the edit about portable version as I posted reply 
Yeah, not really sure what's going on there as the portable version is literally the same EXE file. I expect it just checks for a registry entry and when that doesn't exist it goes into portable mode.
I've asked in the rawdata thread as I don't think it's my fault this time but we will see....
edit: So the problem is the preAmp figure is coming through as zero regardless of what the preAmp is set to on the portable version. It should be coming through as 1 if no preAmp is set. You might think it should come through as zero when the PreAmp is set to 0 but when you are converting things from peak to decibel and back. 1.0 peak is zero decibels so when I'm adjusting the current level by the preAmp value it adjusts it by zero as it should when no preAmp is set.
When the preAmp comes through as 0 it converts to 1.0 and when I adjust the current level by 1.0 it sends the needle to max like you are seeing. It's tricky to check for zero when dealing with decibels as you have to decide if it's zero because of a problem or it's supposed to be zero. In this case there is no reason for PreAmp to ever be zero but it highlighted a problem that might not have been noticed for a while if I had forced preAmp to be 1.0 if it came through as Zero.
I have noticed the upside down left needle on that skin is going off the meter which means I haven't handled the maxAngle thing correctly. The upside down needles make everything backwards! It seems I really suck at testing, and now I'm going to have to start testing installed and portable versions...