Okay, look guys, I screwed up in my post. I didn't mean that I was going to be toying with the range. I record a whole side of a record at a time, then draw the resulting file out and look at all the loud parts for that side. When I find the loudest spot on the record, I adjust the overall volume for that entire record based on that one spot. I do not bring any other parts up, nor adjust anything else down, because I know the peaks will not get higher than that one spot. If it happens in more than one place, then I play back every second of every loud spot and adjust ONLY ONCE to the absolute loudest part. This maintains the integrity.