i'd like to understand this better as well. why is it best to run the R09 at a gain of 8? i would think that whatever setting gets me the desired levels would be optimal. is running above or below 8 going to negatively impact the dynamics of my recording?
To understand this better, you are going through 2 gain stages, your pre has gain, and then the gain of the R-09. If you are using an external pre with cleaner gain than the R-09, setting the R-09 fairly low would be the best, even a little below 8 if the pre remains clean even with the gain cranked up. But many pre's self noise get much more noisy when cranked up, thats why you don't want to attenuate much on the R-09.
Each pre is different, So the best measure would be to record sample files of the self noise of your pre+R-09 at different settings with equal level and compare. The trick to setting equal level is to record a reference signal like a tone at 1khz and then set the combined gain so meter on the R-09 is -6db. For example, pre=high, R-09=4. Then do pre= medium, R-09=8, and pre=low, R-09=20, etc. That is what I did in this thread
http://taperssection.com/index.php?topic=69131.0 and then guysonic did it with the pre he sells. You might be surprised to find that using only the pre in the R-09 by itself is as good or better (self noise wise) as compared to your external pre noise added to the R-09 line in noise. That is if your mics output a high enough level like my AT-853 to be used with the mic-in low gain switch setting which is just as clean as the line-in. The mic-high gain setting gets noisy quickly.
As discussed early in this thread, unity gain is a relative thing since 0 db is a different voltage level on different devices, different op-amps, impedances, etc. Mainly what you want is a low setting on the R-09 where you have some headroom to adjust your gain comfortably with the controls on your pre (if it has a gain knob). Each change of the R-09 setting when between 4-15 makes little difference in the self noise.