Ever perform a preamp bypass comparision test?

I'm still in the throes of trying to evaluate my latest preamp acquisition. Without going into what exactly that is right now, part of my testing lately has been to try bypassing its gain and attenuation sections by taking the output from the processor bypass loop straight into the amps. This sends the unpreamplified signal from the source directly from the preamp's input jacks to the output jacks at unity (zero) gain, preserving the same jack connections and cable runs as are present when using the preamp in the normal way. By setting the volume control to attenuate the regular output so it equals unity gain as well, and swapping the output leads between normal out and processor loop out, I can make volume-matched comparsions of what effects the gain and attenuation stages of the preamp are having on the signal.

Obviously, there is going to be some degradation or changes to the signal revealed in this test, and I am hearing them. At this point in my post, I could go off on a rant about how I fail to understand many print and webzine reviews of preamps that indulge in rhapsodizing over the wonderous benefits conferred upon the music by XYZ preamp - as if a preamp can somehow not only give control over volume and source selection, but also somehow 'improve' the signal coming from the DAC or phonostage - but I will attempt to refrain from this for the time being.

What I am wondering now is how many of you have tried this in your systems, and what were your opinions of what you heard if you did? Has anybody done this and failed to detect a difference? Anybody feel there was actually an improvement of some kind with the preamplification engaged? My own feeling is that if you answered in the affirmative to either of the last two questions, you either have yourself one hell of a magical preamp, or your sources' outputs are not very hardy.

[If you have never tried this test and want to give it a shot, just take care beforehand to judge the resulting volume you will be subjected to when running your source unattenuated straight into your power amplification, because you won't be able to control the resulting volume (unless your DAC has a variable-level output feature) - the level will be determined by what's on the disk and source's own output level. Just move the output leads going from the preamp to the amp from the regular preamp output jacks over to either the processor outs or the tape outs (if unbuffered), and then set the preamp volume control to match that level when listening from the normal attenuator-controlled outputs (unity gain). From there it's just audition and swap, audition and swap, audition and cry...]