How much do I need to spend to get a preamp that sounds better than no preamp?

My AI seems to sound best with the gains at full volume. Is this possible? Never tried it before. Always leave them at 12 o'cock.

Just bought the McIntosh C47 preamp, very similar in performance to the C52 that was reviewed in the latest TAS. Very nice, much better than going straight from my dac to the Classe monoblocks. Plus I needed  preamp since I also have a tt and a cd/SACD player. When Going straight from the dac into the amp, the sound seemed lifeless. The preamp has made the SQ come alive, more spacious and the instruments are more articulated/distinct.

Keep in mind, there are no right or wrong solutions here but rather what the Op likes. We could give scientific measurements from now on (don't think for a minute they are not appreciated) but you already know it really comes down to what the OP perceives as the best sound, at least to his ears. Apparently he likes the dynamics of an active preamp ( like his L1) but again I would consider the McCormack passive preamps as an alternative for greater dynamics.

The audible Illusions L1 offers a really good sound for the price and for some reason much of the competition in its price range don't come close. Yes they can be a bit hard on tubes. When I had mine I ran the same tubes for two years without failure but most of you know that many of the NOS tubes are not durable enough to last without premature failure. So it really comes down to using durable tubes or at least what is recommended by the mfg. Another thing to consider is that even though you might turn the power off the tubes are still lite at a lower voltage .

I forgot to mention that when you purchase an Axiom passive preamp the mfg will ask you what power amp you will be driving so that it can be configured to deliver the best sound. When you buy a used one on the internet it may be designed for a different amplifier and you might not experience the best possibble sound and dynamics you would expect. Just a thought.

@georgehifi 
Wrong yet again. That doesn't even make logical sense. Why would you amplify the signal to it's absolute maximum before attenuating it? It makes a lot more sense to feed the gain stage the least amount of signal to keep the distortion of the gain stage as low as possible instead of as high as possible.