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

Source output impedances in the typical range of about 100 - 500 ohms should easily drive preamps with input impedances of 10K or up....as long as the cables are kept reasonably short.  However, it must then be considered what happens between the passive preamp and the amplifier, how long are the cables, and what is the input impedance of the amplifier?  All of these are factors contributing to the successful use of a passive preamp IMO. 

 I believe it is to simplistic to simply say that if you have enough voltage and reasonable impedances, then a passive will sound good.  I have heard it go the other way too many times.  Based on comparisons between passives and my unity gain buffered preamp, the active buffering (but not necessarily gain) does lead to improved dynamics and bass, even when voltage and impedance rules are met between partnering equipment.

However, it must then be considered what happens between the passive preamp and the amplifier, how long are the cables, and what is the input impedance of the amplifier? All of these are factors contributing to the successful use of a passive preamp IMO.

Nelson Pass addresses this also, as I have always said with the Lightspeed Attenuator passive, interconnect to the amp should be good quality low capacitance, which nearly all are, use <100pf per foot at 2mts or less.

Nelson Pass quote:
"The output impedance needs to be low enough to drive the capacitance of a reasonably long cable. How low does it need to be? I would say that it should be able to drive a 1000 pF load out to 100 KHz.
The worst case output impedance of the Aleph L will drive 1000 pF with a -3dB rolloff at 225 KHz!!."

Cheers George

@georgehifi 
I don't think anyone is trying to call Nelson wrong. But look at his numbers. .2% distortion at a volt? That's not particularly impressive. If I built one, I'd buffer the input. That would get performance way up on it. 

Please give the man a break, he’s earn’t his dues and his credibility, more than you, I, Ralph and everyone else here put together. No wonder he doesn’t come here.

Nelson Pass on the Aleph L preamp. "Less than <0.1% at full volume" (with active gain in the signal path).
This is minuscule and it’s probably 2hd. Knowing Nelson it’s the nice (tube like) distortion .
And it’s only when the active circuit is fully in over 3 o’clock. All active circuits have some small distortion.
At below 3 o’clock when it’s passive it’s probably zero %.

Pass Labs: Printed SPECIFICATIONS Aleph L (volume control full clockwise)
Maximum Gain 10 dB
Freq. Response +0, -1 dB @ 10 Hz and 100 kHz
Distortion < .1 % THD

Give it a rest.
Cheers George

@georgehifi 

Give him a break? You're banging on as if this thing is audio divinity. Like so many things he's designed, it's a quirky component with a narrow appeal. Like the Zen amps. Like the F series. Like the SITs. Pretty much all of those products are ideal examples of something for someone, but their use must be carefully considered. The J2 might be a great amp. I'll never buy one because it would sound like garbage with my Focals. I don't think the Aleph L would be a particularly good pre-amp with my F5. .x% distortion would be an insane amount though that amp.