... how important is Damping factor in determining what amp to buy

Interesting discussion guys.

It is not important at all. It is just another way of expressing the amplifier’s output impedance.

Most SS amps have an output impedance of 0.04 ohms, which is a DF of 200 or 8 divided by 0.04. If you include speaker cable impedance of, say, 0.1 ohm that means the speaker is at 98.3% of the amplifier load and the amplifier’s output impedance is 0.5% of the load, an almost perfect voltage source that is immune from the speaker’s varying load.

If you substitute an amp with a 2000 DF, it has an output impedance of 0.004 ohms and that is 0.05% of the total load. That is only a change of 0.45% with respect to the amplifier’s contribution. The speaker is 98.7% of the load, despite the amplifier being a more perfect voltage source by a factor of 10.

I seriously doubt that 0.45% change is audible. However, a general rule of thumb is that an amp with a DF of 20 or below has sonic consequences. Doing the math, the speaker in this case is 95% of the load -- a considerable change that results in sonic degradation from dynamic load variation since the amplifer is no longer a perfect voltage source .

Almost all of today’s SS amps use global NFB and high quality output transistors. It is the norm to see output impedances ranging from 0.01 to 0.05 ohms. That’s why you don’t have to worry about it.

Another way to say it is it is practically most significant in distinguishing amps when damping factor is low as in <50 or so as is often but not limited to the case  of higher output impedance tube amps.   Speakers that are tube amp friendly will tend to sound more relaxed and perhaps more natural when amp damping is lower or correspondingly output impedance higher.  

Some SS amps designed to sound more like tube amps likely also have lower damping factor.  Two I have owned that attest to that are Carver M4.0t and TAD Hibachi both of which are SS designs made to sound more like a tube amp.  

Gs5556
It is not important at all....
... a general rule of thumb is that an amp with a DF of 20 or below has sonic consequences. Doing the math, the speaker in this case is 95% of the load -- a considerable change that results in sonic degradation from dynamic load variation since the amplifer is no longer a perfect voltage source .

Gs5556, I think you may have worded your post a little more broadly than you intended to.  In addition to these two statements being contradictory, you seem to be saying that any amplifier having DF<20 would cause "sonic degradation," and you seem to be saying that for good results all speakers should be driven by amplifiers that act as voltage sources (i.e., amplifiers whose output voltage is not sensitive to load impedance variations, as long as the amp is operated within its capabilities).

As I'm sure you realize, almost all tube amps have DF<20.  And as I indicated in my earlier comment, differences in damping factor and consequently output impedance within that range can certainly be sonically significant, if speaker impedance varies significantly over the frequency range (as it does with most speakers).  And certainly differences in damping factor/output impedance can contribute to sonic differences between tube amps and solid state amps.

And certainly some speakers are tube amp friendly (meaning they don't necessarily have to be driven by voltage source amps, and in some cases shouldn't be driven by voltage source amps), some speakers are solid state amp friendly, and some speakers (especially those having relatively flat impedance curves, highish impedance, and highish sensitivity) are friendly to both. 

As Mapman alluded to just above, at the low end of the range damping factor does matter.  Statements that it is of no importance are simply incorrect.

Regards,
-- Al 

I seriously doubt that 0.45% change is audible. However, a general rule of thumb is that an amp with a DF of 20 or below has sonic consequences. Doing the math, the speaker in this case is 95% of the load -- a considerable change that results in sonic degradation from dynamic load variation since the amplifer is no longer a perfect voltage source .

The 'traditional' thinking here is that frequency response is the last word on tonality.

It isn't.

It turns out that the human ear/brain system assigns a value to distortions; they are translated to tonality. In fact it now appears that there is a tipping point wherein colorations created by distortion can be favored over actual frequency response errors.

In addition, the use of global negative feedback is well-known to contribute to higher ordered harmonics (see Norman Crowhurst). The higher ordered harmonics are used by the ear/brain system to detect volume levels (rather than the fundamental tones). The result of trace amounts of higher ordered harmonics is called brightness and harshness because of human sensitivity to these harmonics.

This is why many designers don't use feedback and frequently favor tubes- to avoid that particularly annoying coloration. This often results in a high output impedance/low damping factor but this can be easily dealt with giving careful speaker selection.