Reminder: how to tell current from an amp's specs?

P=IE.  Power = Current * Voltage.  

Problem is that amplifier rail voltage varies and is seldom specified.  

Pick an amp that increases output as the resistance is lowered and is stable at loads as low as possible- e.g. 2-ohms.  

This is an interesting exchange that addresses one of the most important aspects of high end audio: how amplifiers and speakers interact. So I'm going to keep going at the risk of being "outgunned," In my world facts and evidence are more important than the reputation of the participant.

If the amplifier is able to act as a voltage source at less than full power then the contray is certainly possible. For example the Osiris amplifier that @akg_ca mentioned can't double power at full power as load impedance is halved, but at lesser powers it can. In a case like that when the current demand is exceeded, rather than changing the FR the amp will simply clip.

I belong to a Krell Forum where there are several people who have very high levels of expertise, including a former CEO of Krell. I have asked this very question, i.e., Does an amp produce more relative current at low power output than full output? In other words, if a 100 watt amp is rated at 100 watts at 8 ohms and 100 watts at 4 ohms, will the amp perform better when it is running at 10 watts? Could it produce 10 watts at 8 ohms and, say, 15 watts at 4 ohms? The answer was clear and unequivocal. The power output as a function of impedance is a ratio that stays steady throughout the amplifiers power range. Running an amp at low volumes does not make it perform better into low impedance loads. In the Stereophile review I see nothing saying that the amplifier performs better at lower power output. If that the case why did JVS hear the reduced bass through his Wilson speakers? He certainly was listening at levels far below full output. Also, why didn't JA mention this in his testing. He clearly said that the amp wasn't appropriate for speakers with impedances below 4 ohms. Why didn't he put a qualifier on that statement explaining that it would be fine at low volumes? If anyone can provide evidence that supports the idea that amplifiers are more capable of driving low impedances when they are run at normal volumes please provide it. I am more than willing to change my mind on the basis of good evidence.

Just because your amplifier can drive to 2 Ohms and is able to double power as it does so does not mean its sounding its best when doing so. All amplifiers make higher distortion when driving lower impedances! If you think that distortion is inaudible think again- the increased distortion is audible as increased brightness, harshness and a reduction of detail (distortion obscures detail) because most of that added distortion is unmasked higher ordered harmonics.

My Krell KSA 300S was measured to have 0.12% distortion at 2 ohms, less than 0.11% distortion at 4 ohms, and slightly less than 0.10% distortion at 8 ohms. If anyone can hear the difference between these distortion figures then you have a lot better hearing than I have. I would consider these differences insignificant but if anyone can provide evidence that listeners could hear these differences please provide it. Again, I'm glad to admit I'm wrong if I encounter facts or evidence that contradicts my understanding of an issue.

There seems to be contempt on the part of some forum participants regarding speakers that have a demanding impedance curve. My view is that this is simply a design decision on the part of the speaker designer. They felt that it was a necessary tradeoff to achieve the sound they were after. I think it's going to be pretty hard to tell Wilson or Thiel owners that they made the wrong choice and that their speakers are full of distortion because their megabuck amplifiers perform poorly at low impedances.

I'm going to state my basic point again. You can run your low impedance speakers with a tube amp all you want. If it sounds good to you then, Terrific! Just understand that you are introducing an unpredictable tone control into your system and that you will not be hearing the speakers sound the way their designer intended. Jim Thiel used big Krell, Levinson, and Threshold amps to voice his speakers and he specifically used a Krell KSA300S to voice the CS6 which is the model I have. I personally want my speakers to sound the way they were intended to sound. I suppose that I could try a variety of tube amps to see if I liked any of them but I would rather buy an equalizer and just change the frequency response to suit my taste. Fortunately I am extremely pleased with the way my system sounds and I have no desire to deviate from the type of amplifier that Jim Thiel used to design my speakers.

Krell amps don't really double the power when impedance is halved, krell understated the 8ohm power ratings, check out stereophiles measurements. I have that same amp you have it' definitely has an overbuilt power supply, but to double it's power with each halving of impedance it would have to have a perfect power supply which doesn't not exist in the real world.

You are correct but my point still stands. Krell amps are better than pretty much anything on the market when it comes to supplying current to low impedance loads and the increase in distortion is negligible at lower impedances.

Krell and other beefy amps may not be perfect but they are lightyears ahead of tube amps when feeding speakers with demanding impedance curves. Their less-than-perfect performance was taken into account by Jim Thiel and David Wilson when they voiced their demanding speakers using these amps.

Some Thiel owners say tube amps sound better, with more natural bass than solid state amps, just goes to show you don't know just going by the numbers, listening is the best policy.