General question about speaker impedance and amps

ZERO autoformers could end up solving your issue. They are worth checking into anyway.

1.) All coil driven loudspeakers vary their impdeance depending on frequency. The single number expressed in ohms on the back of the speaker is of only passing interest.
2.) You want to look at a plot showing impedance v. ferquency for your speaker.
When you look at this plot - look for violent changes in the low bass area ( You may also see some strange stuff happening around the crossover points.) To control the cone at lower frquencies you want current. A amp capable of driving the variable impedance of a real world loudspeaker will double in wattage as the load drops in impedance ( 50 into 8 ohms ,100 into 4 ,200 into 2 ohms for example. This takes power supply components that are not cheap - but they will sound more effortless in the bass area in almost all cases. Such a near ideal bass amp may or may not light your fire in terms of mid or high frequency reproduction.
Tube amps have to be able to send power through output transformers and much as I love the sound of good tube amps - I've never heard one that was simply stunning in deep bass reproduction.
Most true class A biased amps will produce the near perfect power to impedance ratio I describe above. After all - you've got to have a gonzo crazed power supply section to be able to do class A opertion to begin with - making it able to drive low impedances well is nearly a gimmie at that point. Class A's are a lot like tube amps in that a true A's watts seem to have a lot more hair on 'em than even well executed A/B amps. Probably at least part of this is because they can recover from clipping much faster than a A/B that 's been hammered to the wall. In other words a 50 watt class A amp will often walk all over a 150 watt A/B.
Disadvantages of True A biased amps - heat -weight-price - electric bills -patience.In my experience - every A amp I ever listened to - sounded much better once it had a chance to heat up ( usually seemed to take 40 min. to an hour). This is probably because in most amps bias and DC offset are interrelated in terms of adjustment. DC offset is always done once the amp has hit it's optimum temp.
Hope this is some help.

From specs, an yba passion (~200W) has a much heftier PS than its equivalent Gamut (~200W). I'm not saying one's better or worse: ultimately you'll decide if that question arises. However, if 3ohm is the lowest your speakers go, it shouldn't be a problem for any well-designed ss -- and that includes the PS as Audiopile notes above. OTOH speaker EFFICIENCY is usually rated as db (spl) @ 1kHz / 1W. IF your speakers' impedance were 8ohms, you'd need 2,83Volts and 1amp (current) from your amplifier. If your speakers were 4ohms you'd need 2 amperes to achieve this, etc. As speakers' impedance varies with frequency, and music has all sorts of frequencies integrated at any moment, the amp has to be capable of providing required voltage & current in order to drive the speakers adequately: the bass as loud as it should be in correlation to the mids, the highs, etc, i.e. it must amplify the input signal so that the speakers make it sound (as far as they can) like it was at the amp's input.
That's one reason why many people buy ss amps with large wattage -- it immediately means the PS is hefty. OTOH, please note that many SS amps have excellent specs only at FULL power. At the levels we listen to, distortion figures can go up quite dramatically -- another reason (among many) why it's good to listen before you buy

audiopile had some real good points, especially about the class A amps. I have a little 45 watt guitar that indeed walks all over larger A/B amps and is suitable for cooking food with. Ditto for the sound improvement over time though i switch from the stock EL84s to Phillips Miniwatts with very positive improvements in tone and a shorter warm up period.

Now to your original issue.. I have a pair of Dunlavy Athenas and i do not think they are very hard to drive at all, especially compared with speakers like Aerial. I'm using tubes, and have heard them with Pass labs ss amps and the bass looses almost nothing with the tubes while gaining hugely in the mids. Find yourself some amps you like and try them out from a listening perspective rather than getting bogged down in the math.

In responce to gregm comment about high power SS amps having hefty Power Supplies. Maybe /maybe not. The power at various impedance spec is the only real indicator of power supply capability. You can definitly find amps that will do 100 into 8 ohms - maybe only 120-140 into 4 and possibly as little as 150 into 2 ohms. These will definitly have crap power supplies and will tend to clip hard and long when overdriven and have sorta soggy bass on most speakers. A lot of even pretty hefty surround sound recievers and surround sound 5 and 6 channel amps will fall into this . Is the scuttlebut I heard that video/surround amps are being tested for power ratings with only one channel driven at a time - is this true? If it is true this would certainly explain the pathetic sound produced by these products under most conditions - since this makes their power ratings about as real world usefull as a top speed for a car being derived from it's terminal velocity when dropped from a C-130 at 15,000 feet.
I absoulutly agree with gregm - listening is the only thing that really matters.

Audio, I agree that the wattage "doubling" specs can serve as an indication when it comes to SS amps. But there's a limit to how far this doubling can go: the PS and the amplification device. Because, in order to keep energy constant as your amperes go down, you either have to lower the volts or increase the watts (A=W/V). Take a look at a very well presented case here. The VA product of the PS has to be humongous (and very heavy -- PS weight is a good indicator!). Furthermore, specs CAN be misleading (not false, mind you). In the thread above, s/one stated that an amp specs a whopping 72amperes (peak-to-peak)!!! Wow, and they're not lying! looking closely, you notice peak-to-peak: i.e. peak current is 72/2 and rms is 36x0,707= 25A. Excellent -- assuming this is per channel.