Amp Designs That Are Able To Drive 4/2 Ohms Load

I have noticed that some amps like the Krell Evo 302 will output the exact wattage when halving the impedances ie. 300W/600W/1200W into 8/4/2 ohms respectively. On the other hand, some amps like the Bryston 7B-SST does not output the exact wattage when halving the impedances ie. 600W/8 ohms to 900W/4 ohms or the Plinius SA-100 (100W/8 ohms to 180W/4 ohms)

Can someone enlighten as to where the 'missing' watts have gone for amps that do not output the exact 2X the wattage when going from an 8 ohm to 4 ohm load? What are the criterias in amp design that determine which route to adopt, and what are the pros and cons of each design?

Thanks in advance.
Crappy designs can't deliver stable power below 4ohms...even the Krell 400xi can put out exquisite power and control down to 2ohms!!
the critical factor is the power supply. There is a reason a big amp has a huge power supply. also number of output transistors and size of capacitors can be big factors. Great amps also tend to weigh more than lesser amps. (of course there are class d amp-IMO are a complete waste of space)
When an amp can't double its power as the impedance halves, the power supply is to blame. I have a CJ MF-2500A and it doubles down to two ohms.

Watts = voltage x current
Watts = voltage squared / resistance
Watts = current squared x resistance

The max voltage of a amp is limited to its internal DC supply voltage.
If the wattage does not double when the resistance is cut in half then the amp is not able to output the required current.
Starcon, and what class A or class AB are you touting to be superior to class D amps in general, and the large power supply loaded H2O in particular? With the help of it's stable mate, the Fire, this duo can power my 1 ohm speakers with ease, delivering huge dynamic swings along with loads of musical detail.
Post removed 
As Bob noted, inability to drive a 2 ohm load does not reflect "Crappy" design. It reflects inteligent design (sorry about that) for an amp that will be used with 8 or 4 ohm speakers.

If you want to drive low impedance loads look to pro sound amps that are designed to do this.
One thing that came out at the recent CES is that there really is no longer any argument for low impedance speakers anyway. It turns out that transistor amplifiers, while easily able to *drive* a low impedance speaker like 4 or 2 ohms, actually sound their *best* driving 16 ohms.

Steve McCormick, as one example, makes transistor amps that have no problem at all driving 4 ohms and less, but he discovered that the amp *sounds better* when using a set of ZEROs to drive 4 ohms, rather than directly, much the same way MacIntosh did years ago (the ZERO was loading the amp at 16 ohms). By better: smoother, more detailed, more authority, and something easily heard.

So Eldartford and Bob are right- there are many amp manufacturers that don't care if their amp 'plays' a 2 ohm load- more likely is what they are really interested in is how the amp plays 8 ohms.
The original Eagle could almost be used for welding in that no speaker load bothered it.
Doesn't distortion go up as impedance decreases? Why design an amp for a lesser performing speaker? How many speakers actually are a stable 2 ohms, outside of car audio which has much higher distortion tolerances.

My one tube amp is stable at 2 ohms, but the same wattage output as 8 ohms. My other mono block tubes are stable to 1.4 ohms but put out the same wattage as they do at 16 ohms or 70 volts. Does this mean I'm going looking for 2 ohm speakers, NO. I'm looking for the best sounding speakers in my budget but doubt it will be a 2 ohm speaker. 8 ohms will probably sound better, why push my amps to the extreme?
My one tube amp is stable at 2 ohms, but the same wattage output as 8 ohms. My other mono block tubes are stable to 1.4 ohms but put out the same wattage as they do at 16 ohms or 70 volts.

Pardon my ignorance. How do you know your tube amp is stable at 2 ohms? Is the amp capable of driving speakers at 2 ohm load or did the spec specify this? I am just curious on the differences in the output wattages for various amp manufacturers when the impedance is halved. As others have pointed out, the power supply is the critical factor in determining the power output. Now I'm more confused to hear of tube amps that will output a similiar wattage at varying impedances.
It's a little more involved than just the power supply being unable to provide the current to double the power as load is halved. You can design two amplifiers with the same power rating at 8 ohms and have one being able to double power at 4 ohms and the other with the power remaining the same. Yet the amp that doubles the power can sound worse regardless of its heftier power supply and yet have nothing to do with the quality of the components.

When a speaker draws current, the power supply gets taxed. If the power supply serves both the input and output stages, the voltage sag affects both. This will cause the input stage to distort which sends a distorted signal to the output stage which then gets amplified to the speaker.

However, if you split that power supply with two stages of regulation (expensive), then the current demanded by the speaker only goes through the output stages, leaving the input alone. The input signal is left undisturbed by voltage fluctuations but the output power delivery is limited. So that's one answer as to where the "missing power has gone".

Another answer is that the power supply may have the capability, but poor design (such as high impedance) may cause the power supply to waste heat, thus robbing the output stages of the ability to maintain voltage during speaker impedance dips. There are other reasons, but basically the missing power is either diverted for better fidelity, is wasted as heat or is simply a victim of the manufacturer's budget.

Sorry, but designing for compromise is crappy design...period! I've found Krell for example sounds best driving a difficult load. Also, class A (at least in the pre-driver stage) is more musical. Case in point, the new Evo amps (no longer class A), which I owned for a time, sounded cold and flat compared to even the 400xi integrated. One must have an ear for exquisite tonal colors however...otherwise Bose will do just fine! Of course not everyone will agree, but i'll take a "Quantum of Solace" in the knowledge that JValin and Martin Colloms might just agree with me.
Gs5556, thanks for posting. That had clarified most of the doubts I had previously.
Ryder, take a look at

In a nutshell, 'Doubling power' as the impedance is cut in half says nothing about how the amp sounds, in fact there is nothing in the rules of human hearing to suggest that this is helpful.

The result is, there are amplifiers that don't/can't double power as impedance is halved that nevertheless sound better than many amps that can double power, because the former are designed to the rules of human hearing.
I suppose they're are two schools of thought. One that believes that amps should drive speakers and one that believes speakers should accomodate amps. I'm (currently) from the former. No matter how challenging, we have amps that can successfully drive lower impedance speakers. Simply put, in my experience, systems with lower impedance speakers tend to sound better than systems with higher impedance speakers. Of course, YMMV, amplify accordingly.
Ah, the elusive art of design...there still be magic unknown to mere mortals and engineers. Owning alot of systems over the years has afforded me first hand knowledge, not just a theoretical perspective. Even the tube gear I've owned used class A circuitry...none better nor sweeter for the human ear!