In the specifications of the Stellar Strata mk 2 it is stated that the amp’s output power is 100W into 8 Ohm and 150W into 4 Ohm. From working with class A and AB amps in the past (and Ohm’s law) I would expect that an amp that outputs 100W into 8 Ohm, would output 200W into 4 Ohm…
Why does a Class D amplifier like the Strata, output 150W into 4 Ohm? Is the power behaviour of a class D amp not linear in relation to he resistance of the speaker? 
The M700 & M1200 double their power from 8 Ohms into 4 Ohms.
350W into 8, 700W into 4 for the M700’s
600W into 8, 1200 into 4 for the M1200’s
That seems to indicate that the output power of class D amps is linear with regard to speaker to speaker resistance. However, then my question remains: why does then power amp of the Strata mk2 behaves different? Can anybody please shine a light? 
Could be power supply related. When the wattage output is doubled the power supply needs to be engineered commensurately. The Strata power supply is powering a lot more than just the amps and may not have the headroom necessary to maintain the output linearity seen in other amps.
Just guessing…
In a perfect world one would receive a 2x power increase for each halving of impedance; however, it entirely depends upon the amp’s available power supply current and its thermal cooling capacity. It’s also worth noting that only a very few speakers provide a constant load across 20-20K and along with that variability comes phase shifts between voltage and current delivery. The net is that one can’t predict the synergy or dynamic capability from the output specs alone.
Current delivery is not specific to amplifier topology.
Yes, you’re absolutely right. Nevertheless, as @adifferentpaul mentioned the M700 and M1200 amps are specified as 350/700W and 600/1200W. When taking into account the things you guys mention above, and the relatively smaller form factor for an integrated amp, I think there is a big change that the 150W into 4 Ohm is due to a restriction that prohibits excessive heat dissipation or a limitation caused by the power supply.
Actually, the Strata is really a 75W amp into 8ohms and 150W into 4ohms if one limits to the point where the distortion starts to steeply rise per recently published tests so it does double its power.
I don’t believe any amplifier can double maximum power output when impedance is halved, at least not without distortion going up.
We were not talking about maximum power but rated power. Those are very different and the later is up to the manufacturer as to what distortion level its defined at across what frequency range and number of channels driven.
I think you’re right. I only don’t understand why PSaudio specifies the Strata mk2 this way while for all the other amps they just double the power into 4 ohm. But then again, maybe this amp behaves different then the others.
I had a pair of Class D monoblocks for a year or so. I could detect no downside. I was perfectly happy. Then, I switched to amps capable of 100% Class A. I listen to them in pure Class A all the time.
I can detect no downside. I am perfectly happy.
But! On these cold nights, the Class A amp warms the room. It is 6 degrees outside. I have the amp set on KILL. Mmmmmmm.
At least one of Paul’s commentaries on power amps / impedance
Is it never all about the power supply?
Class D amps can perform well with very simple power supplies. They also perform better with better power supplies. How much improvement depends on many things. But a Class D amp doesn’t need spit compared to a Class A or AB. That is why they are small, light, cheap, and run cool. You just have to live with the nagging thought that it all seems to be too good to be true. It is true, but
I encourage everyone to try the latest post filter feedback (PFFB) class D amps. This circuit makes the amplifier completely indifferent to impedance variations of the load.
3e is one such maker that has sold their boards to OEMs for years and now sells a ‘retail’ product. The only downside is the lack of bragging rights. The stereo version is only the size of a paperback and weighs all of 2lbs. They make monos too if one needs way too much power aka bragging rights.
I needed something I can physically move in the future and took a ‘chance’ (very low $ risk). I’m so f*cking impressed it’s simply not right but I’ll get over it. 
I have the A5.
The question with the (PFFB)class D amp is can it supply the current the load requires or does it current limit.
Good question. I suppose you have to know the load characteristics of your speakers well to answer that question but in my experience with ATC SCM11v2 - somewhat inefficient but not terribly hard to drive stand mount, a sensitivity rating of 85dB @ 1W @ 1metre, and a nominal impedance of 8 ohms, - I get more clear dB than I can use in the space all from a 36V 5A GaN power supply that it arrived with. It’s quite responsive to dynamic demands to my ears much better than the 83lbs. hybrid 250W it’s replacing.
Oh, the 3e units have visible clipping indicators for each channel. Nice touch but I don’t plan to find out the level the amp clips.
I have two powered subs, fed high-level into a transformer from the stereo amp (same with both amps in comparison). The ATC are allowed to naturally roll off. The little class D drives the powered subs with much more authority, not gain or volume, but authority. The damping number is very high. >300x.
They do make mono blocks and 48V 10A GaN power supplies to really overkill the situation or for truly hard to drive loads in truly large spaces.