I think I am about to confirm
@amirm test results. Just measured my NC252MP implementation, which has probably the best possible cooling with large, full-size 19" - 3U aluminium box with side heatsinks. I have just made a set of measurements with 2ohm resistive load.
1. First, the noise floor with the forest of mains related lines is similar. Please take into account the graph is in dBV, so 0dBV=1V, in other words the amplitude in dB is related to 1Vrms.
View attachment 290552
2. Max. continuous power into 2 ohm load (but with only 1 channel driven) is 234W, then the amp protection shuts it down.
View attachment 290555
3. Below are the plots of THD and THD+N vs. power into 2ohm at 1kHz. Measuring bandwidth is a conservative 22kHz. It works well with a 2ohm resistor up to some point, like 10-20W, then the noise floor starts to have erratic behaviour and the power is limited, as shown above, to 234W with one channel driven. 5W/2ohm SINAD is 90dB. The amp does not like a complex load together with 2ohm resistive.
View attachment 290556
...... and THD+N vs. frequency into 2ohm at 168W
View attachment 290559
4. Conclusion
The amp is working well with 4-8 load, not so well with 2ohm, however usable with this load with a music signal. However, it is not the amp the could have been called "rock-steady". It is just something we call "value for money".