Finally, The Class D Holy Grail?

[Shadders]

8 hours ago, fas42 said:

 

The basic rule for audio is, or should be, is that the closer one gets to optimum sound, the more the tiniest thing can make all the difference - the slightest variation in the implementation, the way a manufacturer builds in a module, could have a huge impact on the subjective SQ - there is a lot added by the "boxer", .

 

IOW, using a high performance core to a component increases the pressure to make sure every other part matches up to that quality - if one wants to get the best from the overall ... it doesn't get easier, it gets harder!

Hi,

I do not think you can say what an optimum sound is - it is meaningless. Please see the following URL :

"www.slideshare.net/shenpei/class-d-and-linear-power-amplifier-testing"

 

Examine slide 11 and 12. Slide 11 shows the spectral output from a class D amplifier, and slide 12 has typical class D output for a low level signal. It is a shame the signal level is not given in the slide.

 

How can the tiniest thing make all the difference, when you cannot hear the effect of the noise that is permanently on the signal ???

 

If you examine a class A/B signal, it is vastly more cleaner than the class D signal. Also, class D is an approximation to the signal which is filtered/averaged using a filter.

 

The real issue is that your hearing is not as good as you believe it is. You either have to agree that you cannot hear above 20kHz, and any information above 20kHz has no effect on what you hear, or if you agree that 20kHz+ signals make a difference, then why can you not hear the effect of the "mush" on class D amplifiers ?

 

Regards,

Shadders.

[Shadders]

38 minutes ago, psjug said:

Any amplifier output is a scaled approximation of the input.  If it is the modulation in Class D that bothers you, I wonder if you feel the same about sigma delta converters, DSD, etc.

Hi,

The class A/B amplifier is a continuous function, and the output is exactly the same as the input with added noise, whether harmonically related or not. It is not an approximation.

 

A class D amplifier is discrete time and discontinuous before the output filter. It can only be an approximation of the original signal.

 

Regards,

Shadders.