The Future of Audio Amplification


I have recently paired an Audio Research DS225 Class D amplifier with an Audio Research tube preamplifier (SP8 mkii). I cannot believe how wonderful and lifelike my music sounds. The DS225 replaced an Audio Research SD135 Class AB amplifier. Perhaps the SD135 is just not as good as some of the better quality amps that are out there, but it got me thinking that amazingly wonderful sonance can be achieved with a tubed pre and Class D amp. I have a hunch that as more people experience this combination, it will likely catch on and become the future path of many, if not most audiophile systems. It is interesting that Audio Research has been at the forefront of this development.
distortions
Kosst,
Honestly, and truly have you ever sat down and listened at length to ANY Class D amp? Specifically one in your own system? Set in the listening position and show me or more specifically tell me you hear the ringing distortion or switching for that matter that you hear in Class D amps. Although it might be harder than you think with your choice of loudspeakers......

As good as the NC400 module measures it does not sound as good as the slightly less good measuring IceEdge 1200 module.  And this is with the output caps in the NC400 changed to much better sounding modded and correctly oriented Wima caps versus a completely stock IceEdge.

https://www.audiocircle.com/index.php?topic=154910.500 

See post # 509

The switching frequency is at 450kHz, no tweeter could possibly reproduce that... Not that anyone would hear it, not even dogs.
You need to understand better, it’s not the switching frequency "so much" but the "output filtering" needed to get rid of it that ’s problem, which gives phase shift problems right down to 3-5khz (the mids and highs of the audio band). (That's why any complaints of Class-D is the upper mids and highs, never bass or lower mids).

If that switching frequency were to be moved up 3 x higher (like the $20k Technics SE-R1 does with the GaN transistors) then that "output filter" can be also moved up by the same amount, then the phase problems move up as well in the audio accordingly.
I would like to see 5 x higher.

Cheers George
Maybe that is what the class D amplifiers will be in 10 years more. Class D is evolving really fast. It is the future of audio amplifiers indeed.
Oversimplification is usually done by those who have had very little hands on tweaking of amps. I have been designing, modding and tweaking amps, etc. since the late 1970s. When you realize that every single part and execution effects the sound then just going after one measurement or whatever is just plain wrong. Most all of the class D amps till now have switching around 500K........and most use the same type of filtering on the output. However, they all sound different, including the highs. The Merrill probably switches at 500K yet no one says the highs are not great......same with this IceEdge amp.....super clean highs.....and same filtering. Other amps using the same 500K and same filter sound strident or rolled off or compressed in the highs......because there are so many factors that effect the sound. It is not one thing. It is not just the filter effects, it is not just the distortion measurements....it is hundreds of things that make the sound. I am not against wider bandwidth, GaNs, and better distortion figures. But these things will not bring state of the art sound by themselves. For instance, the stock IceEdge modules sounded OK when I first listened to it (all hardwired, no connectors)......but when I changed the fuse to a serious audiophile fuse I said.....Oh Wow....this amp really is good. Want to blow your mind? Damp the heatsinks on your big Class A transistor amps. If you run your fingernail along the heatsink and they sing, then you are adding that sound into the amp. Every person who has done this has heard the same thing.....way less "transistor" sound......which, of course, was really heatsink vibration sound being superimposed on the signal. My Class D amps will have every panel and the modules constrained layer or multiple layer damped.