Expanding the Class D Conversation: How Would You Characterize Their Differences?

@decooney i don’t know, but I’m guessing that the 4 &6 channels are class D; so, it makes sense that they are priced in line with the DIA 400-S.  The pwr-222 are AB and of course one needs 2 of them.

I’ll be interested to see what’s new in Class D, especially if anyone goes to Axpona. There seems to be a lot of new products coming out, and new developments. We know that AGD has at least three iterations of its mono blocks. 

GaN is obviously the biggest recent shift – they switch faster and dissipate less heat than silicon MOSFETs. Interesting explainer about Infineon is here.

I’ve heard Peak unveiled its first GaN-based amplifier module at CEDIA Expo 2025 in Denver

And of course there are recent versions of the Purifi 1ET9040BA (Eigentakt).

Buckeye has come out with a new Hypex Nilai, competing with the Purifi 9040 module.

There’s also this nice write up about Class A vs. Class D here.

This article has lots of generalizations in it. Some choice quotes:

[Ralph of Atmasphere called Class D] “the future of audiophile amplification,” saying his new models sound “every bit as good” as his tube designs. With it, the distortion is handled in a way that preserves a smooth, natural character.

Greg Stidsen, NAD’s Director of Technology, points to the measurements. “The best Class D designs already compete with and beat most Class A and AB amplifiers,” he said. “Using the benchmarks for measured performance, that’s indisputable.”

Paul McGowan of PS Audio:

“You can’t say Class D will wipe out Class A or AB — each has its sweet spot,” he said, “but progress is being made, and we’re all for it.”

Nelson Pass’s comment was interesting to me:

“My longtime business partner Joe Sammut used to say, ‘It’s entertainment, not dialysis,’” he said. “We build what we like, and some audiophiles enjoy living with it.”

Pass respects Class D’s progress but still prefers the simplicity and sound of his Class A amps: “I explored it years earlier… but it didn’t appeal to me.”

The Pass responses and this paragraph – essentially getting as "synergy" between amp and speakers, is resonating with my particular, system-specific experience. 

When Class A Still Makes Sense

Even as Class D dominates the numbers game, Class A still earns its place for listeners chasing a specific character and simplicity. Its fully linear, always-on operation keeps transistors within their most predictable range. And, it produces a natural harmonic structure that some describe as richer or more “alive” with acoustic instruments and vocals.

For high-efficiency speakers, especially horns or single-driver systems, Class A’s modest power output is more than enough, and the low feedback design often pairs beautifully with minimalist setups.

Many engineers and enthusiasts also value its circuit elegance and repairability, as it has no switching stage, no complex control loops, just heat, current, and purity.

In short, Class A remains the right call when the goal isn’t laboratory transparency. But a musical experience shaped by simplicity and tone, where the extra heat and power draw are part of the charm rather than a compromise.

In my experience so far, I would NOT say that the better Class D I’ve heard – AGD, mainly – brings nothing unique to the table. It definitely does, and in ways no Class A does. But overall – with all the factors considered together and adding up to musical engagement and tonal friendliness, it does not. 

This does not rule out, for me, getting another amp – a musical sounding Class D. But it would not displace my Class A. With my speakers, tastes, etc. For perhaps the vast majority of moderately or harder to drive speakers out there, my guess is that Class D is now competing very well with Class A. But, if you get the right speakers first, then I think there is a Class A or A/B amp out there that will sound really good, and you might have to spend a lot to get a Class D to sound as good, and even then, it might be missing the qualities which make you the happiest. 

for any Class D amp that I have heard. Amended: Every Class A amp I have heard with my system has sounded better, overall, than every class D amp. I can name the specific amps if anyone wishes.

@hilde45 +1

FWIW Dept.:

Even as Class D dominates the numbers game, Class A still earns its place for listeners chasing a specific character and simplicity. Its fully linear, always-on operation keeps transistors within their most predictable range. And, it produces a natural harmonic structure that some describe as richer or more “alive” with acoustic instruments and vocals.

Class D amps can be quite simple. Ours has less parts count than an SIT amp that Nelson designed that was a simplified rework of the Sony VFET amp; his version makes about 20 Watts. So simplicity isn't the distinction.

The 'always-on' bit is problematic. One thing that is really nice about a class D amp from a design perspective is that everything about it (slew rate, propagation times and so on) is very knowable, IOW, very predictable. So not a distinction of class A. 

Finally, our class D amp tends to make lower ordered harmonics in much the same way that a tube amp does; 95% of that due to the dead time required to keep the output section running properly. So the last bit about harmonic structure in his comment isn't a distinction either. 

@atmasphere Thanks for your post. If anyone can make this new technology sound good, it’s you. The people I respect most in audio love your tube amps. So, I’m looking forward to trying and comparing it to my Class A stuff.

The nearly $8k AGD monoblocks could not do tonality and presence as well as my $5k XA-25. Given that it is GaN and yours is GaN, I am a little dubious, but as you’ve said elsewhere, implementation is most of the game. For my ears and system, the question is open about whether your implementation (and the inclusion of 2nd/3rd order harmonics) can best the AGD and match the Pass. I hate having a 300 watt beast on all day long, so this trial could really lead somewhere good.

About those harmonics -- is there somewhere we can see an fft plot showing the harmonic profile? 

Out of curiousity, Ralph, beside the Class D, what other solid state amplifiers have you designed and built? I’d love to see or hear them!

Given that it is GaN and yours is GaN, I am a little dubious, but as you’ve said elsewhere, implementation is most of the game.

@hilde45 It is. As you can imagine though, you can have two tube amps that use the same power tube and they can sound quite different. The use of GaNFETs is exactly the same in that manner.