• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Intelligent thoughts on Class D amplification…

skankhobag

Active Member
Joined
Sep 1, 2025
Messages
149
Likes
219
…By a guy who does real amp reviews (not to cast aspersions on what they do here, but elsewhere…).

AudioXpress and Voicecoil magazines provide news and reviews for audio pros and, as such, they never publish subjective reviews. In fact, they rarely editorialize on any subject.

I came upon a review Jan Didden did last March of the Purifi 1ET9040BA Balanced Class-D Amplifier Module:

At the end of the review he had some thoughts about class D amps. Here they are.

Afterthoughts
By Jan Didden

So, looking back on the results of the Purifi 1ET9040BA module, what does it tell us? In an earlier review of the single-ended version of this amplifier [1], audioXpress’ regular contributor Stuart Yaniger commented: “Audio power amplifiers are done. Let’s move on.” Putting this amplifier through its paces left us with similar thoughts. Not that I have any doubts that inventive designers will continue to come up with technically even better performing amplifiers. Personally, I doubt that we will be able to hear any audible improvement between an ET9040BA and the next iteration, whether from Purifi or one of its competitors.

I recently got into an exchange on diyaudio.com with a very talented designer who designed yet another Class-AB amplifier. The design had some very clever circuit details to minimize the usual problems with crossover distortion and Gm doubling. I have no doubt that this amplifier will sound clean and transparent, on a par with the very best out there. Its designer commented that his amp would be a “Class-D killer,” and that is where we disagreed. My very first Class-D amp was a Sony TA-N88 from the mid-1970s. It didn’t test very well, and it didn’t sound very well. Many issues with designs like that were not well understood at the time. So, folklore developed saying that Class-D has no place in high-end audio reproduction. The problem with folklore being, of course, that it persists long after the underlying reasons have disappeared. But high-end audio manufacturers today embrace Class-D more and more in their top products. I remember, many years ago, becoming aware that a pair of $80,000 Jeff Rowland high-end monoblocs were powered by Ncore Class-D modules. And the trend will continue. Class-D has already cornered the professional performance and installation markets, and no doubt the technology will be embraced even more in high-end audio products and particularly with the trend toward active speakers.

For a century, audio amplifier designers have improved on the basic linear amplifier concept. By going to better devices, from tubes to germanium transistors to silicon, silicon-carbide and what have you; by inventing better and better linear circuit topologies. Class-D is an improvement in concept. And contemporary Class-D is such a step up in performance that many listeners come up with terms such as “clinical” and “lifeless.” They are so used to amplifiers that add “warmth,” artifacts, and harmonics, often subtle, that they think something is missing. But life, warmth, envelopment, and the like is what music brings with it, not what a transparent amplifier does. Linear Class A and AB amplifiers will go the way, not of the dodo, but of tube amplifiers. You’ll still be able to buy them, and they will even sound good, but they will become a niche, if they are not already.


Don’t think it comes as a surprise that I agree wholeheartedly.
 
I was there when wall warts were linear regulators

Now USB chargers are switched mode

I was there when CPUs had linear regulators and their big heatsinks on the motherboards

Now VRMs are all switched mode

I was there when headphone amps were class AB

Now...

Sidenote: When you use words like "intelligent" and "real" on the internet, they take away from the credibility of the post, not add to.
 
More than a dozen years ago, I developed a program to compare (subtract) a source signal from result signal (like from an amp). For some demo and for publishment testing, I bought a couple of IcePower class D amp modules off the internet, basically to MAKE FUN of them with my setup. But in my subtraction tests, the Ice amps turned out to have less audible (out-minus-in) difference -- essentially none -- than when doing the same test with a highly modded Hafler or any of several other AB amps I had. Caught me by surprise, there. Been a class D proponent ever since.
 
Back
Top Bottom