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

@yyzsantabarbara    — Your comparison between the Coda and the GaNFET amp is interesting. Your leaning toward the Coda may actually be a determining factor, no? One question worth putting to you directly: given all the Class D amps you've owned and heard, do you think they are genuinely  very different  from one another, or does a family resemblance persist across them?
  @bluethinker    — My observations on the AGD Audion MKiii and the other amps are coming — I just want to avoid first-thought bias before writing anything up. On the "settling in" question: I'm genuinely uncertain whether that phenomenon belongs to the gear or to us. Changes in what I attend to, listen for, the music I select (which I try to keep consistent), time of day, mood — these are all enormous variables. My hunch is that most listeners overweight changes on the physical side and underweight the psychological variation on our side of the equation. We tend to assume we're the stable, non-changing factor in the experiment — but that's exactly the kind of overconfidence that Harman's careful subjective listening controls were designed to correct for. Just my two cents.

I find the AI conversation here a bit off topic but: 

@parkergetdean    — Whether AI can improve on audio recommendations depends heavily on who is curating and improving the training data. There's also a deeper issue that doesn't get enough attention: the datasets used for audio discussion and review are wildly English-speaking dominant. That means what looks like an "objective" answer is already shaped by a particular linguistic and cultural slice of the listening world. For those curious about this bias in LLMs more broadly, there's some good recent work out of Stanford on exactly this problem: https://news.stanford.edu/stories/2025/05/digital-divide-ai-llms-exclusion-non-english-speakers-research

Most Class D: Sterility, brittle, thin sounding/lack of body, fatigue inducing

Some GaN class D: sterility is tamed, less fatiguing, lack of body persists 

re; curating and improving - You can tweak how AI uses the data (tell it that erik_squares is smart, parkergetdean is clueless, Erin is honest, cheapaudioman is shady, etc. ) but until AI can learn to go into hifi shops and listen, I don't know if there is any way to improve the data. That's not to say it can't be useful for beginners, it can work faster in collecting and organizing the data than humans would.

Here is my opinion of the Class D amps that I have owned

• LSA Voyager 350 GAN (modded by Ric Schultz)
  • Very good after the mods with a very clean, detailed, smooth sound. However, the bass was really lacking
• PeachTree GAN1 (modded by Ric Schultz)
  • this amp was much more than an amp. It had a SPDIF digital input. It did not have a volume control, so your streamer had to manage that and it only did that single digital input. I had the Lumin X1 streamer and that is a great streamer into the SPDIF with the LeedH volume control. After the mods this amp was the very best sounding Class D amp I owned and if it had stronger bass, I would still have it.
• PeachTree GAN400
  • the second most powerful Cass D amp but was a bit grainy on top
• D-SONIC M3a-800S
  • Most powerful Class D amp but lack detail like the others.
• NAD M22 V2 
  • I hated this amp and luckily was able to return it to TMR before my demo period ended. The only piece of audio gear I ever returned.

I am playing the Class D GanFet on my Yanaha NS5000 speaker and the imersiv D-1. I played some vocal oriented music, and the amp was really good. However, I put on Led Zeppelin's Live EP and the song Kashmir (Live from Knebworth 1979) my recent go to song for headphone evaluations. Compared to the CODA #16 the sound was much smaller. The pounding drums on this track do don't pound so hard. It sounds OK, but not anything close to the great sound with the CODA #16. My CODA #11 sounds similar to the #16 and it cost me $2400. The top end sound nice and clean and smooth, but everything else sounds limited compared to the #16 and #11. No way I would use the 6.5 amp on the Yamaha, but on the smaller Magnepan Mini it is very good.

Thie issue is that the Yamaha NS5000 is one of the best speakers around and the 6.5 cannot scale to that level. The Magnepan Mini is not a great speaker, it is a fun and very good speaker (like the Magnepan 3.7 but smaller). The 6.5 scales to this speaker nicely. The #16 is better on the Maggies but not much better because the speaker can only do so much.

My listening test is over, not that great.

"@parkergetdean but until AI can learn to go into hifi shops and listen,..."

And/or, not until the future where let’s think ahead how AI engines with probes and sensors are integrated within MY own listening room, [collecting response samples, best it can] taking into account all of my audio components, cables, speakers, room shape/size, placement of my speakers in the room itself - its easy to say at best, as many shared,  can be helpful to "collect and sort vast amounts of data, quickly, whatever is available", at best. i.e. "Garbage in, garbage out". The creating of gourmet food and contents sourced from the trash [above] was an interesting way of putting it. 

More conversations like this thread will continue to inject itself into the results from all of these AI engine/platforms and vast data sources that grow daily.  I’m comparing this years responses to last and next years btw. I bet if we ask the same questions two years from now, we may be quite surprised at the results. For today, we are still best served listening in our room with our gear that matters most imo, just like @hilde45 is doing right now on this thread :) Fun thread.