GaN FET Amps vs. Traditional Class D Amps

 My own experience has been, GANFET amps add "tonal warmth", like tubes, and, without the nasties of tube ownership. I find the GANFETS I have tried in my system, to "soften and round out transients", as I do with tubes. I have also owned designs using MOSFETS, BJTs, JFETS, IGBT. Plus, the number of chip amps I have owned. I do believe, based on my own listening throughout the years, "each" has its own distinct character, whatever the implementation. I will also tell you...in my system, all of these little class d amps using the separate power supply/transformer, sounded like jokes. zero balls. 

I met a DAC designer at a Denver headphone meetup last month. His name is John La Grou and his new DAC is the imersiv D-1. It is really a leap with respect to existing DAC technology. They have come up with a process they call multi-path. I own this DAC.

This same multi-path design can be used with DACs. amps, microphones, etc.. I think this company is targeting a $100M IPO to make this multi-path ecosystem happen,

John told me his next project is to make an amp with this technology using Bruno's Class D, likely Purifi. He said he was friends with Bruno. Not sure what that entails with the design.

I have a feeling this amp is going to be to Class D amps what the imersiv D-1 DAC is too digital. Maybe not just Class D amps.

Home - Imersiv

BTW - that comment on the IPO was from something I read on the internet.

it's easy enough to find or build decent power supplies and input buffer circuits for them. And not every GanFET has a great power supply or input circuit. Just sayin. 

@ericrhodes1 You'd think!!  

I regard it as a sad state of affairs that there are so many amps with good modules that seem to have poor power supplies and/or input buffers. A lot of class D amps in the last 25 years really didn't sound that great, especially in the bass, and as far as I can make out the reason seems to be that since class D is so much more efficient, whoever put the module in the box thought that also meant the amp didn't need a power supply to be as beefy.

But a class D amp power supply has to be able to handle almost no load when the amp is at idle and not sag when the amp is trying to make full power. Its a pretty big ask. On that account we found that if a switching supply is used, if you really want to make it work right it should be tailored for the application. IOW you can't just buy an off the shelf SMPS and expect it to work unless you really vet it carefully. 

WRT to the input buffer I've seen some pretty wonky stuff as well. 

The really variable comments you see about different class D amps using the same module backs this up. Its been my personal experience as well.

@atmasphere Ya, well in the last few years, there’s been some real standouts in this regard. But it requires a keen eye to find, I’ll hand you that.