The Future of Audio Amplification

 Nobody cares about class T?
People consider class T as in the same category with class D?

Yes. 'Class T' is a trademark.

GaN, aside, I definitely recommend people always seek out and consider the latest and greatest technical innovations that can clearly help move things forward, Class D or otherwise, especially before dropping big bucks on older technology alternatives that have not evolved much but now cost more than ever.

Considering an ICE amp? I had this problem with ucore

“One of the difficulties implementing the ICE module is its hungry demands for a powerful input signal. The 700 watt ICE module has low impedance (around 8KΩ) which, if not properly buffered, can cause plenty of poor sound from connected equipment. And what the designer uses to buffer it is critical. In our case, the Analog Cell handles not only this critical buffering task, but sets the synergy between the input and output. Another innovation employed in service of great performance can be found in the power supply design. Here, the Analog Cell received high quality DC through liberal use of regulation and power supply storage through use of capacitance multipliers and generous numbers of low ESR capacitors.”

“One of the difficulties implementing the ICE module is its hungry demands for a powerful input signal. The 700 watt ICE module has low impedance (around 8KΩ) which, if not properly buffered,

Little bit of obfuscation. A lot of Class D OEM modules have both buffered and unbuffered inputs. It is up to the manufacturer to either use the default, or build something else.
This lets you do tweaky things like a tube input buffer, diamond, whatever.

My point is, no end user is really ever going to feel or suffer from the low impedance inputs.

Best,
E

I definitely recommend people always seek out and consider the latest and greatest technical innovations that can clearly help move things forward, Class D or otherwise, especially before dropping big bucks on older technology alternatives that have not evolved much but now cost more than ever.

Correct, all the current big $ or cheaper Class-D’s have not really progressed in a big way, yes they play around with different parameters, inputs ect and keep bringing out different "versions" of the same thing shifted sideways a little.
But it’s still "basically" the same operation/componentry/topology as when it started.