Class D amplifiers. What's the future look like?

Class D is not doing calculations in the strictest sense, so digital is not accurate. It does use fets switching on or off, so a binary state, similar to digital computers. So my thought is that calling them digital amps is a reasonable slang and everybody knows what it is referring to.

With computers, I consider the passage of time increases the value equation. This is due to technology improvements and economies of manufacturing scale. I am seeing class D as the same in this respect, but willing to be educated if I have that wrong. Is it safe to say that product will be moving to the faster switching gan devices? I imagine that industry uses millions of times more switching devices, so audiophile class D will benefit from innovations in the larger market.

I also imagine that the marketplace will consolidate in time. It seems like oem components allow one to enter the market with minimal capital investment. Or might it be remain computers where I can pick the components and build my own? Now atmasphere, and I imagine others, will roll their own, so there are sort of two approaches—like proprietary apple or component pc. Well I am not a very good future guesser, so looking forward to the insights of those closer to this. I will not be buying any more class a/b amps, and see this as the technology where the action will be.

I see these amps as having components: power amp, power supply, case with plugs, some sort of preamp or driver, and maybe adjustable controls. So, many permutations for designers even if they are simply assemblers of oem components.

Interesting, thanks.

All manufacturers have to deal with high risk decisions regarding product life cycle. Do they innovate with product that cuts into sales of cash cow products? Ralph has made his decision. In the last 60 years of computer technology, we see many examples of companies that lost out because they protected existing product. IBM protected their IMS database product and did not innovate with a relational product, and allowed Oracle to take that market. OTOH, they did have enormous success betting the company on system/360.

It sure looks to me like class D will be very disruptive in the amplifier market, but I don’t have a great track record as a future guesser.. ;) 

As to the future, is it just a matter of time until all module makers switch to using ganfet? Is the current state of affairs mature with a much diminished rate of change? Or is there plenty of improvement available in the value equation?

I could go into lots of specifics on ibm, but off topic. My main point is that product life cycle, obsolescence, and innovation require difficult decisions that can determine a company’s future—a general concept far beyond audio devices. Sometimes a bet on the new thing fails, so there is lots of risk and uncertainty. Atmasphere appears to have made a brilliant move here.

Ralph mentioned musical instrument amps. I have a 700 watt class d quilter for bass, and it does the job. I think digital for guitar is very much a debate. I think for clean sounds class d is great. The nuance of an overdriven marshall is the sort of thing folks are not quite there on—but as with all things digital—the technology improves over time. Many bar-playing musicians say the guitar amp emulation is not 100% but its close enough to use on stage and not hump heavy amps around.

As to audiophile, my assessment of Ralph over 40 years is that when he says his amp sounds great, I believe him. Over the years many golden ear types thought his amps sounded good, so he must have a good ear or have access to one.Switching technology is a different paradigm altogether, and its not hard to imagine tubes and linear transistors becoming extremely niche. I wish the best for ARC, and I hope they have this sort of thing in development. 

Given the filters and phase shift, does class d design lend itself to mathematical modeling?