Time to don my "Anne Elk" disguise and put forth a theory... (No - not about the brontosaurus...
)
In the years following the introduction of CD, we saw early players using (and sometimes tweaking) the available DAC chips. Next followed component manufacturers investing and producing their own bespoke DAC circuitry (ladder dacs, ring dacs, etc). Meanwhile, the mainstream digital IC manufacturers like Burr-Brown, ESS, Crystal, Wolfson and co, had upped their game and were producing DAC chipsets that were giving some of the high-end bespoke DACs a good run for their money.
As the Law of Diminishing Returns began to apply to R&D into new and improved DAC design, the high investment required in R&D around new and better DACs made it difficult for the smaller component manufacturers to carry on with in-house development and easier and more cost effective to piggy-back on the R&D spend of the large (and high-volume) DAC chip producers and simply try to value-add in areas such as ergonomics, discrete analogue output stages, better clocks, etc.
This approach allowed these "cottage industry" manufacturers to still market competitive digital components in their various ranges but without the massive R&D overhead associated with keeping pace with DAC performance enhancements being achieved by both the high-volume specialist DAC chip manufacturers and those high-end bespoke DAC developers.
My guess is that, eventually, all DAC chips will come from specialised DAC chip manufacturers who will be only entities who will have the required critical mass to continue the required R&D spend.
And this where the "Anne Elk type theory" regarding Class D amplifiers comes in...
Firstly, Class D technology has some inherent strengths: high efficiency, low heat dissipation (both of which are in line with the increasing focus on "greenness"), compactness, and availability of third-party "modules". Their perceived weaknesses of switching noise levels and brittle upper registers were real in early implementations, but these have been worked on to reduce their impact to the point where today's better Class D amplifiers no longer suffer from any really noticeable artefacts in these areas.
Meanwhile, Class A and Class AB amps have seen little or no quantum improvements in performance - just minor ones due largely to tighter tolerances, better circuit layouts, improved power supplies and regulation and more care in component matching. As a result, R&D costs can be contained and this is what keeps these amp classes going as your smaller "cottage industry" manufacturers can still afford to take their old base designs and focus their R&D spend on minor improvements.
Both (particularly Class A) are still basically old and inefficient designs and still pump out large amounts of heat and require large heat sink areas to dissipate this heat so that running temperatures don't climb too high. Not exactly green in today's world. And herein lies the rub - just how much longer with the evolving market be prepared to spend money on new amplifiers based on old designs whose only benefits lie in central heating savings in winter (sorry - couldn't resist...
).
And here's where the CD and amplifier trends could converge...
It's already started with a groundswell of amplifier manufacturers turning to third party Class D amplifier module manufacturers (B&O, Hypex, Tripath, etc) and in-sourcing the modules so that they can focus on their own core competences to provide differentiation in the marketplace.
At some point I see Class A amplifiers joining the in-line straight 12 cylinder, side-valve 8 litre car engines (at best, in museums and at worst in the waste bin). Class AB amplifiers may outlast Class A but are coming under pressure from the lack of potential for any significant or groundbreaking innovations. The automotive analogy here might be that Class D equates to hybrid petrol/electric cars while Class AB might be the latest small turbocharged OHC, fuel injected petrol/diesel engines...
Tomorrow's audio buyers will have different priorities to today's buyers when it comes to amplifier purchase decisions, so it's not wise to expect them to have the same priorities as us (dinosaurs?).
Amplifier manufacturers who are not aware of (or ignoring) the potential change in market behaviour will not survive - survival will demand an understanding of these changes AND a willingness to adopt or adapt to the changes.
And that, as the saying goes, is my theory and what it is too...
Dave