When Will the DAC Singularity Be Reached?


A humorous title, but wondering if those more in the know have an opinion on either: i) examples today where inexpensive DACs (say under $2500) are comparable or superior to expensive (say over $10K) DACs or ii) can we anticipate that within a relatively few number of years that inexpensive DACs will basically achieve the sound quality of today's expensive DACs? Thanks. 

mathiasmingus

@danager Don't take this personally since, as I said above, the manufacturers are encouraging this misconception.  

But physics and engineering don't support the existence of a digital amplifier. 

indeed you don't need to buy a DAC, since they have incorporated it.  They call this "DACless".  

marketing...much more believable than physics.

Wouldn't it be a great world if we could increase power by manipulating a digital file.  our hard drives would power our house. 

jerry

Borrowed this explanation, paraphrased from threads on ASR.

A class-D amplifier is an analog amplifier. It uses analog circuits, with an analog signal as the input, to turn transistors on and off.

A 1-bit sigma-delta DAC that switches 5V power rails (or whatever they use), is a DAC.

A 1-bit sigma-delta DAC that switches 50V power rails is a digital amplifier. It uses the digital input signal to turn transistors on and off.  To avoid any misconception, there is not a separate DAC chip and class-D amplifier. I assume the digital logic that would be in a DAC chip would need to be in there somewhere, but not the the DAC chip.

Great DACs will always cost more, just like great amps and preamps.

The actual digital to analog conversion of which so much is made is probably the least important component of a DAC.  Great DACs have been made with every type of conversion be they DS chips, R2R chips or discrete, and FGPA chips, which are chips individually programmed, usually partly DS.

Where the expense comes in is in great power supplies and the analog output stages that include I-V conversion, low-pass filters, gain stage and output buffers.   What can easily add up to $5-6000 for a preamp can do the same for a DAC, if you want the very best.

Then there are the things unique to DACs like expensive Xtal clocks and the necessity to totally isolate digital and analog components.

The designer of the DAC I use has written that any competent electrical engineer can make a relatively cheap DAC that will score high with conventional measurements.  But it probably won't sound very good.  There are lots of such DACs around and they sell because people believe their eyes rather than their ears.

Finally I would add that very many DAC purchasers (and many who call themselves critics--and that's easy to do these days because all you need is an iPhone) have never heard music except through loudspeakers.  So they are all too easily satisfied.

@carlsbad2

You’re mistaken - the digital signal is driving the PWM output stage. There’s no DAC because it’s not needed. This is also how the Raspberry Pi onboard audio works. There are details in this article:

https://hackaday.com/2018/07/13/behind-the-pin-how-the-raspberry-pi-gets-its-audio/

To the OP, I think we’re already there. DACs are a mature technology. A well engineered DAC that sells for $100 is going to sound great and possibly better than any other DAC costing more (including thousands of dollars more) in a blind test.

@yage   you are arguing semantics.  The amplification is analog.  You are challenging how it is integrated.