So many variables effect a dac's performance, power supplies, individual rail voltages, coupling, SRC (if used) and dac chip digital grounds, analogue grounds, circuit track distance, dac chip configuration (parallel or dual differential), output stage, whether cathode follower, transformer, fully discrete, op amps etc. Again power supplies to these.

Clocking parameters, multiple clock frequencies, clock power supply, clock linked, external word clock ability, shielding from RFI and EMI, separate tx's for digital and analogue.

Most cd / dac have the ability to up-sample the raw incoming data stream for say 16 bits 44.1Khz to 24bit and 192Khz, however these up-sampler's merely add 8 bits of dither (random noise)and they take the basic sample of 16 and multiple it by 2 /2.5 /4 or 4.5 its original rate.

High resolution files are different PROVIDING they are supplied in the original recorded format and not merely "re-sampled" Stand up HD tarcks!!

A file recorded at a genuine 24 bit and 96Khz resolution will have significantly more information contained within that file than a CD equivalent.

For example CD quality = 1.411Mps

Studio 24/9Khz quality = 4.39Mps

The basic facts boiled down equate to this:-


Recording at 24/96 delivers a greatly increased audio resolution-over 250 times that at 16/44.1 (log scale)
Recording at 24/96 takes up roughly 3 1/4 times the space than recording at 16/44.1

This doesn't mean you will obtain 250 greater sound quality lol!

What it does mean you get a bigger box and more dots on your curve (Ow Matron)


In the last 12 years I have found it fundamentally more desirable to record @ 24 Bits as a raw format it just simply gives you the head room and the ability to recreate the dynamic content with the music being recorded.

Quite fortuitous this thread, John and myself we having a long conversion regarding PC audio and digital audio last night, mainly discussing the software players and machine specs.

However one of the fundamental conclusions we have arrived at:-

The number people do not hear a difference between resolution rates is quite high, they are a few possibly reasons for this:-


1. The genuinely do not hear a difference at all.

2. Their hearing does not allow them to hear much above 9Khz for what ever reason medical or age.

3. Their equipment does not show any perceivable differences or has the ability to do so (this statement is not based on the price of equipment in any way)

4. There is no difference and it all one big trick the reording industry play on us.

Many years ago Wadia used a digital algorithm called digi-master which 'used' several of the 'bit's to make their digital volume control function, so they added a few bits of dither to help smooth things out so a 20 bit song was effectively 18 bits and so on.

Esoteric use up to 48 bits to get around this problem.

However many dsd64/128 and 256 files are encoded at 32 bits, we actually record at 384Khz and 32 bits, can you hear a difference when you are listening to a mastered CD quality file, I truly believe you can.

It is all down to the quality fo the workmanship in the mixing and mastering studios and the personal who are performing these operation.

I have some simply stunning red book masters and some really piss poor 24/192Khz's

All that glitters is not gold, However if done correctly high resolution sounds just 'Real'

Nah. It was a reply to John, there's a wee speeling mistook in his post..

What is the point of 144Db of dynamic range? Let's assume that your system is capable of in - room peaks of 108 Db, would 24 bit dynamic range (144Db?) not be a long way below the threshold of hearing?.

A lot of DACs have a signal to noise ratio of 112 Db or better.. what would be the need for achieving better than 20 bits of resolution?.

To me, the main point of using a 24 bit lpcm encoding scheme is to allow processing in the recording/mastering chain without being affected by excessive quantisation noise. As a playback/distribution format, what is the real need?.