advertisement


Can you hear the difference between 16/96 and 24/96

Igloo Audio said:
Possibly, but for the wrong reasons IMHO, because it seems something has gone wrong with these files.

Firstly, they measure a bit depth of 24, which means they have never been converted (dithered or not) from 16-bit to 24-bit. As mentioned earlier, and equally by Werner, you should achieve a -98dB null. If using dither, you are likely to lose 10 dB or so (Adobe Audition), but the files will still sound identical to all intent and purposes.

Peter
Click to expand...
I am not at all convinced that anything has gone wrong with these files. It would probably best if you either correspond direct with Archimago or simply leave this until after the outcome of the test.

It would be very unfortunate if it were necessary to analyse out exactly whether you were right or wrong. I am concerned that this may involve giving away or encouraging others to work out which file is which.
 
Igloo Audio said:
Possibly, but for the wrong reasons IMHO, because it seems something has gone wrong with these files.
Click to expand...

I don't think so...

Firstly, they measure a bit depth of 24, which means they have never been converted (dithered or not) from 16-bit to 24-bit.
Click to expand...

How are you measuring the bit depth? I'd love to see a program that can tell the real signal apart from dither/noise.
 
adamdea said:
I am not at all convinced that anything has gone wrong with these files. It would probably best if you either correspond direct with Archimago or simply leave this until after the outcome of the test.

It would be very unfortunate if it were necessary to analyse out exactly whether you were right or wrong. I am concerned that this may involve giving away or encouraging others to work out which file is which.
Click to expand...

I agree - can we please leave technical analysis out of the discussion until *after* the test period, in respect for the work archimago is doing?
 
Julf said:
I agree - can we please leave technical analysis out of the discussion until *after* the test period, in respect for the work archimago is doing?
Click to expand...

It'll be interesting to see how this turns... ;)

I'll roast Archimago in the meantime but, generally speaking I have supported his work, I just feel his motive might be a bit misguided on this occasion.

Peter
 
Igloo Audio said:
Why don't you think so?
Click to expand...

Because as far as I know there is no reliable way to detect bit depth in the general case. It is easy when talking about trivial zero-padding, but once the zero-padded result is interpolated in any way, there is no real way to tell.
 
Julf said:
Because as far as I know there is no reliable way to detect bit depth in the general case. It is easy when talking about trivial zero-padding, but once the zero-padded result is interpolated in any way, there is no real way to tell.
Click to expand...

It depends on the coding criteria.

Peter
 
F1eng said:
I did that too...
Click to expand...
Splendid. Sorry about being bossy. Actually he is keeping the test open for ages- June or something. I guess we will all have forgotten about it by the time the results come out.
 
I had a go at this over the weekend and was expecting to have to flit back and forth between tracks. As it turned out, I made clear choices from one straight play through. Very interesting and enlightening. I'm looking forward to seeing the results posted and would encourage as many people to take part as have the time and the equipment to do so. It'll be interesting to see just how many people could tell the difference. Not entirely sure why the survey didn't come with a 'I really couldn't hear any difference between the two', but hopefully the results will shine some light.
 
lovejoy said:
I had a go at this over the weekend and was expecting to have to flit back and forth between tracks. As it turned out, I made clear choices from one straight play through. Very interesting and enlightening. I'm looking forward to seeing the results posted and would encourage as many people to take part as have the time and the equipment to do so. It'll be interesting to see just how many people could tell the difference. Not entirely sure why the survey didn't come with a 'I really couldn't hear any difference between the two', but hopefully the results will shine some light.
Click to expand...

The DAC in my REGA cdp also supports external USB input for downloaded files. REGA has built a quality (may I say "Killer") DAC that operates at 16/44
They shunned away from any manufactured upsampling approach.

MY FINDINGS:

Raw numbers in files are meaningless. Most recordings and audio libraries are 16/44

(1) It matches (at a minimum) and usually bests the higher resolution DACs playing the higher resolution files out there; to at least the $10K price point. (I haven't auditioned enough past that point to offer a strong opinion)
(2) stick with something that does it well in its native format and pass on the illusory upsampling numbers and hype.
(3) the resolution capabilities of the rest of your system (electronics, speakers, cables); and the synergy (or lack thereof) within it matters big-time. This is a cradle-to-grave sum of the parts approach to audio nirvana, not an isolated vignette discussion.
 
AKG said:
Raw numbers in files are meaningless. Most recordings and audio libraries are 16/44
Click to expand...

Very very few recordings are 16/44 and that's been the case for decades if truth be told.

The end product may well be so but the actual recording itself? No.

Whether that matters remains the subject of some debate and it's a shame that this test is not 16/44 vs 24/96. My experience is that it will depend on the processing power and care taken in the downsampling as to whether one is better off with the original studio resolution.

Clearly here, it seems that people are already finding that the effect of downsampling is audible to them.
 
merlin said:
Clearly here, it seems that people are already finding that the effect of downsampling is audible to them.
Click to expand...

No. How could it possibly seem that? At most some people think one file sounds better than another. You don't know whether the file they have identified is the 24 bit one.
 
AKG said:
The DAC in my REGA cdp also supports external USB input for downloaded files. REGA has built a quality (may I say "Killer") DAC that operates at 16/44
They shunned away from any manufactured upsampling approach.
Click to expand...
If they use a digital filter then they are upsampling I'm afraid.
There's no way round this: there is no means of expressing the absence of certain frequencies above nyquist without converting the data to a higher sample rate. If you could do that then you wouldn't need the digital filter.
[apologies if they do use an analog anti alias filter]
 
AAMOI have the people expressing a preference also tried ABX (you can use Foobar) and successfully distinguished between the files?

Tim
 
onlyconnect said:
AAMOI have the people expressing a preference also tried ABX (you can use Foobar) and successfully distinguished between the files?

Tim
Click to expand...
Interesting question but I'd be surprised. Anyway, I would recommend letting everyone have a go.
I would suggest that ABXing them could be left as a subsidiary question (eg if the outcome turns out to be that the overall result is that out of all people about 50% get each test right, and the 12.5% of people who get all 3 right want to find out whether they really do have golden ears).
 
adamdea said:
No. How could it possibly seem that? At most some people think one file sounds better than another. You don't know whether the file they have identified is the 24 bit one.
Click to expand...

Steady on there Adam. No one said one was better, simply that they identified audible differences.

That means (unless they are imagining things) that the process resulted in audible change which is what I stated.
 
You must log in or register to reply here.


advertisement


Back
Top