Phil,
See this post:
Submitted by Miska on Sun, 02/12/2012 - 17:43
Here: http://www.computeraudiophile.com/content/Why-2496-not-24192

Which shows graphically what I mean. Upsampling should make a big
difference to the smoothness of the waveform - if the data points are
interpolated intelligently.

I expect to see smoothed waveforms out of DAC. From a native 16/44 DAC
(my befuddlement yesterday) I expect to see analogue smoothing. From a
24 bit, high sample rate DAC I expect to see smoothing due to the
upsampling (and analogue smoothing at that finer level but this would
not be resolvable probably). You keep waving this away but I think it's
a reasonable expectation.


Phil Leigh;695278 Wrote: 
> erm... the sampling RATE won't make any difference to this...
I think it will (see above)

Phil Leigh;695278 Wrote: 
> 
> Presumably they use 16 bits for this test to represent redbook CD
> replay into the DAC....
> 
Kind of, except I don't think it's exactly equivalent (see further
below).

Phil Leigh;695278 Wrote: 
> 
> ....Had they generated a 24-bit -90dB sine wave file, the result would
> have looked VERY different :-)
> 
Yes of course and they did do that. That is figure 6.

Phil Leigh;695278 Wrote: 
> 
> Just to be crystal clear on this, when Stereophile talk about the 3
> voltage levels... they ARE talking about the AC voltages:
> zero-crossing, peak  and trough for a single bit representing the sine
> wave. They are NOT talking about anything to do with "jagged" DC ladder
> steps or anything like that - because with one bit going on or off there
> would only be 2 DC values anyway (on or off) and those values do NOT
> EXIST outside of the DAC chip/IV stage/Filter... so they can't be
> measured by putting any kind of test equipment across the analogue
> output of a DAC.
> regards
> Phil
No. They measured at the analogue output. Hence my initial puzzlement
about no smoothing.

Phil Leigh;695278 Wrote: 
> 
> I wouldn't want to  speculate too much on how Stereophile created their
> test data, but as you can't easily record a perfect -90dB sine wave from
> any analogue source... (too much noise!) my guess is they generated it
> as a 16-bit wav test tone mathematically on a PC, which is easy enough
> to do.
> 
I proposed what they did in my last post. Please have a read again
carefully:
darrenyeats;695256 Wrote: 
> 
> Now I realise when Stereophile tested the 16 bit undithered waveform,
> probably they used a file at the highest sampling rate and bit
> depth...just mimicking a 16 bit waveform within it. So the DAC doesn't
> treat it to any upsampling goodness...the customer wants a crude
> staircase drawn in hi-rez, the customer gets just that.
> 
IOW I think they created this 16 bit -90db 1kHz sine wave
mathematically and placed it in a 24/96 file. It is already at the
DAC's native bit depth and sample rate so no interpolation is
attempted. The DAC tries to play (as near as possible) a staircase
shaped wave which, as we know, is impossible to do exactly.

Had they played an actual 16/44 file of the same amplitude, I expect it
would have been smoothed. Which is my point.
Darren


-- 
darrenyeats

http://www.amazon.co.uk/gp/richpub/listmania/byauthor/A3H57URKQB8AQO/ref=cm_pdp_content_listmania/203-7606506-5721503.

SB Touch
------------------------------------------------------------------------
darrenyeats's Profile: http://forums.slimdevices.com/member.php?userid=10799
View this thread: http://forums.slimdevices.com/showthread.php?t=94054

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/audiophiles

Reply via email to