1 kHz at 0dB clips - by default?

[Herbert]

Recently I tested my Buffalo II DAC from Twistedpear - a 1kHz sine 0dB test tone.
There is a "squarelike" but faint distortion in the upper frequencies.
But before running a verdict, I should mention that the testtone is from 1982,
from an YEDS-18 Test disc made by Sony for service personnel.

I analysed the file in Audacity , one sample width shows clipping,
whereas 3 samples show no clipping.
No clue what the sample width practically means.
Assuming the sample is error free and spot on with full scale,
clipping needs at least two samples, because the resulting curve
between the two samples overshoots 0dB.
This is what sound engineers call them - overshoots and some software/
plugins check the final mix for it. So normally, the headroom should not
be above -0.5 -0.3 dB
But the test tone on the YEDS 18 it was made while only 16bit ladder resistor DACs
were common, to my knowledge, clipping with them is more audible.

But Audacity also has a tone generator. The generated sinetone made by Audacity also clips
when being analysed with one sample of distance.
So this analyse tool seems to be useless or:
a 1KHz Tone at full scale clips by default, Sony technicians were deaf to notice and especially
at the beginning of the Compact disc, the majority of recordings never exceeded -3dB.

The 60 second sine testtone from the YEDS-18 is too big to be attached.
So here is a downloadlink from Wetransfer, it is valid for 7 days:

https://we.tl/t-z6RtFFQiqf

Would be interesting to check whether the fault is this test-signal from the dawn
of the Digital Age, or 1kHZ at full scale means clipping - though I assume
this is exactly the signal Amir uses, but maybe today with a slight headroom by default...?

 

[pma]

This is usual and often the problem is not the DAC but the generated test tone. Should be investigated first, best by CE Pro or Adobe Audition.

 

[DVDdoug]

I assume that the analog output of a DAC can go over 0dB and I've wondered what happens to "inter-sample overs" in real world DACs... There's no reason the reconstruction filter and analog output has to be limited to whatever voltage 0dBFS converts to.

But Audacity also has a tone generator. The generated sinetone made by Audacity also clips
when being analysed with one sample of distance.
So this analyse tool seems to be useless or:

Audacity isn't "analyzing" the waveform and of course it's not assuming a pure sine wave. I always say it's showing potential clipping. Audacity uses floating-point internally so it can go over 0dB without actually clipping. Or, if you have a clipped waveform and you reduce it below 0dB, Audacity won't "show red".

A lot of MP3s go slightly over 0dB so they will "show red" although they aren't actually clipped. (Of course your DAC will clip if you play it a "full digital volume".)

If you've just made a recording and it "shows red" before any processing, etc., then it's a good indication of clipping. It's also helpful if you've done some processing that pushes the peaks over 0dB. If you "see red" you can reduce the level before exporting.




The Audacity manual says:

A sample is displayed as clipped if it touches or exceeds 0 dB and so touches or exceeds the volume envelope defined by +1.0 to - 1.0...

A run of four or more consecutive clipped samples will also trigger the red clipping indicator in Meter Toolbar...

There is also a Find Clipping feature where you can choose the number of samples at or above 0dB that trigger a clipping indication.

 

[pma]

I assume that the analog output of a DAC can go over 0dB

No. The DAC is almost always oversampling and the interleaved samples are calculated and inserted as digital samples by the interpolating digital filter. The samples value cannot exceed 0dBFS. Only some older DACs used with HDCD in the past could do so.

 

[Herbert]

I just checked the sine linked above with a mastering plugin that has intersample detection, no clipping.
0dB headroom and -3.3RMS btw. As I can hear the square-like distortion very faintly on a 1-bit DAC
(Sony CDP-X5000) as well as on the Buffalo II, I'll blame it on something else for the moment.
Interestingly it seems to be more audible when the ear is between the tweeter and mid/Bass driver.