Marcel, an efficiently communicated, not too heavy, three and a half pages. I would like to offer a suggestion to consider. You specify nonsubtractive dither several times, which makes me wonder whether it may be worth a few sentences to briefly explain to readers what subtractive dither is. Even though, I believe I’ve only ever seen one commercial subtractive dither implementation for DACs, an old Anagram Technologies module. At any rate, nicely done.
Marcel, in reading your description of subtractive dither, I wondered what is your assessment of the effectiveness of Anagram Technologies' 'Sonic Scrambling' approach, which is entirely local to the DAC. See the link down below to an old Anagram data sheet, which opens to page 13 of that document. Just close the ad which floats over the document when you open it. Their Sonic Scrambling technique is shown on pages, 13 and 14.
Essentially, the digital output of the dither generator is summed with the signal data sent to a pair of differentially configured DAC chips. The dither data is summed in mutually inverted phase between the DACs. After d/a conversion, the dither noise now appears as a common-mode signal, which is subsequently subtracted by a following differential amplifier. So, the desired signal is differentially amplified, while the dither noise is rejected as common-mode. All done completely within the DAC box. This seems so neat a subtractive dither solution for DACs, that I've always wondered what I must be overlooking about why it didn't appear to catch on.
S2 1.2 F
Posts about PGGB upsampling software have now been moved to this dedicated thread:
PGGB upsampling software
George
Now that there is an dedicated thread for PGGB related product discussion, located under the Software-Tools sub-forum, ZB is comfortable to directly address your questions posted there himself.
PGGB upsampling software
PGGB upsampling software
Last edited:
Thanks for the clear overview! I had lost track of what exactly you had and hadn't tried, nice to have it all in one report now.
Regarding the Audacity filters, I think the 20 to 30 msec on page 3 should be 2 to 3 ms.
I have to think about that one... It's not the traditional subtractive dither that requires a many-bit DAC just for the dither that is to be subtracted, but obviously some part does get subtracted. Interesting!
Just a warning. No ad with Ublock Origin, but this link crashed my display driver 3 times in 30 second, leading to a blue screen of death. Possible malware in the ad link.
Hi Marcel,
Thx, you're correct, but still waiting for input from others, I'll do the corrections in one go.
Hans
If an ad link allows your os to bluescreen - well, get a proper os! Or firstly remove the malware already on your computer
@Tfive. You know nothing about malware on my laptop. Why you bother to comment?
It is an old Compaq 6720s (Core 2 Duo) running Win7, not protected to the same degree as new PC's (in hardware and OS) from advanced spyware techniques embedded in Web pages. This is one explanation. The other one is that those who created this malware embedded in Webpage tested it successfully only on new computers. Blue screen of death is a proof that it doesn't work well on my laptop.
It is an old Compaq 6720s (Core 2 Duo) running Win7, not protected to the same degree as new PC's (in hardware and OS) from advanced spyware techniques embedded in Web pages. This is one explanation. The other one is that those who created this malware embedded in Webpage tested it successfully only on new computers. Blue screen of death is a proof that it doesn't work well on my laptop.
Last edited:
Ken, Interesting idea.
In an ideal world I would say that when you add someting in the digital domain, in this case dither noise, that is subtracted at a later stage in the analogue domain, the end result should be as if nothing was added in the first place.
But since Dac’s are never the same, a small error remains after removing the common mode part, but whether this trick prevents noise modulation ??
Anxious to see what Marcel’s reaction will be.
Hans
Last edited:
Hans, you've identify the exact question which I'm hoping that Marcel can answer. When dither reduces a DAC's distortion, does such reduction manifest within the DAC, or only externally by error averaging within a spectrum analyzer, or a listener's ears? Sort of like the old Zen question of; If a tree falls in the woods, does it make a sound if there is no one to hear it? If dither randomizes the non-linearity of a DAC, is the distortion actually reduced if there is no one to hear it?
If the Anagram technique were to actually work, it seems that the distortion of any given DAC could be removed by the application of however much dither amplitude is required, then have the original SNR restored by the their dither noise subtraction technique. Which makes me suspect that it may not actually work.
If somebody make pdf or screenshots from this document, I would appreciate.
I was suspecting scrambling technique used in Holo Audio when users reported that they could hear something like sea waves running in parallel with music. Thanks Hans for pointing it out. In this case audible results like sea waves would indicate modulation products of error randomization.
I was suspecting scrambling technique used in Holo Audio when users reported that they could hear something like sea waves running in parallel with music. Thanks Hans for pointing it out. In this case audible results like sea waves would indicate modulation products of error randomization.
from page 14..
Can that D/A Converters section remove the dither in the digital domain..?? Or, the dither removal happens at the I/V Conversion stage?
I think it's the latter. If I'm right, the amount of dither removal will depend on the CMRR capability of I/V Converter...(OP amp) which will vary widely; in other words, that dither will never be removed to a satisfactory degree, which may explain why added dither actually sounds bad.
If I'm wrong, how can two DAC's (independent DAC's within the same die) ever be able to remove common-mode dither??
Can that D/A Converters section remove the dither in the digital domain..?? Or, the dither removal happens at the I/V Conversion stage?
I think it's the latter. If I'm right, the amount of dither removal will depend on the CMRR capability of I/V Converter...(OP amp) which will vary widely; in other words, that dither will never be removed to a satisfactory degree, which may explain why added dither actually sounds bad.
If I'm wrong, how can two DAC's (independent DAC's within the same die) ever be able to remove common-mode dither??