Mnyb wrote: 
> The trend I was referring to with my blanket dismissal of NOS DAC was
> the ones completely without any filter, this was very popular as DIY for
> while so popular that it seems to be the norm :-/

That's it in a nutshell. 

Most recent work with NOS DACs has centered around the Philips TDA 1541
as a more-or-less stand alone DAC with no or minimal to vestigial low
pass filtering. If you study its history it was never really intended by
its developers to be used that way. Information about how the TDA 1541
was supposed to used was discussed in the early days of digital audio
(1981-1983) by professional magazines like as EDN and EE Times (if they
existed them and if they didn't their forebearers), but issues of those
periodicals that old seem to be lost to posterity. There may be relevant
Philips Application notes, but I can't find them either.

The TDA 1541 chip was originally designed to be a component of an
oversampled DAC, most significantly to be followed by a digital filter
chip such as the SAA7030 and its sequels. One fact that I haven't seen
is references to what used to be called Aperture Effect which is the
reason why the output of many DAC chips have an unexpected (to many)
droop at high frequencies.  This issue is discussed in the context of
data acquisition here: http://www.cypress.com/?docID=45630.

When used all by itself the output of a TDA 1541 chip is going to have
that HF-drooping frequency response that is shown in the Cyprus
Semiconductor reference in figure 1.  The digital filter chips such as
the SAA7030  that it was supposed to be used with such as the contained
corrections for this that were in the day called . It is as simple as
that - the TDA 1541 was never intended to be used without some kind of
Aperture Correction, and if you don't provide it, its going to sound a
little soft compared to an accurate DAC. 

There are other potential bad consequences to not having an appropriate
reconstruction filter following the DAC such as IM in following stages
of amplification due to the fairly large amounts of HF noise that is not
filtered out like it should be. The noise itself is > 22 KHz and likely
to not be heard by many if anybody.  The IM doesn't always happen to an
audible degree, but I'm sure given the relatively high amplitude of this
noise it does happen some times. I suspect that there are even people
who like their music with a little aharmonic nonlinear distortion
spurious responses tossed in.


------------------------------------------------------------------------
arnyk's Profile: http://forums.slimdevices.com/member.php?userid=64365
View this thread: http://forums.slimdevices.com/showthread.php?t=93483

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

Reply via email to