DaveWr;503266 Wrote: > Well I suppose this thread could grow and grow, debating the hot topics > of definition, which probably varies by continent. > > In my understanding we have oversampling, which is usually a relatively > simplistic method of generating intermediate samples - you have to > create new points that weren't there originally. > > In up-sampling, there is a digital filter algorithm which can have a > whole variety of mechanisms to generate the new extra samples + > appropriate dither mechanisms. > > The words are used by many people apparently interchangeably. In the > Linn document, I referred to earlier, this whole DAC pre-processing they > call upsampling, indeed their software updates actually change this > whole pre DAC chain. > > In the AK4396, there is also a set of digital filter interpolators to > create the extra samples, what algorithms they use is probably a trade > secret, but I think to just call it oversampling would be a little > unfair. > > The problem I see with all this SOX filter / up-sample is that unlike > Linn and several other vendors you can't fully over-ride the on chip > pre-DAC digital filters. You will get variations because as you change > the sample rate in, the DAC will change its digital filter > characteristics. > > Secondarily a lot of this up-sample work with different filters will > change (improve?) the transient performance of 16 bit 44.1k signals that > had less than optimum ADC recording / translation originally. > > In traditional R-2R ladder type DACs where all bits are used (not that > common in new components) the advantage of up sampling is that you will > tend to get better filter performance from a multi rate DAC. > > Its not the words that counts but what actually happens, and in this > threads case, what the sum of all the processing steps achieve. > > In the excitement over apodizing filters, views seem very mixed. Bob > Stuart of Meridian fame, generally highly regarded, states that the > significant increase in post transient ringing with apodizing filters is > OK, the ear mechanism apparently shuts this out post transients. He is > probably right, but others in the industry seem to disagree. Well > that's always been the audiophile world....... > > Dave
Definitions are tricky... I was under the impression that all delta-sigma DACS oversample (internally). However when I mentioned "upsampling" I was referring to an ASRC process (chip) prior to the DAC chip. For example my TACT uses an ASRC to up everything to 192k, whereas my DAC uses an ASRC to up everything to 384k... My (debateable) theory is that by feeding my TACT 96k exclusively via SOX I have given the TACT chip an easier life. I know that Sean feels this may not be the case. Listening tests are inconclusive. I can't hear any difference between 44.1 upped to 96 by SOX vs raw 44.1. As for the filters - I'll try them and see what happens. Now I have my PC working again I can try and measure the difference with ADM to get a feel for the scale of the impact. -- Phil Leigh You want to see the signal path BEFORE it gets onto a CD/vinyl...it ain't what you'd call minimal... SB Touch Beta (wired) - TACT 2.2X (Linear PSU) + Good Vibrations S/W - MF Triplethreat(Audiocom full mods) - Linn 5103 - Aktiv 5.1 system (6x LK140's, ESPEK/TRIKAN/KATAN/SEIZMIK 10.5), Townsend Supertweeters, Blue Jeans Digital,Kimber Speaker & Chord Interconnect cables Kitchen Boom, Outdoors: SB Radio ------------------------------------------------------------------------ Phil Leigh's Profile: http://forums.slimdevices.com/member.php?userid=85 View this thread: http://forums.slimdevices.com/showthread.php?t=69145 _______________________________________________ audiophiles mailing list [email protected] http://lists.slimdevices.com/mailman/listinfo/audiophiles
