[karohan]

I currently own a Fiio E10 DAC/amp and I the music software I use upsamples music with the iZotope 64-bit SRC. I was wondering how the upsampling to 24bit/96kHZ via the iZotope software upsampler compared with the Fiio E10's upsampling. I've heard it is better to use software upsampling instead of hardware upsampling if the computer is powerful enough to handle it. If I use software upsampling, what else is my DAC doing (and why would it sound better than my Macbook Pro's internal soundcard)?
 

 

[leeperry]

several things at play here:
-most upsampling DAC's using cheapo IC's such as SRC4192, a proper upsampling on the PC will require far more computing and most likely provide better SNR/THD+N/IMD(you can measure them using WaveSpectra)
-most entry/mid level USB/coax implementations provide much lower jitter at 96kHz than at 44.1kHz, especially considering that they never have a 44.1kHz multiple clock....it's usually either 12Mhz or 24.576Mhz(512x48kHz) so yes, 96kHz upsampling has its advantages.
 
You can look up this datasheet: http://www.cirrus.com/en/pubs/appNote/AN339REV1.pdf

 

[karohan]

Thanks for your reply! So it seems like it'd be better to do upsampling via software, and my computer can handle the calculations well enough. However, besides upsampling entirely, do DACs do anything that affects the presentation of the audio? In other words, is my E10 doing something with the sound better than the integrated DAC would be doing on my Macbook Pro. Because if I'm using software upsampling, I'm wondering what other advantages the Fiio E10's DAC has over the integrated DAC, besides the integrated amp on the E10 obviously.

 

[leeperry]

I think you're mixing upsampling and oversampling, this should shed some lights: http://www.audioholics.com/education/audio-formats-technology/upsampling-vs-oversampling-for-digital-audio
 
The built-in soundcard of your Mac is most likely a very cheap ASIC that does everything poorly(much like the Realtek's on PC motherboards) and the Fiio's use better DAC chips but very cheap output stages. Upsampling doesn't "improve" the SQ per se...I personally only enjoy it on lossy movie soundtracks and even though my DAC has an upsampling function, its manufacturer only advises to enable it on lossy audio as well.

 

[karohan]

Thanks for the link; that helped clear up some things about upsampling vs. oversampling. What do you mean by the Fiio having better DAC chips but "cheap output stages"? I tried listening to music without the upsampling actually, and I think the music does sound better this way, so I'm going to keep it like this. Do DACs perform better or worse at different sample rates? I read somewhere here on head-fi that some DACs will have better reconstruction filters (admittedly, I'm not really sure what these are) for their higher sample rates (e.g. 96 kHz) compared to 44.1 kHz. In this case, would it be better to use upsampling to achieve the sample rate the DAC is better with, and do you know if the Fiio is such a DAC?

 

[leeperry]

a cheap DAC could easily sound better with upsampling enabled indeed, for all the reasons explained above. Most DAC chips lower their oversampling rate above 96kHz, so this also has to be taken in account.
 
a DAC outputs a very low signal that needs to be amplified through an output stage, and many ppl think that this is what colors the sound the most....an output stage can go from $0.05 to several hundred dollars, of course they won't sound the same ^^