One thing to consider in your evaluation of DXD PCM vs DSD playback via Hugo… The Chord products are PCM-centric… there is no unfettered 1-bit PDM signal path in the Chord DAC architecture… 1-bit PDM (DSD) is converted to multi-bit PCM… the output circuitry is not a simple low-pass filter… The ESS DAC chipsets follow a similar design dogma.
1 Like
In the past I used upsampling by factor 2, but was never impressed on my then Zodiac Dac and preferred native mode. Since I used an Ifi Neo Stream I tried all kinds of upsampling to the maximum of dsd512. I found on the Ifi dsd512 sounded best. Since a week I own a T+A Dac200, the maximum upsampling over usb is ‘only’ dsd256 over Dop. The T+A is in another league over the Ifi sound wise, but also with the T+A I still prefer upsampling, in this case to dsd256, to native replay by a wide margin.
Yes I’ve read that in discussion forums. My perceptions match a lot of others when it comes to DSD with Chord DAC’s. However the resolution I get with PCM through the TT2 when paired with the Focal Utopia’s is too addictive to want to change anytime soon.
1 Like
There are quite a few DACs that can do an extremely good job reproducing a very satisfying playback experience from both formats… Ultimately, it will always reconcile-down to subjective preference in the final audition. 
1 Like
One thing that has not been discussed here is the affect of changes to the parameters of r8Brain or SoX on the output sound-quality we audition, and the synergy of these parameter settings in concert with any given playback DAC/system…
1 Like
Actual measurements have been done with many of these DAC chips, and the sweet spot for lowest distortion tends to lie with either DSD256 or DSD512.
There’s no particular mystery about upsampling. Virtually every DAC does it internally with non-DSD input, and many DACs, particularly those with ESS chips, will do it with even DSD input. Why? Because it makes good filtering easier, and filtering is necessary in digital audio. (We can get into technical discussions if you want, but that’s the simple explanation.) So the question is not whether to upsample, but where, in the DAC or Audirvāna. Computer CPUs being more capable than DAC chips, I prefer to have it done in the computer by Audirvāna.
2 Likes
This is purely subjective as there are non-linearities that will determine one’s subjective opinion as to this having any value… The main one being the threshold and limitations of the human auditory mechanism and neurological interpretations…
Measurements of noise and distortion may be purely subjective in your view, and that’s fine. You may feel there are other measurements or your own subjective reaction to the sound that are more important. My own personal preference is the settings that minimize noise and distortion figures, but of course everyone is free to choose what they might prefer instead.
1 Like
In the case of up-sampling and modulation of lower bit-depth and sample-rate digital audio file data by r8Brain and SoX, most Audirvana users are going to use their own auditory assessments when making parametric adjustments to the filter operations, subsequently, on playback sound-quality, in a synergistic relationship with the DAC/playback system, etc… Of course, it all boils-down to personal preference in the final interpretive assessment…
Everything is relative to the subjective experience… Isn’t an interpretation of noise and distortion rather ambiguous and related to a context of tangibility, even though quantifiable under certain conditions of measurement? I wonder how many high-end tube-electronics fanatics would agree with a construct of absolute dependency on quantified measurements as a final determination of sound-quality? At some point, cognitive-bias will play a greater role in the interpretation of sound-quality on any given playback system, since most cannot analyze with precision analytical tools, the output quality of their playback system, and the electro-mechanical and acoustical and RF/EMF environment in which these auditory assessments are made… We must look at these things in the context of the real-world of experience… I always work to reduced noise in the signal where this noise can precipitate distortions in the tangible real-world audibility in playback, that I perceive with my own hearing mechanics and neurology on my subjective amalgamation of playback components and tweaks…
I find this a difficult topic to discuss generically on a forum.
If we want to make it ‘objective’, the measured values are never sufficient to interpret observations.
If we make it ‘subjective’ then we will never agree because everyone prefers and perceives something different.
It is also highly complex because it depends on so much. The material you are resampling. The software used, the hardware used, the room, other DSP, etc. The resampling is 1 piece of the entire chain.
With Audirvana and (then) Izotope and a Bel Canto DAC (with expensive cables and a Farad PSU), I spent a lot of time finding settings that were better than standard upsampling of the DAC itself.
When I bought the Hegel Röst, those settings no longer worked and it was better to turn off upsampling. And with SoX I was never quite able to achieve the same results as with Izotope.
Now I have a Heed Abacus and HQplayer to ‘play’ with. There are so many variations possible and I notice that I find it very interesting, but at the same time it also gets in the way of ‘just experiencing music’.
Most of what I read on this forum are users who mainly want to listen to music and want a nice GUI. If you google for HQplayer users, it is about all possible filters and noise shapers. The operation of the DAC and measuring the noise, etc.
Insofar as the forum is representative of the entire Audirvana user group, there seems to be less interest in resamplers here.
As far as I’m concerned, there is no right or wrong. If someone enjoys his music then that’s good and what someone else thinks about it is not that relevant.
5 Likes
The quality of the music we listen to depends on many factors. There are days when, no matter how good the quality is, you will think it is not good. You need a predisposition to listen to music.
There is no magic formula, everyone will find their best way to achieve the best quality.
In my path, the computer with good audio programs is what will provide the most resources for a multitude of adjustments.
Without a doubt, Audirvana is among the best, if not the best.
Today, with a large number of existing devices, it gives the opportunity to choose what suits you best within your financial means.
What matters in the end is the music, good music.
8 Likes
3 Likes
Completely agree with everyone who’s said whatever you enjoy most is best for you. I agree with something Reynaldo alluded to, which is that our preferences can change. For myself, I feel that starting with the lowest measured distortion and noise (and fully recognizing we’re very unlikely to have all the relevant distortion and noise figures) somewhat trains me to listen to the best quality I can get from my system.
2 Likes
Those elements that we cannot measure or are not measured properly in context, are the non-linearities that truly define the final audition…
It’s like pushing on a ballon if not aware of all of the noise related distortion potentials of any given playback system scenario… Many do not have the technical acumen to address the gamut of potentials that affect the sound-quality of their subjective playback experience… Experimentation is key to finding all of that which is imbued in a recording, whether the recording is good or not so good or great… We can attach qualitative meaning and value to just about anything when it comes down to cognitive-biases.
I applaud you and others that take extra measure to realize the most from each and every recording played on their particular amalgamation of components and tweaks.
Utilizing the parametric controls of r8Brain and SoX with insights as to what they do, in concert with output levels, etc, in Audirvana and in concert with any filter settings available in the DAC, allows for very articulate control of the perceived sound-quality of our systems, in the case for up-sampling/modulation of lower-resolution files… However, it can be intimidating for some less patient in dialing in a sound through experimentation.
As audiophiles keep changing their systems for improved listening experience, I guess setlling on one configuration can be a good decision. The brain need time to accommodate to signature of a certain music signal. Not to forget we listen with our brains. Our ears are just the Mic.
But some people including me may from time to time feel board listing to the same stuff. This is why change may be perceived as better. So, the best could be to change until we discover what best suites our taste. But again I need to keep remembering is the music is what matters at the end.
3 Likes
If you have fun endlessly fine-tuning hardware and software (and financing the future of your local audio dealer), who am I to judge.
They sometimes say that audiophiles use music to listen to their audio set and that music lovers have an audio set to listen to music.
I have no problem with someone’s intrinsic reason for doing something. It’s a lot of fun and can be very satisfying to take a audiophile step over and over again, but you’re never done, you’ll always be looking for the next update, and again and again…
5 Likes
I’ve fine-tuned my system a long time ago… and enjoying the quality of sound that it produces…
1 Like
I like a piece of equipment here and there, but I may have cables that are older than some people in this forum, and Audirvāna has been my main music player since Damien was sending out source code and we were experimenting with the sound of binaries made with different compilers.
If it ain’t broke, I’m not “fixing” it.
5 Likes
perfectly said.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.