Audible difference in players? (Audirvana, JRiver, Roon, MusicBee, etc.)

[MediumRare]

I see on various forums that different music players have different SQ. A hardware manufacturer told me the same thing just last week. And personally, I think I hear a benefit to Audirvana. And yet, bits are bits, right?

So, @amirm , please save us from this mystery: Do these players deliver different bits or clock speeds or jitter or "musical ether" to a DAC via USB? Does it matter if you have "good" DAC?

 

[BillG]

While there may be some subtle differences in the sound quality due to their various processing engines, they will be just that when the processing is kept to an absolute minimum.

Yes, having a competently designed, transparent DAC helps, but I wouldn't concern myself with clock speeds, jitter, etc. as the competent design takes care of that for any input into it.

Unless one is willing to do bit level comparisons of the player outputs when rendering the same level matched audio, people saying this and that about the sound quality is just an opinion. And a dealer's would be the last opinion I'd trust to make that determination as their underlying motive is profit.

As for what you think you're hearing, a double blind, level matched listening test between players would be quite revealing...

 

[Rja4000]

Hi
If there are differences, that's more likely to be in streaming stability.
I mean: a bit as Amir demonstrated in his review of the REL wireless relay
(https://www.audiosciencereview.com/...s-of-rel-wireless-subwoofer-transmitter.9165/)
when you look at his distortion vs time graph.

The way the system is correcting for an-avoidable errors (timing, lost packets,...) may make a difference.
For computer data transfer, if a packet has to be sent again, the only difference is that you'll have to wait more for your transfer to be finished.
For audio or video, when buffer is empty, that's just lost data. And then distortion of the signal (and not small amount!)

 

[BillG]

If there are differences, that's more likely to be in streaming stability. I mean: a bit as Amir demonstrated with the revuew of the REL wireless relay

I'm pretty sure the OP is referring to software audio players...

 

[Enkay25]

My personal opinion: not really different.

The only difference which I could confidently detect is when I use WASAPI versus directsound on my surface pro...using the same app (especially foobar).

 

[Rja4000]

I'm pretty sure the OP is referring to software audio players...

A computer is having it's own life. How does it know that audio should have priority if Microsoft tells it differently?
So same thing may apply, even if to a lesser extent, to software players, sound card or DAC driver and so on.

 

[BillG]

A computer is having it's own life. How does it know that audio should have priority if Microsoft tells it differently?

You're speaking in too many generalities for me to want to engage in a discussion with you about this. If you have something much more specific, then we can all talk about it...

 

[Blumlein 88]

I see on various forums that different music players have different SQ. A hardware manufacturer told me the same thing just last week. And personally, I think I hear a benefit to Audirvana. And yet, bits are bits, right?

So, @amirm , please save us from this mystery: Do these players deliver different bits or clock speeds or jitter or "musical ether" to a DAC via USB? Does it matter if you have "good" DAC?

You'd need to be more specific about the players. Most good players will submit the correct bits. They are bit perfect.

Poorly designed DACs might be slightly effected by USB noise from computer activity. Schiit is one example. For the overwhelming majority there is not going to be any jitter effects of computer activity or from various software players. USB DACs generally clock out the data with an on board free running clock. This isn't effected by the software.

So for the most part good players aren't going to sound different.

In your listening comparisons did you match level carefully and get someone to switch without knowing which was which? We all have a tendency no one can ignore to hear differences that aren't there if we know something has changed like which software player is in use.

 

[MediumRare]

My personal opinion: not really different.

The only difference which I could confidently detect is when I use WASAPI versus directsound on my surface pro...using the same app (especially foobar).

Agree, on Windows the differences are huge. I preferred ASIO4ALL over WASAPI as well. But MacBook + Audirvana is consistently better defined and with more bass. It's too hard to A/B back and forth, so I relied on listening to full tracks. I thought was hearing it then - but without data, who knows!?

 

[MediumRare]

You'd need to be more specific about the players. Most good players will submit the correct bits. They are bit perfect.

For specific players, please see the headline on my original post. The key advantages of Audirvana, as far as I can tell are 1. "System Optimization" where it applies "extreme priority" over software running in the background and 2. Loading the full track into memory and playing it back from there. Doubtless there are other claimed technical advantages I don't understand. @amirm or you other professional guys with analyzers, can you do any data-based analysis?

 

[Fluffy]

I personally never heard a difference between the sound of different programs, and I really don't understand why there should be one TBH. I choose what program to use based on features.

And I really don't understand the whole WASAPI vs ASIO vs Direct Sound vs MME thing. If the music is played well it sounds correct, if not you can clearly hear glitches and breaks due to a non-continuous data stream. There are no subtleties or nuances to hear if the program functions correctly.

 

[Thomas savage]

Players that decode MQA will sound the best ..

 

[Kal Rubinson]

A computer is having it's own life. How does it know that audio should have priority if Microsoft tells it differently?

There are many ways to constrain Microsoft and give audio top priority.

For specific players, please see the headline on my original post. The key advantages of Audirvana, as far as I can tell are 1. "System Optimization" where it applies "extreme priority" over software running in the background and 2. Loading the full track into memory and playing it back from there.

1. This can be done for any music player application with commonly available tools.
2. This can be done with JRiver and, I assume, some other players as well.

 

[Veri]

Agree, on Windows the differences are huge.

Define "huge"...

 

[jhwalker]

For the most part, they do sound the same, in my opinion.

That said, I HAVE heard clear, repeatable differences even with "bit perfect" output. For example, there's a track on a very well-recorded Celtic music album (I'd have to find it again) where there's a certain tenor note that CLEARLY distorts when played through iTunes (every time) but is rendered perfectly via Audirvana. And, yes, I confirmed both programs were set for bit perfect (i.e., no processing), the Audio MIDI settings were correct for playback from iTunes, etc.

I also hear clear differences between Foobar (for example) and Roon (or Audirvana or ... ). I think there's more going on than meets the eye, and I don't like it, frankly being mostly a "bits is bits" kind of guy usually.

 

[Veri]

For the most part, they do sound the same, in my opinion.

That said, I HAVE heard clear, repeatable differences even with "bit perfect" output. For example, there's a track on a very well-recorded Celtic music album (I'd have to find it again) where there's a certain tenor note that CLEARLY distorts when played through iTunes (every time) but is rendered perfectly via Audirvana. And, yes, I confirmed both programs were set for bit perfect (i.e., no processing), the Audio MIDI settings were correct for playback from iTunes, etc.

How can it be bit-perfect when one time it distorts/clips your equipment and one time it does not?! what you describe sounds like a software bug, which should not be "bit perfect", then. Seems extremely odd.

 

[Krunok]

A computer is having it's own life. How does it know that audio should have priority if Microsoft tells it differently?
So same thing may apply, even if to a lesser extent, to software players, sound card or DAC driver and so on.

It doesn't, and you can experience posps/clicks with software players because of that. You can use DPC Latency Checker utility to check how it goes with your Windows. It has been writeen by Thesycon, the same company that wrote ASIO drivers for Windows and for many HW USB DACs.

 

[jhwalker]

How can it be bit-perfect when one time it distorts/clips your equipment and one time it does not?! what you describe sounds like a software bug, which should not be "bit perfect", then. Seems extremely odd.

Agreed - as I said, I don't like it, because it shouldn't be that way.

And I think even non-trained listeners s/b able to hear the difference between Foobar and Audirvana, for example - it's pretty blatant.

 

[dc655321]

The key advantages of Audirvana, as far as I can tell are 1. "System Optimization" where it applies "extreme priority" over software running in the background

Audio playback is an undemanding application, primarily I/O-driven. Eg: a 192 ksamples/s stream at 24-bits/sample is only 1.15 MB/s. A CPU is minimally involved (unless DSP is active) in the actual playback of audio - it is a director of traffic.

and 2. Loading the full track into memory and playing it back from there

In order for a CPU to do work on data, it must be in memory (RAM). Whether the whole file is in-memory or not is largely irrelevant as long as the DAC's buffer is sufficiently populated, preventing underflow/starvation. For a buffer on the order of say 64 ksamples, that is about 1/3 of second of playback, a comfortably long time between transfers of data.

From Audirvana's homepage:

By taking control on your computer’s audio flow, Audirvana prevents other applications from interfering during listening and from making hidden changes to audio samples. It reduces CPU activity and stabilizes power supply to minimize digital distortion and possible radio frequency interference. Finally, it optimizes the operation of your system’s digital-to-analog converter – whether integrated or external – by providing it with an adapted and pre-decoded digital stream.

As Ray would say, "Uh-huh..."

Having said all this, it is possible that part of a system's audio stack of software could modify its input in a perceptible way.
IIRC, Windows had problems in the past with a noisy resampler, for example.

@amirm or you other professional guys with analyzers, can you do any data-based analysis?

If you have a soundcard with a recording input, and the will to learn, you can check for audio-player-induced effects yourself.
Search on this site for @Blumlein 88's loopback test and @pkane's null-testing software for ideas about how to proceed.

 

[Fluffy]

I gotta say, this thread got suspiciously subjective compared to the rest of this site. Can anybody here actually produce evidence of differences between players?

And preferably excluding the effects of resampling, that are dependent on the quality of algorithms and such. Just a plain example of two players playing the same 44.1khz/16bit file without any DSP running, sounding different.