• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

VeraVox - a macOS bit-perfect player with measurement tools. Looking for feedback and some tests

Hi Stefano, not planned, and the reason is technical rather than a matter of scope.

No current Mac has an optical drive, so this would apply to external USB drives only. That is the smaller part. The larger part is what reading an audio CD accurately depends on: C2 error reporting, controlled re-reads of doubtful sectors, and correction for the drive's read offset. macOS presents an audio CD as a folder of AIFF tracks, and that path returns what the drive returns, with no indication to the application of whether a sector came back clean. Playing while reading also leaves no time budget for the re-reads that would settle the question.

VV is built on a claim that can be checked: the samples reaching the DAC are identical to the samples in the file, and the built-in bit-perfect test lets you verify that on your own hardware. A real-time optical read cannot be held to that standard, so it stays outside the product.

Playing the disc's contents as files is the supported path, and the CD-specific handling is already there. Declared pre-emphasis is detected and the inverse 50/15 µs curve (IEC 60908) is applied. HDCD encoding is detected and decoded, with dynamic range reported both as stored and as decoded.

R.
Hi Riky, thanks for the reply. Very informative. I'm struggling for long time to find a reliable and most of all bit-wise accurate application to play my CD collection via MacOS. Your explanation can shed light on the reason I cannot find it :-) Apple Music sounds muddy, Foobar for MacOS does not support it, Jriver looked promising but it is terribly buggy: too many disc ejected in the middle of a session.

This aside, VeraVox is a wonderful piece of software, one of the few that pass 32 bit integer to the DAC without messing it. (e. g. RME 32 bit bitperfect test file worked).

It is quite a pain (I'm joking) to see how many hi-res file in my collection are fake or not so hi res... Is that algoritm cosidered stable or can output a possible high number of false positive?

Have you any dealine for production of VeraVox?
 
Hi Riky, thanks for the reply. Very informative. I'm struggling for long time to find a reliable and most of all bit-wise accurate application to play my CD collection via MacOS. Your explanation can shed light on the reason I cannot find it :-) Apple Music sounds muddy, Foobar for MacOS does not support it, Jriver looked promising but it is terribly buggy: too many disc ejected in the middle of a session.

This aside, VeraVox is a wonderful piece of software, one of the few that pass 32 bit integer to the DAC without messing it. (e. g. RME 32 bit bitperfect test file worked).

It is quite a pain (I'm joking) to see how many hi-res file in my collection are fake or not so hi res... Is that algoritm cosidered stable or can output a possible high number of false positive?

Have you any dealine for production of VeraVox?

Thanks Stefano! Good to hear the RME 32-bit file came through intact ;-)

The measurement is solid, the word is not. What the tool measures is deterministic, where the spectrum stops and where the noise floor sits, and it only raises the red verdict when the gap is wide: a container above 48 kHz whose content ends below 25 kHz. Marginal cases do not trip it, so false positives on the measurement itself are rare.

What it cannot measure is why, and that is where "fake" reads more harshly than it should. It is a short label for one condition: the content does not fill the container. Very often that is not deception. An analog tape master never had that bandwidth to begin with. An early digital recording was made at 44.1 or 48 kHz, so a hi-res release of it is an honest transfer of the master that actually exists. Vinyl transfers, noise shaping and some mastering chains leave a similar signature. In all of those cases the file is not defective either, it plays correctly and the music is untouched.

So read a red verdict as "this file has no content above X kHz", which is a fact, rather than "someone cheated", which is a guess. Only the first is something the tool knows.

On timing, beta is closed and VeraVox 1.0 ships tomorrow (hopefully)

R.
 
Last edited:
Hi Asrid, thank you.

On 1 and 2: what you're hitting isn't a VV decision, it's what exclusive mode is at the CoreAudio level. Apple's own docs for the property VV sets say the HAL "will automatically take hog mode" as soon as a device runs in a non-mixable format. Once that happens the system mixer is out of the path, and the macOS volume slider lives in that mixer, so there's nothing left for it to act on. macOS then moves the default output elsewhere, which is why the Sound menu jumps to the Studio Display. From that point the slider is controlling the display, not the 5909.

So yes, exclusive mode assumes attenuation happens after the DAC. Your HDV820 setup is the intended shape, which is why it gave you no trouble.

The bit you've probably missed is that the mode is per device, not global. Open the output picker, select the No 5909, and under PLAYBACK MODE pick Shared Output. Same for the Studio Display speakers, and leave the HDV820 on Exclusive. Each choice is remembered for that device. Shared Output still sets the device to the file's own sample rate so nothing gets resampled, it just goes through the system engine, which is what gives you the volume back.

On a special mode: I won't put a gain coefficient on the samples, that's the one thing the player exists to avoid. Driving a DAC's own hardware volume is a different matter and I'd consider it, but only if the device actually exposes one. Quick check if you're willing: open Audio MIDI Setup, select the 5909, and see whether the output volume sliders are live or greyed out. Your own note that the level freezes at whatever it was last set to suggests there's a real attenuator in there that simply stops getting updates. I'd like to know what you see, for the 5909 and for the display.

On 3 you're right and I won't argue. For now, the app's light/dark button in the top bar is independent of the system, so you can keep macOS in Light and run VeraVox in Dark. The contrast I can fix soon. The type sizes are written literally all over the interface, so a size preference means routing them through a single scale first, and I'd rather do that properly than bolt on a multiplier. After next week launch.

Thank you again,
R.
Thank you for your considered and very informative reply.

Regarding your reply in the 4th paragraph. With Shared Output selected for the 5909 Audio MIDI Setup shows a live Primary volume control. The menu bar volume control functions as well. However, the on headphone volume control does not adjust the volume. First screenshot.
Playing back through the 5909 in exclusive mode is shown in the second screenshot. It does show that VeraVox has a lock on the DAC of the headphones. The sound widget in the menu bar, as you described has switched from the 5909 to the Studio Display Speakers. The sound widget in the menu bar while moveable has no effect.
In the Audio MIDI the lock on the DAC is shown. The Primary Volume control is shown and moveable but has not effect on sound. At no time is any sound heard in the headphones. I hope this is helpful.

I would like to suggest making the Playback Mode control accessible in the VeraVox menu bar under Tools.

Regarding the Waveform Display - it is rather small in height and difficult to make out the waveform at its current size in either ABS or ADP mode. This is the case even when VeraVox is full screen on a 5K MacStudio display. I don't like pop-up displays although that is one possible solution. Sometimes I feel that a Waveform is a distraction while listening while I want a player to be on the display and not in the Dock. Having Waveform Display being controllable On/Off in the VeraVox menu bar under Appearance could give this feature additional flexibility with the default being on so the user initially is aware of this feature.

Thanks again.
 

Attachments

  • Mark Levinson No5905 playback Shared.png
    Mark Levinson No5905 playback Shared.png
    1.2 MB · Views: 44
  • 5909 in Exclusive mode.png
    5909 in Exclusive mode.png
    1 MB · Views: 44
Thank you for your considered and very informative reply.

Regarding your reply in the 4th paragraph. With Shared Output selected for the 5909 Audio MIDI Setup shows a live Primary volume control. The menu bar volume control functions as well. However, the on headphone volume control does not adjust the volume. First screenshot.
Playing back through the 5909 in exclusive mode is shown in the second screenshot. It does show that VeraVox has a lock on the DAC of the headphones. The sound widget in the menu bar, as you described has switched from the 5909 to the Studio Display Speakers. The sound widget in the menu bar while moveable has no effect.
In the Audio MIDI the lock on the DAC is shown. The Primary Volume control is shown and moveable but has not effect on sound. At no time is any sound heard in the headphones. I hope this is helpful.

I would like to suggest making the Playback Mode control accessible in the VeraVox menu bar under Tools.

Regarding the Waveform Display - it is rather small in height and difficult to make out the waveform at its current size in either ABS or ADP mode. This is the case even when VeraVox is full screen on a 5K MacStudio display. I don't like pop-up displays although that is one possible solution. Sometimes I feel that a Waveform is a distraction while listening while I want a player to be on the display and not in the Dock. Having Waveform Display being controllable On/Off in the VeraVox menu bar under Appearance could give this feature additional flexibility with the default being on so the user initially is aware of this feature.

Thanks again.

Hi ASRid, thanks for the screenshots.

The headphone's own volume buttons in shared output is between the 5909 and macOS, not something VV sits in front of. In USB mode the host owns the level, which is what the Audio MIDI Setup slider and the menu bar are doing.

The exclusive result is worth a look, but before reading anything into it, your toolbar shows three non default options active at once: max bit depth, DAC warm-up and sample-rate switch pause. All three act on the DAC in the same narrow window, between VV taking exclusive ownership and the first sample going out, which is precisely where your silence sits. Max bit depth is also why a 16bit mp3 is being sent to the device as 24.

Turn all three off and try exclusive again. If the sound returns, re-enable them one at a time and tell me which one costs you the audio. If it stays silent with all three off, that is worth knowing too: the 5909 is a Bluetooth headphone with USB as its secondary interface, and a device advertising non-mixable formats is not the same as a device that plays them.

During exclusive playback, does the progress bar advance while you hear nothing, or does playback not start at all? If you want, send the diagnostic log (Help > Export Diagnostic Log) and expand ALL PHYSICAL FORMATS in the DAC diagostic panel you already have open.

Playback Mode in the Tools menu is noted, reasonable.

The waveform is drawn at a fixed height, so full screen changes nothing. I'll think about a toggle option.
For a quieter player right now, the compress icon in the top bar switches to the Mini Player.

LEt me know,
thank you,
R.
 
Hi ASRid, thanks for the screenshots.

The headphone's own volume buttons in shared output is between the 5909 and macOS, not something VV sits in front of. In USB mode the host owns the level, which is what the Audio MIDI Setup slider and the menu bar are doing.

The exclusive result is worth a look, but before reading anything into it, your toolbar shows three non default options active at once: max bit depth, DAC warm-up and sample-rate switch pause. All three act on the DAC in the same narrow window, between VV taking exclusive ownership and the first sample going out, which is precisely where your silence sits. Max bit depth is also why a 16bit mp3 is being sent to the device as 24.

Turn all three off and try exclusive again. If the sound returns, re-enable them one at a time and tell me which one costs you the audio. If it stays silent with all three off, that is worth knowing too: the 5909 is a Bluetooth headphone with USB as its secondary interface, and a device advertising non-mixable formats is not the same as a device that plays them.

During exclusive playback, does the progress bar advance while you hear nothing, or does playback not start at all? If you want, send the diagnostic log (Help > Export Diagnostic Log) and expand ALL PHYSICAL FORMATS in the DAC diagostic panel you already have open.

Playback Mode in the Tools menu is noted, reasonable.

The waveform is drawn at a fixed height, so full screen changes nothing. I'll think about a toggle option.
For a quieter player right now, the compress icon in the top bar switches to the Mini Player.

LEt me know,
thank you,
R.
Hi ASRid, thanks for the screenshots.

The headphone's own volume buttons in shared output is between the 5909 and macOS, not something VV sits in front of. In USB mode the host owns the level, which is what the Audio MIDI Setup slider and the menu bar are doing.

The exclusive result is worth a look, but before reading anything into it, your toolbar shows three non default options active at once: max bit depth, DAC warm-up and sample-rate switch pause. All three act on the DAC in the same narrow window, between VV taking exclusive ownership and the first sample going out, which is precisely where your silence sits. Max bit depth is also why a 16bit mp3 is being sent to the device as 24.

Turn all three off and try exclusive again. If the sound returns, re-enable them one at a time and tell me which one costs you the audio. If it stays silent with all three off, that is worth knowing too: the 5909 is a Bluetooth headphone with USB as its secondary interface, and a device advertising non-mixable formats is not the same as a device that plays them.

During exclusive playback, does the progress bar advance while you hear nothing, or does playback not start at all? If you want, send the diagnostic log (Help > Export Diagnostic Log) and expand ALL PHYSICAL FORMATS in the DAC diagostic panel you already have open.

Playback Mode in the Tools menu is noted, reasonable.

The waveform is drawn at a fixed height, so full screen changes nothing. I'll think about a toggle option.
For a quieter player right now, the compress icon in the top bar switches to the Mini Player.

LEt me know,
thank you,
R.
Hello R,

I have tried as you suggested turning off all three settings switches and turning them on sequentially. I am puzzled at the result. Now when playing through the 5909 I hear sound with each successive switch being turned on and with all three on. The volume control does not work while playing, the volume is that which I set before selecting exclusive mode. The Volume widget in the menu bar still does jump from the 5909 setting to the Studio Display Speakers setting as it did before this test.

Prior to this test I had a crash of VV which stopped all audio, I saw a spinning beach ball of death over VV. Even after opened Spotify and tried to play from it did not play any sound. I had to Restart my Mac Studio. Then everything worked once again after restarting, I did not have to shutdown. Unfortunately I have not been able to reproduce the sequence that caused the crash. I have attached to Diagnostic Logs one without all physical formats expanded and the second expanded. Maybe this will shed some light on what happened.
Best, ASRid
 

Attachments

Hello R,

I have tried as you suggested turning off all three settings switches and turning them on sequentially. I am puzzled at the result. Now when playing through the 5909 I hear sound with each successive switch being turned on and with all three on. The volume control does not work while playing, the volume is that which I set before selecting exclusive mode. The Volume widget in the menu bar still does jump from the 5909 setting to the Studio Display Speakers setting as it did before this test.

Prior to this test I had a crash of VV which stopped all audio, I saw a spinning beach ball of death over VV. Even after opened Spotify and tried to play from it did not play any sound. I had to Restart my Mac Studio. Then everything worked once again after restarting, I did not have to shutdown. Unfortunately I have not been able to reproduce the sequence that caused the crash. I have attached to Diagnostic Logs one without all physical formats expanded and the second expanded. Maybe this will shed some light on what happened.
Best, ASRid

Hi ASRid,

logs are gold, thank you.

Max Bit Depth is cleared: your headphones negotiate the identical format with it on and off (24bit packed at 44.1, which is their ceiling there). The other two were never actually tested because DAC warmup only runs on the first exclusive track of a session, so every later track behaved as if it were off, and the sample-rate switch pause only exists when the rate really changes, while all your tracks were 44.1 on a device already at 44.1.

Expanding the physical format list does not reach the exported log, so your two files are the same session twice. A screenshot of that panel is what I actually needed (sorry, my fault not to have specified that). I'll add the list to the export.

The freeze was almost certainly not VV. An app that crashes quits, it does not take Spotify's sound with it. That is the macOS audio server getting stuck. Next time, before rebooting, try the command "sudo killall coreaudiod".

One request, whenever convenient.. in Exclusive Mode play a 44.1 track and then a 96, so the device has to change rate. That is the one case your test could not reach. And if the silence returns, export the log while it is happening, without quitting the app.

Thank you,
best,

R.
 
...HDCD encoding is detected and decoded, with dynamic range reported both as stored and as decoded.

R.
Good on ya for this!
Thou art indeed a stud.

from earlier in the thread
6. Fake hi-res. Not a binary label. File Forensics combines bandwidth collapse, brickwall filter detection, natural rolloff, noise floor, and bit-depth estimation into a contextual verdict.
Apologies if my 68yo grey matter has missed this: is your "fake hi-res" assessment
  • fake = source of claimed hi-res lossless was actually lossy; or
  • fake = source of claimed hi-res lossless was actually only 16/44 (or 16/48) lossless?
 
Good on ya for this!
Thou art indeed a stud.

from earlier in the thread

Apologies if my 68yo grey matter has missed this: is your "fake hi-res" assessment
  • fake = source of claimed hi-res lossless was actually lossy; or
  • fake = source of claimed hi-res lossless was actually only 16/44 (or 16/48) lossless?


As a fifty-one-year-old who is losing neurons because of the heat, I understood the second thing you said.
 
Good on ya for this!
Thou art indeed a stud.

from earlier in the thread

Apologies if my 68yo grey matter has missed this: is your "fake hi-res" assessment
  • fake = source of claimed hi-res lossless was actually lossy; or
  • fake = source of claimed hi-res lossless was actually only 16/44 (or 16/48) lossless?

Hi @emmodad, first of all thank you!

Regarding your question, to be precise, neither: audio analysis tool measures the actual spectral content of the file using FFT. It detects the true bandwidth (where the signal drops off) and the true bit depth (from the noise floor level). If the container claims hi-res but the content shows a bandwidth cutoff below 25 kHz or a 16-bit noise floor, the file is classified as Fake Hi-Res.
 
Hi Ricky,
I'm testing VeraVox with Naim DAC-V1. This old device has 2 exclusive modes:
16 and 24 bit.

1786449738101.png


When I play a 16 bit file, e.g.:

1786449673103.png


The format used is:

1786449820448.png


I have not set Max Bit Depth in settings. Would be possible forcing to deliver 16 bit only stream to the DAC instead of 24 bit, due to the fact that the DAC is capable of managing that?
I know it is a pure "cosmetic" feature with absolutely no benefit audio-wise, but e.g. Foobar 2000 has this feature in audio output.
This also lets the DAC to pass Naim embedded bitperfect tests in 16 bit mode.

Thanks
Stefano
 
Last edited:
Hi Ricky,
I'm testing VeraVox with Naim DAC-V1. This old device has 2 exclusive modes:
16 and 24 bit.

View attachment 550933

When I play a 16 bit file, e.g.:

View attachment 550932

The format used is:

View attachment 550934

I have not set Max Bit Depth in settings. Would be possible forcing to deliver 16 bit only stream to the DAC instead of 24 bit, due to the fact that the DAC is capable of managing that?
I know it is a pure "cosmetic" feature with absolutely no benefit audio-wise, but e.g. Foobar 2000 has this feature in audio output.
This also lets the DAC to pass Naim embedded bitperfect tests in 16 bit mode.

Thanks
Stefano

Hi Stefano,

you're right, your 16bit samples occupy the top bits of the 24bit words and the rest is zero, so the DAC converts exactly the same thing. The wider container is a deliberate default when a DAC offers both, not an oversight.

But it is not only cosmetic. The Naim test is one of the few ways an owner can check bit perfect delivery from the hardware side, and that counts for something. So I'll look at it properly. No timescale though, and I'd rather say that plainly than be vague.. it sits in the most delicate part of the playback engine, and it isn't near the top of the list at the moment.

One thing I can't establish from here is whether your DAC exposes a 16bit format in exclusive mode or only in shared. The bit depths line in the DAC panel counts both, so it doesn't settle it. If you expand "all physical formats" in that panel and send me a screenshot, I'll know straight away whether there's anything to aim at.

Thank you,
R.
 
Hi Stefano,

you're right, your 16bit samples occupy the top bits of the 24bit words and the rest is zero, so the DAC converts exactly the same thing. The wider container is a deliberate default when a DAC offers both, not an oversight.

But it is not only cosmetic. The Naim test is one of the few ways an owner can check bit perfect delivery from the hardware side, and that counts for something. So I'll look at it properly. No timescale though, and I'd rather say that plainly than be vague.. it sits in the most delicate part of the playback engine, and it isn't near the top of the list at the moment.

One thing I can't establish from here is whether your DAC exposes a 16bit format in exclusive mode or only in shared. The bit depths line in the DAC panel counts both, so it doesn't settle it. If you expand "all physical formats" in that panel and send me a screenshot, I'll know straight away whether there's anything to aim at.

Thank you,
R.
Is this one?

1786458971120.png
 

Thanks. your DAC does expose 16bit exclusive formats at every rate, 44.1 included.

One last thing and then I'll leave you alone. If you play a 16/44.1 track and export the diagnostic log (Help > Export Diagnostic Log), there's a line in it recording how the 24bit format is actually laid out on the wire. That single detail decides whether this is a modest change or a delicate one. The panel doesn't show it, which is a gap on my side.

R.
 
Back
Top Bottom