Audio Dandy - How to enjoy High Resolution Audio
It provided a good laugh, if nothing else. I find it incredible you would argue with phofman about how computer-based audio works from a position of total ignorance of the subject.
It provided a good laugh, if nothing else. I find it incredible you would argue with phofman about how computer-based audio works from a position of total ignorance of the subject.
Very true. Then how can change the "audible quality" if it supplies the very same samples and timing is controlled/provided by the soundcard clock?
I think it is because the timing is very important, in not all designs the right solution has been choosen, and then re-clocking or other things will happen
I cannot explain what happend technical in Daphile RT, but I can confirm that the RT version sounds very different then thr normal version
And that is what I do, no voodoo talk, just share my personal experiences
I experiment, test, en listen with my own ears, my reviews are 100% subjective
because they are my personal experiences, I share them because I think that's the way to learn from each other, and to get faster to better quality
thatwe all can benefit
Please try Daphile RT yourself, and let us know what it did for you
P.S: you can call the "Amanero Combo384" an external USB "soundcard" I f you like, I have one ofthe earlier versions, personaly I think the best profit can be made by optimalizing this part (better clock, newer firmware)
This "soundcard" isvery important for the rest of the audio chain,
luckily Daphile can work with it, even when you disable "HD Audio" in the bios of the Daphile computer, by using UAC2 straight from the kernel Daphile can see the external soundcard and pass all audio to it, without touching or change the actual music data
I think it is because the timing is very important, in not all designs the right solution has been choosen, and then re-clocking or other things will happen
I cannot explain what happend technical in Daphile RT, but I can confirm that the RT version sounds very different then thr normal version
And that is what I do, no voodoo talk, just share my personal experiences
I experiment, test, en listen with my own ears, my reviews are 100% subjective
because they are my personal experiences, I share them because I think that's the way to learn from each other, and to get faster to better quality
thatwe all can benefit
Please try Daphile RT yourself, and let us know what it did for you
P.S: you can call the "Amanero Combo384" an external USB "soundcard" I f you like, I have one ofthe earlier versions, personaly I think the best profit can be made by optimalizing this part (better clock, newer firmware)
This "soundcard" isvery important for the rest of the audio chain,
luckily Daphile can work with it, even when you disable "HD Audio" in the bios of the Daphile computer, by using UAC2 straight from the kernel Daphile can see the external soundcard and pass all audio to it, without touching or change the actual music data
Last edited:
Daphile is a regular gentoo linux distribution with some RT patches applied, installed in several layers on your drive (typical for linux appliances). I do not know which RT patches the distribution author selected but he certainly did not write any. Plus it has a custom-made user layer with web interface.
I cannot imagine daphile author would put deliberately some DSP stuff into RT/nonRT to make them sound different.
I do not like the fact that daphile author does not give any credentials to linux, he does not mention it on his webpage, in the installation manual etc. Even though only the web interface is his own in daphile. But OK, that is very likely legal.
Right - timing is the key. Did you change your hardware between tests of daphile and daphile RT? If not, how could timing be different? Software has no effect on timing in audio chain, I have been repeating this for years. Yet people still believe the CPU is somehow timing the samples. Nonsense.
Sorry, but subjective tests have no value to the community. Everyone feels it differently. You had problems with RT installation, eventually you succeeded. Great! So it must sound great, especially when some other people say so. So you hear it. Expectation bias, scientifically proven Psychlopedia - Expectancy Bias
IF you did a blind ABX test on the same hardware and could statistically tell you can distinguish between daphile and daphile RT (i.e. 9/10 answers correct), it would be a sign something is really different - you cannot fake correct guess if you brain does not know the correct answer. But you will not perform the ABX test, nobody claiming to clearly hear improbable phenomenon has ever reported to perform an ABX test, never in the many years I have been following the community. Neither will you.
I am using linux alsa terminology to avoid confusion when troubleshooting over internet problems of other users. Card -> device -> subdevice. If I used the word "device", it could be confused with the term device in alsa meaning - a card can offer several devices (typically analog output/spdif or stereo output/multichannel)
But it is just terminology.
No wonders. HD Audio is the built-in soundcard on your motherboard. Your USB audio device is another soundcard you attach to your computer. Daphile supports USB Audio Class 2 protocol because many people worked hard to write correctly functioning linux drivers for the standard as well as countless changes/hacks fixing implementation errors by various manufacturers History for sound/usb - torvalds/linux * GitHub
And the Daphile author does not even give a single word of recognition.
I cannot imagine daphile author would put deliberately some DSP stuff into RT/nonRT to make them sound different.
I do not like the fact that daphile author does not give any credentials to linux, he does not mention it on his webpage, in the installation manual etc. Even though only the web interface is his own in daphile. But OK, that is very likely legal.
Right - timing is the key. Did you change your hardware between tests of daphile and daphile RT? If not, how could timing be different? Software has no effect on timing in audio chain, I have been repeating this for years. Yet people still believe the CPU is somehow timing the samples. Nonsense.
Sorry, but subjective tests have no value to the community. Everyone feels it differently. You had problems with RT installation, eventually you succeeded. Great! So it must sound great, especially when some other people say so. So you hear it. Expectation bias, scientifically proven Psychlopedia - Expectancy Bias
IF you did a blind ABX test on the same hardware and could statistically tell you can distinguish between daphile and daphile RT (i.e. 9/10 answers correct), it would be a sign something is really different - you cannot fake correct guess if you brain does not know the correct answer. But you will not perform the ABX test, nobody claiming to clearly hear improbable phenomenon has ever reported to perform an ABX test, never in the many years I have been following the community. Neither will you.
I am using linux alsa terminology to avoid confusion when troubleshooting over internet problems of other users. Card -> device -> subdevice. If I used the word "device", it could be confused with the term device in alsa meaning - a card can offer several devices (typically analog output/spdif or stereo output/multichannel)
But it is just terminology.
No wonders. HD Audio is the built-in soundcard on your motherboard. Your USB audio device is another soundcard you attach to your computer. Daphile supports USB Audio Class 2 protocol because many people worked hard to write correctly functioning linux drivers for the standard as well as countless changes/hacks fixing implementation errors by various manufacturers History for sound/usb - torvalds/linux * GitHub
And the Daphile author does not even give a single word of recognition.
Last edited:
BTW these are recent commits supporting DSD on your USB soundcard:
ALSA: usb-audio: Support both DSD LE/BE Amanero firmware versions * torvalds/linux@3eff682 * GitHub
ALSA: usb-audio: add DSD support for new Amanero PID * torvalds/linux@ed993c6 * GitHub
Sent to the alsa-devel mailing list [alsa-devel] [PATCH] Support both DSD LE/BE Amanero firmware versions by the finnish author of HQ Player Jussi Laako (@Signalyst) | Twitter
You can expect Daphile to support DSD on the new firmware of your gear in coming versions (perhaps already in the latest kernel). You can send Jussi a thank-you note 🙂
Another guy fixed your DAC support for big-endian machines (older PowerPC - macs) ALSA: usb-audio: fix Amanero Combo384 quirk on big-endian hosts * torvalds/linux@f83914f * GitHub
ALSA: usb-audio: Support both DSD LE/BE Amanero firmware versions * torvalds/linux@3eff682 * GitHub
ALSA: usb-audio: add DSD support for new Amanero PID * torvalds/linux@ed993c6 * GitHub
Sent to the alsa-devel mailing list [alsa-devel] [PATCH] Support both DSD LE/BE Amanero firmware versions by the finnish author of HQ Player Jussi Laako (@Signalyst) | Twitter
You can expect Daphile to support DSD on the new firmware of your gear in coming versions (perhaps already in the latest kernel). You can send Jussi a thank-you note 🙂
Another guy fixed your DAC support for big-endian machines (older PowerPC - macs) ALSA: usb-audio: fix Amanero Combo384 quirk on big-endian hosts * torvalds/linux@f83914f * GitHub
Thank you Phofman for the links,
I noticed that on the site of AudioLinux they also published new updates:
Index of /ftp/packages/kernel/last
I have no experience with updating the Amanero board, so I sent a email to my
retailer where I bought my DAC if could help me with this, I don't want to brick my device 🙂
What still I don't understand; for example if with this new firmware Amanero can support DSD512 Native, what about the other side? the FPGA in my DAC and the BurrBrown chips don't go so high I believe (32bit 192 KHz max) thr later version of the Master 7 DAc's have faster FPGA's with custom firmware that support 384 KHz)
I also think I need a special flash device/ćable for loading the new firmware in my Amanero?
I noticed that on the site of AudioLinux they also published new updates:
Index of /ftp/packages/kernel/last
I have no experience with updating the Amanero board, so I sent a email to my
retailer where I bought my DAC if could help me with this, I don't want to brick my device 🙂
What still I don't understand; for example if with this new firmware Amanero can support DSD512 Native, what about the other side? the FPGA in my DAC and the BurrBrown chips don't go so high I believe (32bit 192 KHz max) thr later version of the Master 7 DAc's have faster FPGA's with custom firmware that support 384 KHz)
I also think I need a special flash device/ćable for loading the new firmware in my Amanero?
I am afraid I cannot help you with your USB DAC, I do not know anything about it. Please talk to the manufacturer who has received your money for the device.
But the patches look like the existing firmware already supports DSD, that new-firmware patch adds DSD support to alsa for the new firmware which apparently changes the USB Product ID (PID) of your device upon upgrading from 0x071a to 0x0a23. I do not think you have to upgrade your firmware for DSD support.
Thanks,
When I received an answer of my manufacturer I will post it on this topic
It's good to see that the open source community is very bussy with DSD audio related projects.
I'm preparing to install AudioLinux on my passive cooled PC, to test it, so I can compaire Daphile RT, with OpenLinux, after this I will buy the JRiver Id image
so I can install it on my Celeron NUC, it's based on Debian I believe, and it turns my NUC into a JRiver (SACD.ISO playback!)headless streaming solution! exciting!
When I received an answer of my manufacturer I will post it on this topic
It's good to see that the open source community is very bussy with DSD audio related projects.
I'm preparing to install AudioLinux on my passive cooled PC, to test it, so I can compaire Daphile RT, with OpenLinux, after this I will buy the JRiver Id image
so I can install it on my Celeron NUC, it's based on Debian I believe, and it turns my NUC into a JRiver (SACD.ISO playback!)headless streaming solution! exciting!
Just to remind - neither daphile nor jriver ID are open source, only based on open source foundations...
Daphile is free..
I don't think that many people worked hard to write correctly functioning linux drivers worked that for free..
All this talk about RT kernels and whatnot caused me to install the install the 64-rt version this morning, which was a relatively painless process. After configuring my network I listened to a single album from a USB stick and it sounded great. I then went on to load all my personal apps.
A little later I loaded several gigs of music (on the same USB) stick into File Manager, and they show up just like they normally would. But, when I try to play them through Audio Player they won't play. All the artwork, song length etc. shows as normal, but none of the music will play. I've tried reloading several times from formatted stick, uploading through wifi.. but none of them will play in audio player. I can listen to the music by file mgr preview over my UI. All my other apps play music like normal. Not sure how to progress from here. Any ideas?
A little later I loaded several gigs of music (on the same USB) stick into File Manager, and they show up just like they normally would. But, when I try to play them through Audio Player they won't play. All the artwork, song length etc. shows as normal, but none of the music will play. I've tried reloading several times from formatted stick, uploading through wifi.. but none of them will play in audio player. I can listen to the music by file mgr preview over my UI. All my other apps play music like normal. Not sure how to progress from here. Any ideas?
nevermind... working with a reinstall/reconfig. If I had a nickel for how many times I've done that in the last few weeks working up from the 32bit.. 64bit... 64bit rt. 
ok.. time to just enjoy.
ok.. time to just enjoy.
..and back to not playing. "inappropriate ioctl for device" error. But not all songs, a scant handful of songs do play without issue.
Do you experience the problems with non-RT kernel too?
That "inappropriate ioctl for device" message is usually prefixed by name of the failed function with colon, e.g. "tcgetattr: Inappropriate ioctl for device". The name is quite important for troubleshooting.
That "inappropriate ioctl for device" message is usually prefixed by name of the failed function with colon, e.g. "tcgetattr: Inappropriate ioctl for device". The name is quite important for troubleshooting.
This is the first instance of this/these errors between use of 32, 64 and 64-rt.
all look like this:
[17-09-03 12:18:29.5079] Slim:😛layer::Song:😱pen (650) Error: Can't open [file:///srv/mediaserver/music/USB%20Drives/C8B5-2B15/Maneesh%20de%20Moor/Maneesh_De_Moor,Bahramji_-_02_-_Watch_The_Cat.mp3] : Inappropriate ioctl for device
all look like this:
[17-09-03 12:18:29.5079] Slim:😛layer::Song:😱pen (650) Error: Can't open [file:///srv/mediaserver/music/USB%20Drives/C8B5-2B15/Maneesh%20de%20Moor/Maneesh_De_Moor,Bahramji_-_02_-_Watch_The_Cat.mp3] : Inappropriate ioctl for device
[17-09-03 13:13:48.9531] main::main (205) Starting Logitech Media Server scanner (v7.9.1, git55045f0b3, Daphile 17.05-x86_64-rt) perl 5.024001
[17-09-03 13:13:49.4794] Slim::Schema::forceCommit (2149) Warning: Trying to commit transactions before DB is initialized!
[17-09-03 13:13:49.8282] Slim::Music::Import::runImporter (511) Starting Slim::Media::MediaFolderScan scan
[17-09-03 13:13:49.8294] Slim::Utils::Scanner::Local::rescan (181) Discovering audio files in /srv/mediaserver/music
is bolded significant?
[17-09-03 13:13:49.4794] Slim::Schema::forceCommit (2149) Warning: Trying to commit transactions before DB is initialized!
[17-09-03 13:13:49.8282] Slim::Music::Import::runImporter (511) Starting Slim::Media::MediaFolderScan scan
[17-09-03 13:13:49.8294] Slim::Utils::Scanner::Local::rescan (181) Discovering audio files in /srv/mediaserver/music
is bolded significant?
The slim player tries to open a file on some USB drive which fails. Are you sure the USB drive and its filesystem is OK, (including its power supply if it is a regular hard drive)?
The USB is just a 16gb stick. The same issue appears on files loaded into the 160gb sata internal drive in my T610.
[17-09-03 13:29:41.0081] Slim:😛layer::Song:😱pen (650) Error: Can't open [file:///srv/mediaserver/music/Internal%20Drives/Music/Bob_Dylan_-_The_Bootleg_Series_Vol._5__Live_1975_-_The_Rolling_Thunder_Revue__(2002)/Bob_Dylan_-_01_-_It%27s_All_Over_Now,_Baby_Blue.mp3] : Inappropriate ioctl for device
no pun intended..
[17-09-03 13:29:41.0081] Slim:😛layer::Song:😱pen (650) Error: Can't open [file:///srv/mediaserver/music/Internal%20Drives/Music/Bob_Dylan_-_The_Bootleg_Series_Vol._5__Live_1975_-_The_Rolling_Thunder_Revue__(2002)/Bob_Dylan_-_01_-_It%27s_All_Over_Now,_Baby_Blue.mp3] : Inappropriate ioctl for device
no pun intended..
I could not google the error.
On my machine I would open the perl script in debugger, insert a breakpoint, and look at what is going on. The error occurs somewhere before slimserver/Song.pm at public/7.9 * Logitech/slimserver * GitHub . I am afraid I cannot help you remotely.
On my machine I would open the perl script in debugger, insert a breakpoint, and look at what is going on. The error occurs somewhere before slimserver/Song.pm at public/7.9 * Logitech/slimserver * GitHub . I am afraid I cannot help you remotely.