Announce: Enhanced Digital Output app - USB Dac and 192k Digital Ouput

[Triode]

Is it a possibility at anytime in the future that the app will offer support in 24/176/192 for AIFF files?

AIFF should be supported with 7.8 now. There is a bug in the server with 24 bit files though which means they will sound like white noise - I believe this is easily fixed with a small change to the server code. I added a patch to the bug for it somewhere - will look for a reference to it.

As you want AIFF are you using a mac? In this case you should be able to make the modification to the server.

http://bugs.slimdevices.com/show_bug.cgi?id=17163 - post 6 shows a small addition you can make to the AIFF.pm file to change the offset of the audio data by 8 bytes which should make 24 bit files work once you rescan them.

[SBGK]

Absolutely.

There are - out of my memory -

1. Output routing to hw interface
2. Passing by asound.conf routing table
3. 2nd Jive_alsa killing
4. Buffer size modifications
5. Screen totally Off
6. Increase key process priorities

(All this is no rocket science btw. - I do claim I'm the first who put all those tweaks together as package into the context of improving efficiency of the SBT)

The measures I've applied, I obviously put into a slightly different context, which belongs to a different area of this forum.
Triode would of course never admit that he also appreciates associated "side-effects". He intends to survive a bit longer
over here, so better don't touch that hot potato.
I'm pretty sure EDO users are appreciating those side-effects.

Just to mention it. EDO might even work without above listed measures, since the main issues seem to be driver and jive related. I'll stop here...

However.

What's common about the measures is that they intented to increase the efficiency of a computer system. They'd work on any Linux/Android system.


Triode had serious problems to get rid of XRUNS in the early days. Even all my system optimizations wouldn't
stop certain DACs to run into XRUNS. This led him into tweaking the Linux audio drivers.

Just to mention it: USB audio has always been working on the Touch.


Triode put an amazing effort into the subject and has shown a great level of skills to make it all available by putting it into the LMS/Squeezeplay environment. Great job.

I do limit my appreciation to the Squeezeplay/LMS/driver part of the story...

What a week, can it get any better ?

First NASA lands Curiosity on Mars and now Soundcheck gives us the benefit of his knowledge.

If Soundcheck brought out a version of EDO would it be called EGO ?

[Flying Ed]

AIFF should be supported with 7.8 now. There is a bug in the server with 24 bit files though which means they will sound like white noise - I believe this is easily fixed with a small change to the server code. I added a patch to the bug for it somewhere - will look for a reference to it.

As you want AIFF are you using a mac? In this case you should be able to make the modification to the server.

http://bugs.slimdevices.com/show_bug.cgi?id=17163 - post 6 shows a small addition you can make to the AIFF.pm file to change the offset of the audio data by 8 bytes which should make 24 bit files work once you rescan them.

Thanks Triode!.
Yes i am a Mac user.
Thanks for the info i will now try and sort it.
Cheers.
F.E.

[alcarp]

Was wondering if the problem with EDO & USB output to this DAC has been resolved.

[Triode]

Was wondering if the problem with EDO & USB output to this DAC has been resolved.

I don't believe anyone has tried recently. Have there been any firmware updates or has NAD claimed linux support? You could try kernel #9 from the expedients thread if you have access to one, but I have yet to see any debugging of why this is not working..

[guidof]

I run EDO+TT+SBGK. The Touch's USB output is routed to a DSPeaker Dual Core EQ/DAC via a Music Fudelity VLink II USB to SPDIP converter. I noticed that when playing a CD rip, the display of the Dual Core reads 24/44 rather than 16/44 which is the actual resolution of the rip.

According to the VLink II manual, this device does not upsample.

The Dual Core appears to read all other input resolutions normally (24/88, 24/96, etc.).

Any thoughts? Triode?

Regards,

Guido F.

[Triode]

I run EDO+TT+SBGK. The Touch's USB output is routed to a DSPeaker Dual Core EQ/DAC via a Music Fudelity VLink II USB to SPDIP converter. I noticed that when playing a CD rip, the display of the Dual Core reads 24/44 rather than 16/44 which is the actual resolution of the rip.

According to the VLink II manual, this device does not upsample.

The Dual Core appears to read all other input resolutions normally (24/88, 24/96, etc.).

Any thoughts? Triode?

Regards,

Guido F.

The internal processing of usb audio is likely to be 24 bits for all cases and will fill the lower 8 bits with 0s. However if you have replay gain enabled or the volume set to less than 100% then the bits may be altered by the gain setting and this will alter all 24 bits. Some Dacs such as the Audiolab MDAC monitor the number of bits which are changing and use this for the bit depth - people notice that the dac reads 44/16 at 100% volume and then it changes to 44/24 if you change the volume. Is this the case for you, or perhaps the dac is just measuring that there are 24bits per sample being received.

[guidof]

The internal processing of usb audio is likely to be 24 bits for all cases and will fill the lower 8 bits with 0s. However if you have replay gain enabled or the volume set to less than 100% then the bits may be altered by the gain setting and this will alter all 24 bits. Some Dacs such as the Audiolab MDAC monitor the number of bits which are changing and use this for the bit depth - people notice that the dac reads 44/16 at 100% volume and then it changes to 44/24 if you change the volume. Is this the case for you, or perhaps the dac is just measuring that there are 24bits per sample being received.

Hi Triode, thank you very much for your explanation. I don't have replay gain enabled, but I do control the volume at the Touch rather than at the preamp (surprisingly to me, this sounds more transparent -- i run both the Dual Core and the preamp with no attenuation). I'll try reverting to using the preamp volume as a test.

Guido F.

[guidof]

The internal processing of usb audio is likely to be 24 bits for all cases and will fill the lower 8 bits with 0s. However if you have replay gain enabled or the volume set to less than 100% then the bits may be altered by the gain setting and this will alter all 24 bits. Some Dacs such as the Audiolab MDAC monitor the number of bits which are changing and use this for the bit depth - people notice that the dac reads 44/16 at 100% volume and then it changes to 44/24 if you change the volume. Is this the case for you, or perhaps the dac is just measuring that there are 24bits per sample being received.

OK Triode. Just ran the Touch at full volume, and sure enough, it read 44/16. Throttled down to 90% and it read 44/24.

Thanks again.

Guido F.

[Mnyb]

afiak a squeezebox is always 24bit but with the bottom 8 bits as zeros when playing CD resolution at 100% volume ( this will then change with volume adjustment and gain tags ) .