advertisement


MDAC First Listen (part 00100110)

Status
Not open for further replies.
I have done some further testing with the AirMusic device...

- On a Windows machine, the only free way to make it work is to use iTune
- I also got it to work with Foobar and MediaMonkey with an 'airplay' plug-in (see the emilles link above... making sure you also look at the readme! file as installation requires a few steps). Note that, with MediaMonkey, the AirMusic is managed as a separate audio device, which is fairly straightforward. With Foobar, it is managed as a DSP, which means that one can use it in parallel with one's favourite output (WASAPI, ASIO, our a NULL output)... which has some benefits as there are ways to mute each of the outputs individually in practice.
- Airfoil and Tunerblade also work with generic windows application (they basically reroute the windows sound to Airplay compatible devices) and offer a few more options... but they don't integrate well with applications which output audio using WASAPI or ASIO drivers.
- There is a lag in all these applications, which creates challenges with video playback (Tunerblade has a way around it with VLC, and I'm sure one can find ways to delay videos through other routes if required... although getting this to work seamlessly and reliably is challenging)

- Reliability is not that great with any of these applications (save for iTune, which lacks FLAC support unfortunately)... expect a few slow downs on the computer when the apps are working!

- As an aside, I haven't been able to get it to pass the bit-perfect test of the MDAC (under Android or Windows)

So far I haven't been able to get it to be seen by any windows machine as a uPNP/DLNA renderer (my PiCorePlayer and some other windows machines on my home network are visible with at least some applications), this would be the holy grail (with bit perfect, that is)

All in all, I can see that some people may like this 'Wireless playback' feature (especially with apple software or some separate android player). On a PC without installing iTunes, it is not as good a solution (even though it "works"), unless one can find a way to get uPNP to work (which might be possible as it works under Android).
 
EIffel said:
...
Click to expand...
Since AirPlay supports audio/video, did you manage to stream video to an apple-compatible TV (if you have one) and audio to the airmusic device? That could tell us whether the delay is inherent to the technology used (and intentional) or just a "feature" of this device.
 
jirij,
I don't have the gear to test this (nothing else airplay compatible), but I understand a delay is normal with all airplay devices (seemingly by design). This came to me as a surprise as my squeezebox gear plays instantly (and works great... obsolete technology according... to some even though it supports async HD audio in a bit perfect way ;-) )
 
EIffel said:
jirij,
I don't have the gear to test this (nothing else airplay compatible), but I understand a delay is normal with all airplay devices (seemingly by design). This came to me as a surprise as my squeezebox gear plays instantly (and works great... obsolete technology according... to some even though it supports async HD audio in a bit perfect way ;-) )
Click to expand...
Agree with you about sqeezebox.

But the SB Touch is showing its age when you play 192khz and DSD files which is not instant. But I think it is because of the slow CPU.
I have just got a Rasphberry Pi 2 and now it plays the high bit rate files instantly.
piCorePlayer seem to be the best(easiest) way of running squeezelite.
All I had to do was write image to SD card and then setup via a web interface. I would say it takes less than <30 minutes to be up and running.
 
piCorePlayer on the Rasberry Pi 2 is indeed a very good solution (probably a bit overkill), it also works as a uPNP streamer, and can be easily upgraded (if one wishes to... although is safer to do with easy access to the micro sd card). Spares will be available for a long time one way or another.

A Pi2 does cost a bit more (after adding a cheap micro sd card and a wifi device) and takes more space / generates bit more heat.

While having a Pi integrated would have some benefits (no need for a power supply, or to tie up the only external USB connection on the MDAC, possibly direct Ethernet connection), I do see the merit of the AirMusic device (smaller footprint, slightly easier initial setup, native apple support)
 
Maybe we shoud consider the Arietta G25 (25€) + WiFi Module (7€) too. It is an ultra small Linux based SoC with all the necessary ports. And on top it would give us the freedom to freely configure it, hopefully to the degree, that it could stream bit-perfect audio over WiFi.

http://www.acmesystems.it/arietta

Even Sorftware distribution could be done easily throu a MicroSD card, which costes peanuts. Once a stable and satisfying software configuration has been found, it coud be loaded to the SD card and inserted into the board. (It would be even neater, if that SD card could be accessed from the rear panel of the MDAC2)

wifi_banner.jpg
 
Even more possibilities come to mind, when considering that it can communicate using i2c. For example a remote app for tablets and smartphones. This could eliminate the need for buying the remotes, for which John has to place an order of at least 1000pcs. Given everyone has a smartphone. ;)
 
NuTron said:
Even more possibilities come to mind, when considering that it can communicate using i2c. For example a remote app for tablets and smartphones. This could eliminate the need for buying the remotes, for which John has to place an order of at least 1000pcs. Given everyone has a smartphone. ;)
Click to expand...
It is a shame it only has a single core 400Mhz CPU a dual core would be preferable.

And there has to be existing apps for smartphones/tablet that can be used and SD card images like piCorePlayer that works with it.
Can't imagine John want to get involved in the software for a streamer.
It should be our own responsibility.
 
As I understand, the MDAC2 is a ever evolving project software wise. So this would be the most flexible solution for WiFi connectifity I can think of. The Idea of sitting on my listening spot and tune the MDAC2s DSP with my iPad one day, realy is appealing to me.
 
sam_cat said:
Why not the original thoughts of the MicroSOM board from CuBox?

http://www.solid-run.com/products/micro-som/

I have the CuBox i4 at home and its a great bit of kit. :)
Click to expand...
Have you got that running LMS and squeezelite stably? I got one to play with in anticipation of it being part of the project and I found the squeeze on arch distribution very frustrating. It seemed to be a mixture of problems with wifi and difficulties getting lms to update in arch. IIRC LMS for debian wouldn't run
on the cuboxi last time I tried. It seems to me that perhaps the pi2 is a better idea because it has a huge user base and already seems to work with a fairly user friendly installation under max2play
http://forums.slimdevices.com/showt...ients-(Audioplayer-Squeezeserver-Airplay-XBMC)
 
No, I run a dedicated PC in living room as a media box, Plex, Kodi, netflix, prime streaming and ambibox
 
fusion5 said:
Why not keep it simple?
Click to expand...
I believe that was John's intention - to have something optional, extra, on top of MDAC2 (on top of L2/L3, which are already extensions of the original MDAC2 idea) for convenience in some use cases. This was maybe misunderstood by multiple people, incl. me and John - I was trying to help the idea become more useful by making it more general, which was seen as "complicating" it, and not worth for the low module price and (already) delayed time scales. The problem is that "works for my iphone" is maybe too simple.

So my (I'm not John) take on this is - as long as we, the community, can provide a plug-and-play default SD card OS for whatever board fits the size / price / power requirements that would behave similarly to the proprietary Atheros solution (first boot = default wifi SSID with basic user-friendly configuration, not command line stuff), John can wire the board up on the digital PCB.
The "more advanced" people can then modify the OS however they want or install their own, the point is that the default solution doesn't require this. No additional MDAC-specific stuff promised - eg. no volume/interface control, no iPhone/Android custom app, no custom kernel driver for DSP, nothing beyond the basic playback that the Atheros thing would do.

That would keep it simple (hopefully) for John / Dominik while being easily extensible in the future (software-wise, by this community).

The problems are
  • who is going to do the software work? can we trust the person to fulfill the promise and not give up in the process? .. is there already existing distro for such task, ready to be used? :)
  • wouldn't the price for the module be higher? the price of the microsd card?
  • possibly more complex interface on the digital PCB (SPDIF vs I2S) - maybe slightly more complex power requirements? .. size? .. mounting?

(Despite all this, I still don't have a use for it, so I don't want it, even though it sounds cool. :) )
 
jirij said:
I believe that was John's intention - to have something optional, extra, on top of MDAC2 (on top of L2/L3, which are already extensions of the original MDAC2 idea) for convenience in some use cases. This was maybe misunderstood by multiple people, incl. me and John - I was trying to help the idea become more useful by making it more general, which was seen as "complicating" it, and not worth for the low module price and (already) delayed time scales. The problem is that "works for my iphone" is maybe too simple.

So my (I'm not John) take on this is - as long as we, the community, can provide a plug-and-play default SD card OS for whatever board fits the size / price / power requirements that would behave similarly to the proprietary Atheros solution (first boot = default wifi SSID with basic user-friendly configuration, not command line stuff), John can wire the board up on the digital PCB.
The "more advanced" people can then modify the OS however they want or install their own, the point is that the default solution doesn't require this. No additional MDAC-specific stuff promised - eg. no volume/interface control, no iPhone/Android custom app, no custom kernel driver for DSP, nothing beyond the basic playback that the Atheros thing would do.

That would keep it simple (hopefully) for John / Dominik while being easily extensible in the future (software-wise, by this community).

The problems are
  • who is going to do the software work? can we trust the person to fulfill the promise and not give up in the process? .. is there already existing distro for such task, ready to be used? :)
  • wouldn't the price for the module be higher? the price of the microsd card?
  • possibly more complex interface on the digital PCB (SPDIF vs I2S) - maybe slightly more complex power requirements? .. size? .. mounting?

(Despite all this, I still don't have a use for it, so I don't want it, even though it sounds cool. :) )
Click to expand...

More or less what I had thought and of course with more options there can be more problems. I'm not asking John or Dominik to do any programming or such, but the metioned board can do what the Atheros board can. Plus the freedom for upgrading or configure it as everyone wishes and not to rely on a ventor. That's all. If John desides to make some functions (title info on display, set filters, selecting inputs, dsp config, aso) available throu i2c or an other way, is up to him alone! The digital pcb is not designed yet, so there is still some space to give some ceative input.
 
I may be late to the project, but it does seem to be going the way of Nimrod. I too would like to keep it simple, with ultimate SQ and ability to play / upgradeable to play all formats available. I know it would be easier for John to produce just one digital board (especially on such a small run) But I would not like to think that SQ could be compromised in any way.
 
I've got my Fusion MDAC running naked. Sounds brilliant, best upgrade ever. Only need to build a box now. As has been said before the Audiolab box is precision engineering and not so easy to replicate for old hackers with blunt saw :(:eek:
 
Status
Not open for further replies.


advertisement


Back
Top