Featured Articles
Announcing Pi
Technical Specifications
Raspberry Pi FAQs
Win a Pi
Raspberry Pi Wishlist
Actions
Forum Thread Details
  • Replies 4 replies
  • Subscribers 665 subscribers
  • Views 822 views
  • Users 0 members are here
Related

Report of DSP audio quality problem

morgaine

This is the first I've heard about this DSP issue, but it appears to be researched technically and evidence is presented:

 

"Can't get good sound on RPi. Power problems" -- http://slashdot.org/comments.pl?sid=3070945&cid=41116889

 

"I bought two Raspberry Pi(es) to use as audio servers and have been disappointed by the sound quality.  The on-board audio out's DSP has limited bandwidth so sound is down-sampled to 11 bits.  Scratchy.  It's not advertised so that was a let-down.

 

Using a USB AUDIO dongle is no-go either, because of the crappy USB drivers. Stutters non-stop. Here are oscilloscope grabs of two music samples and a 1Khz tone: http://imgur.com/a/rVR99 . The flat parts shouldn't be there. The only way to get good sound now is to use rather expensive USB soundboards or the HDMI output, but extracting line-level audio signals from that isn't a simple or cheap proposition."

 

 

I find it hard to believe that it's appeared first on Slashdot and not on the RPF forums.  Have any of you come across it before?

 

Morgaine.

  • I think this active thread is related:

     

    http://www.raspberrypi.org/phpBB3/viewtopic.php?f=28&t=14199

     

    p.s.

      completely off topic, but SCO was converted from Chapter 11 to

    Chapter 7 today, with a new interim trustee to be appointed.

  • in reply to Former Member

    Yes, thanks, the post at the bottom of the first page of the RPF thread you linked (by odokemono, Thu Aug 16, 2012 6:15 pm) appears to be the same or equivalent to what was submitted to Slashdot, as it contains the same oscilloscope images.

     

    For reference the RPF forum thread is at Troubleshooting->"How to get high quality audio from R-Pi?"

     

    There are at least 3 different issues mixed up in that user's post:  analogue audio, USB issues, and power.

     

    I don't think anyone expected very high quality on the 3.5mm connector, but 11 bits of depth is pretty aweful even for analogue in 2012, not even as much as FM radio, and with virtually no dynamic range headroom.  I guess it may suffice for some genres of highly compressed pop, but that's about it.  16, 20 and 24-bit depths are the norm today.  11 bits is ... unusual.

     

    His other audio problem (stutter) may be resolved if USB is resolved, but that remains to be seen.

     

    The only other audio output is through HDMI, and since that is typically plugged directly into a display, in effect you have no means of tapping the high quality audio directly.  This may be adequate for A/V applications if the display provides digital audio out, but if it doesn't then quality audio is just not available.  And for non-video applications, the HDMI output is more a liability than an asset.

  • in reply to Former Member

    Wow SCO the Linux that would not die. Or should that be Unix. I can't seriously believe that anyone would risk even a single $ in that company unless they have a huge "idden patent portfolio. image

     

    My sound is fine through the HDMI. But I have yet to try any other methods. Are we saying that ALL Pi's have these problems or that some have a manufacturing issue that make the sound design probs worse? image

     

    Ray

     

    Perhaps I should state that I am almost totally deaf so Pi-Fi sound is not that important to me. But I can hear some audio with hearing aid in one ear. (I usually aim the other ear at the wife) image

  • in reply to Former Member

    It's a problem with all the Pi's. They simply use a pwm output to create the sound. The signal runs on a 100 Mhz clock. So, at 48 kHz this gives you roughly the equivalent of an 11 bit DA converter. Someone on the Rpi forum also discovered that the clamping diodes in the audio output are placed behind the dc decoupling capacitors instead of before them. As long as the signal stays below 0.5V, that shouldn't be responsable for the distortions.