Skip to contentSkip to conversation

rme 9632 crackling when sound > 0db

DISCUSSION Ed Wildgoose 2 participants 4 messages Apr 12, 2004 – Apr 13, 2004

lore
Share & tools
lore
#0 lore raw

Hi, I'm running kernel 2.6.5-mm1 with built-in alsa drivers. Also, v 1.04 of the various alsa utils, etc.

I'm playing (for example) Dido - No Angel, track "Isobel", and according to the hdspmixer tool, at the end of the track the drums have transients that go greater than 0db, and this causes nasty metallic crackles on my RME 9632 (tested with both Alsa and OSS interfaces). Without looking at the code, I'm wondering if there is an overflow occuring, ie something unsigned going into a signed value and causing this... However, it doesn't seem that likely, for example if you turn up the middle layer of mixer controls and then turn down the last layer of output mixer controls then as long as the red clip light isn't on then everything is still ok, ie the crackle isn't occuring as a result of the first layer of software mixing, it purely seems to be something happening *after* the last mixer control

You can also simulate this simply by turning up the mixer controls so that normal music starts to clip. Same effect.

Two points to note, first I don't remember getting this on an alsa snapshot from about 1-2 months ago, so I'm thinking it's a driver issue, rather than a feature of this card? Secondly, before now I couldn't use the card at all with many apps, I would just get constant over-runs and other problems. So thankyou for fixing those problems!

Just a heads up, this card still doesn't work with mplayer 1.0 pre3. The sound appears to be playing too slowly, and several times a second there are "xruns" reported (which makes sense). I will try and rip out the mplayer audio code and make up a test app so that I can experiment further with this.

Thanks to everyone who has made Alsa what it is today!

Ed W

-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click

#1 lore raw
Quoted 3 lines

You can also simulate this simply by turning up the mixer controls so that normal music starts to clip. Same effect.

At risk of creating noise. Further experimentation shows that you need to set the output levels to +4db in hdspconf for this crackling to occur. So in fairness this could really be an overdriven output...?

The -10dbv output level is significantly quieter than any other equipment that I own (I know it's what they call consumer output device level, but still...)

Ed W

-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click

#2 lore raw

The good news: you don't have a driver problem! But it sounds like you have a Wordclock problem. Namely, your rme 9632 has a different clock than your playback device (D/A) and they are not in Sync. Check the setting for who is (and should be) Wordclock Master and take adjust accordingly. You can set the mode of synchronisation for you card by editing /etc/asound.state and the running "alsactl restore".

-joachim

Quoted 23 lines · Ed Wildgoose
Quoted 3 lines

You can also simulate this simply by turning up the mixer controls so that normal music starts to clip. Same effect.

At risk of creating noise. Further experimentation shows that you need to set the output levels to +4db in hdspconf for this crackling to occur. So in fairness this could really be an overdriven output...?

The -10dbv output level is significantly quieter than any other equipment that I own (I know it's what they call consumer output device level, but still...)

Ed W

-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel

Signature · 10 lines
Joachim Deguara				Email: [email protected]
Fraunhofer IDMT, Langewiesener St. 22, 98693 Ilmenau, Germany
Phone: +49 3677/69-4851 		Fax:   +49 3677/69-4399


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
#3 lore raw
Quoted 1 line · Joachim Deguara

The good news: you don't have a driver problem! But it sounds like you have a Wordclock problem. Namely, your rme 9632 has a different clock than your playback device (D/A) and they are not in Sync. Check the setting for who is (and should be) Wordclock Master and take adjust accordingly. You can set the mode of synchronisation for you card by editing /etc/asound.state and the running "alsactl restore".

Hi, Thanks for the thought, but I don't see how this could be the issue? I am reading input from the HD, writing it to the card via either oss or alsa, and the output is analogue to a pre-amp and then on to an amplifier. The crackle only occurs when I use +4db as the output level, and then either find a song which clips anyway, or crank up the mixer so that the output clips. It could actually be some sort of electrical issue at the pre-amp to be fair, it's just that I don't remember it happening before? It doens't seem to be a problem if I drop the output level to the -10bdv setting for example.

The main thing I would prefer to debug right now anyway is why mplayer doesn't work with this card. A cursory glance at the mplayer output code suggests that everything looks fine... Oh well, will need to break out that code and study it a bit more - most other apps are working now though

Thanks

Ed W

-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click