gmane.org.netlabs.uniaud.general

Re: Test build using ALSA 1.0.24

Doug Bissett — 2011-09-15T14:50:02

On Thu, 15 Sep 2011 13:52:53 UTC, "Andy Willis" 
<abwillis< at >geocities.com> wrote:



I didn't hear the "hum" (or anything like it) with anything else, 
including a couple of other streams from the same web site, but a 
different server. I didn't get it with VPC either, but I was testing 
with FLASH, and FLASH was blowing up before it was likely to cause 
another problem (perhaps related, but FLASH is so flakey anyway that I
didn't pursue it any further).

The only rel difference between the problem stream, and the other 
streams (that I know about), is that the problem stream is in mono, 
while the others are in stereo.

I suspect that these problems might have something to do with your 
"unknown code", since they don't happen (for me), with the most recent
version (except for yours).

Re: Test build using ALSA 1.0.24

Andy Willis — 2011-09-15T13:52:53

On Thu, 15 Sep 2011 03:50:21 UTC, "Doug Bissett" 
<dougb007!SPAM< at >telus.net> wrote:


Interesting.. I can reproduce a problem under PM123 every 33-34 
seconds that normally happens once every other day system wide.  I 
don't get any hum but I do get a "vibration" I guess is the best way I
could describe it, almost like the last sound is repeating over and 
over very very very quickly causing just  a zdzdzdzd sound.  I can 
pause PM123 and restart it and it is good for another between 33 and 
34 seconds.  When it happens with VPC I have to kill and restart VPC 
because it has exclusive access, other things I can do similar to 
PM123 but nothing else will reproduce it so reliably.  
Andy

Re: Test build using ALSA 1.0.24

Doug Bissett — 2011-09-15T03:50:21

On Sat, 10 Sep 2011 01:59:50 UTC, "Andy Willis" 
<abwillis< at >geocities.com> wrote:


This works with VPC, but it seems to have a problem where VPC is 
feeding sound with small gaps in it, sometimes. That is a minor 
problem that some system tweaking (both VPC and the host system) might
fix.

A more serious problem occurs when i attempt to listen to the on air 
radio stream from JAZZ.FM, using PM123. Use a file called 
JAZZFM91.PLS, and put this:
=========================
[playlist]
NumberOfEntries=1
Title1=JAZZ.FM91 - Canada's Premier Jazz Station
File1=http://streamerepsilon.jazz.fm:8000
=========================
into it (minus the delimiter lines). Then double click that to start 
PM123 (I assume you have it properly installed, with the associations 
set up).

On my system it plays properly for 5 to 10 seconds, then starts to 
hum, very much like the old 60 hz buzz. The previous versions of 
UniAud do not do that.

Hope this gives you a clue...

Thanks...

Re: Test build using ALSA 1.0.24

Andy Willis — 2011-09-14T18:56:30

On Wed, 14 Sep 2011 18:20:21 UTC, Peter Brown <losepete< at >ntlworld.com> 
wrote:

I haven't tried mplayer yet, could be that the problem section for VPC
is used by mplayer but the main purpose f this build was to verify 
where the problem is at for VPC which seems to be confirmed.
Running mplayer now with -ao kai:dart well... without this I get no 
output but I believe that is a known issue with the current status of 
uniaud.dll.
Andy

Re: Re: Test build using ALSA 1.0.24

Peter Brown — 2011-09-14T18:20:21

Hi

Peter Brown wrote:


Guess I posted too soon.

Sound works in VPC but audio does not work with mplayer and vlc now.

When I start playing a video both mplayer and vlc start fine but after 
about 15 seconds both show video becoming very jerky and audio degrades 
to white noise.

Regards

Pete

Re: Re: Test build using ALSA 1.0.24

Peter Brown — 2011-09-12T22:16:48

Hi Andy

Andy Willis wrote:


Yes, working fine here. Looks like you've identified where the problem is.

Thanks  :-)

Pete

Re: Sound lost when updating MPlayer

2011-09-12T20:06:08

Are you using an editor that puts an EOF marker at the end?  Many programs
don't like that.


"quiet" is unrelated to volume, it's for suppressing some of the verbose
console output.

Re: UniAud "complete" package contains UniAud.DLL...

David Azarewicz — 2011-09-12T04:11:03

It goes in \MMOS2\DLL. It is the interface for the uniaud API.

David

On Sun, 11 Sep 2011 23:33:33 -0400 (EDT) Carl Gehr wrote:

-----
David Azarewicz
http://www.88watts.net

UniAud "complete" package contains UniAud.DLL...

Carl Gehr — 2011-09-12T03:33:33

What should be done with the UniAud.DLL file?

The install/update instructions say to copy the two *.SYS
files [UniAud16 and UniAud32] to x:\MMOS2 directory.  But
there is no mention of the DLL file.  I assume it's there
for something, but what?

Re: Sound lost when updating MPlayer

Carl Gehr — 2011-09-12T03:15:43


Well, I installed the recommended UniAud*.sys levels and added
the ao=kai:dart to the Config file.  Now, when i try to run
MPlayer, I get the following:

    parse error at line 11

My Config file contains:
*  Lines 1 - 7 are all comments
*  Lines 8 - 10 contain:
   000008 quiet=yes
   000009 vo=kva:dive
   000010 ao=kai:dart
There is no "Line 11"...

I thought maybe it needed a blank 'Line 11' but that just
moves the 'parse error...' to Line 12.

As a last resort, I just deleted the 'config' file...

Amazingly, files that had not played before, are now
playing.  Even the one "*.3GP" file that was the reason
I updated MPlayer in the first place.  [The older MPlayer
did not recognize that file type. It was a video generated
by a 'Droid SmartPhone.

I should mention that the quiet and vo lines would produce
some very low volume sound from the *.3GP file.  But, now
without anything for the config file, it plays at an acceptable
volume.  I only put those two lines in because there was some
comment in the MPlay

Re: Sound lost when updating MPlayer

David Azarewicz — 2011-09-11T17:47:56

The correct place to look is:  http://svn.netlabs.org/uniaud

David

On Sun, 11 Sep 2011 12:54:06 -0400 (EDT) Carl Gehr wrote:

-----
David Azarewicz
http://www.88watts.net

Re: Sound lost when updating MPlayer

Carl Gehr — 2011-09-11T16:54:06


THANK YOU very much.  I do think that a part of the various
problems that I've had were memory leaks.  But, this is the
first time anyone has so clearly pointed to a particular
level as a likely source of the problem.  Maybe it's just
that some of the talk about other factors went by me.

The NetLabs site really need some help... :-(

The link to "the latest" version results in a message:
    "Cannot change directories..."
So, I dropped the *.ZIP from the link:
...and this did take me to the directory of downloads.  In fact,
there is no "uniaud_latest.zip" in the directory.  So, it's
not surprising the link from the NetLabs site is broken.

There is no name that is clearly  V1.9.26.  The closest that
I see is:
That also appears to be the latest date; except for:

I guess I'll download both and try to decipher what's different.

Also, I do not see any level marked as "v1.9.5" either on the
sit, nor any prior download that I have.  I do see a file:
But, there is also a '212' from a month or so before.

It wo

Re: Test build using ALSA 1.0.24

David Azarewicz — 2011-09-10T21:52:29

You are correct.  Whoever put that code there obviously put it there for a reason. It still has to 
be determined what it was supposed to do and why, and why it doesn't work for VPC.  Until 
then, it can't be called a fix.

However, finding exactly where a problem is is usually the most time consuming part of 
debugging. I, for one, appreciate the time and effort.  When I get some time I'll figure out 
exactly what is supposed to be happening there and implement a real fix. That is, unless one 
of the other developers gets there first :)

David

On Sat, 10 Sep 2011 21:35:32 +0000 (UTC) Andy Willis wrote:

-----
David Azarewicz
http://www.88watts.net

Re: Sound lost when updating MPlayer

David Azarewicz — 2011-09-10T21:42:51

The "uniaud" output feature of mplayer is known to be broken.  Use DART instead.


This is very old and extremely dangerous to use as it has known memory management 
problems that are known to cause widespread system problems. The only safe version of 
Uniaud16 is 1.9.5.


You can use any version of Uniaud32 you want. However, the newer versions are much 
better, especially for ICH7 hardware.  I use Uniaud16 v1.9.5 and Uniaud32 v1,9.26 on my 
T-60.

David

-----
David Azarewicz
http://www.88watts.net

Re: Test build using ALSA 1.0.24

Andy Willis — 2011-09-10T21:35:32

This just commented out a couple sections of code... it really isn't a
fix but it shows where the problem is.  Unfortunately, I am not really
certain what that code is even doing so I went as far as I was able to
go, I found it through checking the logs and comparing it to changes 
that occurred between the last working version and the first that 
didn't.  Much trial and error before I found the right section but 
hopefully the real cause can be found and resolved more easily.
Andy
On Sat, 10 Sep 2011 17:46:04 UTC, "David McKenna" 
<davidmckenna< at >comcast.net> wrote:

Re: Sound lost when updating MPlayer

Steve Wendt — 2011-09-10T21:28:56


The UniAud output for MPlayer has never worked for me.  Try using DART 
instead (ao=kai:dart).

Sound lost when updating MPlayer

Carl Gehr — 2011-09-10T21:06:38

I recently updated MPlayer to the level contained in:
    mplayer-SVN-r33491-OS2-4_4_5.zip
because I had some videos that would not play at all.

The most recent date/time in that ZIP is:
     29-May-2011   12:44a
The ChangeLog indicated:

The prior MPlayer was designated in the ChangeLog as:

After the upgrade, a number of videos that would not play
with the older version will now play.  The BAD NEWS is that
the sound on ALL videos is now either gone or the volume is
so low that it is barely detectable.

I have tried several levels of UniAud**.DLL, but they made
no difference.  I am currently running:



I know there are later levels, but some have cause system
instability.  This is the one that I've picked, mostly without
any factual basis; only that it seems to cause the fewest
problems.  

BTW, my base platform is:
    ThinkPad T-60
    eCS 2.0 RC2, but I'm really not sure that matters a
                 lot since the two key pieces seem to be
                 MPlayer and UniAud**.

PCI.exe reports [a pa

Re: Test build using ALSA 1.0.24

David McKenna — 2011-09-10T17:46:04

Andy,

On Sat, 10 Sep 2011 01:59:50 +0000 (UTC), Andy Willis wrote:

-snip-


  I tried this on my new Intel HDA based hardware and now I have sound in
VPC... thanks! I hope this makes it into the 'official' versions...

Regards,

Dave McKenna

Re: Test build using ALSA 1.0.24

Andy Willis — 2011-09-10T01:59:50

On Tue, 2 Aug 2011 16:42:39 UTC, Peter Brown <losepete< at >ntlworld.com> 
wrote:


I found a section of code that had changed that I reverted and it is 
now working on my pc.  I have uploaded to netlabs incoming as 
uniaud32-19226-vpc.zip.  It is not a real fix, however, it seems to be
working on my system at least.  If you want to give it a try and see 
if it is also where your problem is at or if you have another.
Andy

Re: Test build using ALSA 1.0.24

Andy Willis — 2011-09-06T14:00:13

On Tue, 6 Sep 2011 02:39:36 UTC, "Andy Willis" 
<abwillis< at >geocities.com> wrote:

Re: Test build using ALSA 1.0.24

Andy Willis — 2011-09-06T02:39:36

On Sat, 3 Sep 2011 15:58:32 UTC, "Andy Willis" 
<abwillis< at >geocities.com> wrote:


OK, I forgot that 1.9.4 did not work in VPC so what I have tested thus
far is:
1.9.2  Good
1.9.3  Good
1.9.4  Bad
1.9.6  Good
1.9.8 - 1.9.15 Bad

Here is a link to the changes that took place between them:
<http://svn.netlabs.org/uniaud/changeset?reponame=&new=385%40GPL%2Fbra
nches%2Funiaud32-2.0&old=364%40GPL%2Fbranches%2Funiaud32-2.0>
There seems to be only one change that affected only affected hda as 
far as I could tell and that deals:
Prevent interrupt handler form calling uniaud16 driver when not 
necessary
<http://svn.netlabs.org/uniaud/changeset/376>
with the latest trunk code I backed out what was left of it but it did
not make a difference for VPC.
Adding Followup to uniaud developers to see if someone can give me an 
idea what I may be missing.
Andy