Hello, everybody.

On Sun, Jan 04, 2026 at 12:29:00 +0000, Alan Mackenzie wrote:
> Hello, Jack.

> Thanks for the reply!

> On Sat, Jan 03, 2026 at 15:33:43 -0500, Jack wrote:
> > On 2026.01.03 11:44, Alan Mackenzie wrote:
> > > On Fri, Jan 02, 2026 at 15:44:06 +0000, Alan Mackenzie wrote:

> [ .... ]

> > > > I think I am just going to buy a new drive.  At ~25 Euros, it's
> > > > just not worthwhile trying all these things on the current one.

> > > Well, I've bought and installed a new drive (made by ASUS) and it
> > > hasn't helped in the slightest.  :-(

> > > Maybe modern DVD drives just aren't capable of reading audio CDs  
> > > properly.

> > It's been a while, but I'm pretty sure I've successfully read audio CDs  
> > with a DVD reader.  Can you find any documentation on the specific  
> > drive model you just bought?

> Well, things have taken a turn for the better!  I've discovered the
> program xine, which plays audio CDs faultlessly.  :-)  Maybe I'm kidding
> myself, but the sound quality seems better even than the sections without
> crackle on deadbeef.  So it's not my hardware which is at fault.

> Looking at the C source code for xine, it seems it uses direct ioctl
> calls to the kernel to read data from the CD.  deadbeef instead uses
> dev-libs/libcdio for this.  Maybe there's some incompatibility between
> libcdio and my hw/sw setup.  I will be trying to pin this down in the
> coming days/weeks.

> > As a temporary workaround, you might consider copying the entire  
> > content of the CD to a folder, then play from local storage instead of  
> > directly from the disk.  I've done this in the past with VLC, but just  
> > using a file browser should show each track as some sort of audio file.

> I haven't tried this yet, but I intend to do so this afternoon (European
> time).  Thanks for the suggestion.

I've just tried this, using xfce's thunar (file manager) which
recognised the tracks on a CD as .wav files.  I copied them to my home
directory, then played them back in xine.  There was a high level of
crackle through the entire length of both tracks I copied.  This was on
an original non-ripped CD recorded around 1990.

So, perhaps this copying used the same utility as aqualung.

Maybe deadbeef takes ~6 seconds to fill its buffer sufficiently at the
start of a CD after which it uses error correction of some sort.  But it
fails to maintain this buffer at track boundaries, hence ~2 seconds of
crackle at the start of every track.  Maybe.

> [ .... ]

> > Jack

-- 
Alan Mackenzie (Nuremberg, Germany).

Reply via email to