| View previous topic :: View next topic |
| Author |
Message |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Sat Sep 25, 2010 2:32 pm Post subject: |
|
|
| well I burned an audio disc with a 0sec pregap and ripped it back on windows. Everything was fine. The CD also plays perfectly fine with media player. I don't know what's going on, maybe an os x related bug ? I don't know |
|
| Back to top |
|
 |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Sun Sep 26, 2010 11:53 am Post subject: |
|
|
| Do you think the audio glitch is only between the first and 2nd track or do you hear it between every track ? |
|
| Back to top |
|
 |
djmastera2000
Joined: 30 Jul 2010 Posts: 12
|
Posted: Tue Sep 28, 2010 2:38 am Post subject: |
|
|
between every track  |
|
| Back to top |
|
 |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Tue Sep 28, 2010 2:01 pm Post subject: |
|
|
| Do you hear the glitch on a desktop/car/hifi set CD player ? Or just when playing it on the mac ? |
|
| Back to top |
|
 |
djmastera2000
Joined: 30 Jul 2010 Posts: 12
|
Posted: Sat Oct 16, 2010 10:22 am Post subject: |
|
|
Sorry, was not at home.
It is noticable on every player (PC, HiFi, Car, etc.) |
|
| Back to top |
|
 |
edgar
Joined: 03 Dec 2010 Posts: 4
|
Posted: Fri Dec 03, 2010 11:14 am Post subject: |
|
|
| I can confirm this bug, when burning a cue+mp3 as a continuous mix. I played the cd in a car and there were noticable skips between tracks. |
|
| Back to top |
|
 |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Sun Dec 19, 2010 11:26 am Post subject: |
|
|
| I just have no idea why there is that glitch between the tracks. I've tried burning two sine waves (440Hz and 880Hz) and rip the burned CD and clearly the ripped tracks are not what they should be. But I don't know why. |
|
| Back to top |
|
 |
djmastera2000
Joined: 30 Jul 2010 Posts: 12
|
Posted: Sat Jan 01, 2011 1:53 pm Post subject: |
|
|
Hm, no more ideas?
PS: Happy New Jear!  |
|
| Back to top |
|
 |
edgar
Joined: 03 Dec 2010 Posts: 4
|
Posted: Sun Jan 02, 2011 11:22 pm Post subject: |
|
|
Happy new year all!
I suspect that the last 2 seconds of every track is missing. When I compare the burnt CD to playing the cue&mp3 in Vox/VLC, the beginning of tracks are right but they all end too soon. Not sure if it exactly is 2 seconds ... |
|
| Back to top |
|
 |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Mon Jan 03, 2011 9:37 pm Post subject: |
|
|
| Ok I'll post detailed results in 2-3 days. If I don't, remind me. |
|
| Back to top |
|
 |
edgar
Joined: 03 Dec 2010 Posts: 4
|
Posted: Tue Jan 04, 2011 12:45 am Post subject: |
|
|
It took some time to puzzle out. I ripped a CD, converted to mp3 & cue and written the it again to a virtual device using Virtual CD/RW (shareware unfortunately). Then used bchunk to convert the generated bin/cue to separate WAV files so I could use a tool to measure track lengths.
Ok, here's the CUE for the first 5 tracks (and the 6th):
TRACK 01 AUDIO
TITLE "Track01"
PERFORMER "Unknown Artist"
INDEX 01 00:00:00
TRACK 02 AUDIO
TITLE "Track02"
PERFORMER "Unknown Artist"
INDEX 00 00:07:25
INDEX 01 00:07:27
TRACK 03 AUDIO
TITLE "Track03"
PERFORMER "Unknown Artist"
INDEX 00 01:45:05
INDEX 01 01:45:10
TRACK 04 AUDIO
TITLE "Track04"
PERFORMER "Unknown Artist"
INDEX 00 02:00:34
INDEX 01 02:00:35
TRACK 05 AUDIO
TITLE "Track05"
PERFORMER "Unknown Artist"
INDEX 01 03:27:60
TRACK 06 AUDIO
TITLE "Track06"
PERFORMER "Unknown Artist"
INDEX 00 03:43:61
INDEX 01 03:43:63
Seems a bit like a mess to me with micro gaps. Some tracks have only index 00 while others have index 00 and 01. Ohwell, the original seems to play fine.
In Audacity, opening the first 3 tracks, this results in the following:
Track 1 starts fine, but ends at 00:07:0266
Track 2 ends at 01:44:0533
Track 3 ends at 01:59:0799
Track 4 ends at 03:26:1066
Track 5 ends at 03:42:1333
Now, if I'm not mistaken that is, the CUE sheet shows frames after the seconds and an audio CD uses 75 frames per second.
So, the actual lengths would be:
Track 1 = 00:07:02
Track 2 = 01:37:02
Track 3 = 00:15:02
Track 4 = 01:27:02
Track 5 = 00:16:02
Trying to figure out a pattern, it would be round down to the nearest second and then adding 2 frames? |
|
| Back to top |
|
 |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Wed Jan 05, 2011 10:48 am Post subject: |
|
|
Awesome, thank you.
| edgar wrote: |
Now, if I'm not mistaken that is, the CUE sheet shows frames after the seconds and an audio CD uses 75 frames per second. | Correct.
Here are some of the files I'll be burning/ripping. They are 440Hz and 880Hz sin waves. The duration is exactly 20 seconds (1500 frames):
http://maconnect.ch/download/forums/880.aif.bz2
http://maconnect.ch/download/forums/440.aif.bz2
As a side note, you should not use sound tracks shorter than 6 seconds (450 frames), because then you'll be facing another bug (which seems to affect several burning softwares). |
|
| Back to top |
|
 |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Wed Jan 05, 2011 12:33 pm Post subject: |
|
|
Allright this is what I have so far.
Disc 1
Track 1 : 440Hz sine 1500 frames, 2s pregap
Track 2 : 880Hz sine 1500 frames, 0s pregap
Burned using Toast
Extracted with dd and The Finder and analyzed with Audacity
dd: sizes = both 1500 frames
finder + audacity: track 1 has prepended blank audio, but only 1.13 frames instead of the expected 150 frames (2s pregap)
track 2 has ~2 frames blank audio + ~2 frames 440Hz audio, the rest is the actual 880Hz track
Conclusion: everything is wrong. Track1 duration should be 1650 (1500 + 150) and have 150 frames of blank data. Track2 should not have any prepended blank data and no 880Hz audio parts.
I can hear the 440 to 880Hz transition in the 2nd track if I read it with QuickTime. But with iTunes I don't. Maybe there is something worth exploring here.
------------------------------------
------------------------------------
Disc 2
Track 1 : 440Hz sine 1500 frames, 2s pregap
Track 2 : 880Hz sine 1500 frames, 0s pregap
Burned using Express Burn for Windows XP
Extracted with Window Media Player (WAV), dd and analyzed with Audacity
Window Media Player:
Track 1: duration = 22sec (ok), but the blank data is appended (which makes no sense, really) rather than prepended. The track also has ~2 frames a prepended blank data.
Track 2: duration = 20sec (ok). The track has ~2 frames a prepended blank data. This implies that the track is truncated.
dd:
Track 1: size = 1652 frames (should be 1650)
Track 2: size = 1502 frames (should be 1500)
Conclusion: everything is wrong, again.
I have no idea what's going on here. The result depends on which software was used for the burn and which was used for the extraction. The end result is always wrong. |
|
| Back to top |
|
 |
edgar
Joined: 03 Dec 2010 Posts: 4
|
Posted: Thu Jan 06, 2011 2:47 am Post subject: |
|
|
This is going to be a cool puzzle to solve!
At first sight, Toast seems right to me.
The pregap is missing, but can the first track of a CD have a pregap at all?
Are you using an mp3 as audio source? It is a known issue that mp3 files always starts with a blank frame and thus have a pregap (something around 25ms, which is near 2 CD frames). If you'd convert this to WAV and cut it exactly as the CUE sheet describes, it would explain why a little bit of track 1 moves to track 2.
I'll run some experiments too next evening! |
|
| Back to top |
|
 |
Fabio Administrateur - Site Admin
Joined: 01 Jul 2006 Posts: 358
|
Posted: Thu Jan 06, 2011 11:00 am Post subject: |
|
|
I'm using the mentioned AIF files, which are raw uncompressed bytes.
| Quote: | At first sight, Toast seems right to me.
The pregap is missing, but can the first track of a CD have a pregap at all? | The first track always has to have a pregap for historical reasons. This was necessary because old laser systems were not very accurate. Nowadays, some recording devices still require a pregap of 2s for the first track.
And still, the 2nd track has a few frames from the 1st track in it. So there definitely is something wrong. |
|
| Back to top |
|
 |
|