Ripping hidden tracks before index 01, Hardware-based? What to look for? Options

[BoraBora]

The detection of the ability to read into lead-in must be based on read errors only.
The program asks the drive to read before track 1. If the drive returns a read error, the program says "this drive can't overread", and if the drive returns music, or silence, the program says "this drive can overread into lead-in".
The program is not supposed to know if the audio before track one is actually silent or not !

Maybe you misunderstood me. When I write "Can rip index 00 but only silence", I mean: "seems to do the job, but actually rips only silence instead of music".

I read your post twice and I think I understand what you say, but I'm still not clear on the main point:

Some drives can't read before track 01 index 01.
Some drives can, but can't overread into "lead-in". Which is a wrong statement, because it actually means that they can't read the compulsory pregap, or in other words, can't read negative audio sectors (from -150 to -1).
Some drives can read all along from Track 00 Index 00 (sector -150, Absolute time 00:00:00:00). They may even overread into the actual lead-in if you tweak the registry.

So far, we have listed in this topic 3 types of ripping ability with a hidden track (ie: an entire song, not 2 seconds of music):

1 - The drive can't rip it at all.
2 - The drive can rip it (or seems to), but there's only silence instead of music.
3 - The drive properly rips the music.

Which drive in your list could produce the second type of ripping, ie producing silence instead of ripping music? Those which can't read before track 01 index 01? If so, why the difference with those which can't rip it at all?

This post has been edited by BoraBora: Nov 7 2005, 21:12

[Cosmo]

Why couldn't a particular drive read Track 01 Index 00 - as long as the data is in the audio session and not the lead-in area?

Is such a Track 01 Index 00 pregap somehow represented in the TOC?
(the only TOC readout i'm able to get doesn't indicate pregaps, so I wonder if the drive firmware simply refuses to acknowledge any data prior to the Index 01 starting point...)

This post has been edited by Cosmo: Nov 7 2005, 22:28

[Pio2001]

Which drive in your list could produce the second type of ripping, ie producing silence instead of ripping music? Those which can't read before track 01 index 01? If so, why  the difference with those which can't rip it at all? 

I consider drives ripping silence instead of music as unable to read the hidden track. Thus for me, they can't read before track 01 index 01.
About the difference between reading silence and not reading, ask the drive designer

[Alex B]

Do the drives have integrated logic for reading and interpreting the TOC? Do the software components like EAC and the OS level driver just kindly ask the drive to do the job instead of giving the drive direct orders like: go to the absolute position xxx and start reading? I mean: Can this be depended on the software used?

Edit: Oops, It wasn't ready...

This post has been edited by Alex B: Nov 7 2005, 23:08

[BoraBora]

Lite-On SOHW-832S
Drive can read max. 139 sectors (-00:01.64) from the Lead-in
Drive can read max. 6749 sectors (-01:29.74) from the Lead-out

Plextor PX-W5224A
Drive can read max. 75 sectors (-00:01.00) from the Lead-in
Drive can read max. 6755 sectors (-01:30.05) from the Lead-out

Mine:

Plextor PX-712A
Drive can read max. 75 sectors (-00:01.00) from the Lead-in
Drive can read max. 6755 sectors (-01:30.05) from the Lead-out

Plextor PX-740A (rebranded BenQ)
Drive can read max. 143 sectors (-00:01.68) from the Lead-in
Drive can't read lead-out

Strangely enough, CD-DVD Speed also reported the offset of the Plextor as 0 bytes/samples.

Same here. Weird.

I consider drives ripping silence instead of music as unable to read the hidden track. Thus for me, they can't read before track 01 index 01.
About the difference between reading silence and not reading, ask the drive designer

You mean... not, this can't be... you mean I asked a question you can't answer? Oh boy! I'll remember this day forever!

OK. So the conclusion would be:

1 - When tracks are hidden at the beginning of the CD, the drive must be able to rip between absolute time 00:00:02:00 and Track 01 Index 01. This is not related to the overreading in lead-in feature.

2 - Neither CDSpeed or the Plextools can tell you if your drive is able to rip between absolute time 00:00:02:00 and Track 01 Index 01. DAE tests published on specialized sites and web databases are useless too. You'll have to own such a CD and try it yourself.

Thanks to your explanation, I'm not puzzled by this problem anymore. Still, I'm amazed: this feature has never been properly commented and listed for each drive on the various DAE-oriented sites and listings for the bit-exact anal-retentive rippers. I mean... we're talking WHOLE SONGS not ripped here. Not a 2352 octets offset or even a 2 seconds gap.

[Alex B]

1 - When tracks are hidden at the beginning of the CD, the drive must be able to rip between absolute time 00:00:02:00 and Track 01 Index 01. This is not related to the overreading in lead-in feature.

2 - Neither CDSpeed or the Plextools can tell you if your drive is able to rip between absolute time 00:00:02:00 and Track 01 Index 01. DAE tests published on specialized sites and web databases are useless too. You'll have to own such a CD and try it yourself.

These conclusions seem to be logical.

It's surprising that some drives obviously cannot access the identically located physical audio data if the beginning of the data is marked as INDEX 00 instead of INDEX 01. If I have understood correctly the only small difference between the discs is in the TOCs.

I wonder if anyone could explain how the physical data structures on the CDs would differ if the two different "Hendrix" cue sheets I posted earlier are used. In case this matters, the total length of the source wave file is 42741720 samples (about 16 minutes). That would be very interesting information.

Edit: Edited a bit. Hopefully it's easier to understand now...

This post has been edited by Alex B: Nov 8 2005, 08:43

[Synthetic Soul]

Many thanks to BoraBora for so doggedly pursuing this topic, and to Pio2001 for his informed posts.

It's just a shame that we are no closer to being able to determine a drive that will do this from the specs!

[Pio2001]

You mean... not, this can't be... you mean I asked a question you can't answer? Oh boy! I'll remember this day forever! 

I'm not sure about the information that you are looking for. If the designer wants the drive to read before track 1 index 1, the drive can, and if he doesn't want, then the drive can't.

It's surprising that some drives obviously cannot access the identically located physical audio data if the beginning of the data is marked as INDEX 00 instead of INDEX 01. If I have understood correctly the only small difference between the discs is in the TOCs.

The index information is repeated all along the audio, in the subcode channels, together with the track number, absolute time, SCMS, pre-emphasis, etc

[Egor]

Probably it would be useful to create a list of drives that can properly extract "track 01 index 00" information. Here is mine little contribution to it, I tested the functionality using self-made audio cd with hidden track and EAC. Minus sign means that there were only silence in extracted waveform instead of a hidden track.

Pioneer DVR-110D: +
NEC ND-3520AW: -
Toshiba SD-R5112: -
Samsung SM-332B: -
Toshiba SD-R1202: -
Mitsumi CR-4804TE: +

[BoraBora]

Probably it would be useful to create a list of drives that can properly extract "track 01 index 00" information.

Absolutely.

I'll update the list in the first post with your information and clean it a bit (no need to mention the overreading feature since this is not related). If moderators are OK with that, I'll crosspost it on this pinned topic with a request to include these informations in the database: List of DAE Drive Features.

If anyone knows people writing for www.cdrinfo.com or www.cdfreaks.com or any other site doing DAE tests with optical drives, it could be nice to request information on this very subject when they're doint tests.

[southisup]

The drive on my HP ZE4315 laptop reports itself as QSI CDRW/DVD SBW-241 and I just used it to rip the hidden track at the beginning of David Gray's White Ladder album.

I ripped by range but EAC wouldn't let me put a negative number in for the start frame and the rip audio starts suddenly so I assume a bit is missing at the start.

It looked like standard Windows error trapping preventing negative numbers being entered - is a negative frame number something that just can't exist?

[.halverhahn]

I tested some of my Drives.

Drive - Hidden Track with Music=Yes, Silence=No

Plextor PX-40TS - YES
Teac DV-516G - YES
NEC ND-3520 - NO

[Pio2001]

I ripped by range but EAC wouldn't let me put a negative number in for the start frame and the rip audio starts suddenly so I assume a bit is missing at the start.

The only way to rip earlier is to set the lowest possible read offset correction value in the drive properties.
And if this is not enough, you can tweak the registry entry for the read offset correction in order to set even lower values than EAC can accept in its GUI.

[BoraBora]

List updated.

[evereux]

Thanks for updating the list there BoraBora. I've been going through it comparing that to what I have at daefeatures.co.uk.

Here are the discrepencies.

CODE

Man. Model        : HA reported (author) | Daefeatures (author)

Lite-On SOHW-1673S: NO  (langoustator)   | Lead In (Thundik81)
Mitsumi CR-4804TE : YES (Egor)           | None    (The Coaster Factory)
Plextor PX-712A   : YES (BoraBora)       | Leadout ( liekloo)
Samsung SD-616E*  : NO  (Synthetic Soul) | Lead In (The Coaster Factory)

*daefeatures lists a 616F, I need to check to see what the differences are

What I'll probably do soon is edit the coaster factory entries and mail the persons who submitted the other two drives and get them to check their entries (if possible). I'll document these changes somewhere too. Also, I'll provide a link to this thread on the site so others can learn from the information here.

Thanks again BoraBora.

edit: Thanks to Pio2001 for his explanation too.

This post has been edited by evereux: Nov 9 2005, 12:29

[BoraBora]

Thanks for updating the list there BoraBora. I've been going through it comparing that to what I have at daefeatures.co.uk.

Don't thank me, you're the one who does the hard work with daefeatures.co.uk.

Here are the discrepencies.

Actually, these are not discrependencies. The YES/NO listed is the ability to rip between absolute time 00:00:02:00 (after the compulsory 2 seconds pregap noted by Pio2001) and Track 01 Index 01. This is not related to the overreading feature. We're talking a whole new kind of feature. Well... not exactly new, but never listed before on DAE databases.

So far, we don't have tools to test this feature. Neither CDSpeed nor the Plextools mention this feature. We have to own or make (like Alex B did) a CD with sound before Track 01 Index 01 and test and listen ourselves.

Hiding songs before Track 01 Index 01 seems to become a trend. A very stupid one, if I may add, but unfortunately we can't do much about it. So adding listings for this feature to daefeatures.co.uk would probably be a welcome upgrade.

[evereux]

So, to summarize. The list details drives that will return audio in the region between TRACK 01 INDEX 00 and TRACK 01 INDEX 01.

Drives that can overread into the lead-in, some will return audio, some will not (even the disc may contain audio within the hidden track)?

edit: It might be an idea to create an image + cue sheet for people to download so they can test this feature. I could possibly host this at daefeatures and it's something I'll certainly consider adding to the database.

This post has been edited by evereux: Nov 9 2005, 17:33

[Drenholm]

Lite-On LTR-52327S

Rapidly fills single lines of error correction at a time for entire pregap, result is silence.

[southisup]

I ripped by range but EAC wouldn't let me put a negative number in for the start frame and the rip audio starts suddenly so I assume a bit is missing at the start.

The only way to rip earlier is to set the lowest possible read offset correction value in the drive properties.
And if this is not enough, you can tweak the registry entry for the read offset correction in order to set even lower values than EAC can accept in its GUI.

Thanks.

The lowest EAC accepts is -11,760, which gave me a fraction more, but still not all.

Editing the registry entry to -50,000 seems to have done the trick.

-100,000, which EAC's GUI couldn't display correctly, gave me timing errors when ripping the audio. I did get a bit of silence extra, but the audio start was the same as that for -50,000.

[madorangepanda]

Tested my two drives on Bloc Party-Silent Alarm and both work
LG-Hitachi HL-DT-STCD-RW GCE8487B
Sony DVD-ROM DDU1615
Ill check if they can overread the lead-in later.

[Cosmo]

repeat:

Lite-On LTR-52246S : NO
BCD E520C : YES

[BoraBora]

List updated.

@Cosmo: Sorry I missed your post the first time.

So, to summarize. The list details drives that will return audio in the region between TRACK 01 INDEX 00 and TRACK 01 INDEX 01.

Drives that can overread into the lead-in, some will return audio, some will not (even the disc may contain audio within the hidden track)?

Exactly.

edit: It might be an idea to create an image + cue sheet for people to download so they can test this feature. I could possibly host this at daefeatures and it's something I'll certainly consider adding to the database.

That's a good idea, I'm gonna look into this.

[BoraBora]

OK, I created a wav + cue with two songs, including one hidden.

The songs

index 00: Paul Westerberg / Looking Up in Heaven
index 01: Gilberto Gil / Oslodum

I downloaded both songs on the Wired site. They're released under a Creative Commons Licence and noncommercial copying and distribution is authorized.

The files

hidden_song_test.cue
hidden_song_test.exe (Wavpack self-executable).

Double-click hidden_song_test.exe to decompress it to wav. Load the .cue in your burn program and burn it.

The test

Rip the CD as an image, then listen to the wav. If you have 3'08'' of silence instead of "Looking Up in Heaven" by Paul Westerberg, your drive can't rip track 01 index 00. If you hear the song, rejoice. Either way, listen to Oslodum by Gilberto Gil, begginning at 03:08:25, it's a nice song.

P.S. : I hadn't burned a CD-A for years, so I'm not sure I did it the best way. Feel free to modify anything or do a better test CD.

P.P.S.: This is a crappy free personal site so I probably won't be able to keep it online. If anyone can mirror it, it would be appreciated. Or maybe shared on a P2P network?

[Pio2001]

-100,000, which EAC's GUI couldn't display correctly, gave me timing errors when ripping the audio.

This is not surprising. You are then ripping before the audio track, that begins around -88200, in the real lead-in. And I'm not sure if the lead-in features any audio data (we can check in the ECMA-130 specs).

[evereux]

Plextor PX-708A - YES

Mirrors:
hidden_song_test.exe
hidden_song_test.cue
sfv
md5