I received several CDs from a friend that...have seen better days. (He just had them all loose in a satchel, and it looks like some of them have food stains on the encoded side...)
Anyway, I'm trying to extract what I can via EAC. I'm not at all concerned with how long the extraction takes; I only want to get the best (i.e. most accurate) copy possible.
When you're dealing with badly damaged CDs, will you usually get a better (more accurate) copy in "High" error mode - causing EAC to reread the same sector up to 80 times - or in Burst Mode, or in some other mode? Does the answer depend on how badly damaged that particular track is, i.e. how many sectors can't be read, or on the "quality" rating (83.3%-100%) given by EAC in "High" error mode?
and it looks like some of them have food stains...
Of course, you can clean the bottom side. I've got a gizmo the polishes-out scratches, but I don't think it has ever helped when simple cleaning didn't work.
on the encoded side...
In case you don't already know, the data is on the top/label side, and the CD is read through the entire thickness of the polycarbonate.* So CDs are more-easily damaged from the top-side, and I assume that's why polishing rarely helps. If you attempt to polish the top, you will do even more damage to the data-layer.
I'm sure someone else can help with the EAC settings. I'm not and expert on EAC, but when I can't read without getting reported-errors, I try another computer/drive. If all else fails, I just
listen for the error and if I can't hear I try to forget about it! (That's assuming I can read the file well-enough to get a WAV file.)
If I get an audible error there a few things I might do, depending on how important that particular song is to me:
- Buy another new or used CD.
- Purchase the MP3.
- Try to repair the sound with an audio editor or noise reduction software. "Ticks" can often be removed with vinyl "click" and "pop" reduction software. If you have a skip or a stutter, there's usually not much you can do.
- Try an "analog rip". (Often, the error-hiding features in the player will mask the error.)
- Delete that file/song. (Although, I hate to have an album with a missing song.)
* On a DVD, the data-layer better-protected is in the middle of a polycaroate sandwich.
Of course, you can clean the bottom side. I've got a gizmo the polishes-out scratches, but I don't think it has ever helped when simple cleaning didn't work.
Yeah, I've cleaned every one of them using dish soap and a soft cloth, moving in to out (rather than in spirals) as is recommended for cleaning CDs. I was able to get rid of a lot of the food stains but some still remain - I suspect the plastic itself is damaged.
In case you don't already know, the data is on the top/label side, and the CD is read through the entire thickness of the polycarbonate.*
Heh, I'll have to admit that I'd forgotten that. It's been a while since I've used CDs! Most of the stains and heavy scratches are on the bottom (i.e. reading) side, not on the label side. Fortunately I haven't tried cleaning the tops of any of the discs.
I'm sure someone else can help with the EAC settings. I'm not and expert on EAC, but when I can't read without getting reported-errors, I try another computer/drive. If all else fails, I just listen for the error and if I can't hear I try to forget about it! (That's assuming I can read the file well-enough to get a WAV file.)
If I get an audible error there a few things I might do, depending on how important that particular song is to me:
- Buy another new or used CD.
- Purchase the MP3.
- Try to repair the sound with an audio editor or noise reduction software. "Ticks" can often be removed with vinyl "click" and "pop" reduction software. If you have a skip or a stutter, there's usually not much you can do.
- Try an "analog rip". (Often, the error-hiding features in the player will mask the error.)
- Delete that file/song. (Although, I hate to have an album with a missing song.)
All good suggestions. I routinely use 2 different drives to see which one I can get a better read from. The type of damage seems to dictate which one's the better choice.
Some of these CDs are rare/indies, so getting a replacement CD or even an MP3 is unlikely. (It's a real shame my friend took such poor care of these and didn't keep the plastic case - many would be valuable otherwise.)
I had considered trying to repair the sound myself, but I'm enough of a noob I figured I'd do more harm than good. EAC will do some autorepair of WAV files but not MP3s. Do you have any suggested/preferred software?
Hmm...an analog rip? I wonder if EAC's even capable of that, but it's worth investigating. I don't believe that's the same thing as Burst Mode, at any rate - I think Burst Mode just ignores errors.
Finally, deleting that file/song isn't viable for me, unless I have a suitable replacement copy from another source (i.e. a compilation or "greatest hits" album). Still, some of these are so badly damaged that large segments of the data appaer to be unrecoverable, and EAC records silence.
By analog rip, DVDdoug meant play the CD on a regular CD player and connect its analog out to your sound card's analog in. Then use some program to save the digitized signal to a file.
EAC will do some autorepair of WAV files but not MP3s.
Rip to WAV, repair, then encode to MP3... There are quite a few free LAME based-MP3 encoders around. MP3 is lossy compression, and you will have to decompress for any editing/processing. (All "normal" audio editors decompress when you open the file, and then you have the option of re-compresing to MP3, or saving-as a different format when your'e done.) . For the best possible quality, you should compress
once as the final step. (Ripping "directly" to MP3 is fine, if you are not going to edit/process.)
Do you have any suggested/preferred software?
For "vinyl-like" defects, I use
Wave Repair (http://www.delback.co.uk/wavrep/) ($30 USD). For general audio editing (say to cut-out a few milliseconds of silence) or for "regular" noise reduction, I use
GoldWave (http://www.goldwave.com/) ($50 USD), or sometimes
Audacity (http://audacity.sourceforge.net/?lang=en) (FREE!!!). But realistically, if you have anything other than short-duration "ticks", there is often nothing you an do.
Hmm...an analog rip? I wonder if EAC's even capable of that, but it's worth investigating.
Technically, digital-to-analog-to-digital... Last time i did that, I used
Audiograbber (http://www.audiograbber.org/) (FREE!!!). If you have a desktop computer (with audio line-in) you can connect line-out/headphone-out to line-in, or plug-in a CD player and record with an audio editor (or other recording software).
This feature is getting rare, but depending on your soundcard/driver you may have the ability to record "What-U-Hear" or "Stereo Mix" (with the appropriate recording software). I'm not sure if Audiograbber requires your soundcard to support this for analog ripping. Or,
Total Recorder (http://www.totalrecorder.com/) ($18 USD & up) has it's own built-in virtual driver to record whatever is coming from your computer's speakers.
The mic input on a laptop has too much gain for a line-level signal, and they are generally noisy and low-quality. If you have a laptop, you can get a line-level to USB interface for less than $50 USD. Don't get a regular "USB soundcard", because they usually have only mic-in and headphone-out like a laptop.
I don't believe that's the same thing as Burst Mode, at any rate - I think Burst Mode just ignores errors.
I think your right, but I'm not an EAC expert.
Finally, deleting that file/song isn't viable for me, unless I have a suitable replacement copy from another source (i.e. a compilation or "greatest hits" album). Still, some of these are so badly damaged that large segments of the data appaer to be unrecoverable, and EAC records silence.
Well... A deleted file is perferable to a silent file. In other cases, it's your choice to live with the best you can get, or nothing.
I'll second DVDdoug's suggestion of an "analog rip" if all else fails.
In my case, I had a badly scratched CD-R that caused skipping on various parts of the disc, no matter how I ripped it in EAC. I ended up trying my standalone DVD player (and in this case, I was able to record its S/PDIF output, so it was still an all-digital capture), and the disc played completely glitch-free (although I'm sure there was quite a bit of sample interpolation going on in the player - sounded just fine, though).
In my case, I had a badly scratched CD-R that caused skipping on various parts of the disc, no matter how I ripped it in EAC. I ended up trying my standalone DVD player (and in this case, I was able to record its S/PDIF output, so it was still an all-digital capture), and the disc played completely glitch-free (although I'm sure there was quite a bit of sample interpolation going on in the player - sounded just fine, though).
I'd strongly prefer interpolation over silence or clicks, if those are the only two options! Thanks for the idea.
A related question - what types of "resurfacing" or related devices have any of you used to attempt repairs to badly damaged CDs? I'm not interested in investing tons of money into this, but if there's a decent, reasonably priced option, I'll look into it.
A related question - what types of "resurfacing" or related devices have any of you used to attempt repairs to badly damaged CDs? I'm not interested in investing tons of money into this, but if there's a decent, reasonably priced option, I'll look into it.
There's a video rental store where I live that has a commercial resurfacer. I took a badly beaten up Supertramp CD to them out of curiosity and it worked quite well. Of course it could not fix the pin holes.. But it came out looking almost new! Maybe there is something like that in your area?
A related question - what types of "resurfacing" or related devices have any of you used to attempt repairs to badly damaged CDs?
I have one like
this (http://www.aleratec.com/dvdireplpa24.html). I'm not at home right now and I think I might have a different brand, but it looks similar (like a portable CD player) and the donut polishing pads look the same.
I also have some
plastic polish (http://www.tapplastics.com/product/repair_products/plastic_repair/novus_plastic_polishes/428) that you can use with just a cloth.
I've seen
this one (http://www.amazon.com/gp/product/B0015ACUKC/ref=pd_lpo_k2_dp_sr_1?pf_rd_p=1535523722&pf_rd_s=lpo-top-stripe-1&pf_rd_t=201&pf_rd_i=B00005B9W6&pf_rd_m=ATVPDKIKX0DER&pf_rd_r=14MZWR1B47RBY55MF35E) in quite a few stores.
And, I've heard of people using toothpaste...
Cali Games in Hawthorne CA has a machine that will resurface discs and it does not grind away part of the disc. It adds material and buffs it out. I know it works great on both CDs and DVDs. I'm sure other game stores have similar machines. It will not fix damage to the label side. If you scratched away too much of the reflective layer you're probably out of luck.
G²
Toothpaste is known as a mild abrasive. I have also bought that screen renewer thing they use on plastic screens on boats. I had tenths of troublemaking CDs, so it was worth the $8 or whatever, but do not expect miracles.
Anyway:
* more drives = more chances to find one rip that is good or at least sounds less annoying.
* copying from SPDIF out should not be any better than burst mode, except that your audio CD player is a different drive
* dBpoweramp and EAC use somewhat different addressing to grab the bits. You might want to try both, and why not try CUERipper as well. dBpoweramp has a free trial (you might want to wait until you have a couple more troublemaking discs).
* as a general rule, re-reading (i.e. more secure modes) improves your chances of getting more frames out OK, and then an interpolation over unrecoverable samples could be necessary. I do not know if drives interpolate different (there shouldn't be much they could do!) or if EAC does it different. (Warning: dBpoweramp has a checkbox for interpolation over entire frames, and that should be avoided as far as you can.)
* if there are not too many errors, then the CUETools database might provide a repair.
(Secure rippers could actually improve if they would implement a feature logging the troublesome frames, and use that log upon re-ripping with another drive. Then they could pick whatever passes C1&C2 on at least one of the rips.)
If they cause errors, Brasso (i.e. polish) the non-label-side scratches and stains away. There are threads on this - do a search.
If there are still errors, connect the SPDIF output of a stand-alone CD player to the digital input of a sound card, and capture that way, as described above.
EAC and dBPowerAmp have different ways of coping with errors - I prefer dBPowerAmp because it uses AccurateRip first, then two complete track rips if the disc it not in AccurateRip, and only resorts to multiple individual re-reads in the places where those two methods fail. More often than not, if dBPowerAmp resorts to this (or EAC gets stuck for ages doing its individual re-reads), it's time to break out the brasso or play the disc on an external player. While the individual re-reads are sometimes successful, as often as not they get nowhere. Generally the more trashed the disc the less chance you stand, but there are surprises, so it's worth at least trying.
With care and luck, you can rip a cue sheet (don't need the audio, just the cue sheet), record the CD from an external player to one .wav file, and then use the cue sheets to auto-split (and tag) the wav file.
Cheers,
David.
Since this question has been overlooked in favour of other potential solutions:
When you're dealing with badly damaged CDs, will you usually get a better (more accurate) copy in "High" error mode - causing EAC to reread the same sector up to 80 times - or in Burst Mode, or in some other mode? Does the answer depend on how badly damaged that particular track is, i.e. how many sectors can't be read, or on the "quality" rating (83.3%-100%) given by EAC in "High" error mode?
Damage might make a sector return inconsistent sets of data on repeated reads. If you use Burst Mode and therefore only perform a single read, you cannot guarantee that it gathers the correct data. Repeating the read numerous times enables a sort of statistical assessment of how accurately that sector can be read. That’s what the percentage describes: how many of the attempts obtained the same (most common) set of samples.
I don’t know exactly how badly damaged these CDs are, but if you haven’t already, it would probably be best to try the available secure reading modes before starting ‘physical therapy’ to the discs. If you cannot get worthwhile levels of confidence from such modes, or the process seems to be highly stressful to the drive, then other attempted solutions may be necessary.
I would be careful letting the retries go on for two long. I have used EAC in High mode a number of times. On a two CD's it ran for an hour before I shut it down. I had problems with two drives it did it on ever since.
EAC's "statistical" method amounts to little more than picking data that is most consistent and presenting an error when this data does not occur in at least 50% of the time by the end of the user-allotted 16-reread sets.
Setting the (poorly named to the point that it is misleading) error recovery level high is no guarantee that the result will be any more accurate than setting it to something lower than high. Another (more disturbing) way to view this is setting it higher can actually gives EAC more chances to give you bad data an not report an error. This is because consistent data is not in any way whatsoever a guarantee that the rip was error-free.
That’s what the percentage describes: how many of the attempts obtained the same (most common) set of samples.
This is not quite correct.
From http://www.exactaudiocopy.de/en/index.php/...tion-questions/ (http://www.exactaudiocopy.de/en/index.php/support/faq/extraction-questions/)
It is the ratio between the number of minimum reads needed to perform the extraction and the number of reads that were actually performed. 100% will only occur when the CD was extracted without any rereads on errors.
The next sentence, however,
ONLY when there are suspicious positions reported, there are really uncorrectable read errors in the resulting audio file.
is complete bullshit.
If all else fails, check your country laws. I've downloaded disk images from shady sites for CDs that were so badly damaged I couldn't rip and repair them through CueTools. Also, you may borrow the disk from someone and rip it, as long as you keep original disks, it's legal here where I live.
A lot of good suggestions, everyone. Thank you; please keep them coming.
I do use AccurateRip on EAC when ripping my FLAC copies; but I'm doing an MP3 rip first, and since I delete leading/trailing silence on my MP3s, AccurateRip isn't available there. So as long as AccurateRip returns good results on the FLAC I'm not too overly concerned that the MP3 was faulty.
(I know many will question my ripping of each disc twice - once per format. But I haven't found any other way to preserve all of the metadata, including album covers, that I've painstakingly entered by hand. It's lost in my FLAC to MP3 processing, but there's probably something simple I'm missing.)
I like the dbPowerAmp / CueTools suggestion. I've also thought about using Brasso, toothpaste, car wax, or that hard plastic filler that's used to even out scratches and pits in windshields etc. But I suspect the best solution would be to find a cheap professional resurfacing machine/service...anything that would be less than $1 per disc.
I agree that running EAC for too long on a badly damaged disc can be cause for concern. Right now, I'm trying to rip a single 7:30 track, and EAC is going on 24 hours for that one track. The drive's cooling down 20 minutes for every 80 of reading, but still...
One other thing I've found: updating my drive's firmware seemed to help a little.
Finally, greynol, I was aware of EAC's weakness in grabbing data from damaged tracks (its 8 out of 16 reads threshold), but I have no experience with dbPowerAmp, CueTools or any other CD ripping tool. Could you point me to a thread, or state your own opinion, on which software, in general, is most likely to result in the most accurate reading on a badly damaged CD? This is making two assumptions: (1) that a track passing the AccurateRip check can never be generated from the disc; and (2) that a certain piece of software will be superior on ALL drives. (And don't worry, I'll be searching the forums for answers too )
There is no software that will work the magic you request. Spending more than 10 minutes on a ~4 minute track is a complete waste of time.
You will probably not get anything audibly better than if you simply rip in burst mode, which is what you should be using in instances like this.
There is no software that will work the magic you request.
Heh, I have no delusions that any software (or for that matter, resurfacing etc.) will work magic. But if one piece of software would result in 80% accuracy, and another 90%, I'd prefer to take the 90%.
I'll ask it a different way. In your opinion, which of the available software packages has the most rigorous error-checking procedure in the absence of AccurateRip usage?
And yes, I probably should be using Burst Mode. I'll give that a try tonight.
In this situation it simply does not matter; hence the suggestion to use burst mode which basically levels the playing field for all ripping programs.
(I know many will question my ripping of each disc twice - once per format. But I haven't found any other way to preserve all of the metadata, including album covers, that I've painstakingly entered by hand. It's lost in my FLAC to MP3 processing, but there's probably something simple I'm missing.)
There have been these discussions about foobar2000 losing album art in this conversion, but try the following with Mp3tag (freeware!):
- Create mp3s from your FLACs, using precisely the same folder/filenames. That ensures they are sorted the same way incoming.
- Dump the folder(s) containing the FLACs into mp3tag
- Ctrl-a Ctrl-c
- Dump the folder(s) containing the mp3's into mp3tag
- Ctrl-a Ctrl-v
I agree that running EAC for too long on a badly damaged disc can be cause for concern. Right now, I'm trying to rip a single 7:30 track, and EAC is going on 24 hours for that one track. The drive's cooling down 20 minutes for every 80 of reading, but still...
24 hours for one track, that is more than even I have done. I've let a bad disc in overnight though. It wears the drive, but I have collected a few from end-of-life computers.
I disagree on the argument that secure will not improve over burst for a bad disc. You cannot recover the unrecoverable, but if you have a disc damaged to the point of one unrecoverable error, then chances are that there are lots of recoverables too. For a given disc and drive, only trial and error can tell.
chances are that there are lots of recoverables too
Of course, but do you honestly think this will end up in an audibly better enough result to warrant the trouble? I do not.
If it's just a couple of minor spots that can't be extracted correctly then sure I'm all in favor of it, especially if CTDB can come to the rescue, but for a track that takes an hour or longer?!? Hell no!!!
Don't kid yourself, we are neck-deep in irrational OCD on this one. For those who really care about tiny pieces of good data in an otherwise hosed rip, your time is better spent looking for an alternate copy. If this is an extremely rare or one-of-a-kind recording, I do sympathize.
So as long as AccurateRip returns good results on the FLAC I'm not too overly concerned that the MP3 was faulty.
Don't you worry that there might be a random read error that gives the right data when you rip to FLAC, but the wrong data when you rip to MP3?
Don't kid yourself, we are neck-deep in irrational OCD on this one.
LOL! Truer words were never said.
pdq - I recognize that's a real possibility, but considering how badly damaged these discs are, I figure it's going to be impossible to reconstruct the original tracks. And to be honest, my hearing isn't good enough to hear random bad reads, unless they take the form of pops/skips.
That’s what the percentage describes: how many of the attempts obtained the same (most common) set of samples.
This is not quite correct.
From http://www.exactaudiocopy.de/en/index.php/...tion-questions/ (http://www.exactaudiocopy.de/en/index.php/support/faq/extraction-questions/)
It is the ratio between the number of minimum reads needed to perform the extraction and the number of reads that were actually performed. 100% will only occur when the CD was extracted without any rereads on errors.
Of course. I was rambling as though the figure was for a single sector at a time, but obviously EAC provides only one figure for the entire range (CD, track, whatever). I should try thinking before I post. Thanks for the important correction.
(I know many will question my ripping of each disc twice - once per format. But I haven't found any other way to preserve all of the metadata, including album covers, that I've painstakingly entered by hand. It's lost in my FLAC to MP3 processing, but there's probably something simple I'm missing.)
Indeed, there should be a simple way to perform the conversion and include metadata automatically with no trouble at all.
There have been these discussions about foobar2000 losing album art in this conversion
Do you mean from the past when it could not transfer any art, or are there problems even nowadays that I’ve missed?
So as long as AccurateRip returns good results on the FLAC I'm not too overly concerned that the MP3 was faulty.
Don't you worry that there might be a random read error that gives the right data when you rip to FLAC, but the wrong data when you rip to MP3?
It happening the other way around is the scenario that’s worth worrying about, I would think. But then, in any case, how likely is it that an exactly matching checksum would occur once by chance from an inconsistent CD? (That’s a genuine question, not a trick one. )
Anyway, ripping twice is massively wasteful and should be easy to avoid using either of the methods that have already been suggested and probably many more. That would make such questions purely academic at best.
There is no software that will work the magic you request. Spending more than 10 minutes on a ~4 minute track is a complete waste of time.
I agree entirely - from experience, it just doesn't make things better.
Whereas brasso, and stand-alone CD players,
do. Plus they don't take 24 hours, or destroy your PC's CD drive in the process.
Making EAC re-read data that simply isn't there any more (label side damage), or is completely obscured (bad under side scratch) is pointless. It only makes sense on damage that
sometimes causes a read error, but mostly does not cause a read error. That scenario is very damage and drive dependent. It can help sometimes - but just because it's the default fall-back in some beloved software does not mean it should be the default fall-back when we humans encounter damaged CDs. On some discs, brasso, stand-alone CD player, CTDB, and/or just manually fixing a burst rip, will be preferable.
Cheers,
David.
chances are that there are lots of recoverables too
Of course, but do you honestly think this will end up in an audibly better enough result
Audibly better? If the count of scratch sounds is the yardstick, then sometimes
yes – if all tracks with at least one are deemed equally useless, then
very rarely (only a very few times in thousands of CDs, and likely no full CD was “saved” [edit: until CTDB arrived]). Five years ago I had quite a few rips I could have uploaded seconds from, but for obvious reasons I deleted the least tolerable ones. That said, switching drive (and maybe between dBpoweramp and EAC) was more likely to improve.
to warrant the trouble?
The “trouble” of pushing a button and going to sleep? I'm not OCD'ed enough to sit there staring at it :-o
but for a track that takes an hour or longer?!? Hell no!!!
FWIW, I just opened a few of my biggest .log files, and found an Accurate-verified track that actually took five hours. (Not the best drive though, but I did not know at the time.)
...and I have seen a rip that took hours in secure more which only took minutes in burst mode and was extracted accurately. Lots of things are possible, especially when you are willing to entertain phenomena that exists less than 1% of the time.
Also, I wouldn't be so quick to imply that errors which are benign enough to be ripped accurately in secure mode will necessarily be audible when ripped in burst mode.
...and I have seen a rip that took hours in secure more which only took minutes in burst mode and was extracted accurately.
I was about to go “who does that when the burst rip is good?” but then, EAC can still not be configured to (automatically) follow the dBpoweramp-esque workflow of burst -> AR check -> secure?
Lots of things are possible, especially when you are willing to entertain phenomena that exists less than 1% of the time.
Well this discussion does deal with troublesome discs.
Also, I wouldn't be so quick to imply that errors which are benign enough to be ripped accurately in secure mode will necessarily be audible when ripped in burst mode.
OK, maybe “FWIW” wasn't a sufficient disclaimer, but still I think you turned the “warrant the trouble” argument on its head. If I can get an AR-verified rip in my sleep, why should I instead spend an hour listening carefully through a doubtful track? Except if I am worried over a few dimes' worth of drive wear?
EAC can still not be configured to (automatically) follow the dBpoweramp-esque workflow of burst -> AR check -> secure?
That's correct, it's either burst mode where no re-reads will be performed or secure mode where re-read sets will be performed if triggered. Also, EAC does not make any workflow decisions based on AR results.
why should I instead spend an hour listening carefully through a doubtful track?
How would you then know if a track ripped with secure mode that didn't pass AR would sound any better than a track ripped using burst mode that didn't pass AR?
Except if I am worried over a few dimes' worth of drive wear?
Your drive might not make it through the night, in which case it wasn't exactly a few dimes' worth of wear.
So... I was just looking around online a bit and I found CleanDISCPro (http://www.cleandiscpro.com/). You mail your discs to them (Miramar Florida) and for $3 (or less depending on quantity) they will resurface them. The good news is they won't charge you if they can't repair the disc. Like I said, I didn't have much luck with my little polishing machine, but the service seems worth a try if you have an important irreplaceable disc. (Actually they say they don't play the disc, so I guess you have to ask for a refund if they fail.)
why should I instead spend an hour listening carefully through a doubtful track?
How would you then know if a track ripped with secure mode that didn't pass AR would sound any better than a track ripped using burst mode that didn't pass AR?
Eh ... please read again the part of the sentence that you cut away from the quote.
Except if I am worried over a few dimes' worth of drive wear?
Your drive might not make it through the night, in which case it wasn't exactly a few dimes' worth of wear.
If a drive breaks down after thousands of hours, does it seem appropriate to allocate the cost to the wear from the single last track it was in the process of reading?
Once a drive is in the state where it cannot stand eight hours, isn't it then likely that you would have to replace it very soon anyway?
Eh ... please read again the part of the sentence that you cut away from the quote.
The part about you having a crystal ball telling you that your decision to let the rip go overnight would pay off?!? Yes, I read that.
isn't it then likely that you would have to replace it very soon anyway?
Not if you are to believe what Spoon says about drive stress:
http://www.dbpoweramp.com/secure-ripper.htm (http://www.dbpoweramp.com/secure-ripper.htm)
Eh ... please read again the part of the sentence that you cut away from the quote.
The part about you having a crystal ball telling you that your decision to let the rip go overnight would pay off?!? Yes, I read that.
OK, so I was wrong in assuming good faith. Not the first time.
Not if you are to believe what Spoon says about drive stress:
http://www.dbpoweramp.com/secure-ripper.htm (http://www.dbpoweramp.com/secure-ripper.htm)
The data there proves the other thing that is mentioned previously in this thread: If you leave your ripping program thrashing your drive for minutes/hours, and after lots of work it finally reports an error-free rip, that rip might
still have errors in it.
I've long suspected this, because I have discs where I can get an accurate rip match with one burst read pass on every track on the CD
but one. That track clearly has errors, but after ten minutes effort re-ripping sectors I'm told the rip is now secure - but still doesn't match accurate rip. It seems obvious to me that this track has (on this disc, in this drive) unrecoverable errors, my rip is not correct, and labelling it "secure" is misleading at best.
In that context, it's pretty obvious what's gone wrong, and that something
has gone wrong. Whereas on a disc which has loads of initial errors and/or no accurate rip match, who can
really say whether the final result is accurate?
Cheers,
David.
Well that is just it, drives cannot detect errors 100% of the time (c2 or re-reading), an external source is required. The more re-reads that are done, the greater the chance of a consistent error (c2 can help to an extent with re-reads).
(Secure rippers could actually improve if they would implement a feature logging the troublesome frames, and use that log upon re-ripping with another drive. Then they could pick whatever passes C1&C2 on at least one of the rips.)
I have been trying to get Spoon to add that for YEARS. Maybe Gregory will add it to CueTools ripper...
If the problem is scratches (vs stuff stuck on the plastic) I'd go with filler rather than removing material. Ideally something (usually a wax) with the same index of refraction as the polycarbonate, but in a pinch I've even had saliva work temporarily (obviously it won't permanently affect the disk) and in that case I'd specify a slow rip speed so it doesn't just fly off.