Inaccurate rips – what are those and how to deal with?

Hi these 'inaccurate' rips would typically occur where you have either a damaged disc or a falling CDROM drive, or the CDROM drive is working outside its tolerance. The latter are likely to show up as errors towards the end of the CD on extended play CDs, and the only real answer is to get a higher quality CDROM drive. I believe the  dbpoweramp keep a directory of the top performing models.

Now dbpoweramp is, or can be, set up that if Accurip doesn't get a checksum match, it will re read the CD track using a slower and different access method. If the the two rips match you get a little green secure tick by the rip, if they don't, then dbpoweramp will then try a brute force attempt over the error area.  this, depending on size of error, can take a very long time.. and will either end with a green secure or red insecure rip. Even if insecure unless the CDROM drive is completely unreliable or the CD damage is extensive the error is usually subtle and may result in a small skip or a tiny segment of silence.

in practice I have about one insecure track rip every 500 to 600 or so. Of those out of my collection of over 1200 plus CDs there were  only two CDs from memory that were unusable through damage and unacceptable audible errors.

The real benefit of Accurip is that it speeds up ripping and bypasses the need of doing a slower validation rip.

You are more likely to hear the errors in digital masters than in the rip.. and these tend to show up as little soft occasional ticks. In the early days of CD, these used to be indexed on the inlay card with many classical music CDs.. it seems to have fallen out of fashion now.

Simon

Hi Osprey, I use dB on a rather old Novatech PC, I went through the set-up and benchmarking as laid out by dB, then got on with ripping 1000+ CDs to wave format.

16 had track(s) that had over the 'threshold' number of frame errors (is it 50?), I didn't keep count of those that had errors but could be recovered using the database info., but I think it was a similar number. Only 1 CD was totally rejected by dB and I made a bog-standard 'insecure' copy of that using WMP as it was one of my wife's, and she would be none the wiser!

As I understand it if you only get a few frame errors the software will patch up your copy to the info. held on it's database. If the CD is new to dB, as approx. 30 of mine were (mostly avant-garde, noise and minimal style stuff [like Hafler Trio!] then the software has to assume the disc is ok after it's 7x ultrasecure read passes. As there are no comparatives, a stand-alone bit perfect rip is not really much use for confidence, but is better than nothing!

Having kept a record of the tracks that are insecure rips I did one of two things:

1) Went to the website of the label looking for individual track downloads at 44.1k/16bit to replace my dodgy track rips.

2) Sourced the CDs concerned from friends, and ripped the relevant tracks again then 'pasted' over the inaccurate rips.

As a good proportion of the inaccurate rips were from the same 'label' it was reasonably easy to get most rips fully sorted.

Hi Osprey -

Have ripped about 3300 CDs using an old HP laptop running MediaMonkey. About 1 in 50 rips reported at least one song as inaccurate, but in most cases, re-ripping fixed the problem.

In a few instances, when re-ripping did not help, the inaccuracy turned out to be noise at the end of the final track, and I was able manually edit that out using Audacity.

I did have to give up on (and replace) a few CDs, but no more than a handful.

ATB.

Hook

Not sure if i am just lucky, but have ripped 7000 albums now, and can only remember having 2 or 3 errors throughout, but hat was using the HDX, which may be the difference !

Osprey.

I use dbpoweramp.  I have kept detailed records so that I can revisit some of the "problem" rips at a later date if necessary. It is important to note that most of the ones that are classed as not accurate - rip secure and insecure sound fine.

However results of rips of 734 CDs

11 not in accuraterip at time of ripping

16 whole CDs classed as secure

30 additional individual tracks classed as secure

17 individual tracks insecure

4 CD rips aborted

2 individual tracks not ripped

So two conclusions: one is that I seem to have high error rate and it might be sensible to look at my CDROM drive and the second is that I have CD collection envy based on the numbers that other posters have.

Hi Osprey,

I use dbpoweramp to rip 1800 cds in wave on a rarely used 2 year old pc, I would say about 1 in 30 have inaccurate tracks so would seem I have a higher than average fault rate, however in playback through the asset network on my SuperUniti 250.2 setup there seems no audible degradation on the said tracks. I think there have been 2 cds where discs would not rip at all.

Rob.

Using data posted to accuraterip (214 million CDs ripped) the average is a 2% error rate, so 1 disc in 50. Many of those will be CDs which look perfect to the naked eye, yet the CD has a manufacturing defect.

Modern CD drives interpolate (silence) errors, that is why someone can rip 5000 CDs with iTunes, and not hear the errors, because it can be difficult to hear these small silence sections.

@Hook thanks for sharing your experience. Using Audacity to trim the problem tracks is a good tip also. So far I have only recorded vinyl by it but of course it can be used for this purpose too.

@Paul you have been lucky indeed   but HDX might have helped also...

@Forester you have very good records on your results. I have thought that I might try to re-rip some of the problematic CDs with another PC to see if that gives better result and it would also help to determine if I need to consider a new CDROM drive (or laptop). I have a similar size collection as you (and growing only slowly) and sometimes I can be a bit envious too when I see very large ones. However, I rarely listen to my records as background music (I use radio for that) and since most of them are still in rotation I do not have to hear the same songs/performances too often (once or twice a year).

@Rob T your experience seems to be closer to mine regarding the frequency of the inaccurate tracks and also the fact that besides one CD I cannot hear the degradation either. I have not come across an item yet which would not rip at all though.

@mrspoon thanks for sharing the official statistic and also for the further information about the technical background - it is appreciated.