Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Where can you check drive speed and accuracy? (Read 3456 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Where can you check drive speed and accuracy?

There use to be a few pages out there where you could compare your cd drive to see how fast, if it had C2 error checking and hos accurate it was ripping. I have access to a bunch of drives and wanted to check them out to see which was best. I imagine sata drives are faster than ide.


Where can you check drive speed and accuracy?

Reply #2
http://daefeatures.co.uk ?



I did check there, I have TSST corp drive TS-HS653T and was wondering how I could test it.  I also have HP GH80N HP and want to test these guys out.

BTW, some of these columns confuse me.  HTOA, Overread, sample offset and write offset.  What is good and what is considered bad?

Where can you check drive speed and accuracy?

Reply #3
BTW, some of these columns confuse me.  HTOA, Overread, sample offset and write offset.  What is good and what is considered bad?

The ability to read ‘Hidden Track One Audio’ from the INDEX 00 of the first track, which you’d hear if you rewound it in ye olde style player, is good – in case you have any of the relatively minor number of CDs that include such a region.

The ability to overread into the lead-in or lead-out only matters if your drive has a negative or positive read offset, respectively – in which case it’s good. Edit: iDerp: Most drives do, so overreading is pretty much a universal positive.

A read offset exists when your drive does not actually read audio from exactly where it’s supposed to. (How the CDDA format was allowed to be this ambiguous is another matter entirely!)

A write offset is like a read offset but for writing.  Correcting for this is done in order to ensure that burned CDs, when read in another drive, do not have an offset; compensating for it in that case would require knowledge of the writing drive’s offset as well as the reader’s.

Have you read any guide to these basic features of DAE, such as the FAQ for Exact Audio Copy? I can’t quite tell whether you were asking for definitions or just some other method of ranking these features (which doesn’t really make sense); if the former, ignore me, I guess!

Where can you check drive speed and accuracy?

Reply #4
HTOA is the drive's ability to read "hidden track one audio"—audio that comes in track 1's "pregap", which you'd have to "rewind" from the beginning of track 1 in order to ever hear. There aren't very many albums which have HTOA. See the list on Wikipedia.

All drives, when told to go to a certain place on the disc, actually go to the approximate location, due to manufacturing tolerances. As a result, the audio samples are "offset" by a certain tiny amount. That is, when the drive is told to read the first 588-sample block of track 2, it actually returns, say, the last 6 samples of the last block of track 1, plus the first 582 samples from the first block of track 2. That's the right number of samples, mostly from the right location, but they are "offset" by -6. So the rip's track boundaries will be 6/44100ths of a second too soon, as compared to how they were mastered on the disc. The ripping software can correct for the offset by reading extra blocks and shifting the samples as needed. The correction offset in this example would be +6. However it's unlikely the CD was mastered with sample-perfect boundaries anyway; one run of pressings might use more than one master, each with its own offset.

Offset is also introduced when CDs are mastered or when CD-Rs are burned. For the latter, this is the "write offset" and only affects you if you are burning CD-Rs and are trying to make the burned CD have no offset vs. the original. The burning software might allow you to compensate for the drive's write offset.

The de facto standard used for offset correction was later determined to be slightly off, itself, so when you "correct" the offset, it's still not quite right, but we're talking about a few milliseconds, so you needn't worry about it. For purposes of comparing your rips to other people's via AccurateRip, you have to use the de facto standard.

Overreading is reading beyond what the drive thinks is the beginning or end of the CD. When there's an offset that you're "correcting", ideally the drive will be told to read at least one block beyond the audio boundaries. Many drives can't do this, so the ripping software may give you the option to fill in the unreachable samples with silence (zeroes), which is probably what they were anyway.

So:
  • The amount of offset a drive has doesn't matter. The biggest offsets are only like 1/10th of a second, usually much less.
  • Not being able to read HTOA only really matters if you have a CD with HTOA.
  • Not being able to overread doesn't really matter; all CDs are mastered with plenty of silence at the beginning; and at the end, if there's no silence, at worst, you lose a tiny bit of the fade-out. (By tiny bit, I mean, probably less than 20 milliseconds).

The only thing along these lines that you might care about is the drive's tendency to have read errors on clean, unscratched, ordinary CDs. This isn't monitored very well, but you might be interested in Spoon's list. We don't know the condition of discs that people were ripping, or the condition of the drives, but the list reveals that the worst drives only had problems on 10% of the CDs, and most other drives only had problems on ~5% or less.

However I wouldn't even worry about that much. You don't need to test your drive. Just rip some clean, popular CDs with a ripper like EAC, CueRipper, or XLD. If these kinds of CDs consistently rip without errors and have matching rips in AccurateRip or CTDB, then your drive tests out OK.

As for speed, there's an upper limit to the read speed, and ripping generally goes a certain degree slower than that. Don't worry about it.

Where can you check drive speed and accuracy?

Reply #5
The de facto standard used for offset correction was later determined to be slightly off, itself
Yes, by 30 samples, so…

Quote
so when you "correct" the offset, it's still not quite right, but we're talking about a few milliseconds
Nope, only 0.68!

 

Where can you check drive speed and accuracy?

Reply #6
Hey I really appreciate both the responses.  Really helps a lot