Cache and C2 pointers on HL-DT-ST DVDRAM GT30N

[LarsG]

I have tried to find out whether my drive caches audio and supports C2 pointers but is increasingly confused. Hope somebody here can help me clarify this.

Audio Cache
I have used DBpoweramp to detect the cache with 3 different CD’s. In all cases the cache is detected to be zero. According to http://www.daefeatures.co.uk the cache seems to be 37 KB however and according to DBPoweramps informations on CD ROM Technical Details, the cache is 1 MB – this is also what I find on the web, when searching for info on the drive’s cache (but this is possibly data cache – and something different?)
So, what is the audio cache? And should I just use the “nobrainer” solution in setting up cd ripper, and set cache to 1024 KB?

C2 Error Pointers
I have tested my drive as described in DBpoweramps CD ripper setup guide (using a CD with a black triangle, http://www.dbpoweramp.com/cd-ripper-setup-guide.htm ). According to the test and according to http://www.daefeatures.co.uk my drive do support C2 error pointers. However, according to DBPoweramps informations on CD ROM Technical Details, the drive do not support C2 error pointers. So which information is correct?

Best regards
Lars

[Spoon]

Re: Cache and C2 pointers on HL-DT-ST DVDRAM GT30N

>So, what is the audio cache? And should I just use the &*8220;nobrainer&*8221; solution in setting up cd ripper, and set cache to 1024 KB?

Yes

If the test detects c2 pointers part way through the scan, then they are supported.

[eaglescout1998]

Re: Cache and C2 pointers on HL-DT-ST DVDRAM GT30N

So, what is the audio cache? And should I just use the &*8220;nobrainer&*8221; solution in setting up cd ripper, and set cache to 1024 KB?

That's what I do.

I have tested my drive as described in DBpoweramps CD ripper setup guide (using a CD with a black triangle, http://www.dbpoweramp.com/cd-ripper-setup-guide.htm ). According to the test and according to http://www.daefeatures.co.uk my drive do support C2 error pointers. However, according to DBPoweramps informations on CD ROM Technical Details, the drive do not support C2 error pointers. So which information is correct?

In my experience with HL-DT-ST drives, I have to keep C2 pointers off. The reason is because enabling C2 error detection produces too many false positives. When you performed the C2 test, did it detect a C2 error almost immediately?

[LarsG]

Re: Cache and C2 pointers on HL-DT-ST DVDRAM GT30N

Spoon: Thanks for your answer - I'm impressed by your clarity and speed

eaglescout1998: I have tested again with two different CD's. C2 errors is detected after 8% and 16% of the test.

If I enable C2 error pointers and it is not properly supported, I risk to many false positives. What is the consequence of not enabling C2 error pointers? As I understand it, it means that the drive will have to read every block twice and compare them. Does this mean, that reading a CD will take at least twice as long?

[Porcus]

Re: Cache and C2 pointers on HL-DT-ST DVDRAM GT30N

What is the consequence of not enabling C2 error pointers? As I understand it, it means that the drive will have to read every block twice and compare them. Does this mean, that reading a CD will take at least twice as long?

No, it doesn't. C2 error pointers tells dBpoweramp where there are errors, and dBpoweramp will act accordingly.
If implemented badly, it could tell dBpoweramp that there are errors everywhere, and dBpoweramp will start re-reading all over the place to eliminate these errors.

Now if they are well supported but turned off, some errors might go unnoticed. To compensate for this, dBpoweramp (secure mode) uses two burst passes rather than one, in an attempt to hit an AccurateRip match -- provided that this is necessary. Also, it tries to combine results to get an AR match. If no AR match is found, it will start re-reading. And here, the absence of C2 pointers means that errors might go unnoticed. Finally, if it isn't satisfied, it will re-read individual frames until it gets 10 matches or hits maximum; if C2 is enabled, results flagged erroneous will be dropped (and not qualify in the "10"), while the absence of C2 pointers may again leave errors unnoticed.

So if there is no AR match, dBpoweramp will compensate for lack of C2, by one more burst. Later on, no C2 may mean that it fails to detect errors (and those errors would, if detected, of course take more time).

[LarsG]

Re: Cache and C2 pointers on HL-DT-ST DVDRAM GT30N

Thanks again Porcus, I see that it's more complicated, than I thought.

I understand now that what you meant by "false positives" in your first post was reported errors, when there are none. I thought it was rips incorrectly described as correct.

I'll try to spell out the consequences to see, if I understand you correctly.

If I have an AR match for a song, the rip of this song is OK - and I need not worry about C2 or not C2 - as far as rip quality is concerned.

If C2 error detection is enabled it could mean (1) DBpoweramp reporting a lot of errors when there are none (badly implemented), or (2) with no AR match a more time-consuming ripping proces, that however detects errors, which would be incorrectly described to be correct with C2 not enabled (C2 error pointers correctly implemented)

Is this correct?

[LarsG]

Re: Cache and C2 pointers on HL-DT-ST DVDRAM GT30N

Another thing: If I do not enable C2 pointers, I could inadvertently submit rips with errors to the Accurate Rip database?

I guess there so many different possible errors, that this will not mean, that anyone else could get a false AR match, even if I try ripping several times.