IPB

Welcome Guest ( Log In | Register )

2 Pages V   1 2 >  
Reply to this topicStart new topic
Which drives are caching audio data?
norton
post Sep 1 2004, 14:35
Post #1





Group: Members
Posts: 45
Joined: 21-August 04
Member No.: 16423



EAC => What about the 'caching audio data'!?


I have read almost everything on this forum about this subject.

No doubt about it: If the Drive you are using to rip with EAC is caching audio data, the 'Drive caches audio data' Box (Drive options... / Extraction method) have to be checked (ticked).

But how can you know if your Drive is caching audio data?

Of what I have read on this forum, you should run 'Detect Read Features' (Drive options... / Extraction method / Detect Read Features). If the result is: Caching: No => You're OK! If the result is: Caching: Yes => You have to checked the 'Drive caches audio data' Box.

Well... The problem is, many people are saying that the 'Detect Read Features' is NOT accurate!!! (How do they know that??) That means even if you perform 'Detect Read Features', you are still not sure if your Drive is caching audio data, or not.


MY QUESTION IS: How can I know if my Drive is Caching audio data, or not?


Thanks for helping,


norton

This post has been edited by norton: Sep 1 2004, 16:21
Go to the top of the page
+Quote Post
Pio2001
post Sep 1 2004, 18:49
Post #2


Moderator


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73



The EAC detection feature is not accurate because sometimes it reports caching, sometimes not, for the same drive.
Sometimes, EAC reports no caching while Feurio does. We have 2 or 3 examples of this, but none of the opposite. It is too few to draw any conclusion of Feurio's detection accuracy against EAC.

To know if your drive caches :
Ask EAC. If it says yes, then it caches.
If it says no, try Feurio. If it says yes, then it caches. If it says no, try the following :

If your drive supports C2

Enable C2, uncheck cache, and find a CD lightly damaged, that triggers error correction (red lights), but still give you no errors on a range which triggers the first red bar at least 20 times.
Test and copy this range.

If the CRC are the same, your drive doesn't cache.
If the CRC are different,

Check cache and try again.

If you can't get "no error" anymore, it is likely that your drive caches. Whatever, leave it checked, since it obviously make the secure mode work properly.
If you still get no errors, but CRC are OK, your drive caches.
If you get no errors and CRC still mistmatch, find a less damaged CD, or try a shorter range.

If your drive doesn't support C2,
...But also valid if it supports C2.

Check cache, and try to find a CD lightly damaged, so that it reports no errors, but which the first red bar filled at least 20 times, and CRC are OK.
Uncheck cache. Test and copy.

If you get errors, find a less damaged CD or range.
If you get 100 % quality (no red bars) and CRC mismatch, your drive caches.
If you get 100% quality and CRC are OK, find a more damaged range or CD and try again.
If you still get red bars, but CRC mismatch, your drive caches.
If you still get red bars and CRC are OK, your drive doesn't cache.
Go to the top of the page
+Quote Post
markanini
post Sep 1 2004, 18:59
Post #3





Group: Members
Posts: 552
Joined: 22-December 03
From: Malmö, Sweden
Member No.: 10615



Pio2001, that's the best explaination I've seen on how to set cache and C2 options in EAC. Should be added to the FAQ IMO
Go to the top of the page
+Quote Post
norton
post Sep 1 2004, 22:04
Post #4





Group: Members
Posts: 45
Joined: 21-August 04
Member No.: 16423



QUOTE
To know if your drive caches :
Ask EAC. If it says yes, then it caches.
If it says no, try Feurio. If it says yes, then it caches. If it says no, try the following :


Tested many times:

EAC said: NO
Feurio said: NO (It seems that your device isn't able to cache audio data!)


QUOTE
If your drive doesn't support C2,
...But also valid if it supports C2.

Check cache, and try to find a CD lightly damaged, so that it reports no errors, but which the first red bar filled at least 20 times, and CRC are OK.
Uncheck cache. Test and copy.

If you get errors, find a less damaged CD or range.
If you get 100 % quality (no red bars) and CRC mismatch, your drive caches.
If you get 100% quality and CRC are OK, find a more damaged range or CD and try again.
If you still get red bars, but CRC mismatch, your drive caches.
If you still get red bars and CRC are OK, your drive doesn't cache.


My drive doesn't support C2.

I found those steps a little bit confusing (sorry), but I'm trying to understand those right at this moment.

RIGHT NOW, I am working to find a CD that is enough scratched to fill red bars at least 20 times, but not too scratched, so that it reports no errors. Well... I don't remember having ripped that kind of CD (250 CD+). When I ripped a CD that filled red bars more then 20 times, there was always 'errors'! BUT, I'm trying to find one at the moment.

But I have to say, I'm kind of thinking that, it's a BIG waste of time!... I mean, I can't figure wasting many hours to find a good CD for that test (2 hours now), while I could buy another Drive that I KNOW (now) that caches audio or not. The only reason I keep trying to find out If my Drive caches audio or not, it is beacause I have already ripped 250 CDs and I want to know if they have been ripped OK.


I was wondering... After the EAC test done, and the FEURIO test done, with both NO like answer, how much confident should I be? 80%? 90%? 99.9%?


Thanks again for helping me,


Regards,

norton
Go to the top of the page
+Quote Post
norton
post Sep 1 2004, 23:00
Post #5





Group: Members
Posts: 45
Joined: 21-August 04
Member No.: 16423



Well... I QUIT!!

I have spent the last 12 hours to figure out if YES or NO my Drive (ASUS CD-S520/A4) is able to cache audio data. I still don't know. headbang.gif

If someone comes to me with the answer of my question (with proof), I'm sending to this someone 20 bucks right now through Paypal. mad.gif


Again, thanks again for helping me.


Regards,

norton
Go to the top of the page
+Quote Post
Pio2001
post Sep 1 2004, 23:06
Post #6


Moderator


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73



QUOTE (norton @ Sep 1 2004, 11:04 PM)
I was wondering...  After the EAC test done, and the FEURIO test done, with both NO like answer, how much confident should I be?  80%?  90%?  99.9%?
*


I'd say more than 95 %
Go to the top of the page
+Quote Post
dreamliner77
post Sep 2 2004, 00:03
Post #7





Group: Members
Posts: 2150
Joined: 29-June 02
From: Boston
Member No.: 2427



QUOTE (norton @ Sep 1 2004, 06:00 PM)
Well...  I QUIT!!

I have spent the last 12 hours to figure out if YES or NO my Drive (ASUS CD-S520/A4) is able to cache audio data.  I still don't know. headbang.gif

If someone comes to me with the answer of my question (with proof), I'm sending to this someone 20 bucks right now through Paypal.      mad.gif


Again, thanks again for helping me.


Regards,

norton
*



I'd say be safe and just check "drive caches...". it shouldn't have any bad side effects anyway.


--------------------
"You can fight without ever winning, but never win without a fight." Neil Peart 'Resist'
Go to the top of the page
+Quote Post
norton
post Sep 2 2004, 00:32
Post #8





Group: Members
Posts: 45
Joined: 21-August 04
Member No.: 16423



QUOTE
I'd say be safe and just check "drive caches...". it shouldn't have any bad side effects anyway.


I have already ripped 250 CDs with the 'Drive caches audio data' Box unchecked. I tought that my drive was not able to cache audio data, until a member on this forum told me that he have the same drive, and it DOES cache audio data.

I want to know if my 250 CDs already ripped, have been ripped OK (!??)

crying.gif
Go to the top of the page
+Quote Post
kwanbis
post Sep 2 2004, 03:20
Post #9





Group: Developer (Donating)
Posts: 2362
Joined: 28-June 02
From: Argentina
Member No.: 2425



just listen to them ... or a random sample


--------------------
MAREO: http://www.webearce.com.ar
Go to the top of the page
+Quote Post
norton
post Sep 2 2004, 03:48
Post #10





Group: Members
Posts: 45
Joined: 21-August 04
Member No.: 16423



QUOTE
just listen to them ... or a random sample


Well... I will not listen 250 CDs! But for the last 2 days, while looking on forums about the answer of my question (and others), I'm listening (random) those files... I didn't heard anything wrong about them. But I have listened what, 15-20 CDs?... This is not very very accurate! wink.gif


norton

This post has been edited by norton: Sep 2 2004, 03:48
Go to the top of the page
+Quote Post
VolMax
post Sep 2 2004, 05:26
Post #11





Group: Members
Posts: 60
Joined: 24-February 04
Member No.: 12234



Caching or not, it doesnt matter for me now.

The quality of optic and internal device error correction is much more relevant.

I have Teac-CDW548E, ASUS DVD-ROM E616, and have access to some NEC RW drives. Because TEAC caches audio(old NEC and asus - not) EAC tells me that
TEAC<<NEC<ASUS
in terms of quality.

I conducted some test (Nero CDSpeed) on specially scratched test audio cd. And got WERRY intresting results:

ASUS DVD-ROM E616
-----------------------------------------------------------------------------------------------
On The Fly Copying Simulation
-----------------------------------------------------------------------------------------------
Copying at 1 X
OK
Copying at 2 X
data errors
Data Errors: 1
Copying at 4 X
OK
Copying at 6 X
data errors
Data Errors: 16541
Copying at 8 X
data errors
Data Errors: 57
Copying at 10 X
data errors
Data Errors: 20
Copying at 12 X
OK
Copying at 16 X
too slow

The ONE speed working correctly is 12x. Lower or higher speeds produced unrecoverable errors. I tried all speeds to ~40x without much results. Firmware of ASUS seems to be flawed. NEC drive could not read this disc without errors at all.

Teac-CDW548E
-----------------------------------------------------------------------------------------------
On The Fly Copying Simulation
-----------------------------------------------------------------------------------------------
Copying at 1 X
OK
Copying at 2 X
OK

Copying at 4 X
OK
Copying at 6 X
OK

Copying at 8 X
OK

Copying at 10 X
OK

Copying at 12 X
OK
Copying at 16 X
too slow


Later i scratched cd to make unrecoverable errors for TEAC. I lost result of various test with binary comparision, but i clearly remember TEAC even in burst mode have more than TWICE LOVER ERROR RATE! It even slowdouns from 20x to lower speed on scratches even in BURST mode! I was shocked. Because I bought asus drive after teac for better extraction results :Z

Later testing shows me using cache overreadind lowers error rate insignificantly. But slowdowns from ~20x to <2x. Also i found that using С2 for error correction more significantly lowered error rate. Its weird that latest eac doesnt support this feature.

My current preferences:
NEC<<ASUS<TEAC
Now i ripping with TEAC (EAC_09b4 settings:
Accurate stream: ON (for quality reasons)
Drive caches audio data: OFF (for speed and security reasons)
Drive is capable of retreiving C2 ON (for quality reasons)
Use C2 for error correction YES (for quality and speed reasons)
)

This is ONLY for this particular situation, DO NOT use without rigorous testing on your drive!

What the point overriding cache and trying to re-read sector if drive make everything it can fo recover errors? It could even damage drive...

P.S. Test takes 3 full days...
Go to the top of the page
+Quote Post
Pio2001
post Sep 2 2004, 12:05
Post #12


Moderator


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73



Your results are interesting, but they are all about quality, not secureness.

QUOTE (VolMax @ Sep 2 2004, 06:26 AM)
The quality of optic and internal device error correction is much more relevant.


It's not my opinion. I've got a Teac too (540e), that outsmarts my other drives in reading quality, but it's not secure (because of a firmware bug). I get glitches in the wave files that it produces. With other drives, it is worse, but at least the secure mode tells me about the problem.

QUOTE (VolMax @ Sep 2 2004, 06:26 AM)
Its weird that latest eac doesnt support this feature.


I was the only one to have tested it properly, but it didn't work on my drive. Since no one, when it was out, did understand what it was for, and I reported it not to work, it was given up. It also generated a mess of a newbie confusion in the EAC forum.

QUOTE (VolMax @ Sep 2 2004, 06:26 AM)
What the point overriding cache and trying to re-read sector if drive make everything it can fo recover errors? It could even damage drive....


In order to detect the remaining ones, of course.

QUOTE (VolMax @ Sep 2 2004, 06:26 AM)
P.S. Test takes 3 full days...
*


I'm not surprised.
Go to the top of the page
+Quote Post
VolMax
post Sep 2 2004, 15:54
Post #13





Group: Members
Posts: 60
Joined: 24-February 04
Member No.: 12234



QUOTE
Your results are interesting, but they are all about quality, not secureness.

What is secureness? Capability of producing CONSTANT good results? Error-free? Yes, a some bug in firmware (in C2 reporting, in error correction, etc.) could result problems (unnoticed errors in stream). But i tested on one normal CD, result was bit-identical. Better results on scratched CDs and possibly not very bad on normal was sufficient for me... But sacrifacing good drive for cache override...
Do you think that this particular drive will perfom better if cache will be removed from firmware? It will break burst mode perfomance (now im not so scared to use even cdex in burst mode!). It will probably produce errors in stream if there is not enough CPU resources...

Cache is not really bad feature, but it is good only with correct firmware and good error correction... It will be better IMHO if drive have big cache and reread bad sectors internally... Dreams... sad.gif

Do the situation with ASUS better? It could be used only with grabbers supporting ForbidReducingSpeed feature... And speed will be ~4x.

I do not even talk about NEC.

Even if drive (say MEC) could report with 100% probability that it COULD NOT CORRECT error somewhere what is good for? Rereading will result same error msg. With TEAC i most probably cannot be sure what CD was ripped without errors, but i could be sure i will get much more errors using that 100%C00l MEC drive...


QUOTE
I get glitches in the wave files that it produces.

Whoops! Was this particular or usual situation? My drive could suffer from same probl... How you detected that?


QUOTE
I was the only one to have tested it properly, but it didn't work on my drive

Enabling this feature decreased quantity of errors not more than 10%
Go to the top of the page
+Quote Post
norton
post Sep 2 2004, 16:07
Post #14





Group: Members
Posts: 45
Joined: 21-August 04
Member No.: 16423



About ASUS CD-S520/A4...

On the ASUS website, in the specs of the drive, we can read:

Data Buffer : The drive has a data buffer, which is implemented for look-ahead cache. This buffer has a size of 128K bytes.

I was thinking that maybe THIS is an answer of my question. Does that say that this drive is caching audio data?

I'm not sure.

Someone wrote this the other day (when answering about my question about if yes or no, my drive caches audio data):

This shouldn't be counfused with the drive's cache/buffer. This is to reduce the jitter and protect against buffer underruns.


Someone can confirm, if what I have read on the Asus website isn't related to the 'Caching audio data' question???



Thanks again for helping,


norton
Go to the top of the page
+Quote Post
Never_Again
post Sep 2 2004, 16:50
Post #15





Group: Members (Donating)
Posts: 698
Joined: 31-March 04
From: NYC
Member No.: 13152



QUOTE (VolMax @ Sep 2 2004, 10:54 AM)
What is secureness?


Extraction errors reporting.

QUOTE
It will be better IMHO if drive have big cache and reread bad sectors internally... Dreams... sad.gif


This makes no sense. If a drive reads the sectors correctly the first time, there is no need to re-read. If it doesn't, no amount of re-reading from the cache will help.

If interpolation is what you meant, have a look at the Plextor Premium. It does that in PlexTools Pro, though I don't know if its 8MB of cache has anything to do with that.

edit:
P.S. VolMax and Pio: what feature are you two talking about?


This post has been edited by Never_Again: Sep 2 2004, 16:50
Go to the top of the page
+Quote Post
minix
post Sep 2 2004, 23:53
Post #16





Group: Members
Posts: 328
Joined: 19-January 03
Member No.: 4639



QUOTE (norton @ Sep 1 2004, 11:04 PM)
I was wondering...  After the EAC test done, and the FEURIO test done, with both NO like answer, how much confident should I be?  80%?  90%?  99.9%?

Make sure DMA is enabled for your drive.
Feurio test will report no cache for my LG burner if DMA is disabled.
With DMA enabled, it will report 24KB of audio cache.
Go to the top of the page
+Quote Post
norton
post Sep 3 2004, 01:31
Post #17





Group: Members
Posts: 45
Joined: 21-August 04
Member No.: 16423



QUOTE
Make sure DMA is enabled for your drive.
Feurio test will report no cache for my LG burner if DMA is disabled.
With DMA enabled, it will report 24KB of audio cache.



I have to say, I didn't tought about that. But I just checked, and DMA is enable.


norton
Go to the top of the page
+Quote Post
VolMax
post Sep 3 2004, 04:58
Post #18





Group: Members
Posts: 60
Joined: 24-February 04
Member No.: 12234



QUOTE
VolMax and Pio: what feature are you two talking about?

Drive coud report for ripping program AFAIK C1 (always unrecoverable) and C2 errors. C1 reported by any drive and if are present, there are serious problems in quality of resulting file. Drive have access to some redundant data on CD and if detected that C2 error occured, tries to correct it internally using special algorithms (NOT interpolates, with small errors reconstruction of original information IS possible without rereading, with help of renundancy). Or even try to PHYSICALLY reread from surface of CD (depends on quality of realisation such algos in firmware). Does error corrected perfectly or was interpolated, drive will report C2 error. Program could tell drive to reread sector (program does not know, have error been corrected already). Drive will reread from CACHE (???) if cache overriding is off. (WHY decreased error rate?!)

"Use C2 for error correction" changes algorythm of error detection from "rereading sectors ALWAYS twice and reread (3rd,4,..) if unequal" to "Reread only if drive reported c2 error"

This is less secure IMO, but SPEEDYYYY. However that may be without cache overriding on, rereading twice to detect error is completely useless (they always be equal). [ADD]It should be THE option to copy khm... such special US CDs that even being new copying with clicks... This clicks for some reason are mastered and rereading returns equal but not correct information... IMHO its the real reason to remove it with some other features wink.gif[/ADD]

Note that most drives (possibly and my TEAC) reports C2 inaccurate.


QUOTE
This makes no sense. If a drive reads the sectors correctly the first time, there is no need to re-read. If it doesn't, no amount of re-reading from the cache will help.

Drive have access to redundant information on CD. It 100% sure, occured C2 or not. And it have more capabilities to restore original than whatever else. But good error correction is hardly implemented because this is not a selling point for consumers. Maybe plextors... but they aren't cheap wink.gif

This post has been edited by VolMax: Sep 3 2004, 05:29
Go to the top of the page
+Quote Post
liekloo
post Oct 17 2004, 18:39
Post #19





Group: Members
Posts: 456
Joined: 22-March 02
From: Belgium
Member No.: 1596



QUOTE (Pio2001 @ Sep 1 2004, 06:49 PM)
Uncheck cache. Test and copy.
...
If you still get red bars, but CRC mismatch, your drive caches.
...

I fully agree with the rest of your instructions, but here is a sentence that contradicts with what I have in mind: I don't see how a caching drive with cache unchecked, can cause any red bars.
Here is my reasoning: AFAIK the red bars are only triggered due to a difference in the 2 audio sector reads. However the caching drive with cache unchecked will always return the same cached data, without any differences and thus without triggering the red bars.

Am I overlooking something?


--------------------
"E S S E N T I A L" Guide for E A C :

http://users.fulladsl.be/spb2267/
Go to the top of the page
+Quote Post
Never_Again
post Oct 19 2004, 08:34
Post #20





Group: Members (Donating)
Posts: 698
Joined: 31-March 04
From: NYC
Member No.: 13152



QUOTE (VolMax @ Sep 2 2004, 11:58 PM)
Drive coud report for ripping program AFAIK C1 (always unrecoverable) and C2 errors. C1 reported by any drive and if are present, there are serious problems in quality of resulting file.
I think you got your terminology mixed up. C1s are not unrecoverable. Do you mean CUs?
QUOTE
Drive have access to redundant information on CD. It 100% sure, occured C2 or not. And it have more capabilities to restore original than whatever else.

I cannot make any sense of what you are saying here, sorry.
QUOTE
But good error correction is hardly implemented because this is not a selling point for consumers. Maybe plextors... but they aren't cheap ;)
*

Plextors, definitely. Worth every penny, too. =)
edit: quotes trimmed

This post has been edited by Never_Again: Oct 19 2004, 08:35
Go to the top of the page
+Quote Post
Pio2001
post Oct 23 2004, 03:02
Post #21


Moderator


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73



QUOTE (liekloo @ Oct 17 2004, 07:39 PM)
the red bars are only triggered due to a difference in the 2 audio sector reads. However the caching drive with cache unchecked will always return the same cached data, without any differences and thus without triggering the red bars.

Am I overlooking something?
*


It is only true if the cache is bigger than 62 kB. If it is smaller, only a part of what EAC reads is cached (EAC reads by bursts of 62 kB).
Go to the top of the page
+Quote Post
rutra80
post Oct 23 2004, 04:57
Post #22





Group: Members (Donating)
Posts: 810
Joined: 12-September 03
Member No.: 8821



QUOTE (Pio2001 @ Oct 23 2004, 04:02 AM)
It is only true if the cache is bigger than 62 kB. If it is smaller, only a part of what EAC reads is cached (EAC reads by bursts of 62 kB).
*

Does cache = drive's buffer? If my DVD writer has a 2MB buffer, and my DVD-ROM drive a 512KB buffer, then if they were caching, would they cache 2MB and 512KB of audio data?

This post has been edited by rutra80: Oct 23 2004, 05:03
Go to the top of the page
+Quote Post
Pio2001
post Oct 24 2004, 00:11
Post #23


Moderator


Group: Super Moderator
Posts: 3936
Joined: 29-September 01
Member No.: 73



No, the audio cache can be smaller than the data cache.
Go to the top of the page
+Quote Post
mai9
post Oct 24 2004, 01:51
Post #24





Group: Members
Posts: 100
Joined: 17-March 03
Member No.: 5542



QUOTE (dreamliner77 @ Sep 2 2004, 12:03 AM)
I'd say be safe and just check "drive caches...".  it shouldn't have any bad side effects anyway.
*

that's what I always thought.

If you don't disable it and your drive does cache it, then you're doing it wrong. if you disable it and it doesn't cache then nothing happens.
Go to the top of the page
+Quote Post
rutra80
post Oct 24 2004, 05:17
Post #25





Group: Members (Donating)
Posts: 810
Joined: 12-September 03
Member No.: 8821



QUOTE (Pio2001 @ Oct 24 2004, 01:11 AM)
No, the audio cache can be smaller than the data cache.
*

How can I determine how much audio does my drive cache (if it does)?
Go to the top of the page
+Quote Post

2 Pages V   1 2 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 30th October 2014 - 15:16