IPB

Welcome Guest ( Log In | Register )

AccurateRip - Future Direction
spoon
post Feb 21 2008, 19:14
Post #1


dBpowerAMP developer


Group: Developer (Donating)
Posts: 2748
Joined: 24-March 02
Member No.: 1615



It has been brought to my attention that the CRC used in AccurateRip is not doing its job propperly, in laymans terms the Right Channel rolls out of the CRC Calculation every 1.5 seconds (that is 1st sample right channel is used 100%, by the 65535 sample it is not used, 65536 sample it is used 100% again, this repeats over and over). It is estimated that effectively 3% of the data is not getting into the CRC (at a 97% coverage, I stand behind AccurateRip @ 97% is better than most (? all) c2 implementations). Going back over the early AccurateRip code it seems the design of the CRC is fine, just the implementation (L and R channels were supposed to go in seperately, but were optimized to both go in without bringing down the upper 32 bits).

Steve will post his findings in detail on his discovery.

It is a relatively easy fix (detailled below), however this presents an opportunity, which was not around when AccurateRip was first implemented (the understanding of different CD pressings and how they were implemented was almost non-existing).

----------------------------
1. Fix: Fix the algorithm so all the data is used, both new and old CRC are calculated, new checked first, old second (with less Accuracy). New submissions would effectively appear as different pressings in the database.
----------------------------
2. Fix : Change the CRC algorithm to something like CRC32, the reason it was not used in the first place, was tracks 2 to x-1 would match the CRC presented in EAC, but 1 and last would never, causing confusion, the CRC could be XOR'd to avoid this confusion.
----------------------------
3. Fix & Additional Development: Use CRC32 and the old CRC (there is lots of data in the database), new CRC32 would go into a parallel 2nd database, increasing the strength of the CRC to almost 64 bits (not taking into account the flaw). Back end there is little changes to be made, both databases are the same design.
----------------------------
4. Fix & Additional Development: Use a different hash, MD5, sha-1, these would increase storage of the database by 5x (160bits of sha-1).
----------------------------
5. Brainstorm a method of having a hash which would be resistant to pressings, yet still be feasable for a CD ripper to rip track rather than whole CD based (and not have the need to read outside of the track).
----------------------------
6. ???

Bear in mind the existing database before construction takes up some 14 GB.


--------------------
Spoon http://www.dbpoweramp.com
Go to the top of the page
+Quote Post

Posts in this topic
- spoon   AccurateRip - Future Direction   Feb 21 2008, 19:14
- - Steve Gabriel   My information about the implementation of the Acc...   Feb 21 2008, 20:48
|- - Steve Gabriel   Reading Spoon's initial post to this thread I ...   Feb 21 2008, 21:05
- - Fandango   Concerning spoon's point #5: Considering that ...   Feb 22 2008, 00:01
- - sbooth   I personally vote for using the MD5 hash of the PC...   Feb 22 2008, 01:01
|- - Steve Gabriel   I lean toward option 4, using MD5 as well, but I w...   Feb 22 2008, 04:52
- - spoon   In the DB are stored comp ident - big long OLE typ...   Feb 22 2008, 10:45
|- - flacflac   Hi Spoon, great to see you asking for features o...   Feb 22 2008, 11:37
|- - Steve Gabriel   This idea sounds pretty good, but I don't unde...   Feb 22 2008, 18:14
|- - Steve Gabriel   QUOTE (spoon @ Feb 22 2008, 02:45) [DB2] ...   Feb 22 2008, 23:21
|- - Steve Gabriel   QUOTE (spoon @ Feb 22 2008, 02:45) In the...   Feb 23 2008, 22:15
- - spoon   >Are all the other, better AR results not being...   Feb 22 2008, 12:18
- - spoon   The offset detetion is simply a CRC (checksum) of ...   Feb 22 2008, 22:31
|- - Steve Gabriel   QUOTE (spoon @ Feb 22 2008, 14:31) The of...   Feb 22 2008, 23:05
- - Eli   some other ideas have been expressed over at dbpow...   Feb 23 2008, 16:27
- - Eli   If we are able to have one rolling checksum that w...   Feb 24 2008, 01:03
- - spoon   >DB1-OD field were computed with ARCS and not A...   Feb 24 2008, 19:11
|- - Steve Gabriel   QUOTE (spoon @ Feb 24 2008, 11:11) They a...   Feb 24 2008, 19:48
- - spoon   >DB1-OD's purpose is solely offset detectio...   Feb 24 2008, 22:25
|- - bilbo   QUOTE (spoon @ Feb 24 2008, 16:25) Yes ea...   Mar 10 2008, 16:04
- - Cerebus   Is there ANY way that we can avoid using the DISCI...   Mar 8 2008, 22:11
|- - Cerebus   QUOTE We used everything from the CD TOC to cut th...   Mar 10 2008, 15:25
- - spoon   We used everything from the CD TOC to cut the numb...   Mar 9 2008, 10:11
- - skamp   QUOTE (spoon @ Feb 21 2008, 19:14) 4. Fix...   Mar 9 2008, 12:11
- - spoon   How would the system know if it is a good submissi...   Mar 10 2008, 16:08
|- - Eli   QUOTE (spoon @ Mar 10 2008, 10:08) How wo...   Mar 10 2008, 16:16
- - bilbo   Using EAC for an example, if the report to the AR ...   Mar 10 2008, 16:24
- - Eli   T&C does not account for consistent errors   Mar 10 2008, 17:31
- - greynol   QUOTE (Eli @ Mar 10 2008, 09:31) T&C ...   Mar 10 2008, 19:41
- - bilbo   I am just concerned with the confidence levels. Sa...   Mar 10 2008, 22:50
|- - Eli   QUOTE (bilbo @ Mar 10 2008, 16:50) I am j...   Mar 11 2008, 00:45
- - greynol   <quoting myself for the hundredth time> A co...   Mar 10 2008, 23:24
- - greynol   Anyhow, as far as checksums go, I think CRC32 is f...   Mar 10 2008, 23:53
- - bilbo   @Eli I brought this up because op Spoon's stat...   Mar 11 2008, 00:56
|- - Eli   QUOTE (bilbo @ Mar 10 2008, 18:56) @Eli I...   Mar 11 2008, 02:34
|- - greynol   QUOTE (Eli @ Mar 10 2008, 18:34) Especial...   Mar 11 2008, 03:19
- - CoolHandZeke   MD5! Widely used, 128-bit hash value... Used...   Mar 16 2008, 00:58
- - Jean Tourrilhes   Hi, I already posted that on the dBpoweramp forum...   Mar 16 2008, 07:02
- - rpp3po   Has this been fixed in the April 2009 rewrite? Whi...   Jan 30 2010, 16:03
- - greynol   No and it's not going to be fixed. If it used...   Jan 30 2010, 22:46
- - spoon   There is no reason why such a fix could not go in ...   Jan 31 2010, 10:10
- - greynol   Change the hash calculation to CRC32 or something ...   Jan 31 2010, 20:44
- - spoon   The existing algorithm can be fixed and yet mainta...   Jan 31 2010, 21:20
- - greynol   What about CUETools and XLD? You know they don...   Jan 31 2010, 21:24
- - spoon   No they do not, they hammer around 4000 CRCs looki...   Jan 31 2010, 22:26
- - spoon   Here is the proposed new CRC calculation, which sh...   Jan 31 2010, 22:33
|- - Gregory S. Chudov   QUOTE (spoon @ Feb 1 2010, 00:33) Here is...   Feb 9 2010, 08:57
- - Skybrowser   I'm slightly new and have found this topic qui...   Feb 1 2010, 01:45
|- - greynol   QUOTE (Skybrowser @ Jan 31 2010, 16:45) F...   Feb 1 2010, 03:39
|- - Skybrowser   Is TOS #9 American Legislation? Because I live in ...   Feb 1 2010, 10:45
- - Akkurat   QUOTE Is TOS #9 American Legislation? No. Already ...   Feb 1 2010, 13:05
- - greynol   Actually a CUE sheet does little good with discs t...   Feb 1 2010, 17:54
- - Akkurat   Ahh, brain fart, of course logfile, not cuesheet, ...   Feb 1 2010, 18:02
- - spoon   I do not think there is a need for parallel calcul...   Feb 9 2010, 10:22
- - hellokeith   Spoon, This is a suggestion. I do realize what a...   Feb 14 2010, 08:39
- - viktor   about using md5 or sha1: http://en.wikipedia.org/...   Feb 14 2010, 12:19
- - Gregory S. Chudov   There's no need for a strong hash algorithm in...   Feb 14 2010, 13:31
- - spoon   Also if you wanted extra bits, you could potential...   Feb 14 2010, 16:01
- - hellokeith   QUOTE (spoon @ Feb 14 2010, 10:01) Not su...   Feb 15 2010, 02:44
- - radu   So... In what stage of developing is this? When ca...   Feb 16 2010, 15:30
- - spoon   Different pressing detection code is already in be...   Feb 16 2010, 15:40
- - spoon   Details on the current AR CRC calculation routines...   Feb 16 2010, 23:38
- - Gregory S. Chudov   Why create another CRC that is not much stronger t...   Feb 17 2010, 09:45


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: 20th September 2014 - 16:37