DiscImageCreator (Page 49 of 142)

I have a suggestion regarding the handling of Safedisc protected discs.

I have a disc that shows random errors in the Safedisc protected areas. Because of the /sf parameter, the entire Safedisc sector range will receive no C2 error correction or re-reads. I think this is some kind of design flaw, because currently it is not possible to dump Safedisc protected games in a reliable way - you'll never know if your disc has a C2 error caused by Safedisc or a "real" one that could be fixed by re-readings.

Therefore I propose the following change to the Safedisc protection handling:

Detect the protection as always and store the Safedisc error range somewhere in memory. Read the entire disc and then try to correct C2 errors by re-reading. When DIC detects a C2 error inside the previously detected Safedisc error range, it should re-read it multiple times (user defined). If a C2 error can be fixed by re-reading: Congrats, we fixed an unintentional C2 error. If it can't be fixed by re-reading, we probably hit a Safedisc error. In this case, DIC shouldn't quit with an error as it now does, but simply skipping this sector and replace it with 0x55.

This would make dumping Safedisc discs more reliable in my opinion. What do you think?

But when DIC hits a "real" Safedisc sector, it will quit after re-reading this sector multiple times because it was unable to fix it, right?

Thanks sarami for looking in to this! I really appreciate it!

I think DIC only works with D8-capable drives, giving the messy data tracks on the rest?

http://forum.redump.org/topic/15076/add … ification/
I don't have this drive, but he reported this drive worked about scrambled and lead-in.

Try using the latest firmware.

Uploaded the test version. I haven't tested the protected disc except safedisc yet.

How much is ripping time increased because of this new feature?

I did a few more runs on that problematic disc - it seems that it is in a condition that makes it unusable for proper dumping. I get not only random C2 errors (that get fixed by the re-readings, so the new feature works as expected), but also sync issues. Damn.

sarami, thank you for the fast implementation of this feature request!

sarami, thanks for the new test version.

Just trying with another known good safedisc disk - I'll see if checksums match with "old" stable version and latest test version. Hold on - logs will follow. Then I'll see if the new test version can do something about the bad safedisc disc.

I'll post the logs of the bad disc too... maybe I can finally get a good dump without any unintentional C2 errors.

At least for one good safedisc disc I can confirm that *all* sectors with C2 errors are 312 bit.

Most of the sectors of my bad safedisc disc have 312 bit too, the random ones have random values. Logs of the bad disc attached. I'll try to confirm more good safedisc disks.

Logs of the bad disc after a full redump: https://www.sendspace.com/file/9i1ne5

Thanks. Uploaded the latest test.
added: Skip rereading for safedisc 312 errors
              => The problem is that dic skips rereading if unintentional error is also 312.

dic gets the c2 (294 byte=2352bit). If error exists, a bit stands and coutinue checking this 2352 times.

how high is the chance that an unintentional error is exactly 312?

Unintentional error changes every times CD is read, but I don't know how high rate (depending on CD).
Anyways, I always recommend dumping at least twice, especially bad condition disc.

Sync is 12byte (00 FF ... FF 00)

========== LBA[002832, 0x00b10]: Main Channel ==========
       +0 +1 +2 +3 +4 +5 +6 +7  +8 +9 +A +B +C +D +E +F
0000 : 00 FF FF FF 7F FF FF FF  FF FF FF 00 01 B9 A3 61   ...............a
========== LBA[002832, 0x00b10]: Main Channel ==========
       +0 +1 +2 +3 +4 +5 +6 +7  +8 +9 +A +B +C +D +E +F
0000 : 00 FF FF FF FF FF FF FF  FF FF FF 00 01 B9 57 61   ..............Wa

Fixed 0x7f to 0xff.

Check _EdcEcc.txt. I don't know the error of this disc is truly 584 because I don't have this.