Jiri Zita
-
Posts
20 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by Jiri Zita
-
-
I have tested 2.5.1.0 version of ImgBurn and the results are:
The above-mentioned LITE-ON DVDRW SOHW-1693S KS0B now can read all the ISRC codes properly. MCN is unfortunately still bad.
The above-mentioned TSSTcorp CD/DVDW SH-S183A SB03 now can read all the ISRC and MCN codes properly.
Good work:-) Thanks.
-
Yes, I know that we probably found the way only to ignore ONE kind of bugs in their firmwares and the rest were left uncorrected, but for practical reasons it is more helpful to us to do it that way. We have perfect readability with C2 error tests and both ISRC and MCN codes for almost all discs (with several exceptions of badly burned CD-Rs). We use(d) Samsung drives from SH-S182,183,202,222,223 and LiteOn drives from SOHW-1633,1673,1693. Even our old Toshiba DVD-ROM drives returns good values, but these work also with ImgBurn well.
I am trying to solve this problem now because I was asked by several our customers to recommend them a free utility for preparing disc images - reliable, compatible and well tested. And for me the best solution was ImgBurn so I started to write a tutorial for them. But I cannot be sure now that they will supply us with a good 1:1 image because I don't know if their drive reads all the info properly or not.
ImgBurn isn't perfect and maybe has more bugs but even those one-click big-icon hundred-MB applications for trained apes have some. We use it mainly for burning of DVDs and for this task it is superior. But proper reading of ISRC, MCN, CD-Text and indexes is also important to us and our customers too. All these information can be retrieved with ImgBurn so I would want to recommend it to them. And your Enhanced CD (CD-Extra) CUE support/improvement is also one of the advantages :-) over other utilities.
If you want we can send you our source code and you can test it if it works for you or not. To be honest, for us is easier and less work to help you with your worldwide known ImgBurn than to create a new separate application for reading discs to images.
-
I talked with our chief programmer today morning and he told me that he had found the problem with wrong subchannel reading several years ago and that he had to change our software to more secure and wider compatible way. I've seen its source code so if it can help you, here is the algorithm and info how it can be done or we can send you the complete source code:
- there's a cycle for all tracks - for every track we read first 100 sectors (the minimum for ISRC and MCN according to specification)
- we use READ CD MSF command (using of READ CD command only is not secure in all situations - addressing problems sometimes, READ SUBCHANNEL can lead to problems too)
- if the "Formatted Q sub-channel data" function is supported (Sub-Channel Selection bits 010b), the function will return Q channel only, otherwise "RAW" subchannels (backup for older drives, bits 001b according to ftp://ftp.seagate.com/sff/INF-8090.PDF, but not in latest MMCv6) are extracted and then decoded
- if non of the above commands are supported, only then the READ SUBCHANNEL should be processed but with warning that codes can be incorrect (for user to verify codes with supplied documentation)
We processed several thousands pressed and burned CDs with ISRC and MCN codes so our way is tested very well:-))
-
I have several CD audio discs with properly burnt ISRC codes and MCN codes but when I use Read mode and "View Media Information" feature, there are wrong or no ISRC codes and wrong Media Catalog Numbers. I tried to do several tests and it seems to be a problem of my drives or the method ImgBurn tries to read from these drives.
Drives:
LITE-ON DVDRW SOHW-1693S KS0B (wrong MCN, no ISRC)
TSSTCorp CD/DVDW SH-S183A SB03 (good MCN, wrong ISRC)
both very good, reliable and C2 capable drives with latest firmware version
Tested on:
ImgBurn 2.5.0.0
Windows Vista Business SP2, with both SPTI and Nero ASPI
Windows XP Professional SP3, with both SPTI and Nero ASPI
DiscJuggler 6 Free edition
When I created CDI image of the CD using DiscJuggler software (RAW mode with subchannels), mounted by Daemon Tools 4.12.3 and used Read mode and "View Media Information" feature in ImgBurn, all information were correct. The ISRC and MCN were verified and tested by our premastering software which also had burnt the CD from DDP fileset. Similar results from pressed CDs with ISRC codes and MCNs... so it is not problem of burning.
I don't know the method you use for reading ISRC and MCN but it seems to be incompatible with these drives and maybe with many more. I think that reading of subchannels (for drives supporting it) will be much more accurate and I hope you will include this method into next releases :-))
Images attached: LiteOn, Samsung, PremasteringSW, Mounted CDI image created by DiscJuggler
-
Thanks mmalves.... We asked what their requirements are and thats what they replied with. We'll definitely push for the ISO support as that usually is industry standard nowadays. They'd be loosing big business if they don't
and we might have to look for yet another vendor to replicate our master DVDsYes, look for another replicator... maybe a small hint: www.gzdm.cz :-))
DDP, CMF and also...
Files containing DVD image:
Files, which can be used without any adjustments for burning of the master disc.
Recommended formats:
• ISO, IMG - images of DVD discs (user data in sectors of 2048 bytes)
we recommend to use ISO+MDS for double layer DVD images; these files can be created with ImgBurn for example
• NRG (Nero)
Acceptable formats:
• BIN/CUE
• MDF/MDS (Alcohol 120%)
• IMG/CCD (Clone CD)
• IBP/IBQ (IsoBuster)
• CDI (DiscJuggler)
• CIF (Easy CD Creator)
• BWT/BWI, B5T/B5I, B6T/B6I (BlindWrite)
• DMG (Apple Macintosh) - disc images for Apple Macintosh
• TOAST, DVDR (Toast, Apple Disc Utils)
• ISZ (UltraISO)
• UIF (MagicISO)
-
I did reading of Audio CD into CUE+WAV+CDT image with MD5 option enabled and it created MD5 text file with "Image MD5" but it was referenced to "Image.wav" file. It is not correct because when somebody tries to calculate the MD5 checksum for WAV file (as it supposed to be from the meaning of the text file, e.g. in HashCalc, FreeCommander, MD5sum etc.), the calculated MD5 is different. It works perfect for CUE+BIN+CDT probably because the BIN file doesn't contain WAV header.
I think that the text file should contain the correct MD5 checksum for WAV file so everybody can calculate the MD5 checksum by himself and verify the WAV file. The "Image MD5" should be written elsewhere in the text file so it would be clearly visible which one is for WAV file and which one is for burned CD or binary image.
-
Look at the actual image, how big is it?
The leadin is in negative lba ranges so it wouldn't actually show up in that info... in any case, it's definitely written.
Nero might have just stuck 2 seconds worth of zeroes on the end of the track.
I think that ImgBurn SHOULD (as a default) add 150 sectors of zero data at the end of data tracks in BUILD mode for ISO images to be fully compatible with Yellow Book specification (for mode 1 discs it is obligatory but for mode 2 it isn't). Of course for Read mode it is not recommended to do it automatically if the user wants to have a perfect copy of his source (non-postgap) CD.
We here in premastering department of our pressing plant have sometimes problems when adding the obligatory postgap to the customer's ISO image and then he asks us why the MD5 checksum of the pressed CD-ROM differs from the MD5 checksum of his ISO image and we have to explain him the reason why. It also makes our internal verification procedures more complicated...
So I believe Nero prepares ISO images well (in this case:-)) and ImgBurn not.
I cannot accept the explanation written by LIGHTNING UK! in the following old thread:
http://forum.imgburn.com/index.php?showtop...&hl=postgap
Any comments or disagreements?
-
Isn't that a replication master image? No, you can't burn those with ImgBurn.
Hi,
it is possible to burn a disc from data files of supplied DDP files BUT only in several specific cases:
- IMAGE.DAT or similar main data file must start at sector 150 or later (for files starting before sector 150 you can use a hex-editor to remove the pregap)
- it is always better to work with one big file instead of several separate tracks (CD-Audio tracks, Video CD tracks)
- it is not easy to burn multisession DDP without proprietary software or some modifications
- only DDP without any protection (CSS, ...)
1) CD-ROM with 2048 bytes/sector or single layer DVD-ROM, DVD-Video - just burn IMAGE.DAT or the largest file of DDP set - similar as burning an ISO image
2) CD-ROM with 2352 bytes/sector
prepare a CUE sheet with something like
FILE "IMAGE.DAT" BINARY
TRACK 01 MODE1/2352
INDEX 01 00:00:00
- burn the CUE sheet and verify the disc
3) CD-Audio - it is more difficult to burn from DDP because you have to create track times for CUE sheet from times specified in PQDESCR text file (subtract 2:00 etc.), but it is possible too.
4) Dual layer DVD - more complicated but can be done
- use "Create DVD MDS File" option from "Tools" menu
- insert main data file for Layer 0 first!! and then!! the data file for Layer 1
- layer break field should be filled with number of sectors of Layer 0 (the length of the file /2048)
- you should know if the DDP files are prepared for the OTP or PTP disc and should choose the proper one
- save the MDS file to the parent folder of both layers (not to L0 or L1 folder)
- burn the MDS file and verify the disc
I hope it will help you with the DDP files. Ask more if needed.
Regards,
-
Hi,
I suggest to add two media profiles for burning DVD discs which can be used as masters for pressing plants:
DVD-5 - 2,294,912 sectors
DVD-9 - 2,084,960 sectors per layer
It would be also nice to receive single layer ISO (and double layer ISO+MDS) images prepared for mastering.
Maybe there should be a warning when ImgBurn tries to build an image with the layer break over sector number 2084960. Or at least some option to do such a test even for DVD+R DL or DVD-R DL profile.
Sometimes we receive recordable DVDs with more than allowed number of sectors burned and every time we have to ask the customer if he wants to:
1. change his order to DVD-9 (in case of full DVD+-R SL supplied)
2. confirm production of non-standard DVD and confirm that he accept that we will not guarantee proper playback of such a disc in every player
3. let us change the layer break (if it is possible)
Jiri Zita
Premastering department
-
..........
I've done the relative folder thing now - providing the images are in folders that are subfolders of the one where the mds is.
Great
You are soooo fast that I am not able to think out and write new suggestions and requests for you :-) Thanks.
But wait! .... Yeah! I've got one. I will write it in Suggestions forum soon.
Jiri Zita
-
The MMC specs say that for OTP media, the physical format info for each layer is just a copy of the 1st, so I'm not seeing any reason for the 0xDDFBD0 value anywhere.
On PTP media (the ones i've tried anyway), the 'First Physical Sector of Data Area' is still always 0x30000 for both of the layer descriptors.
There is no way to store relative paths within the MDS in ImgBurn, it wasn't designed to support it.
It's either an absolute path or just the file name - mainly because I saw no reason for the type of paths you're saying you need. I'll see about implementing it for the next release.
Hi,
If the MMC specs say it then it is true for sure. They are the authority :-) I've only tried several combinations to make it work (reading of mounted MDS) so that our app wouldn't crashed and I've found one which works. Probably 0x30000 will work too, I will test it and correct the MDS. It had already worked before with that wrong number.
For mastering we need both layers in DDP format to be in separate folders. The best thing to test final DDP master before production is to mount the DDPs. We use CUE sheets for most other formats but for DVD9 we need MDS files with relative paths to subfolders. The MDS file will be saved into parent folder of both layers folders and then we have a possibility to mount the whole DVD9 title either on the local hardisc, either on the network. We will implement this feature in our own app so don't bother yourself with it if you have more important things to do. I thought it could be interesting also for someone else.
Jiri Zita.
-
I only care about the layer break position so I haven't made any effort to make anything else the same.
Yes, I understand, but then the MDS file is usable only in your software. If ImgBurn shows that information from OTP DVD9 disc then it also should be correctly saved in MDS. So we can emulate the disc properly in Daemon Tools.
The proper calculation:
IF
First Physical Sector of Data Area: 196
-
What does the disc info look like in the window on the right in ImgBurn with that MDS mounted? (copy + paste it all)
There must be *some* reason why ImgBurn doesn't think 2070432 is acceptable.
MDS+ISO were probably created with Read mode directly from DL disc.
Mounted MDS info follows:
-----------------------------------------
UM7404C IMS525C 1.01 (ATAPI)
Current Profile: DVD-ROM
Disc Information:
Status: Complete
Erasable: No
Sessions: 1
Sectors: 3
-
Do it at the start of any VTS, if possible.
If you can't find a good LB point, apart from splitting cells, you can remap the titlesets. Do this with PgcEdit. It's almost a one step process - just decide which VTSs should be in layer 0 and move them around.
Regards
Thanks for your suggestion, I think it will surely help and maybe I will try it in this case, but I have to ask the customer for confirmation to make this change in his supplied master. I cannot make any change without his written permission.
I really need to know if his ISO+MDS masters ARE wrong or NOT - for me not to be considered panic-monger
I have one more screenshot from analyzing of that ISO. I was told that every good layer break should have these "No previous VOBU" and "No previous possible VOBU" messages and here they are. I have no DVD+RW DL so I have to waste one or more DVD+R DL to check the behaviour of the disc in DVD player. I will post the playback results then ...
Regards,
Jiri Zita
-
Hello,
and what do you think about following attached screenshot of large mounted ISO. It is taken from that problematic ISO image. Our software recommends to put the LB in the place of the yellow line - sector 2070432. Supplied MDS file (I don't know its origin) shows also 2070432, but according to ImgBurn in "Display IFO Layer Break Information" mode, there is none in VTS_16, VTS_17 and VTS_18. VTS_18 is their multiangle bonus clip. So who should I believe?
Jiri Zita
-
Ok, that's easily done
I will change it to use the real data when read in Read mode but I can use your calculation when I create my own from Build mode.
I'd never have guessed at the XOR'ing!
Many thanks
Hello,
If you want I can do some betatesting with several MDS files and DL DVDs we have here. I can also compare the results with our DDP creating software if it shows the correct numbers for layers. The only thing I need for that is the new version with better MDS support you mentioned. You can find my email in my profile. Thanks for your support to ImgBurn.
Jiri Zita.
-
Hello all,
I have received one ISO+MDS image with multi-angle VOBs inside. ImgBurn cannot find a proper layer break in ISO image (in IFO). But our premastering software found a layer break at the same position as it is written in the MDS file. It seems that these ISO+MDS files were made (Read) from a pressed DVD and not with Built mode.
Have you any experience with layer breaks in multi-angle VOBs? And any theory behind it?
Jiri Zita
-
I only care about the layer break position so I haven't made any effort to make anything else the same.
Yes, I understand, but then the MDS file is usable only in your software. If ImgBurn shows that information from OTP DVD9 disc then it also should be correctly saved in MDS. So we can emulate the disc properly in Daemon Tools.
The proper calculation:
IF
First Physical Sector of Data Area: 196
-
Hi,
I think there is a problem with MDS file creation in ImgBurn (using 2.4.1.0 but probably in every older too). It contains improper information for DVD9 OTP discs so our premastering software crashes when it tries to read the mounted MDS image (through Daemon Tools 4.12.2) of OTP DVD9 disc. Please see attachments and you will find the differencies regarding the "Last Physical Sector of Data Area".
MDS+ISO files were created by Read function in ImgBurn from properly prepared OTP DVD9.
Probably there is a wrong calculation somewhere during the creating of MDS files.
Of course we can mount supplied ISO images but then we have to manually change layer breaks in our DDP premastering software. MDS+ISO could be better solution.
I hope this post will help you improve your great piece of software. We like using it :-)
Best regards,
Jiri Zita
Premastering department
GZ Digital Media Lodenice
Czech Republic
Bad or no ISRC and MCN codes from CD audio discs
in ImgBurn Bugs
Posted
Hi,
I'm only confirming that the new version 2.5.2.0 works well regarding the above-mentioned issues. Good work:-)
I will recommend the new version to our customers and will send them the PDF guide for creating disc images suitable for mastering.