ImgBurn 1.1.0.0 Verify Fails But Nero Doesn't

[karangguni]

Hi, I'm posting here because I've pretty much tried everything I know. This is very detailed but I'll try to keep it short.

 

I have a system which has an odd problem : burning an ISO with ImgBurn 1.1.0.0 always fails at the verify stage. By fail I mean a mismatch, not a read error. The odd part is that the same ISO burned with Nero 6.6.0.16 on the same machine produces a burn which is identical to the ISO. I don't trust Nero's verify option, so I've been using CDCheck on another system to do MD5/RIPEMD160 sums of both the original disc and the copy. I've run CDCheck on both the copies made by Nero and ImgBurn. The Nero copies always come out identical. The ImgBurn ones (which already failed ImgBurn's verify phase) are full of mismatches.

 

What is worse is that I tried burning with DVDDec as well and the copies made also fail CDCheck's checksum comparison. I've been using DVDDec and ImgBurn for a while now, and evidently this has been happening all along. I just didn't know because until ImgBurn 1.1.0.0, the verify option did not do a bit by bit comparison.

 

This is a brief description of my hardware :

 

AMD Athlon 1200

Asustek A7V133 (VIA KT133A chipset)

512MB PC-133 RAM

Primary channel : master-IBM 30GB 7200RPM HD, slave, WD 60GB 7200RPM HD

Secondary channel : master-BenQ DW-1640 writer (BSLB RPC1 firmware), Pioneer 16x DVD-ROM (can't remember the exact model)

WinXP SP2

 

I do not have any other DVD burning or region protection software on the system, apart from Nero 6.6.0.16 OEM which was bundled with my BenQ writer. No AnyDVD, no DVD43. I don't even have software to watch DVDs on that system.

 

Discs used : Ricoh 8x DVD+RW (RICOHJPNW21). Also tried Ricoh 16x DVD+R and Mitsubishi 4x DVD+R

 

This is a list of things I have tried which have not worked :

 

Software :

- installed a new version of ASPI from Adaptec and checked that ASPI layer was ok with ASPICHECK

- set ImgBurn to use the system's own WNASPI32.DLL (installed above) instead of the standard SPTI

- copied Nero's WNASPI32.DLL to ImgBurn's directory and set ImgBurn to use that to burn

- shutdown every other application running in the background

- disabled all services which were not needed (I even killed the print spooler)

- disabled windows own ability to burn on my writer via explorer

- copied the ISO to the other hard drive on my system and burned from there

- reinstalled a fresh copy of the OS (Win2k SP4 this time) on a different HD and disabled all non-essential services

- made sure both the BenQ writer and the Pioneer DVD-ROM drive were running at UDMA2 via Device manager

- uninstalled VIA 4-in-1 IDE drivers and used XP's default

- installed version 1.35 of VIA 4-in-1 IDE drivers

- autorun is disabled

 

Hardware :

- ran MemTest86 to verify that my RAM was working properly

- changed the IDE cable from a 66 wire calbe to a 33 wire cable

- changed the power unit from a 300W power unit to a 350W power unit

 

I'd post a log from ImgBurn, but I don't know that it would be of much help. I'll post any information or log files needed.

 

Please, any hints would help. I am at my wits end.

[mlg]

Nero only tests if its possible to read the disc or not, it doesnt really verify what it has written.

Edit: Reread your post and I saw that you had already taken care of that part.

 

Are you sure you dont have AnyDVD or DVD43 running in the background? They changes the files on the disc on-the-fly while READING. (That is, the written DVD is indeed correct but it looks different.)

 

Otherwise i have no idea, but posting a log is never a bad idea

Edited December 8, 2005 by mlg

[cornholio7]

how old are your hard discs and how defragmented are they?

[LIGHTNING UK!]

Run ImgBurn's verify again, but this time press F6 before you start it to enable debug mode.

 

Then when you get a mismatch, the entire sector's worth of data will be added to the log.

 

Copy + paste all of that log (or save to file + attach it).

 

I noticed you said 'I have A system' - does that mean it's ok on another one you have?

 

I've done lots of verifying since that area was improved to check bit by bit and I've never had a failure yet!

[Cynthia]

Nero only tests if its possible to read the disc or not, it doesnt really verify what it has written.

Not true.

 

According to the 'User Guide for Nero Burning ROM 7' section 11.5.3.

 

all the files in the compilation are compared whether the files have been written correctly.

[LIGHTNING UK!]

That does say 'in the compilation' though and we're talking about burning image files here.

 

I can't imagine Nero would break down the ISO file and parse it on a file by file basis - which it would have to be doing for that documentation to be correct (in it's current wording).

 

More likely, it does a bit for bit compare over the entire disc area (ignoring the actual filesystem contents) - like how ImgBurn now does it.

 

It's still a very odd problem though. I can't imagine for the life of me where this corruption is getting into the burn process!

[mlg]

Nero only tests if its possible to read the disc or not, it doesnt really verify what it has written.

Not true.

 

According to the 'User Guide for Nero Burning ROM 7' section 11.5.3.

 

all the files in the compilation are compared whether the files have been written correctly.

 

 

Oh, could someone please remember me to not talk about things I have no idea about?

 

Well, in Nero 5.5 (Later versions are way to bloated for me to install them ) it does not verify against the original as there is no(t much) file reading.

[karangguni]

Are you sure you dont have AnyDVD or DVD43 running in the background? They changes the files on the disc on-the-fly while READING. (That is, the written DVD is indeed correct but it looks different.)

 

Otherwise i have no idea, but posting a log is never a bad idea

 

Nope. I don't have either of those installed on that system at all. Both the BenQ 1640 and the Pioneer 16x DVD-ROM drive are flashed with hacked firmware to RPC1. So I have no reason to use either of those two programs. I would need something like DVD Genie still, but I don't use that system to watch DVDs, so I have no DVD playing software installed.

[karangguni]

how old are your hard discs and how defragmented are they?

 

Horribly old. One of them is 5 years old and the other is 4 years old. Fragmentation should not be a problem though, because I just installed the OS on that system from scratch in Aug. So that should account for the HD with my OS on it. The other HD is virtually empty. It's split into two partitions, both of which are 90% empty. It's main purpose is to hold disc images and process large video files. I did install win2k on one of the two partitions to see if my OS was causing the problem, but that is about it.

 

Anyway, fragmentation should not be the cause of the problem. I wrote ISO files to both hard drives to see if the hard drive was causing my problems. Same result. Also, what I have been doing is to create an ISO image by reading a disc using DVDdec. Once saved, I used the same image to burn in both ImgBurn and Nero. If it was something to do with the image, harddrive or fragmentation, the burns with Nero should have failed verification.

[karangguni]

Run ImgBurn's verify again, but this time press F6 before you start it to enable debug mode.

 

Then when you get a mismatch, the entire sector's worth of data will be added to the log.

 

Copy + paste all of that log (or save to file + attach it).

 

I noticed you said 'I have A system' - does that mean it's ok on another one you have?

 

I've done lots of verifying since that area was improved to check bit by bit and I've never had a failure yet!

 

Ok, will go get that log file with debug data done.

 

And yes, I have another system - the one I used to do CRC checks with CDCheck. That system has a different writer (LG GSA-4163B) and ImgBurn burns and verifies fine on that. The chipset on the motherboard is the same (KT133A) but it is a different brand of motherboard (MSI). I did not think that info was worth putting down, as I felt it might make things more confusing.

[LIGHTNING UK!]

It's often nice to hear when people have it working in one place but not in others. I find it makes them WANT to get the other one working too, rather than just casting the software aside

[karangguni]

It's often nice to hear when people have it working in one place but not in others. I find it makes them WANT to get the other one working too, rather than just casting the software aside

 

lol, I think that if I try any harder, I will probably go insane. Amazingly enough, this is not the most frustrating computer thing I have had to do. That award goes to the time I spent 1 whole week installing and reinstalling WinNT 4.0 server (and all the service packs and patches) on an old Pentium II 400 which was faulty and locked up for no reason.

 

Anyway, attached is the log file with debug mode turned on. Thanks for all the help from all of you. Much appreciated!

ImgBurn2.log

[LIGHTNING UK!]

D 00:59:29 Device Sector Dump...

D 00:59:30 5B 6F B9 A6 60 A6 9A 92 63 0B 5C 18 D5 97 48 0D

D 00:59:30 69 ED C1 83 9E 6A 48 F7 AD 70 8A CD 15 00 D6 A6

 

D 00:59:30 Image File Sector Dump...

D 00:59:30 5B 6F B9 A6 CD 2F AF BF F3 BB CD 9F AB 6B 72 EE

D 00:59:30 2D 29 AB EE FF 6A B9 F1 9F BD 6B EB 5B A6 F7 85

 

Verify that disc again using Verify mode... does it error out in the exact same place? (look in the debug log to check the contents again).

 

Also, burn it again and see if the same info gets burnt - again, compare the 'miscompare' logs!

 

It's very odd that 98% of the sector data is ok. I don't work on a level any smaller than that, and for data to be ok before AND after the problem bytes... well.... your guess is as good as mine!

[karangguni]

Verify that disc again using Verify mode... does it error out in the exact same place? (look in the debug log to check the contents again).

 

Also, burn it again and see if the same info gets burnt - again, compare the 'miscompare' logs!

 

It's very odd that 98% of the sector data is ok. I don't work on a level any smaller than that, and for data to be ok before AND after the problem bytes... well.... your guess is as good as mine!

 

Hmm, I didn't read this properly and did not realise you wanted me to verify the disc again in verify mode. I won't be able to do that for a few days because the system with this idiotic problem is currently packed away into a corner and disconnected. Space constraints in my room, unfortunately don't allow me to have that system setup and ready to run all the time.

 

However, I did a few more burns and the verify errors happen in totally different and random places each time. Here are two more log files of two different burns of the same file as the one I used for the log I posted yesterday.

 

Will try and rerun the verification part as soon as I can and post results here. Thanks again for all the help!

ImgBurn2.log

ImgBurn3.log

[karangguni]

Ok, I've done a burn with verify and then used verify mode twice. The mismatch occurs at exactly the same sector and offset each time. I've also noticed that each time, the mismatch starts with the 5th byte of the lines of data that do not match. The 1st to 4th bytes match, but the 5th onwards do not. I am not sure if this means anything but I noticed the trend, so I thought I'd point it out.

 

Attached are 3 log files, the first is of the burn with verify, the second and third are of the next two verification processes I ran. It's originally one log file but I chopped it into three to make it easier to compare. I did the same thing for the logs of the three burns I did above.

 

Also, you will notice a time discrepancy from when the program was started and when I started the burn. That is because I did one burn and verify run, then realised I forgot to enable debug mode. So I had to do that again. I edited out the first burn and verify process from the logs, so that they do not confuse you.

 

Hopefully this will be of some use to you. If there is anything else I can do, let me know.

 

Thanks!

ImgBurn.log

ImgBurn2.log

ImgBurn3.log

[LIGHTNING UK!]

Can you use something other than Nero to make the images - if you've not already tried that. Just make sure that program makes REAL ISO images and not it's own format of ISO images.

 

I don't have the Nero specs for NRG images so maybe there is something weird in your files that I'm not accounting for.

 

You're making an MDS file for them which leads me to believe they're multipart images too (The MDS shouldn't be necessary btw) - multipart images are even less likely to work.

[karangguni]

Can you use something other than Nero to make the images - if you've not already tried that. Just make sure that program makes REAL ISO images and not it's own format of ISO images.

 

I don't have the Nero specs for NRG images so maybe there is something weird in your files that I'm not accounting for.

 

You're making an MDS file for them which leads me to believe they're multipart images too (The MDS shouldn't be necessary btw) - multipart images are even less likely to work.

 

The images were not made with Nero. They were made with DVDD. I only used Nero to burn the images.

 

No, they were not multipart images. There was just one single ISO file. DVDD created the MDS file by default.

[LIGHTNING UK!]

Oh... according to the filesystem, the image was originally created using Nero. I just assumed you were therefore burning NRG files that you'd created an MDS for.

[karangguni]

Oh... according to the filesystem, the image was originally created using Nero. I just assumed you were therefore burning NRG files that you'd created an MDS for.

 

How did you figure that out from the filesystem? I definitely created those images with DVDD. How is it managing to report that the image was created using Nero?

[zacoz]

Um, maybe from this extract ...

 

I 03:33:01 Source File: D:\burning\MORE_CRAP.MDS
I 03:33:01 Source File Sectors: 2,292,112 (MODE1/2048)
I 03:33:01 Source File Size: 4,694,245,376 bytes
I 03:33:01 Source File Application Identifier: NERO BURNING ROM
...
I 03:40:57 Image File: D:\burning\MORE_CRAP.MDS
I 03:40:57 Image File Sectors: 2,292,112 (MODE1/2048)
I 03:40:57 Image File Size: 4,694,245,376 bytes
I 03:40:57 Image File Application Identifier: NERO BURNING ROM

Edited December 18, 2005 by zacoz

[kevdriver]

@ Zac,

Edited December 18, 2005 by kevdriver

[spinningwheel]

[karangguni]

Um, maybe from this extract ...

 

I 03:33:01 Source File: D:\burning\MORE_CRAP.MDS
I 03:33:01 Source File Sectors: 2,292,112 (MODE1/2048)
I 03:33:01 Source File Size: 4,694,245,376 bytes
I 03:33:01 Source File Application Identifier: NERO BURNING ROM
...
I 03:40:57 Image File: D:\burning\MORE_CRAP.MDS
I 03:40:57 Image File Sectors: 2,292,112 (MODE1/2048)
I 03:40:57 Image File Size: 4,694,245,376 bytes
I 03:40:57 Image File Application Identifier: NERO BURNING ROM

 

No, that was not a mistake on my part. This is explorer showing the image in question and the MDS file :

 

Nero does not produce disc images of type ISO and MDS. The images it produces are in NRG format.

 

The original disc that the image was made from, was probably burned using Nero. My guess is that that is what the logs are indicating. I don't know for a fact what program was used to burn that disc originally, because I wasn't the one who burned it.

Edited December 18, 2005 by karangguni

[cornholio7]

[weAponX]

Judging from this:

 

AMD Athlon 1200

Asustek A7V133 (VIA KT133A chipset)

512MB PC-133 RAM

Primary channel : master-IBM 30GB 7200RPM HD, slave, WD 60GB 7200RPM HD

Secondary channel : master-BenQ DW-1640 writer (BSLB RPC1 firmware), Pioneer 16x DVD-ROM (can't remember the exact model)

WinXP SP2

 

The only problem I see in there, is SP2- Take it out! If anyone were to make a list of stuff that needs to be immediately disabled after upgrading to SP2, large books will be made! So, I avoid it when at all possible.

 

installed a new version of ASPI from Adaptec and checked that ASPI layer was ok with ASPICHECK

- set ImgBurn to use the system's own WNASPI32.DLL (installed above) instead of the standard SPTI

- copied Nero's WNASPI32.DLL to ImgBurn's directory and set ImgBurn to use that to burn

- shutdown every other application running in the background

- disabled all services which were not needed (I even killed the print spooler)

- disabled windows own ability to burn on my writer via explorer

- copied the ISO to the other hard drive on my system and burned from there

- reinstalled a fresh copy of the OS (Win2k SP4 this time) on a different HD and disabled all non-essential services

- made sure both the BenQ writer and the Pioneer DVD-ROM drive were running at UDMA2 via Device manager

- uninstalled VIA 4-in-1 IDE drivers and used XP's default

- installed version 1.35 of VIA 4-in-1 IDE drivers

- autorun is disabled

 

Krikes! You ought not to have to do all that! Also, there should be a newer Via 4-in 1... But go to the Asus site, and get theirs!

 

But your system is the same as the one I do most of my heavy burning on, except yours 200 MHz faster. Are you sure you are making coasters? Check your IDE cables, get new ones if you have to.

 

NO burn is ever a perfect, 100% exact copy of the original, even in the process of making the image, a few bits are dropped here and there. Another thing to think about is... That Pioneer DVD reader. I'm not familiar with BenQ? But Pioneer, I forget what they really are underneath the logo, but they are a pretty cheap brand. Sometimes, you just have to replace your equipment.

 

Just some alternate stuff to think about, but I have this thing in my gut that you oughtta test that Pioneer unit, make sure it is being read right.

 

Are both of those Units UDMA 33? Cos newer drives run at UDMA 66- Mine does...

Edited December 19, 2005 by weAponX