Roman,
I am using clrmamepro 3.114a for auditing my MAME 125 set that I got from retroroms. I use a logiqx dat. Windows XP SP2. I use full zip structure scan compressor option, all checksum options.

The issue I'm having is that some sets come up wrong crc32 or sha1 sometimes, other times they don't. The actual error msg is 'full archive integrity check reported errors, or a header crc32 mismatch was found.' What does that mean? I have read the documentation. Redownloading the entire set from pdome or retroroms, known reliable sources, do not correct the problem.

What am I missing? What factors can contribute to crc32/sha1 corruption or mismatch? Maybe I'm using clrmamepro wrong? I have been into roms and using your great program since 97 though. Any help would be reatly appreciated!

> reported errors, or a header crc32 mismatch was found.' What does that mean?

It means your zipfile is most likely corrupt. cmpro decompresses the archive to memory, recalculates the crc32 values and compares it against the one in the datfile *AND* against the data found in archive header.

There are NO reliable sources. Bad zips are a common thing.

If you got the problem that one and the same archive produces such a problem randomly (i.e. one scan it's fine, the other scan it's bad, then fine...then bad..). Then your system may have a problem. This can be either a different program interfearing (e.g. Virusscanners) or RAM/HD is bad.

What you can do is: simply mail me such a zip and I'll have a look (if mail doesn't work, put it somewhere where I can grab it).
..or simply open the archive in question with a common zip program (winzip/winrar/7zip/etc..) and run a test operation.


Roman Scherzer

> > reported errors, or a header crc32 mismatch was found.' What does that mean?
>
> It means your zipfile is most likely corrupt. cmpro decompresses the archive to
> memory, recalculates the crc32 values and compares it against the one in the
> datfile *AND* against the data found in archive header.
>
> There are NO reliable sources. Bad zips are a common thing.
>
> If you got the problem that one and the same archive produces such a problem
> randomly (i.e. one scan it's fine, the other scan it's bad, then fine...then
> bad..). Then your system may have a problem. This can be either a different
> program interfearing (e.g. Virusscanners) or RAM/HD is bad.
>
> What you can do is: simply mail me such a zip and I'll have a look (if mail
> doesn't work, put it somewhere where I can grab it).
> ..or simply open the archive in question with a common zip program
> (winzip/winrar/7zip/etc..) and run a test operation.
>
>
> Roman Scherzer
>

Thanks for the advice Roman... I have a feeling the problem does indeed relate to my system - I forgot I had realtime virus scanning going on at the same time. I am also going to do a scan for errors on my external drive which the roms are on. I'll try the test operation with winzip as well.

I did a chkdisk on my HD and I disabled my antivirus for the scan, and I was sure that there would be no prob. I did a scan, redownloaded the sets, then I cleared the cache, reloaded the dat, and scanned again. It reported totally different sets this time with 'wrong sha1'

???
here's an example of what type of thing comes up a lot:

2020 Super Baseball (set 1) [system: Neo-Geo - folder: 2020bb - size: 8mb]
wrong sha1: 030-c2.bin [wrong: 1f43363e52d2e07b669bf6d01f01c149ee10478f] [right: 0920cc580d7997fcb0170dd619af2f305d635577]

what's also wierd is mostly neo geo games have this unique issue.

I tell ya I do a scan and it seems like different sets come up as full archive integrity check reported errors or wrong SHA-1
I guess my memory is bad? does using an external usb 2.0 HD make a difference?

> I guess my memory is bad? does using an external usb 2.0 HD make a difference?

Well...copy the set in question to a different HD (if available) and scan there. And/Or calculate the sha1 value for the set/rom in question. For example, drag'n drop the set into cmpro's about window. Compare it against the cmpro scan report. Winzip/Winrar don't calculate sha1 values and maybe you got a crc32 collision ( = different sha1, while crc32 is identical for 2 different files) here (however this should be a very rare effect and I doubt this happens for more than 1 neogeo file)

Of course there is a little chance that RAM is bad or the USB driver has some issues in timing....but maybe the file is simply bad.

If the error happens randomly (as mentioned before, sometimes the file is reported, then sometimes the same file isn't reported, then it is reported...), the chance is higher that maybe the USB HD has some timing issues or other system stuff got problems. I remember very old VIA boards had major issues...but that's was like 8 years ago ;)

And of course third party tools are bad, especially:
- virusscanner which hook in and try to scan the zip before the application does (I use AVG and got no problems, while antivir locked the archives too long sometimes)

- zip tools which also provide some system hook. One user reported an issue when he had 7zip installed (however I never was able to repeat that).


Roman Scherzer

Whatever was causing a problem - it had to do with my system be it the external HD, the memory, or something else.

I brought my work laptop home & loaded the roms on there and clrmamepro reported a 100% clean set. I'm glad it wasn't the case that I was using the program incorrectly.

Thanks again