Home Page Forums MAME MAME Extracting raw data from a chd....

Hi All,

I have a CHD file (RM nimbus hard disk), that contains a DOS FAT16 filesystem. The filesystem contains some files that I want to extract so I use the following :

chdman extractraw -i windows.chd -o windows.raw

Then under linux (on WSL2) I mount it with losetup :

losetup -f windows.raw -o 0x400
mount /dev/loop0 /mnt2 -r -t msdos

All appears well, however there are files on the windows chd that are *NOT* on the extracted raw file. So where are they?
Is this the right way of geting a full dump of what's currently on the hard disk?

Cheers.

Phill.

Was the CHD created with -c none so it's not compressed? Otherwise you probably have an associated overlay image in your diff/ directory, and you'll need to merge that with the CHD.

Ahh there is indeed a chd in the diff directory, so how do I go about merging them?

More importantly is there a way of getting it to not write a diff, but write it to the original?

Cheers.

Phill.

You have to convert it to an uncompressed CHD, applying the diff. It will write directly to uncompressed CHDs.

Indeed, but what chdman commands do I need to do that?

Right it seems that :

chdman extractraw -i windows.diff -ip windows.chd -o windows.raw

Did what I needed to merge and extract. Now to create an uncompressed chd from that.....

Cheers.

Phill.

First up, you can actually use raw images with most MAME hard disks devices now (I put it in I think over a year ago now). .hd, .hdv, or .hdi are all accepted to mean "raw image", although for IDE it will try to auto-compute the geometry which may not be correct. SCSI disks don't have that problem.

To create an uncompressed CHD, you do the same thing as for a regular CHD but add "-c none", meaning compression type none.