All GPT does is chop a disk up into partitions. It has absolutely nothing to do with filesystems or permissions.
Just look it up and examine the header and partition table formats - it's self-explanatory.
I thought it had something to do with it but silly me. I found the original disc image copy of it in 2019 before I copied it to a SDD stripped stuff out of it.
I did convert it just now and got the same error message with trying to delete the system files.
I can make folder and copy stuff but not delete certain things.
Obviously it is something else 80070780.
Taking ownership does nothing.
Actually on further looking it is permission protected despite taking ownership, it is has been set for every account except trustedinstaller so basically locked out.I can't remember what I did to solve that.
I believe I found the issue just now but not what I remember.
I noticed "read only" attribute in c:\windows\ folder
Trying to removing this attribute leads to "THE TARGET OF THE SYMBOLIC LINK DOES NOT EXIST"
https://level400.org/2019/06/18/how-to-deal-with-the-target-of-the-symbolic-link-doesnt-exist/Luke Edson Posted on June 18, 2019
How to deal with The target of the symbolic link doesn’t exist
The other day I came across a Hyper-V host server where an OS crash had occurred. A new OS install was made on the C: drive. After the rebuild, the D: drive’s VMs (VHDX’s etc.) were visible and appeared to be fully available, but when an attempt was made to import the existing VMs or even access the files, the following error message was displayed:
"The target of the symbolic link doesn't exist"
Trying to access the with Windows Explorer gave an Error 0x80070780: The File cannot be accessed by the system.
Attributes APLI
To disable DeDuplication:
If DeDuplication is disabled, it doesn’t actually undo the work that was done, and if it is disabled, garbage cleanup commands can’t be run.
It’s VERY Important that Deduplication is left enabled, but leave the entire drive Excluded.
Once that’s done, run the following two commands (which will take quite a bit of time depending on how much data there is).
I'll have a go at that later.
Once again sorry about that.