As you know, because a misunderstanding about a paper by Peter Gutmann referring to 1990,s hard drives, a wrong idea about the 35 passes supposedly being needed to erasing completely a hard drive is commonly hold by many people.
However, I am asking that if Gutmann was referring to old hard drives, would this method be valid or had any sense in completely and safely cleaning hard drives made before 2000 year?
As you know, because a misunderstanding about a paper by Peter Gutmann referring to 1990,s hard drives, a wrong idea about the 35 passes supposedly being needed to erasing completely a hard drive is commonly hold by many people.
However, I am asking that if Gutmann was referring to old hard drives, would this method be valid or had any sense in completely and safely cleaning hard drives made before 2000 year?
Short answer No.
Long answer No, It wasnt valid even in the 90s, one wipe would be sufficient.
This is what i have heard from a data recovery company (IBAS) that had looked into it
The (really) old harddrives that was to be wiped with the old DoD standard had interleaving up to 17, that means that it had to spin up to 7 times to read/write all from one track. Because of this, DoD made it a standard to overwrite all harddrives 7 times to make sure that any such harddrives would be wiped clean. Modern drives from the 90s and onwards rarely used interleaving. I've only seen interleaving being a factor on an old 128 MB harddrive i had a loooooong time ago.
Gutmans proposed wipe claimed that it took 7x5=35 passes to wipe a drive clean (which i never read any explanation or research for). It was more paranoia than science in the 90's.
Gutmans proposed wipe claimed that it took 7x5=35 passes to wipe a drive clean (which i never read any explanation or research for).
See his paper 'Secure Deletion of Data from Magnetic and Solid-State Memory', which can be found on the net, and particularly section 3. (http//www . cs . auckland . ac . nz/~pgut001/pubs/secure_del.html)
That paper makes it clear it's not 7x5 – it's 22 patterns designed for various combinations of MFM/RLL, and additional passes for other forms of encoding.
And the added 'Epilogues' seem to make it quite clear that those 35 passes were for the case where you did not know the encoding method used. If you you knew for certain the disk was a EPRML disk, a few passes (pass 1-4) with random data would be enough. It would be superfluous to use patterns intended for different types of MFM/RLL encoding.
As for the OP, I think the Epilogues in the cited work answer your question.
The method was aimed at MFM-RLL hard drives only. Moreover, it wasn't even officially proved that so many wiping passes with different patterns are really necessary.