Referring to the article:
https://en.wikipedia.org/wiki/Logical_block_addressingWhat is not clearly stated is that the CHS scheme can bit hard. It might not be an issue or it can. This is not ambiguity on my part. The scheme lends to mistakes. Put another way, you need to format the drive with the machine that is going to use it. That will insure there is no alternative interpretation in the plan.
I am just trying to help.
I disagree. I don't think LBA could be an issue here; LBA support is part of the original IDE specification (and later, ATA-1).
One would have to go back to ESDI or ST-506 interfaces to find drives which did not provide LBA support. I do not think a 10GB drive that has Windows 95 on it is likely to be either an ESDI or an ST-506 'winchester' drive.
I think the formatting issue you mentioned is unrelated to CHS addressing, but the result of older Hard Drives (we're talking the original drives for the 5150 and 5160) using Stepper Motors to control the read/write head, which meant that the stepper motor degrading or even the drive being at a different temperature could cause disk read failures.
Alternatively, I suppose later drives could become unreadable if the incorrect parameters were put into a BIOS that required Hard drive parameters to be entered manually- they could also seemingly work if the drive was formatted with those incorrect parameters (but it would degrade over time) However, that was a BIOS limitation- not a drive one. I had a 42 MB Western Digital IDE drive from 1987 which was auto-detected without issue on more modern systems, despite the system it was original in requiring me to enter the drive parameters into the CMOS setup manually.