[PLUG] hard drive geometry...

plug_0 at robinson-west.com plug_0 at robinson-west.com
Wed Jan 31 08:03:58 UTC 2007


Back around 1995 or so, we started to see geometry
translating hard drives.

I'm wondering why the registers that the old CHS standard
depended on weren't increased in size?  Where an 8 bit
register was used, replace that with a 16 bit register,
etcetera.

I've received a comment that the register size can't be changed
because: dos, Windows 95, and every other OS from that far back
will break.  Someone also decided to throw in anti Linux fud by
commenting that BSD is better because it isn't "cobbled" together
with third party software, mostly from GNU, but I'm digressing.

Doubling the size of the respective registers so that CHS will
still work today with our larger hard drives seems to me to be
a more straightforward solution.  I guess I don't see where the
software problems lie.  Is it the size of on processor registers
that are creating the: 16 heads, 1023 cylinders, 63 sectors
limitation?  Are these registers in memory?

Hard drive translation is extremely confusing to me.  I find
it a very hard topic to understand.  Maybe I just lack a good
explanation that I can follow.

Beyond hard drive geometry discussions, the idea that partition hiding
is a bad hack was proposed.  Is the idea of hiding fat partitions
really a hack?  I think there is a special partition type for hidden
fat.  XP and 2000, perhaps NT too, don't seem to honor hidden fat.

With Vista only running on machines that are less than a year old, Microsoft
could demand a change to the way hard drives are accessed on PC's.   
Microsoft doesn't have to worry about Vista being run on computers  
before the change
is made, because it won't run on them.  Sucking up processor power just
to impress me visually doesn't make an operating system better in my
opinion.  From what I've heard, it sounds like Microsoft likes the OEM
channel and that Vista will give OEM hell a whole new meaning.

Do Apple computers have far less legacy support than PCs in order to
keep them simple and theoretically produce a more stable platform?

      --  Michael C. Robinson

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.




More information about the PLUG mailing list