On Thu, 2008-10-16 at 05:22 -0400, Bryan J. Smith wrote:
> While virtually all other BIOS manufacturers solve the 256 head problem
> by just using 255 in an 8-bit field (1111111), IBM-Levono does this
> funky mess.  The resulting 240 value (11110000) is completely divergent
> from the rest of the industry.

Ugh, I was reading further, this isn't correct.  My original post was
more correct.  It's been awhile since I looked at this, but I knew I was
on the right track (I documented the whole issue long ago).

It's that all other BIOS manufacturers now always use the 8-bit/255
heads (where its 0-255 = 256) and IBM-Levono continue to force the old,
5-bit/16 heads setting, period.

Again, goes back to the '90s 24-bit versus original PC/AT 21-bit field.
Both are ignored if you use newer Int13h disk services, but even IBM
does not in its own utilities at the factory.  ;)

As I mentioned, always forcing 255 heads at the factory solves it,
because both NT (as of 4.0 SP4+) and Linux always read the partition
table geometry itself to determine the actual geometry it is
using/assuming.  It must still be 0-255 heads, or less, in the case of
NT though (through 5.1 at least, possibly 6.0/6.1 too?).


-- 
Bryan J  Smith              Professional, Technical Annoyance
mailto:[EMAIL PROTECTED]  http://www.linkedin.com/in/bjsmith
-------------------------------------------------------------
           Fission Power:  An Inconvenient Solution

_______________________________________________
lpi-examdev mailing list
[email protected]
http://list.lpi.org/cgi-bin/mailman/listinfo/lpi-examdev

Reply via email to