> Thanks for the obvious answer. I wrongly assumed that LBA had properly
> addressed this repeating problem.
>
> My comment about stupid software comes from the fact that this is at
> least the 3rd or 4th time this has happened. You'd think that we would
> have learned by now and done exactly what you suggest, have a standard
> that supports the predictable and obvious growth of storage and the
> need to address it.

        I don't think it's as big a problem as you make it sound. By the time most
users want a drive that outstrips their current hardware and OS they are
upgrading those components anyways. Personally I've never hit a limit: each
time I bought a drive that was too big for the machine I had I was upgrading
the machine anyways.

        Remember too that you NEVER get something for nothing. Adding unneeded bits
for addressing reduces performance, and GREATLY increases hardware support.
If we had drives with 128bit sector addressing alot of the simple devices
out there would be far more expensive then they currently are.

> At least the IP standard was overly generous
> for it's day, but as a result it lasted over 10 years before
> requiring rework (IPV6).

        IP got lucky. They looked FAR more ahead then others at the time, and trust
me, it wasn't such a good decision in the beginning. Remember how EXTREMELY
expensive routers were only a decade ago? Remember how slower computers just
couldn't handle the TCP/IP stack? That was a result of IP being so large.
Now, in hindsight, it probably was the better idea, but for many years the
only form of "internet" people had were the dumb terminals at a university.

> With SATA (serial ATA) becoming available, the opportunity to have
> variable width addressing should have been addressed.

        I just don't see it as enough of a benefit. By the time we outstrip the
48bit LBA currently being employed new hardware and software will be around.

> And making sectors bigger than 512 bytes would have greatly
> improved throughput and media efficiently. Back in the old decades
> of CP/M, floppy drives could be formatted with 256/512/1024/2048 bytes
> per sector.

        Sector size IS something that could be increased, however there is again
something you pay for bigger sector sizes, and that is efficiency of
storage. Now, with drives as large as these days this isn't as much of an
issue, but again, think of the smaller devices. Say you set sector sizes to
4k, all of a sudden every embedded device would have to have, at minimum, at
4k buffer just the deal with the drive. This would raise costs, and I just
don't see the benefit being enough to outweigh that, at the moment.

> So once again we're back to having to load driver programs for our
> bigger hard drives. Sheesh.

        I haven't had to. My hardware and software is modern enough to handle it.
TTYL

----------------------------------
Herbert's PIC Stuff:
http://repatch.dyndns.org:8383/pic_stuff/

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.