[TriLUG] OT: hd speed with 33MHz and 100MHz controller are the same

[Warren Myers]

seek time is just to find the first sector - if the data is more-or-less
sequential, the next seek is pretty darn quick

none-the-less, if the system bus is 33 Mhz, it won't matter if you put a 100
Mhz controller on the bus, it still can't push data from itself to the CPU
faster than at 1/3 it's speed.

WMM

On 6/13/07, Jim Ray <jim at neuse.net> wrote:
>
> > > How can the bus be the bottleneck when the reciprocal of the bus
> frequency
> > > is an order of magnitude faster than the hard disk access time?
> > >
> >
> > Is it because the seek is only done once per block, and then the entire
> > block is read? That first byte takes a long time, but the next 511 bytes
> > after that are pretty quick.
> >
> > Alan
>
> [Jim Ray pontificates] doesn't average access time account for seek and
> read?  I'm thinking a 10 ms access time >> 1/(66 MHz)
>
> --
> TriLUG mailing list        : http://www.trilug.org/mailman/listinfo/trilug
> TriLUG Organizational FAQ  : http://trilug.org/faq/
> TriLUG Member Services FAQ : http://members.trilug.org/services_faq/
>



-- 
http://warrenmyers.com
"God may not play dice with the universe, but something strange is going on
with the prime numbers." --Paul Erdős
"It's not possible. We are the type of people who have everything in our
favor going against us." --Ben Jarhvi, Short Circuit 2