🚀 go-pugleaf

In article <1993Jan25.192638.1458@julian.uwo.ca> wlsmith@valve.heart.rri.uwo.ca (Wayne Smith) writes:
>
>Are current IDE drives able to handle local bus speeds?
>
Nope... and you obviously found that out.  The drives are limited by
the interface chip and until recently they only needed to handle the
8MHz bus timing.  Those interface chips have to time slice the 8-bit
data buffer SRAM (on the drive) or the n-bit cache RAM (on the drive)
between the serializer logic for the drive data and the 8-bit (XT) or
16-bit (AT and Local-bus) IO port.  

Some drives can handle very fast transfers, but the majority are
relatively slow.  There is a move on, initiated by Tandon and other
companies, to have the ATA standard changed to include the buffer
access speed.  It is already in there for some timing, but not for
processor IO time on the local bus.  Until then, the BIOS folks have
to hardcode drive tables to program the local-bus IDE interfaces for
optimum performance.

The fastest drives that I've seen are from Seagate (Yes, Seagate... I
wouldn't have believed it myself and it took a damned long time before
I was convenience that they were any good... But, they are!), Conner, and
WD.  I think that the Seagate ones were further along, but the WD and
Conner seemed to have fairly good access times, too.

These drives, without any cache, on the local bus are as fast or
faster than cached drives.  Furthermore, the potential for the local
bus interface is such that for a single head/platter the bottleneck is
just getting the data off of the platter.

-- 
|Tom Barrett (TDBear), Sr. Engineer|tom.barrett@amd.com|v:512-462-6856 |
|AMD PCD MS-520 | 5900 E. Ben White|Austin, TX  78741  |f:512-462-5155 |
|"No is yes, And we're all free" ---Tracy Chapman, "Why?"              |
|My views are my own and may not be the same as the company of origin  |