Re: ATA anomaly Question

[Bill Hacker <wbh@xxxxxxxxxxxxx>]

Matthew Dillon wrote:
> :>>Fine with me - so long as it doesn't actually downshift I/O rate-wise.
> :>>(which IIRC, the FreeBSD ones actually *did* do, at least early-on.
> :>>Ancient history now...)
> :> 
> :> 
> :> 	It certainly doesn't I've seen well over 40MB/s out of a SATA
> :> drive with the limited to UDMA33 message.
> :> 
> :
> :Sounds good! - Few single drives can sustain even that in 'real world'
> :use- regardless of the interface. Burst-mode is, of course, another matter.
> :
> :Thanks,
> :
> :Bill
>     
>     35-50 MBytes/sec is fairly typical for a modern drive these days.  The
>     linear transfer rate goes up with density.  The head seek times, on the
>     otherhand, haven't changed much over the last few years.
>
> 					-Matt
> 					Matthew Dillon 
> 					<dillon@xxxxxxxxxxxxx>

ACK.  Hence 'real world' comment.....

If you want to see astonishing I/O benchmarks, format a ZIP as hpfs-386.

Eject the media.

Run the I/O benchmark against the empty drive.....

Magic!

. ..absent media, the OS/2 version runs the bench against the 100 MB (-) 
cache ...

Which is in system DDRAM.....

Tilt!  ;-)

Bill