D. Hugh Redelmeier wrote:
That seems like a very reasonable time.  My desktop (PATA, HP
nForce-based AMD 64, a couple of years old) gets 1002.12 MB/sec.  My
newer HP desktop, with SATA, gets 1288.27 MB/sec.

I would worry if it were an order of magnitude slower.

| I don't know how useful this timing test is because my Sony with a SATA hard
| drive tested about the same.

In theory, this test should maximize the difference between SATA and
PATA.  It *only* tests the buffer/cache of the drive and the channel
bandwidth.  SATA should come out faster, I think.

But the difference should not matter.  For real performance, the
characteristics that matter are the disk-to-buffer bandwidth, the
rotational latency, and the seek time.

I think that hdparm -t, not -T, would be more usful.  On my older
desktop, I get 55.26 MB/sec (I get only 36.94 MB/sec on the newer
one!).  Note: this really only tests sequential reading (from the
actual disk, not the buffer).  You need another test program to deal
with access time issues.  Or you could read the spec sheets.

Ok, on the Compaq R3000Z:

# hdparm -t /dev/hda

/dev/hda:
Timing buffered disk reads:  106 MB in  3.10 seconds =  34.14 MB/sec

On my Sony Vaio with a SATA drive:

# hdparm -t /dev/sda

/dev/sda:
Timing buffered disk reads:  110 MB in  3.03 seconds =  36.29 MB/sec


Once again, very minimal difference.  The Vaio certainly "feels" way faster.

--
Greg Gulik                                 http://www.gulik.org/greg/
greg @ gulik.org


_______________________________________________
LinuxR3000 mailing list
[email protected]
http://lists.pcxperience.com/cgi-bin/mailman/listinfo/linuxr3000
Wiki at http://prinsig.se/weekee/

Reply via email to