On Thu, 29 Jun 2000, you wrote:
> On Wed, 28 Jun 2000, you wrote:
> 
> > 
> >   Is anybody getting substantially better than 142 mb/sec and 22
> > mb/sec from 'hdparm -tT' ??   ATA/33 is capable of that 'cause
> > those are my numbers, and one (slave) HHD's a WD Caviar 8.4.  ...
> > and is it worth all that hoops y'all seem to be jumpin thru to get
> > it workin? .......... reliab'ly?  and is it reliable ?
> > 
> >   just askin what I seem to never hear about in conjunction with
> > ata/66 and performance.  seems a better motherboard would do all
> > y'all /66'rs more good
> > 
> I'm not even getting THAT...at least not on my EIDE drive
> ('course it's one that's been around a couple years, so
> it's not optimized for UDMA...only an 850 meg drive <G>)
> Now, just for giggles, I"m gonna see what I get on my UW
> SCSI drive... 
> IDE:
> [root@slave1 /root]# hdparm -tT /dev/hda
>  
> /dev/hda:
>  Timing buffer-cache reads:   128 MB in  3.18 seconds = 40.25 MB/sec
>  Timing buffered disk reads:  64 MB in 36.08 seconds =  1.77 MB/sec 

/dev/hda:  
Timing buffer-cache reads:   128 MB in  0.88 seconds = 145.45 MB/sec
Timing buffered disk reads:  64 MB in  2.88 seconds = 22.22 MB/sec
  that's an IBM-DPTA-371360 7200 rpm, 2mb cache on ata/33 IDE with
hdparm -m64 -c1 -u1 -d1 -k1 -a128 /dev/hda   in rc.local

/dev/hdb:
Timing buffer-cache reads:   128 MB in  0.85 seconds =150.59 MB/sec
Timing buffered disk reads:  64 MB in  6.19 seconds = 10.34 MB/sec   
   that's a WD AC38400L 5400rpm, 256k cache on ata/33 IDE with
hdparm -m16 -c1 -u1 -d1 -k1 -a128 /dev/hdb   in rc.local 

   I just ran those, many runs yields about 142+ for both HDD's.
The 22.2 for the IBM is normal and constant. For the life of me I
can't get the WD to run any better (info hdparm suggests a lower
-m<value> for WD's), it's disc reads vary from 10.3 to 11.1

   Motherboard's a Soyo 6ba+III (ata/33) oc'd at 135mhz FSB, but I
doubt the oc enhances HDD performance.  The pci bus is 135/4 (damn
near spec of 33.3mhz), so I'm sure it doesn't hurt HDD performance
either.  Single p3-450 cpu at 4.5x135 (608mhz)

 > 
> SCSI:
> Same exact specs....must be my motherboard can't handle
> anything faster. :-/ Heh, course I'm only using a dual-PPro
> system here... :-)
>       John

   I know I must sound sort'a snippish 'bout this, but I really
think the best way to run ata/66 is to install an old 40 wire IDE
cable and disable it ;-)  It's been shown that ata/66 only improves
burst speeds and that 'real world' performance is more times hurt
more than the little bit /66 helps.  Maybe you can guess i've never
been a scuzzy fan either  ;)

 'bout the only time ata/66 performance is impressive is when
running HDtach on Winblows, all it does is measure burst rates

  -- 
~~   Tom Brinkman    [EMAIL PROTECTED]

Reply via email to