[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: determining UDMA rate



hi ya george

On Sat, 18 Jan 2003, George Georgalis wrote:

> I have a 40 pin ribbon on an ide drive (which supports udma100, at
> least)

40 conductor cables does NOT electrically support anything faster
than ata33 -- ( too much crosstalk at faster speeds ) 
they really should have used a different ide connector
for it but oh well..
	
> I know the board only goes up to udma66. do I need an 80 pin ribbon for
> that?

ata66/ata100/ata133 should be using the 18" or less 80-conductor ribbon
cable ... if you use the longer 24" cables, you're gambling
that you're not on the leading edge of disaster ( usually not )
since nobody runs their systems at the max rated specs
- some ide chip drivers can drive a bigger capacitive (cable) load 

> looking at hdparm, I cannot see how to determine the present udma rate.
> how is that done?

the entry with the "*" next to th4e output is your current udma setting
	hdparm -iv | grep udam

to change the value... you're gambling that its supported
	hdparm -X  $(64 + udma_mode_number)

	more ata speed info
	http://www.Linux-1U.net/Disks

> BTW - this is a scsi box, just adding and ide for extra space... I can
> do some hdparm experimenting at this point.

have fun to your hearts content .. see which is faster ??

ide_partition#	time  dd if=/dev/zero of=/raid_partition/test/2G  
		bs=1024 count=2000000

scsi_partition#	time  dd if=/dev/zero of=/raid_partition/test/2G  
		bs=1024 count=2000000

c ya
alvin 



Reply to: