Re: How to change from UDMA5 to UDMA6



"Mondo" <hmk@xxxxxxxxxxxxxxxx> wrote in message
news:51CNg.53189$ns1.26278@xxxxxxxxxxxxxxxxxxxxxxxxxxx
Thank you guys for your help.

I know for regular IDE drives the Western Digital's came with a
floppy that you could set the UDMA
with. You may want to check to see if they are actually at 6.

Dave my WD does not come with a disk and could not find any tool to
change the UDMA on WD site, only data Lifgaurd tools is there, its
only for copying and partitioning the drive and giving infos about
it, it says the drive supports UDMA6 and current is UDMA5.

If they are then maybe the motherboard needs to be checked to see
if it indeed supports the 6 speed. The specs of the board
from the manufacturer should tell drive speed support.

Nothing is told about that in the manual nor on Gigabyte's site,
thats why I was asking if there is anykind of a testing tool that
will give me such info.

You will not notice the difference between
the two if you are able to switch to 6.

I get it that there is no difference in the speed from UDMA5 to
UDMA6? If so how come I noticed a big difference between lower UDMA
modes and UDMA5? Please explain more because if so then no need to
bother trying to switch.


Because the limiting factor after UDMA 5 is the speed the bus system can
deliver the data. There is no discernible difference betweeh UDMA5 and
UDMA6. Before UDMA 5, you're limiting not only the cached speed, but also
the sustained throughput speed by selecting a lesser mode for data movement.
You won't see this in msdos, it uses throughput period, UDMA setting in the
bios makes no difference. And may sometimes be visibly faster than in XP
environment with current hard drives as there is no or little overhead in
msdos.

Do a little research on the hard drive cache, and how often its used in
everyday use. Throughput is the standard for everyday usage monitoring.
And throughput is limited by the bus system after UDMA5 mode. UDMA6 is
hype, not results.

Also on sandra file system benchmark I'm getting very low results
(50+Mb/s) compared to other similiar systems (100+Mb/s), what could
be the cause of this?

The first may be a dried up (data not applicable on the cache to the SiSoft
test) cache or not used, the latter is using the cache on the hard drive.
The latter is burst speed, not throughput. Then there is also a concern for
overhead, hardware monitoring software and so forth that can affect the
results.

I did alot of tweaking in the past. Hitting deadends maxxing what I had,
and buying faster stuff. In the end, I decided to just use and enjoy the PC
rather than spin my wheels and spend money that doesn't really make a whole
lot of difference for me.
--
Jonny


.