should I change how my drives are cabled?

[Talal Itani]

Hello,

I have in my PC two hard drives, and two CD-ROM drives. The two hard drives
are connected to the same IDE cable. The two CD-ROM drives are connected to
the same IDE cable. Do I speed things up if I connect a hard drive, and a
CD-ROM drive to the same cable? My computer is running XP.

Thanks.
T.I.

 

[Anna]

Hello,

I have in my PC two hard drives, and two CD-ROM drives. The two hard
drives are connected to the same IDE cable. The two CD-ROM drives are
connected to the same IDE cable. Do I speed things up if I connect a hard
drive, and a CD-ROM drive to the same cable? My computer is running XP.

Thanks.
T.I.

As you shortly will discover, you're probably going to get a number of
conflicting responses recommending this or that configuration of your
IDE-connected devices. All I can tell you is that based upon my own
experience and tests the computer facility I was associated with conducted a
few years ago on this very issue -- in virtually every case, when working
with modern equipment, aside from connecting one's working PATA HDD as
Primary Master, it really didn't matter performance-wise how the remaining
drives (hard drives & optical drives) were connected on the two IDE
channels. Nearly all of our tests were conducted with connecting two hard
drives and two optical drives - a CD-ROM & a CD-DVD burner.

Note I said in "virtually every case" there were no significant performance
differences regardless of the IDE device configuration., There were,
however, some rather rare situations where it *did* matter with respect to
HDD connections/configurations. This usually involved the encoding/decoding
of extremely large video files (gigabytes in size) so this was an issue that
would involve only a extremely small percentage of PC users.

Also, again in some very rare instances, where the process involved copying
CDs (we didn't use DVDs at the time of these tests) from one optical drive
to another optical drive, there were some instances (rare as they might be)
where the configuration of the optical drives *did* matter in terms of
performance. Strangely enough, in that situation we were unable to come up
with a hard & fast rule as to the best configuration of the optical drives.
In some cases we found better, i.e., faster, data transfer rates when both
optical drives were connected on the same channel. In other cases we found
it was best to connect each on a separate IDE channel. And we could find no
correlation involving the make/model of these optical drives. It was quite
puzzling. But let me emphasize that these were relatively rare exceptions.
As I previously stated, we generally found *no* significant performance
differences regardless of how the optical drives were connected/configured.

I would add one additional thing. If both HDDs are bootable devices and the
user has occasion to boot to one or the other drive, then it might be
necessary to connect/configure the second HDD as Secondary Master, rather
than as a Slave to the Primary Master or a Slave on the secondary IDE
channel. We came across a number of motherboards that balked at booting to a
potentially bootable HDD that was connected in the Slave position. But, in
general, it simply didn't matter.

But do this. Experiment for yourself in determining the precise
configuration of your IDE devices. In this instance don't rely on my advice
or anyone else's advice. Try different configurations of your IDE devices
and run simple speed tests based on your normal & usual day-to-day
activities with the computer, e.g., accessing programs, moving/copying
files, burning CDs, etc. See if you can determine any performance difference
depending upon how this or that device is connected, and thus determine the
best setup for your particular needs should there be any significant
differences.
Anna

 

[Bill Ridgeway]

As you shortly will discover, you're probably going to get a number of
conflicting responses recommending this or that configuration of your
IDE-connected devices. All I can tell you is that based upon my own
experience and tests the computer facility I was associated with conducted
a few years ago on this very issue -- in virtually every case, when
working with modern equipment, aside from connecting one's working PATA
HDD as Primary Master, it really didn't matter performance-wise how the
remaining drives (hard drives & optical drives) were connected on the two
IDE channels. Nearly all of our tests were conducted with connecting two
hard drives and two optical drives - a CD-ROM & a CD-DVD burner.

Note I said in "virtually every case" there were no significant
performance
differences regardless of the IDE device configuration., There were,
however, some rather rare situations where it *did* matter with respect to
HDD connections/configurations. This usually involved the
encoding/decoding of extremely large video files (gigabytes in size) so
this was an issue that would involve only a extremely small percentage of
PC users.

Also, again in some very rare instances, where the process involved
copying CDs (we didn't use DVDs at the time of these tests) from one
optical drive to another optical drive, there were some instances (rare as
they might be) where the configuration of the optical drives *did* matter
in terms of performance. Strangely enough, in that situation we were
unable to come up with a hard & fast rule as to the best configuration of
the optical drives. In some cases we found better, i.e., faster, data
transfer rates when both optical drives were connected on the same
channel. In other cases we found it was best to connect each on a separate
IDE channel. And we could find no correlation involving the make/model of
these optical drives. It was quite puzzling. But let me emphasize that
these were relatively rare exceptions. As I previously stated, we
generally found *no* significant performance differences regardless of how
the optical drives were connected/configured.

I would add one additional thing. If both HDDs are bootable devices and
the user has occasion to boot to one or the other drive, then it might be
necessary to connect/configure the second HDD as Secondary Master, rather
than as a Slave to the Primary Master or a Slave on the secondary IDE
channel. We came across a number of motherboards that balked at booting to
a potentially bootable HDD that was connected in the Slave position. But,
in general, it simply didn't matter.

But do this. Experiment for yourself in determining the precise
configuration of your IDE devices. In this instance don't rely on my
advice or anyone else's advice. Try different configurations of your IDE
devices and run simple speed tests based on your normal & usual day-to-day
activities with the computer, e.g., accessing programs, moving/copying
files, burning CDs, etc. See if you can determine any performance
difference depending upon how this or that device is connected, and thus
determine the best setup for your particular needs should there be any
significant differences.
Anna

I don't want to detract from your clear explanation. However, although you
say there are rare occasions when the configuration does matter you haven't
commented on the significance of the configuration was when it was found.
The instances of configuration making any differences may be rare but if,
when it does occur performance is also insignificant makes it even less
worth worrying about.

Regards.

Bill Ridgeway

 

[RalfG]

There is a potential issue due to cabling. High speed harddrives can require
80 wire data cables for best performance and sometimes won't work at all on
40 wire data cables. The existing data cable for the optical drives might
only be 40 wires.

 

[Anna]

I don't want to detract from your clear explanation. However, although
you say there are rare occasions when the configuration does matter you
haven't commented on the significance of the configuration was when it was
found. The instances of configuration making any differences may be rare
but if, when it does occur performance is also insignificant makes it even
less worth worrying about.

Regards.

Bill Ridgeway

Bill:
It's really difficult, if not truly impossible, to precisely quantify (in
this particular instance or issue) what is "significant" or "insignificant"
in terms of a level of performance. As an example, for some users a
difference of say - 20 seconds in the total copying time of one CD to
another CD would be considered an enormous difference, i.e., a "significant"
difference - for others this was a trifling difference and not worth
worrying about.

As I tried to explain, in our view we considered whatever differences we
encountered re performance (speed) based on this or that IDE configuration
were, in the main - "insignificant". It was simply a value judgment we made
on how we considered the vast majority of users would look upon these
differences.

That is why I emphasized in my final paragraph that the user experiment for
him or herself in setting up their IDE devices in this or that configuration
and determine if he or she encounters what they would consider any
"significant" difference in performance due to one or another type of
configuration.
Anna

 

[David B.]

You already have good replies, so as simple an answer as I can provide, if
your secondary channel cable going to the CD drives is a 40 conductor and
not an 80 conductor, leave your setup as is, otherwise you can experiment
but you won't see any noticeable gains.