Transfering Data from one hard drive to another

Thanks Barry, - I think! g

Not precisely sure quite what you mean in your opening sentence :-

The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.

Perhaps you meant :-

The reasons for not putting a drive (of a DIFFERENT type) on ....

Illustrated by your informative explanation and summarized in your later
sentence :-

Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.

with which, I totally concur, citing my own setup as an example.

Or perhaps you even meant only ONE device per channel (controller) bg

However, Rick T's statement was :-

.. don't put hard-drives onto the same IDE channel.
.. if you can avoid it though, .... etc.,

which begs the question - Under 'ideal' conditions, are read/write
operations BETWEEN two disks on the SAME IDE channel faster or slower than
read/write operations between those same two disks, each on a DIFFERENT IDE
channel?

Likewise, for copying CD's, does the same argument apply? Should you put
each CD-ROM on a different channel?

Muddy water?? Hence, my question of 'best practice' vs. 'myth'.

I'm not trying to be controversial - I genuinely don't know!

Mart


"BarryG" wrote in message
...

Interesting comment Rick. I'm not sure if there's a 'best practice'
explanation to your suggestion, or whether it's just another computing
"myth". Sure, I've heard the 'so-called' advantages of separating HD's
and
CD-ROM's etc., to speed-up read/write operations but...



Mart,

The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.

First up, CD-Rom drives are slow! I believe ATA-33 is about as fast as
they
go.
Late model ATA hard disks are capable of ATA-100/133 under the following
conditions.
1. The motherboard supports ATA-100/133, and BIOS is set to AUTO for the
bus
speed.
2. An ATA-100/133 IDE bus cable is used. It has 80 wires, instead of the
usual 40 wires. Still only 40 pins, but 80 wires. Every second wire is
earthed to provide shielding for the higher bus speed.
3. The hard disk drive is capable.
4. ALL devices on the bus are ATA-100/133 capable.

So, the bottom line is:-
A bus will only run as fast as the slowest device on the bus.
Put a CD-ROM on the same bus as an ATA-100/133 hard disk, and all you get
is..... ATA-33!!!!!
Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.
If
you must put a hard disk on the same bus as the CD-Rom, don't expect
maximum
performance from, (unless of course, the HD is an old slow one)

BTW. For CS (Cable Select) to work you must use an 80 wire ATA-100/133
cable, the master drive goes on the end plug, and the slave drive goes on
the
middle plug. The connectors should be labelled Primary, Secondary, and
System
Board.

Good Luck,
BarryG

I'm (reasonably) sure that each channel buss uses the top speed for
whatever device is being used (so even though you have a slower optical
device on the channel, that slower speed will only happen for xfers
to/from the optical device).

Data transfer is *not* direct between devices; the data still has to go
through Windows(or other) before going to the other device (think
different filesystems; the peripheral doesn't know anything about them,
that's the OS's work).

If you only have data going one way at a time on each channel you get
the full bandwidth... if data has to xfer back and forth on the same
channel you only get half each way.

So if you were running an MP3 shop you might have 2 hard drives on one
channel and the CD and the CDR on the other because the dataflow would
be from the CD to the 2nd HD and from the 2nd HD to the CDR.


Rick

(I thought CS cables were 'Y' cables... you say they look like the
standard cables?)




Mart wrote:
Thanks Barry, - I think! g

Not precisely sure quite what you mean in your opening sentence :-


The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.


Perhaps you meant :-

The reasons for not putting a drive (of a DIFFERENT type) on ....

Illustrated by your informative explanation and summarized in your later
sentence :-


Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.


with which, I totally concur, citing my own setup as an example.

Or perhaps you even meant only ONE device per channel (controller) bg

However, Rick T's statement was :-


.. don't put hard-drives onto the same IDE channel.
.. if you can avoid it though, .... etc.,


which begs the question - Under 'ideal' conditions, are read/write
operations BETWEEN two disks on the SAME IDE channel faster or slower than
read/write operations between those same two disks, each on a DIFFERENT IDE
channel?

Likewise, for copying CD's, does the same argument apply? Should you put
each CD-ROM on a different channel?

Muddy water?? Hence, my question of 'best practice' vs. 'myth'.

I'm not trying to be controversial - I genuinely don't know!

Mart


"BarryG" wrote in message
...

Interesting comment Rick. I'm not sure if there's a 'best practice'
explanation to your suggestion, or whether it's just another computing
"myth". Sure, I've heard the 'so-called' advantages of separating HD's
and
CD-ROM's etc., to speed-up read/write operations but...



Mart,

The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.

First up, CD-Rom drives are slow! I believe ATA-33 is about as fast as
they
go.
Late model ATA hard disks are capable of ATA-100/133 under the following
conditions.
1. The motherboard supports ATA-100/133, and BIOS is set to AUTO for the
bus
speed.
2. An ATA-100/133 IDE bus cable is used. It has 80 wires, instead of the
usual 40 wires. Still only 40 pins, but 80 wires. Every second wire is
earthed to provide shielding for the higher bus speed.
3. The hard disk drive is capable.
4. ALL devices on the bus are ATA-100/133 capable.

So, the bottom line is:-
A bus will only run as fast as the slowest device on the bus.
Put a CD-ROM on the same bus as an ATA-100/133 hard disk, and all you get
is..... ATA-33!!!!!
Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.
If
you must put a hard disk on the same bus as the CD-Rom, don't expect
maximum
performance from, (unless of course, the HD is an old slow one)

BTW. For CS (Cable Select) to work you must use an 80 wire ATA-100/133
cable, the master drive goes on the end plug, and the slave drive goes on
the
middle plug. The connectors should be labelled Primary, Secondary, and
System
Board.

Good Luck,
BarryG

Thanks for your further input Rick.

See in-line

"Rick T" wrote :-
I'm (reasonably) sure that each channel buss uses the top speed for
whatever device is being used (so even though you have a slower optical
device on the channel, that slower speed will only happen for xfers
to/from the optical device).

Me too (reasonably g). That was my 'assumption' with the caveat that it
didn't apply to 'older' kit, but can't find a definative article. So does
this support or negate you previous assertion?
.. don't put hard-drives onto the same IDE channel.
.. if you can avoid it though, .... etc.,

Data transfer is *not* direct between devices; the data still has to go
through Windows(or other) before going to the other device (think
different filesystems; the peripheral doesn't know anything about them,
that's the OS's work).

Yes, I can see that makes sense.

If you only have data going one way at a time on each channel you get the
full bandwidth... if data has to xfer back and forth on the same channel
you only get half each way.

Mmm.. not too sure - I thought that IDE 'acted' like full duplex at full
bandwidth .. but ?

So if you were running an MP3 shop you might have 2 hard drives on one
channel and the CD and the CDR on the other because the dataflow would be
from the CD to the 2nd HD and from the 2nd HD to the CDR.

But from your first paragraph, "uses the top speed for whatever device"
which of course brings back the original point - 'best practice' vs. 'myth'.
I can't find anything on the web. I've found a couple describing IDE
Controllers but none explaining this issue.


Rick

(I thought CS cables were 'Y' cables... you say they look like the
standard cables?)

The 'original' was a 12 inch 80 way ribbon blue ended, labelled 'MoBo'
connector and a black ended labelled 'Master' connector. It also has a piece
of wire 60 (I think) cut from the ribbon at the blue end. I don't know its
significance.

The replacement (Belkin) 18 inch 80 way ribbon blue ended, black ended and
intermediate grey connector about 6 inches from the black end. (No cut in
wire 60). The package details gave connection instructions.

Mart



Mart wrote:
Thanks Barry, - I think! g

Not precisely sure quite what you mean in your opening sentence :-


The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.


Perhaps you meant :-

The reasons for not putting a drive (of a DIFFERENT type) on ....

Illustrated by your informative explanation and summarized in your later
sentence :-


Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.


with which, I totally concur, citing my own setup as an example.

Or perhaps you even meant only ONE device per channel (controller) bg

However, Rick T's statement was :-


.. don't put hard-drives onto the same IDE channel.
.. if you can avoid it though, .... etc.,


which begs the question - Under 'ideal' conditions, are read/write
operations BETWEEN two disks on the SAME IDE channel faster or slower
than read/write operations between those same two disks, each on a
DIFFERENT IDE channel?

Likewise, for copying CD's, does the same argument apply? Should you
put each CD-ROM on a different channel?

Muddy water?? Hence, my question of 'best practice' vs. 'myth'.

I'm not trying to be controversial - I genuinely don't know!

Mart


"BarryG" wrote in message
...

Interesting comment Rick. I'm not sure if there's a 'best practice'
explanation to your suggestion, or whether it's just another computing
"myth". Sure, I've heard the 'so-called' advantages of separating HD's
and
CD-ROM's etc., to speed-up read/write operations but...



Mart,

The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.

First up, CD-Rom drives are slow! I believe ATA-33 is about as fast as
they
go.
Late model ATA hard disks are capable of ATA-100/133 under the following
conditions.
1. The motherboard supports ATA-100/133, and BIOS is set to AUTO for the
bus
speed.
2. An ATA-100/133 IDE bus cable is used. It has 80 wires, instead of the
usual 40 wires. Still only 40 pins, but 80 wires. Every second wire is
earthed to provide shielding for the higher bus speed.
3. The hard disk drive is capable.
4. ALL devices on the bus are ATA-100/133 capable.

So, the bottom line is:-
A bus will only run as fast as the slowest device on the bus.
Put a CD-ROM on the same bus as an ATA-100/133 hard disk, and all you get
is..... ATA-33!!!!!
Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.
If
you must put a hard disk on the same bus as the CD-Rom, don't expect
maximum
performance from, (unless of course, the HD is an old slow one)

BTW. For CS (Cable Select) to work you must use an 80 wire ATA-100/133
cable, the master drive goes on the end plug, and the slave drive goes on
the
middle plug. The connectors should be labelled Primary, Secondary, and
System
Board.

Good Luck,
BarryG

Rick, a further snippet CS cables and 'Y' connectors.

http://www.pcguide.com/ref/hdd/if/ide/confCS-c.html
See Universality and the 'STOP' warning near the bottom of the page.

PS. FWIW - I think its wire 65 and not 60 that's been cut.

Mart

Mart,

The reason for not putting a fast hard disk, and a slow CD-Rom on the same
bus is that the bus can only run at the speed of the slowest device on the
bus. You will cripple the transfer speed to your lovely fast hard disk by
putting a CD-Rom on the same bus with it.
The transfer speed on the bus is set at bootup time, when the BIOS
identifies all the devices, and their speed capabilities. If the BIOS IDE bus
speed is set to AUTO, it will use the fastest speed possible on the bus,
which the speed of the slowest device.

For a description of the workings of IDE bus, here's the specification.
http://ssdl-delta.stanford.edu/pascal/standards/ATA-4_(d1153r18).pdf
Section 4 is the part on the electrical and physical interface.
Appendix A, table A1 and A2 show the comparison of a 40 pin 40 wire cable,
and a 40 pin 80 wire cable.

Good luck.
BarryG

"Mart" wrote:

Thanks Barry, - I think! g

Not precisely sure quite what you mean in your opening sentence :-

The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.

Perhaps you meant :-

The reasons for not putting a drive (of a DIFFERENT type) on ....

Illustrated by your informative explanation and summarized in your later
sentence :-

Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.

with which, I totally concur, citing my own setup as an example.

Or perhaps you even meant only ONE device per channel (controller) bg

However, Rick T's statement was :-

.. don't put hard-drives onto the same IDE channel.
.. if you can avoid it though, .... etc.,

which begs the question - Under 'ideal' conditions, are read/write
operations BETWEEN two disks on the SAME IDE channel faster or slower than
read/write operations between those same two disks, each on a DIFFERENT IDE
channel?

Likewise, for copying CD's, does the same argument apply? Should you put
each CD-ROM on a different channel?

Muddy water?? Hence, my question of 'best practice' vs. 'myth'.

I'm not trying to be controversial - I genuinely don't know!

Mart


"BarryG" wrote in message
...

Interesting comment Rick. I'm not sure if there's a 'best practice'
explanation to your suggestion, or whether it's just another computing
"myth". Sure, I've heard the 'so-called' advantages of separating HD's
and
CD-ROM's etc., to speed-up read/write operations but...



Mart,

The reasons for not putting a drive (of any type) on the same bus as any
other dirve have to do with the bus speed.

First up, CD-Rom drives are slow! I believe ATA-33 is about as fast as
they
go.
Late model ATA hard disks are capable of ATA-100/133 under the following
conditions.
1. The motherboard supports ATA-100/133, and BIOS is set to AUTO for the
bus
speed.
2. An ATA-100/133 IDE bus cable is used. It has 80 wires, instead of the
usual 40 wires. Still only 40 pins, but 80 wires. Every second wire is
earthed to provide shielding for the higher bus speed.
3. The hard disk drive is capable.
4. ALL devices on the bus are ATA-100/133 capable.

So, the bottom line is:-
A bus will only run as fast as the slowest device on the bus.
Put a CD-ROM on the same bus as an ATA-100/133 hard disk, and all you get
is..... ATA-33!!!!!
Keep your fast hard disks on one bus, keep your CD-Roms on the other bus.
If
you must put a hard disk on the same bus as the CD-Rom, don't expect
maximum
performance from, (unless of course, the HD is an old slow one)

BTW. For CS (Cable Select) to work you must use an 80 wire ATA-100/133
cable, the master drive goes on the end plug, and the slave drive goes on
the
middle plug. The connectors should be labelled Primary, Secondary, and
System
Board.

Good Luck,
BarryG

Mart wrote:
Thanks for your further input Rick.


Thanks for pulling me back into it.


Rick

Rick wrote :-
Thanks for pulling me back into it.

You're welcome Rick - I'll take that as a "Load balance" then g

Mart


"Rick T" wrote in message
...
Mart wrote:
Thanks for your further input Rick.


Thanks for pulling me back into it.


Rick

Thanks Barry - and for the 339 pages of homework g Sheesh! They should
keep me quiet for a while.

Seriously though, thanks again for the info. As I keep saying, you learn
something every day on these NG's.

Mart


"BarryG" wrote in message
...
Mart,

The reason for not putting a fast hard disk, and a slow CD-Rom on the same
bus is that the bus can only run at the speed of the slowest device on the
bus. You will cripple the transfer speed to your lovely fast hard disk by
putting a CD-Rom on the same bus with it.
The transfer speed on the bus is set at bootup time, when the BIOS
identifies all the devices, and their speed capabilities. If the BIOS IDE
bus
speed is set to AUTO, it will use the fastest speed possible on the bus,
which the speed of the slowest device.

For a description of the workings of IDE bus, here's the specification.
http://ssdl-delta.stanford.edu/pascal/standards/ATA-4_(d1153r18).pdf
Section 4 is the part on the electrical and physical interface.
Appendix A, table A1 and A2 show the comparison of a 40 pin 40 wire cable,
and a 40 pin 80 wire cable.

Good luck.
BarryG

"Mart" wrote:

Thanks Barry, - I think! g

Not precisely sure quite what you mean in your opening sentence :-

The reasons for not putting a drive (of any type) on the same bus as
any
other dirve have to do with the bus speed.

Perhaps you meant :-

The reasons for not putting a drive (of a DIFFERENT type) on ....

Illustrated by your informative explanation and summarized in your later
sentence :-

Keep your fast hard disks on one bus, keep your CD-Roms on the other
bus.

with which, I totally concur, citing my own setup as an example.

Or perhaps you even meant only ONE device per channel (controller) bg

However, Rick T's statement was :-

.. don't put hard-drives onto the same IDE channel.
.. if you can avoid it though, .... etc.,

which begs the question - Under 'ideal' conditions, are read/write
operations BETWEEN two disks on the SAME IDE channel faster or slower
than
read/write operations between those same two disks, each on a DIFFERENT
IDE
channel?

Likewise, for copying CD's, does the same argument apply? Should you
put
each CD-ROM on a different channel?

Muddy water?? Hence, my question of 'best practice' vs. 'myth'.

I'm not trying to be controversial - I genuinely don't know!

Mart


"BarryG" wrote in message
...

Interesting comment Rick. I'm not sure if there's a 'best practice'
explanation to your suggestion, or whether it's just another computing
"myth". Sure, I've heard the 'so-called' advantages of separating HD's
and
CD-ROM's etc., to speed-up read/write operations but...



Mart,

The reasons for not putting a drive (of any type) on the same bus as
any
other dirve have to do with the bus speed.

First up, CD-Rom drives are slow! I believe ATA-33 is about as fast as
they
go.
Late model ATA hard disks are capable of ATA-100/133 under the
following
conditions.
1. The motherboard supports ATA-100/133, and BIOS is set to AUTO for
the
bus
speed.
2. An ATA-100/133 IDE bus cable is used. It has 80 wires, instead of
the
usual 40 wires. Still only 40 pins, but 80 wires. Every second wire is
earthed to provide shielding for the higher bus speed.
3. The hard disk drive is capable.
4. ALL devices on the bus are ATA-100/133 capable.

So, the bottom line is:-
A bus will only run as fast as the slowest device on the bus.
Put a CD-ROM on the same bus as an ATA-100/133 hard disk, and all you
get
is..... ATA-33!!!!!
Keep your fast hard disks on one bus, keep your CD-Roms on the other
bus.
If
you must put a hard disk on the same bus as the CD-Rom, don't expect
maximum
performance from, (unless of course, the HD is an old slow one)

BTW. For CS (Cable Select) to work you must use an 80 wire ATA-100/133
cable, the master drive goes on the end plug, and the slave drive goes
on
the
middle plug. The connectors should be labelled Primary, Secondary, and
System
Board.

Good Luck,
BarryG

Mart wrote:
Rick wrote :-

Thanks for pulling me back into it.


You're welcome Rick - I'll take that as a "Load balance" then g

Mart

was this that thread?... should just start saying "it's magic" or
something I suppose.

On days when I peruse these NGs prior to/during morning coffee, I
usually blip over the new posts half-asleep unless something catches my
eye. "Rick said..." pretty well did it.

Check your DeviceManager/IDE Controller properties. Win2K shows what
"current usage" for xfer protocol is for each device. Dunno if WinME does.


Rick

"Rick T" wrote in message
...

Mart wrote:

Thanks for your further input Rick.


Thanks for pulling me back into it.


Rick