DVD Burner Problem

patrickp wrote:
|
| A while ago I was having trouble with a mobo that disk drive transfers
| weren't working too well on (no, nothing to do with PCI Latency,
| although it was a VIA chipset).
|
| Anyway, one thing I noticed then was that flat 'ribbon' type 80 wire
| cables seemed to give greater reliability than rounded ones. I have
| no idea why.
|
| Perhaps these 80 wire cable problems are with rounded cables - might
| be interesting to see if things are any better with flat ones? Which
| was your client using?
|
| Patrick

Hi Patrick -

With flat 80-wire high-speed IDE cables, every other wire is a ground to
help eliminate signal cross-talk between the lines.

What happens when those 80 wires are all bundled in a rounded cable?

Could be those signal lines really like chattering with their neighbors,
resulting in the IDE channel dropping all the way, step-by-step, to PIO mode
as error thresholds are reached.

This phenomenon can be verified by deleting the IDE controllers and
rebooting (this resets the error thresholds). On reboot, if the IDE channel
is back where it's supposed to be either rounded cables are suspect -or- the
cable is somehow otherwise faulty.

Lurking Rat in 'Da Hood® -
Jef

"Lurking Rat in 'Da Hood®" wrote in message
. ..
..
This phenomenon can be verified by deleting the IDE controllers and
rebooting (this resets the error thresholds). On reboot, if the IDE channel
is back where it's supposed to be either rounded cables are suspect -or- the
cable is somehow otherwise faulty.

Are you talking about Windows? At least on Win2K, it never
sets UDMA mode automatically. You need to go in after
installation and change it from PIO manually. Not sure about
WinXP.

Peter wrote:
|
| Are you talking about Windows? At least on Win2K, it never
| sets UDMA mode automatically. You need to go in after
| installation and change it from PIO manually. Not sure about
| WinXP.
|
|

Hi Peter -

Actually I'm talking (specifically Windows XP) about Windows throttling
*down* transfer speed based on errors exceeding pre-determined thresholds.

Faulty IDE cables and cheap rounded cables are often responsible for this.

My response was to Patrick saying "I have no idea why" when he posted a
suggestion to try a flat IDE cable.

Lurking Rat in 'Da Hood® -
Jef

I'd agree with that. (I had started an overly complicated reply which
basically said the same thing and hadn't yet sent it - thanks for saving me
a lot of time)

Nick

.. --------------------------------------
Mike Richter, were you born with
"Scam Artist" emblazoned on your face?
--------------------------------------
http://tinyurl.com/6eldj

(No Mikey S-lickers have been able to prove ANY of the above is a LIBEL)
( )
( -- despite Mikey claimed to have proof of misquotes !! )
'

NickM wrote:

OK. This is a long thread and it's some days old. I've checked all of it
and understandably nowhere has anyone suggested trying a 40 wire cable.
Recently I had a similar experence with a client who'd installed a new and
relatively expensive Plextor DVD-RW drive. We tried new drivers, setting
the drive jumpers to master and ensuring it was the only device on the cable
and also we tried a known good Pioneer DVD drive to check against the
Plextor. The results were similar to those obtained with the Plextor - i.e.
unreliable.

The customer had used an 80 wire IDE cable, but according to Plextor whom he
phoned, he should have used a 40 wire cable. Hey presto! when a 40 wire
cable was installed, all his problems went away and he can now reliably burn
DVDs. Their reasoning was that the 80 wire cable actually caused more
interference (crosstalk), which is the complete opposite of what I
understood 80 wire cables were for (i.e. to reduce crosstalk). Whatever,
although it sounds stupid and contradicts what we 'know', it worked.

Read somewhere that 80-wire cable is NOT compatible with 40-wire Cable
Select cable. Don't know why that is so, and don't know how to replace
the 40-wire CS cable when 80-wire cable is required.

Also, don't think the round 80-wire cable is made by simply rolling-up
the flat 80-wire cable. If indeed that's case, however, the crosstalk
won't be present. It would be too asinine for
manufacturers/specification not to have considered the issue.

That's right in most cases. If a 40 wire cable is specifically manufactured
as a cable select cable there will be a deliberate break in one of the wires
near the motherboard connector (OK on a standard 40 wire cable the
connectors are all the same, but I refer to the motherboard end as the
connector n the longer 'half' of the cable). The break simply looks like a
small hole in the cable. I have a couple of 80 wire cables with the same
type of 'hole'. With a standard 40 wire cable there is no particular
preference electrically as to which connector plugs in where. However with
a standard 80 wire cable it is very important that the connector at the end
of the longer 'half' of the cable is plugged into the board, the connector
at the other end is plugged into the master drive, and the one in he middle
is plugged into the slave drive.

Nick

On Sat, 17 Sep 2005 11:15:28 +0100, smh wrote:

NickM wrote:

OK. This is a long thread and it's some days old. I've checked all of it
and understandably nowhere has anyone suggested trying a 40 wire cable.
Recently I had a similar experence with a client who'd installed a new and
relatively expensive Plextor DVD-RW drive. We tried new drivers, setting
the drive jumpers to master and ensuring it was the only device on the cable
and also we tried a known good Pioneer DVD drive to check against the
Plextor. The results were similar to those obtained with the Plextor - i.e.
unreliable.

The customer had used an 80 wire IDE cable, but according to Plextor whom he
phoned, he should have used a 40 wire cable. Hey presto! when a 40 wire
cable was installed, all his problems went away and he can now reliably burn
DVDs. Their reasoning was that the 80 wire cable actually caused more
interference (crosstalk), which is the complete opposite of what I
understood 80 wire cables were for (i.e. to reduce crosstalk). Whatever,
although it sounds stupid and contradicts what we 'know', it worked.

Read somewhere that 80-wire cable is NOT compatible with 40-wire Cable
Select cable. Don't know why that is so, and don't know how to replace
the 40-wire CS cable when 80-wire cable is required.

"Cable select cable"? I thought ANY cable worked with cable select. I always use the cable select option and have never bought special cable.

Also, don't think the round 80-wire cable is made by simply rolling-up
the flat 80-wire cable. If indeed that's case, however, the crosstalk
won't be present. It would be too asinine for
manufacturers/specification not to have considered the issue.



--
http://www.petersparrots.com http://www.insanevideoclips.com http://www.petersphotos.com

How come the dove gets to be the peace symbol? How about the pillow? It has more feathers than the dove, and it doesn't have that dangerous beak!

On Thu, 27 Oct 2005 21:12:25 +0100, "Peter Hucker"
wrote:

On Sat, 17 Sep 2005 11:15:28 +0100, smh wrote:

NickM wrote:

OK. This is a long thread and it's some days old. I've checked all of it
and understandably nowhere has anyone suggested trying a 40 wire cable.
Recently I had a similar experence with a client who'd installed a new and
relatively expensive Plextor DVD-RW drive. We tried new drivers, setting
the drive jumpers to master and ensuring it was the only device on the cable
and also we tried a known good Pioneer DVD drive to check against the
Plextor. The results were similar to those obtained with the Plextor - i.e.
unreliable.

The customer had used an 80 wire IDE cable, but according to Plextor whom he
phoned, he should have used a 40 wire cable. Hey presto! when a 40 wire
cable was installed, all his problems went away and he can now reliably burn
DVDs. Their reasoning was that the 80 wire cable actually caused more
interference (crosstalk), which is the complete opposite of what I
understood 80 wire cables were for (i.e. to reduce crosstalk). Whatever,
although it sounds stupid and contradicts what we 'know', it worked.

Read somewhere that 80-wire cable is NOT compatible with 40-wire Cable
Select cable. Don't know why that is so, and don't know how to replace
the 40-wire CS cable when 80-wire cable is required.

"Cable select cable"? I thought ANY cable worked with cable select. I always use the cable select option and have never bought special cable.

The original 40 conductor IDE cables had full connections going to
both drive connectors. Those won't work with cable select.

The old 40 wire CS cables had a gap in conductor 28, between the two
drive connectors. The middle drive thus sees a connection and
declares itself master, and the one past the gap sees what looks like
a ground, so declares itself slave. That's how cable select works.
Yeah, I know, that's backwards from the normal master/slave
configuration -- keep reading before you nitpick.

Some were wired in a "Y" configuration, with a complete set of
connections going to the end drive (thus making it master) and a
branch leading to the middle connector. This middle branch had
conductor 28 broken, making the middle connector CS slave. This was
more in keeping with the traditional drive orientation.

Later cables were wired in the "Y" configuration but put the break
into the middle connector itself, so there was no external indication
that there was any difference between it and a "regular" cable. Since
pin28 is dedicated to cable select chores, it doesn't affect manual
master/slave configurations.

The 80 conductor cables are all supposed to be CS cables wired up in Y
configuration as described in that previous paragraph, so the
conductor 28 stuff should be invisible. However, it's conceivable
that someone could have just given a full connection to both
connectors. It wouldn't be the first time that I've seen an IDE cable
on the shelf that wasn't 100% conformant in some way (usually it's
just something silly like the colour of the connector plastic).

---------------------------------------------
Thanks.


MCheu

On Thu, 27 Oct 2005 23:19:36 +0100, MCheu wrote:

On Thu, 27 Oct 2005 21:12:25 +0100, "Peter Hucker"
wrote:

On Sat, 17 Sep 2005 11:15:28 +0100, smh wrote:

NickM wrote:

OK. This is a long thread and it's some days old. I've checked all of it
and understandably nowhere has anyone suggested trying a 40 wire cable.
Recently I had a similar experence with a client who'd installed a new and
relatively expensive Plextor DVD-RW drive. We tried new drivers, setting
the drive jumpers to master and ensuring it was the only device on the cable
and also we tried a known good Pioneer DVD drive to check against the
Plextor. The results were similar to those obtained with the Plextor - i.e.
unreliable.

The customer had used an 80 wire IDE cable, but according to Plextor whom he
phoned, he should have used a 40 wire cable. Hey presto! when a 40 wire
cable was installed, all his problems went away and he can now reliably burn
DVDs. Their reasoning was that the 80 wire cable actually caused more
interference (crosstalk), which is the complete opposite of what I
understood 80 wire cables were for (i.e. to reduce crosstalk). Whatever,
although it sounds stupid and contradicts what we 'know', it worked.

Read somewhere that 80-wire cable is NOT compatible with 40-wire Cable
Select cable. Don't know why that is so, and don't know how to replace
the 40-wire CS cable when 80-wire cable is required.

"Cable select cable"? I thought ANY cable worked with cable select. I always use the cable select option and have never bought special cable.

The original 40 conductor IDE cables had full connections going to
both drive connectors. Those won't work with cable select.

The old 40 wire CS cables had a gap in conductor 28, between the two
drive connectors. The middle drive thus sees a connection and
declares itself master, and the one past the gap sees what looks like
a ground, so declares itself slave. That's how cable select works.
Yeah, I know, that's backwards from the normal master/slave
configuration -- keep reading before you nitpick.

Some were wired in a "Y" configuration, with a complete set of
connections going to the end drive (thus making it master) and a
branch leading to the middle connector. This middle branch had
conductor 28 broken, making the middle connector CS slave. This was
more in keeping with the traditional drive orientation.

Later cables were wired in the "Y" configuration but put the break
into the middle connector itself, so there was no external indication
that there was any difference between it and a "regular" cable. Since
pin28 is dedicated to cable select chores, it doesn't affect manual
master/slave configurations.

The 80 conductor cables are all supposed to be CS cables wired up in Y
configuration as described in that previous paragraph, so the
conductor 28 stuff should be invisible.

And there was me thinking that the drives were clever enough to detect (by signal bouncing?) whether they were in the middle or at the end!

I've connected drives in CS mode successfully on very old machines (486 era) with the original old cables. How old would a cable have to be to not be a CS cable?

However, it's conceivable
that someone could have just given a full connection to both
connectors. It wouldn't be the first time that I've seen an IDE cable
on the shelf that wasn't 100% conformant in some way (usually it's
just something silly like the colour of the connector plastic).

You mean they don't work so well if they're the wrong colour? ;-)

--
http://www.petersparrots.com http://www.insanevideoclips.com http://www.petersphotos.com

We've all heard that a million monkeys banging on a million typewriters will eventually reproduce the works of Shakespeare.
Now, thanks to the Internet, we know this is not true.

On Thu, 27 Oct 2005 18:19:36 -0400, MCheu wrote:

On Thu, 27 Oct 2005 21:12:25 +0100, "Peter Hucker"
wrote:

On Sat, 17 Sep 2005 11:15:28 +0100, smh wrote:

NickM wrote:

OK. This is a long thread and it's some days old. I've checked all of it
and understandably nowhere has anyone suggested trying a 40 wire cable.
Recently I had a similar experence with a client who'd installed a new and
relatively expensive Plextor DVD-RW drive. We tried new drivers, setting
the drive jumpers to master and ensuring it was the only device on the cable
and also we tried a known good Pioneer DVD drive to check against the
Plextor. The results were similar to those obtained with the Plextor - i.e.
unreliable.

The customer had used an 80 wire IDE cable, but according to Plextor whom he
phoned, he should have used a 40 wire cable. Hey presto! when a 40 wire
cable was installed, all his problems went away and he can now reliably burn
DVDs. Their reasoning was that the 80 wire cable actually caused more
interference (crosstalk), which is the complete opposite of what I
understood 80 wire cables were for (i.e. to reduce crosstalk). Whatever,
although it sounds stupid and contradicts what we 'know', it worked.

Read somewhere that 80-wire cable is NOT compatible with 40-wire Cable
Select cable. Don't know why that is so, and don't know how to replace
the 40-wire CS cable when 80-wire cable is required.

"Cable select cable"? I thought ANY cable worked with cable select. I always use the cable select option and have never bought special cable.

The original 40 conductor IDE cables had full connections going to
both drive connectors. Those won't work with cable select.

The old 40 wire CS cables had a gap in conductor 28, between the two
drive connectors. The middle drive thus sees a connection and
declares itself master, and the one past the gap sees what looks like
a ground, so declares itself slave. That's how cable select works.
Yeah, I know, that's backwards from the normal master/slave
configuration -- keep reading before you nitpick.

Some were wired in a "Y" configuration, with a complete set of
connections going to the end drive (thus making it master) and a
branch leading to the middle connector. This middle branch had
conductor 28 broken, making the middle connector CS slave. This was
more in keeping with the traditional drive orientation.

Later cables were wired in the "Y" configuration but put the break
into the middle connector itself, so there was no external indication
that there was any difference between it and a "regular" cable. Since
pin28 is dedicated to cable select chores, it doesn't affect manual
master/slave configurations.

The 80 conductor cables are all supposed to be CS cables wired up in Y
configuration as described in that previous paragraph, so the
conductor 28 stuff should be invisible. However, it's conceivable
that someone could have just given a full connection to both
connectors. It wouldn't be the first time that I've seen an IDE cable
on the shelf that wasn't 100% conformant in some way (usually it's
just something silly like the colour of the connector plastic).

Actually, some 80 wire cables are wired with the centre connector as
master. I have a couple somwhere.

Patrick

- take five to email me...