Sign in with
Sign up | Sign in
Your question

Western Digital + LiteOn master/slave problems

Last response: in Systems
Share
Anonymous
a b B Homebuilt system
September 6, 2004 10:01:09 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

I am trying to get WD800JB and LiteOn LTR 48246S to work together as
secondary master and secondary slave respectively. The problem is that
the auto-detection of the drives during POST just hangs forever (or for
a long time at least). If I switch the drives so that LiteOn is master
and WD is slave, everything works. I think I have tried all possible
combinations of jumper settings.

Does anybody know a solution?

/David
Anonymous
a b B Homebuilt system
September 7, 2004 12:14:37 AM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Mon, 06 Sep 2004 18:01:09 +0200, David Rasmussen
<david.rasmussen@gmx.net> wrote:

>I am trying to get WD800JB and LiteOn LTR 48246S to work together as
>secondary master and secondary slave respectively. The problem is that
>the auto-detection of the drives during POST just hangs forever (or for
>a long time at least). If I switch the drives so that LiteOn is master
>and WD is slave, everything works. I think I have tried all possible
>combinations of jumper settings.
>
>Does anybody know a solution?
>
>/David


Sometimes two drives just won't get along, but that is very
rare these days. More likely you simply need to try a
different jumper configuration.

Did you try jumpering both as Cable Select?

Why not leave the LiteOn as master? Cable routing is
probably easier like that too, since HDDs are usually lower
in the case than the optical drive. If your HDD is not in a
case position to do this, and you MUST have it as master for
some reason, can you simply relocate the drive?

Try other combinations of jumpers for either and both drives
again, and inspect the jumpers... while it is very rare,
sometimes jumpers are defective.
Anonymous
a b B Homebuilt system
September 7, 2004 3:06:48 AM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

As a general rule you should NOT combine a modern harddrive and an optical
drive on the same IDE channel. The harddrive is much faster than the
optical drive in data transfer normally, but combining them on the same IDE
channel makes the harddrive run at the same slow speed of the optical drive.
Harddrives should be together, and optical drives should be together.

--
DaveW



"David Rasmussen" <david.rasmussen@gmx.net> wrote in message
news:413c89c6$0$173$edfadb0f@dtext02.news.tele.dk...
>I am trying to get WD800JB and LiteOn LTR 48246S to work together as
>secondary master and secondary slave respectively. The problem is that the
>auto-detection of the drives during POST just hangs forever (or for a long
>time at least). If I switch the drives so that LiteOn is master and WD is
>slave, everything works. I think I have tried all possible combinations of
>jumper settings.
>
> Does anybody know a solution?
>
> /David
Related resources
Anonymous
a b B Homebuilt system
September 7, 2004 5:28:39 AM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

DaveW wrote:
> As a general rule you should NOT combine a modern harddrive and an optical
> drive on the same IDE channel. The harddrive is much faster than the
> optical drive in data transfer normally, but combining them on the same IDE
> channel makes the harddrive run at the same slow speed of the optical drive.
> Harddrives should be together, and optical drives should be together.
>

Yeah, but I have 3 hard drives and 1 optical drive. And 2 controllers
with one master and one slave each. So...

/David
Anonymous
a b B Homebuilt system
September 7, 2004 5:28:40 AM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Tue, 07 Sep 2004 01:28:39 +0200, David Rasmussen
<david.rasmussen@gmx.net> wrote:

>DaveW wrote:
>> As a general rule you should NOT combine a modern harddrive and an optical
>> drive on the same IDE channel. The harddrive is much faster than the
>> optical drive in data transfer normally, but combining them on the same IDE
>> channel makes the harddrive run at the same slow speed of the optical drive.
>> Harddrives should be together, and optical drives should be together.
>>
>
>Yeah, but I have 3 hard drives and 1 optical drive. And 2 controllers
>with one master and one slave each. So...
>
>/David

Well it's not true regardless, DaveW just keeps repeating
this nonsense with no evidence of it.

Benchmarks bear out that it isn't true, anyone can test this
for themselves... doesn't even need to be a thorough
benchmark, just the first moment you get a burst rate it can
be seen that the "same slow speed" is NOT being used.
September 7, 2004 12:48:25 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Have you considered adding another controller card? This would take the
optical drive off the same channel as the hard drive.
"David Rasmussen" <david.rasmussen@gmx.net> wrote in message
news:413c89c6$0$173$edfadb0f@dtext02.news.tele.dk...
>I am trying to get WD800JB and LiteOn LTR 48246S to work together as
>secondary master and secondary slave respectively. The problem is that the
>auto-detection of the drives during POST just hangs forever (or for a long
>time at least). If I switch the drives so that LiteOn is master and WD is
>slave, everything works. I think I have tried all possible combinations of
>jumper settings.
>
> Does anybody know a solution?
>
> /David
Anonymous
a b B Homebuilt system
September 24, 2004 3:22:35 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Strictly speaking, he is right. But since even modern drives are unlikely
to fill an ATA-100 bus and certainly won't fill an ATA-133 bus (if you know
different and have a soruce to show me, I'll recant) it's no great loss to
stick it on the same channel as an ATA-66 optical. If the optical is
ATA-100 or better (not seen any first-hand that are, but inevitably will
happen sooner or later) great!

If you want the comfort of knowing that your hardware can potentially fill
the bus at a full (S)ATA 150 rate then by all means keep them separate (I
do) but it WON'T come anywhere near doing so in reality.

John
Anonymous
a b B Homebuilt system
September 24, 2004 5:53:02 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Fri, 24 Sep 2004 11:22:35 +0000 (UTC), "Gaidheal"
<some.one@some.isp.net> wrote:

>Strictly speaking, he is right. But since even modern drives are unlikely
>to fill an ATA-100 bus and certainly won't fill an ATA-133 bus (if you know
>different and have a soruce to show me, I'll recant) it's no great loss to
>stick it on the same channel as an ATA-66 optical. If the optical is
>ATA-100 or better (not seen any first-hand that are, but inevitably will
>happen sooner or later) great!
>
>If you want the comfort of knowing that your hardware can potentially fill
>the bus at a full (S)ATA 150 rate then by all means keep them separate (I
>do) but it WON'T come anywhere near doing so in reality.
>


No, strictly speaking or speaking any way you like, it's not
right, and hasn't been for a decade.
Anonymous
a b B Homebuilt system
September 24, 2004 9:20:47 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Depends greatly on what you think we are arguing over.....

The entire channel WILL drop to the lowest spec supported by both devices.
That simple. As for "can't mix optical with non-optical" - that obviously
is not the case. Wasn't for very long, either, though it tended to be
sub-optimal.

John
Anonymous
a b B Homebuilt system
September 24, 2004 9:29:40 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Fri, 24 Sep 2004 17:20:47 +0000 (UTC), "Gaidheal"
<some.one@some.isp.net> wrote:

>Depends greatly on what you think we are arguing over.....
>
>The entire channel WILL drop to the lowest spec supported by both devices.
>That simple.

No, it won't.
It switches to lower speed ONLY when accessing the slower
interfaced device. This has been proven time and time
again, stop wasting our time and test it for yourself.

The interface of faster device will operate in fastest mode
supported by the device or the controller, whichever is
lower.

In other words, given an ATA100 channel with an ATA33 CDROM,
an ATA133 HDD, the CDROM will operate in ATA33 mode but the
HDD in ATA100 mode. This is VERY easy to test, prove.



> As for "can't mix optical with non-optical" - that obviously
>is not the case. Wasn't for very long, either, though it tended to be
>sub-optimal.

The overhead of having ATAPI device on a channel is very,
very slight... not noticable in use.
Anonymous
a b B Homebuilt system
September 24, 2004 10:28:58 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Well, last time I did test it, which is years ago, that was definitely not
the case. If it has improved (and I don't doubt you, but might have to see
it for myself anyway!) then that's good. But I do certainly remember
demonstrating to a customer why I had installed their drive as a master and
moved an optical device to make it slave to a another optical device on the
secondary channel. There was CONSIDERABLE improvement in performance
overall, not only data throughput.

I do accept however, that the last time I did any such thing was about 4
(IMS) years ago.

John

P.S. I never said the overhead was significant for having an ATAPI device
on the channel... what made you raise that strawman?
Anonymous
a b B Homebuilt system
September 24, 2004 11:07:00 PM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Fri, 24 Sep 2004 18:28:58 +0000 (UTC), "Gaidheal"
<some.one@some.isp.net> wrote:

>Well, last time I did test it, which is years ago, that was definitely not
>the case. If it has improved (and I don't doubt you, but might have to see
>it for myself anyway!) then that's good. But I do certainly remember
>demonstrating to a customer why I had installed their drive as a master and
>moved an optical device to make it slave to a another optical device on the
>secondary channel. There was CONSIDERABLE improvement in performance
>overall, not only data throughput.
>
>I do accept however, that the last time I did any such thing was about 4
>(IMS) years ago.
>
>John
>
>P.S. I never said the overhead was significant for having an ATAPI device
>on the channel... what made you raise that strawman?
>

All contemporary chipsets of 4 years ago also supported
independant device timing, a board that didn't would have to
have either A) buggy bios B) been much older at the time
C) 80 conductor cable wasn't used D) CDROM was in PIO
mode.

I mentioned the ATAPI issue because it is related, nothing
more or less.
Anonymous
a b B Homebuilt system
September 26, 2004 4:40:50 AM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Now that you mention it; would have been the cables - most of the builds I
saw used the older cables until about 2 years ago, actually, maybe 3.

As for the ATAPI question.. I really don't think it was relevant, actually,
but whatever ;¬)

John

P.S. My 'arguing' with you should not be taken as a sign I don't value your
knowledge / input - just in case I gave that impression. I simply state my
opinion and what I base it on... if I am in error I'll take the flak.
Anonymous
a b B Homebuilt system
September 26, 2004 5:41:34 AM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Sun, 26 Sep 2004 00:40:50 +0000 (UTC), "Gaidheal"
<some.one@some.isp.net> wrote:

>Now that you mention it; would have been the cables - most of the builds I
>saw used the older cables until about 2 years ago, actually, maybe 3.
>
>As for the ATAPI question.. I really don't think it was relevant, actually,
>but whatever ;¬)
>
>John
>
>P.S. My 'arguing' with you should not be taken as a sign I don't value your
>knowledge / input - just in case I gave that impression. I simply state my
>opinion and what I base it on... if I am in error I'll take the flak.
>

The ATAPI issue was of relevance after introduction of the
"Optical" drives to the discussion, which are ATAPI.
Anonymous
a b B Homebuilt system
September 27, 2004 1:26:11 AM

Archived from groups: alt.comp.hardware.homebuilt (More info?)

The optical drives were mentioned right at the start by the original poster.
Raising the question of overhead in a reply to my posts was an utter
irrelevance - as I said.

John
!