Can IDE cable's end connector be left idle ( w-o affecting data transfer ) ?

[TE Cheah]

Thanks for your advice.
Does this restriction apply to floppy drive cables ?
This web page I quoted from used to say that hdd & cd drive must not
share the same channel, is this restriction correct too ?

| If your question derives from a desire to keep down the amount
| of ribbon cable in the PC's interior
Yes, I rolled the idle end toward the middle connector.

| try "round" cables.
No, they're costly [ii] all their 40 active wires are tied together,
can their signals not cross talk to 1 another ?

 

[chrisv]

Which is zip.

LOL

 

[Arno Wagner]

Thanks for your advice.
Does this restriction apply to floppy drive cables ?
This web page I quoted from used to say that hdd & cd drive must not
share the same channel, is this restriction correct too ?

After I was wrong on the termination issue, I should probably not
answer this, but what the hell: There may be problems when you have
a CDROM and a HDD on the same channel. Now I also understand why: If
the CDROM is nit UDMA capable, it will not have the termination
needed for UDMA mode. However it may work fine even in this case.

| If your question derives from a desire to keep down the amount
| of ribbon cable in the PC's interior
Yes, I rolled the idle end toward the middle connector.

| try "round" cables.
No, they're costly [ii] all their 40 active wires are tied together,
can their signals not cross talk to 1 another ?

They are twisted-pair (except some very cheap ones). The
twisting of a signal and a ground line prevents crosstalk.
It is sort of a "magnetic shielding", just as the ground wires
between the signal wires in 80 pin cables are sort of a
"capacitive shielding". This is only an approximate explanation
though. Note that the rounded cables are out of spec, but often
work fine. My personal experience is that I had no problems
with HDDs for rounded cables up to 60 cm, but serious problems
with two 90 cm cables I tried. I have a CD writer (which
incidentially seems not to be UDMA capable), that has problems
with rounded cables of any length.

Arno

 

[Beemer Biker]

SNIP

. I have a CD writer (which
incidentially seems not to be UDMA capable), that has problems
with rounded cables of any length.

Arno

I had the same problem on an old i815 chipset genuine intel motherboard. We
still have that computer and it is running the latest XP SP2 with a 1.3mhz
P3. I just fixed it only last week when I went to intel's site and
downloaded some "intel architecture driver . Before doing that I tried
setting UDMA in bios and in system, but every time it rebooted it picked PIO
mode. Both the burner and the hard drive were on PIO mode and it was
terribly slow. My wife uses that computer, it never runs games, and I only
got a complaint when she had trouble burning some music. Even though I had
those round cables and there was only one device on each cable, it was stuck
in PIO mode with the lastest microsoft driver. I had to find that intel
driver before it handled UDMA. When I originally set it up for her I
checked for yellow bangs in system, everything looked o.k. and I didnt
bother checking for pio mode.

That is not too bad, we still got an IBM PS2 model 80 with 486 planar and it
only runs its single ide port in compatibility mode. You dont want to run
in compatibility mode.


--
=======================================================================
Beemer Biker (e-mail address removed)
http://TipsForTheComputingImpaired.com
http://ResearchRiders.org Ask about my 99'R1100RT
=======================================================================

 

[Folkert Rienstra]

I looked.

There is no bus termination,

Yes, Babble, that's what he said in the paragraph above.

but apparently a device termination was added.
How this plays out if you have a UDMA and
a non-UDMA capable device on the same bus,

I have no clue.

Period.

The clue, Babble, is in the spec where it says that the device with the
termination goes to the end of the cable.

 

[Folkert Rienstra]

Your comments are confused.

And your's are any better?

There's no such thing, in SCSI or IDE, as a terminated or unterminated device.

Well, there is in the physical sense, as a device that provides the termination.

There's a device, a terminator, or no device.

The terminator is a cheap way of pretending some sort of device is connected,

Uh no. It is pretending that the cable is endless.

and it's important for SCSI mainly because of the cable lengths involved.

In part.
Without any termination a SCSI bus doesn't function at all, nomatter how short.

It doesn't really matter for IDE because the impact of an unterminated
connector on the signal quality (ringing, overshoot) is much less with the
shorter lengths involved,

In part.

and the IDE interface can cope with it.

No, it *can't*. It *could* when interface speeds were still (s)low and
the *interface* wasn't yet acting like a *bus*, ie a *transmission line*.
Changes were necessary to keep it working with the higher UDMA modes.
One of those came to be known as series termination.

 

[Folkert Rienstra]

SCSI is complicated today.

No it isn't, just complicated for you.

U2W cables are actively terminated.

No, they aren't. Ultra SCSI and LVD buses are actively terminated.

Older standards allow terminated devices (i.e. devices with internal
termion) as well as external terminators.

You got that backwards too.

As it turns out UDMA capable devices has some sort of half-assed
device terminaton,

No, they have series termination.

i.e. _every_ device on the bus has some pull-up or pull-down resistirs.

Quite inelegant IMO.

Quite elegant actually and no one cares OYO. Least T13.

 

[Folkert Rienstra]

After I was wrong on the termination issue,

(And what not).

So much for your:
"If anybody says different, then they do not understand the subject matter"

I should probably not answer this, but

Afterall, you are babblemouth, you have a reputation to hold high.

And if some one is so stupid to ask the wrong person ...

what the hell:
Exactly.

There may be problems when you have
a CDROM and a HDD on the same channel.

Now I also understand why:

We'll see. Not.

If the CDROM is nit UDMA capable, it will not have the termination
needed for UDMA mode.

However it may work fine even in this case.

It does if the non UDMA capable device is in the middle.

Yes, I rolled the idle end toward the middle connector.

No, they're costly [ii] all their 40 active wires are tied together,
can their signals not cross talk to 1 another ?

They are twisted-pair (except some very cheap ones). The
twisting of a signal and a ground line prevents crosstalk.

It is sort of a "magnetic shielding", just as the ground wires
between the signal wires in 80 pin cables

are sort of a "capacitive shielding".
Nonsense.

This is only an approximate explanation though.

As in "bullshit".

Note that the rounded cables are out of spec,

As in 'not in the spec'.

but often work fine.

My personal experience is that I had no problems with HDDs for round-
ed cables up to 60 cm, but serious problems with two 90 cm cables I tried.

I have a CD writer (which incidentially seems not to be UDMA
capable), that has problems with rounded cables of any length.

Bable, babble, rant.

 

[Timothy Daniels]

Does this restriction apply to floppy drive cables ?

Floppy drive cables only have two connectors and
they operate at lower frequencies, so they are less
sensitive to signal reflections. Since round floppy drive
cablescome in various lengths, I use a round floppy drive
cable, too, to keep the interior cabling manageable.

This web page I quoted from used to say that hdd &
cd drive must not share the same channel, is this
restriction correct too ?

Not always. But it's a good restriction for newbies.
As I interpret Folkert's current tirade, you can pair a
highspeed device (one with UDMA) with a lower speed
device (e.g. a CD drive) if the highspeed device is on
the end connector since the highspeed device has an
interface that is designed to supress reflections better.

Yes, I rolled the idle end toward the middle connector.

That's still a no-no since the empty end connector
still acts to reflect signals just as if the unused length
weren't rolled up.

| try "round" cables.
No, they're costly

$3 plus shipping is too costly? Take a look
at the SVC.com website again:
http://svc.com/cables-ata-100-133-round-cables.html

[ii] all their 40 active wires are tied together,
can their signals not cross talk to 1 another ?

Most of the round cables nowadays have 80 wires -
40 signal wires and 40 ground wires (colored white).
Each of the 40 signal wires is twisted together with a
ground wire to reduce crosstalk and susceptibility to
EMI. The kind with metal braid (aluminum or copper)
further suppress EMI. Although this is not according
to the very exacting ATA specifications which describe
ribbon cable, most people (including myself) find that
they work just fine, even with ATA/133 HDs. And it
reduces air congestion inside the case and so adds
to the longevity of interior components - such as HDs.

*TimDaniels*

 

[Timothy Daniels]

As I interpret Folkert's current tirade, you can pair a
highspeed device (one with UDMA) with a lower speed
device (e.g. a CD drive) if the highspeed device is on
the end connector since the highspeed device has an
interface that is designed to supress reflections better.

It's probably more accurate to say that interface of the
lowspeed device on the middle connector doesn't produce
the degree of signal reflections that it would if it were on the
end connector, so when using a UDMA device (e.g. high-
speed HD) with a lowspeed device (e.g. CD drive) put the
lowspeed device where it does no harm - on the middle
connector.

*TimDaniels*

 

[Rod Speed]

TE Cheah wrote

Floppy drive cables only have two connectors

They often only have one now.

and they operate at lower frequencies, so they are less sensitive to
signal reflections.

And one floppy drive goes on the end connector
anyway because floppy drives specify the letter
the drive gets by the connector its plugged into.

Since round floppy drive cablescome in various lengths, I use a round
floppy drive cable, too, to keep the interior cabling manageable.

More fool you.

Not always. But it's a good restriction for newbies.
As I interpret Folkert's current tirade, you can pair a
highspeed device (one with UDMA) with a lower speed
device (e.g. a CD drive) if the highspeed device is on
the end connector since the highspeed device has an
interface that is designed to supress reflections better.

That's still a no-no since the empty end connector still acts to reflect
signals just as if the unused length weren't rolled up.

try "round" cables.

No, they're costly

$3 plus shipping is too costly? Take a look at the SVC.com website
again:
http://svc.com/cables-ata-100-133-round-cables.html

[ii] all their 40 active wires are tied together,
can their signals not cross talk to 1 another ?

Most of the round cables nowadays have 80 wires -
40 signal wires and 40 ground wires (colored white).
Each of the 40 signal wires is twisted together with a
ground wire to reduce crosstalk and susceptibility to
EMI. The kind with metal braid (aluminum or copper)
further suppress EMI. Although this is not according
to the very exacting ATA specifications which describe
ribbon cable, most people (including myself) find that
they work just fine, even with ATA/133 HDs. And it
reduces air congestion inside the case and so adds
to the longevity of interior components - such as HDs.

And the flout the standard so only fools bother with them.

 

[Rod Speed]

Timothy Daniels wrote

It's probably more accurate to say that interface of the
lowspeed device on the middle connector doesn't produce
the degree of signal reflections that it would if it were on the
end connector, so when using a UDMA device (e.g. high-
speed HD) with a lowspeed device (e.g. CD drive) put the
lowspeed device where it does no harm - on the middle
connector.

Utterly mangled all over again.

 

[Peter]

Floppy drive cables only have two connectors

They often only have one now.

I normally discard cables with only one connector. ;-)

 

[Rod Speed]

I normally discard cables with only one connector. ;-)

Down, boy |-)

 

[Folkert Rienstra]

It's probably more accurate
Nope.

to say that interface of the lowspeed device on the middle connector
doesn't

produce

Signal reflections aren't produced, they occur or don't occur.

the degree of signal reflections

Actually it does, on two ends even, if it 'speaks'.
But since it is a 'slow' device the effect doesn't occur.

that it would if it were on the end connector,

Since it is at the end it doesn't actually cause signal reflections at
that end. The other end (hostbus controller) has series termination.

 

[TE Cheah]

| check out Maxtor's installation instructions

I bought a Maxtor hdd in 4-05, a WD hdd in 12-03, a Seagate hdd in
12-01, a Sony dvd drive in 9-02, no such warning came with any of
these.

 

[Jeff Richards]

But that's the point. While the device might incorporate a terminator, that
doesn't make it a terminated device. If the terminator is enabled then it
becomes a device plus a terminator - exactly the same as a device with a
pass-through connection and a terminator plugged into the end of the chain.
It's a fine distinction, but unless you make it clear then you will make the
mistakes apparent in your comments, confusing what a device does in terms of
line impedance etc with what a terminator does. They are different things
and they have different effects.

 

[Arno Wagner]

But that's the point. While the device might incorporate a terminator, that
doesn't make it a terminated device. If the terminator is enabled then it
becomes a device plus a terminator - exactly the same as a device with a
pass-through connection and a terminator plugged into the end of the chain.

Unfortunattely that is not true.

It's a fine distinction, but unless you make it clear then you will make the
mistakes apparent in your comments, confusing what a device does in terms of
line impedance etc with what a terminator does. They are different things
and they have different effects.

A device-integrated terminator is at the end of the "T". A
cable attached or integrated terminator is at the end of the bus.
The two things coincide if the cable end-connector is attached
to the terminated device. Otherwise they do not.

And, yes, a device-integrated terminator turns a device into a
terminated device. A cable attached or cable integrated terminator
creates a terminated cable. It is a difference and it may be important
for fast buses. It is critical if there are terminated devices in the
middle of the bus, like ATA UDMA termination does.

Arno

 

[Folkert Rienstra]

But that's the point.

What point, topposter.
Not only do you not have a point you allowed babblemouth another opportunity to add to the confusion once again.

While the device might incorporate a terminator, that doesn't make
it a terminated device. If the terminator is enabled then it becomes
a device plus a terminator - exactly the same as a device with a pass-
through connection and a terminator plugged into the end of the chain.

It's a fine distinction, but unless you make it clear then you will make
the mistakes apparent in your comments, confusing what a device does
in terms of line impedance etc with what a terminator does.

They are different things and they have different effects.

And since you didn't explain what the difference is no point was made.

 

[TE Chea]

| Floppy drive cables only have two connectors
I have old ( made in '90 - '93 ) floppy cables with 5 connectors.

| you can pair a
| highspeed device (one with UDMA) with a lower speed
| device (e.g. a CD drive) if the highspeed device is on
| the end connector
In this setup ( using VIA kt266a's south bridge & 80 wire cable,
all drives using cable select ), my Seagate ata33 hdd ( @ the end
connector ) seems to work fine ( can pass Disk Checker 2000 )
with my LG ata33 dvd*drive ( @ the mid-connector ), but * can
not pass BCM diagnostic ( in Win98se, using win98se cd in * )
unless * is @ end connector & hdd is absent ( i.e. no sharing of
this 2ndary channel with a hdd ), but a Toshiba PIO mode 3 cd
drive @ mid-connector does not affect * passing BCM
diagnostic.
My experiment today indicates that this web page's warning is
real.