cable length specs?
Ron Reaugh
Plethora86 wrote in message
<19990727011745...@ng-ch1.aol.com>...
>a few questions:
>
>how long are IDE cables officially spec'd at?
>
>how long can they be and still work consistently?
>
>are ATA-66 cables different as far as length requirements go?
>
>thanks!
Ron Reaugh
Plethora86 wrote in message
<19990727011745...@ng-ch1.aol.com>...
>a few questions:
>
>how long are IDE cables officially spec'd at?
18 inches.
>how long can they be and still work consistently?
<=18 inches.
>are ATA-66 cables different as far as length requirements go?
Up to 24 inches.
Plethora86
how long are IDE cables officially spec'd at?
how long can they be and still work consistently?
are ATA-66 cables different as far as length requirements go?
thanks!
Plethora86
>
>
ok, so it's always a gamble with cables longer than this? this is what i would
think, but then i see people selling 36" cables and i'm forced to wonder how
those cables would work at all...
Ron Reaugh
Plethora86
>would
>>think, but then i see people selling 36" cables and i'm forced to wonder
>how
>>those cables would work at all...
>
>
>In UDMA33 mode they don't.
sorry to bother you again, but is this 18" limit only a DMA limit, or would PIO
devices (zip drives, etc) not work either?
Ron Reaugh
Plethora86 wrote in message
<19990727121850...@ng-bj1.aol.com>...
On a HX mobo where everything is limited to DMA2/PIO4 then one might be able
to run a little longer.
Dave M.
UDMA even being proposed, let alone actually implemented. It has nothing to
do with DMA in any manner. 18" length was re-emphasized in the UDMA33
specification, and is much more important with this specification due to
data speeds.
The specification for the 18" total length limit has been time and again
been abused by the ribbon cable makers. As long as they got 18" or less
from the ide connector to the first connector on a dual drive cable, they
assumed they were within specifications. In the case of the 36" length, 18
+ 18 =36. So, the standard was abused or misunderstood or maligned
depending on your view of things. The standard was set to prevent data
transportation delivery problems under arduous conditions.
Will a long ide cable work for pio mode devices, depends on the case, whats
in the box, orientation of devices and cables, how much data is sent over a
given time and how much data.
Large cases like towers tend to disperse external influence on ide ribbon
cables due to their size, but at the same time may require a longer ide
cable depending on where the ide device is in the case, which increases
chances of external influence on the ribbon cable. Catch22. An over 18"
cable may work on your PIO only device in this case.
Small cases like mini/micro towers tend to inhibit dispersal of external
influence on ide ribbon cables, but at the same time may require a longer
ide ribbon cable depending on where the ide device is in the case, which
increases in some cases external influence on the ribbon cable. This is
where you should be more heavily influenced to get an ATA66 cable. An ATA66
cable was designed to ground data noise before it hits the data lines on the
ribbon cable. Usually, an 18" or less total length ide ribbon cable should
be used instead.
To compound the problem, the higher bus/cpu frequencies used today are more
likely to create havoc than the slower bus/cpu speeds. Due to the decreased
wavelength of the RF noise, the ribbon cable is more likely to be a multiple
of the length of it. Any wavelength multiple will cause current flow.
My use of external influence means anything inside the case that creates a
frequency noise that will "walk" up and down the data lines. Proximity to
this noise is more of problem in small cases than large expansive types.
The specification was set to prevent data read/write problems on ide
devices. Intermittent failures are seldom detected. A missed bit here or
there can be meaningless or render operating system failures. Such
incidents can be misdiagnosed as hard drive problems, memory problems, or
cpu overheating.
Dave
Plethora86 <pleth...@aol.com> wrote in message
news:19990727121850...@ng-bj1.aol.com...
Ron Reaugh
Dave M. wrote in message <7nl2ph$r6t$1...@nntp3.atl.mindspring.net>...
>18" has been the TOTAL length limit for ide ribbon cables, even prior to
>UDMA even being proposed, let alone actually implemented. It has nothing
to
>do with DMA in any manner. 18" length was re-emphasized in the UDMA33
>specification, and is much more important with this specification due to
>data speeds.
>
>The specification for the 18" total length limit has been time and again
>been abused by the ribbon cable makers. As long as they got 18" or less
>from the ide connector to the first connector on a dual drive cable, they
>assumed they were within specifications. In the case of the 36" length, 18
>+ 18 =36. So, the standard was abused or misunderstood or maligned
>depending on your view of things. The standard was set to prevent data
>transportation delivery problems under arduous conditions.
>
>Will a long ide cable work for pio mode devices, depends on the case,
NO, it has nothing to do with the case.
>whats
>in the box, orientation of devices and cables, how much data is sent over a
>given time and how much data.
Well, not really.
>Large cases like towers tend to disperse external influence on ide ribbon
>cables due to their size,
Nope.
> but at the same time may require a longer ide
>cable depending on where the ide device is in the case, which increases
>chances of external influence on the ribbon cable. Catch22. An over 18"
>cable may work on your PIO only device in this case.
The case is not relevant.
>Small cases like mini/micro towers tend to inhibit dispersal of external
>influence on ide ribbon cables,
Huh, does this have to do with abduction and lost time?
> but at the same time may require a longer
>ide ribbon cable depending on where the ide device is in the case, which
>increases in some cases external influence on the ribbon cable.
Cable routing can in some cases have an effect but only at distances of less
than a cm. for any reasonable situation.
> This is
>where you should be more heavily influenced to get an ATA66 cable. An
ATA66
>cable was designed to ground data noise before it hits the data lines on
the
>ribbon cable.
Not primarily.
> Usually, an 18" or less total length ide ribbon cable should
>be used instead.
Never use a standard EIDE cable where an ultracable is needed.
>To compound the problem, the higher bus/cpu frequencies used today are more
>likely to create havoc than the slower bus/cpu speeds. Due to the
decreased
>wavelength of the RF noise, the ribbon cable is more likely to be a
multiple
>of the length of it. Any wavelength multiple will cause current flow.
Not of first order relevance.
>My use of external influence means anything inside the case that creates a
>frequency noise that will "walk" up and down the data lines. Proximity to
>this noise is more of problem in small cases than large expansive types.
No.
>The specification was set to prevent data read/write problems on ide
>devices. Intermittent failures are seldom detected.
Huh?
> A missed bit here or
>there can be meaningless or render operating system failures. Such
>incidents can be misdiagnosed as hard drive problems, memory problems, or
>cpu overheating.
Or simply EIDE cable CRC retries.
Eric Lim
Ultra 66 cable, or can I just use the one that comes with the motherboard?
Ron Reaugh <Ron-R...@worldnet.att.net> wrote in message
news:7nl8dl$ap8$1...@bgtnsc02.worldnet.att.net...
Ron Reaugh
Eric Lim wrote in message <7nqgua$310$1...@fir.prod.itd.earthlink.net>...
>I'm getting a Western Digital Expert and an AOpen AX6BC Pro. Do I need a
>Ultra 66 cable, or can I just use the one that comes with the motherboard?
Just use the one with the mobo.