Ultra DMA (80-Conductor) IDE/ATA Cables
There are a lot of issues and problems associated with the original 40-conductor IDE cable, due to its very old and not very robust design. Unterminated flat ribbon cables have never been all that great in terms of signal quality and dealing with reflections from the end of the cable. The warts of the old design were tolerable while signaling speeds on the IDE/ATA interface were relatively low, but as the speed of the interface continued to increase, the limitations of the cable were finally too great to be ignored.
In the ATA/ATAPI-4 standard that introduced the Ultra DMA transfer mode set, a new cable was introduced to replace the old standby: the 80-conductor IDE/ATA cable. The name is important: the new cable has 80 conductors (wires)--it does not have 80 pins on each connector, though, just 40. This means that the new cable is pin-compatible with the old drive. No change has been made to the IDE/ATA connectors, aside from the color-coding issue (see below).
The obvious question, of course, is this: what's the point of adding 40 extra wires to a cable if they aren't connected to anything? :^) Well for starters, the 40 wires are connected to something, just not their own pins on the interface connectors. The extra 40 wires don't carry new information, they are just used to separate the "real" 40 signal wires, to reduce interference and other signaling problems associated with higher-speed transfers. So the 40 extra conductors are connected to ground, interspersed between the original 40 conductors of the old cable. Any stray signals that would "cross-talk" between adjacent wires on the 40-conductor cable are "absorbed" by these extra ground wires, improving signal integrity. The extra ground wires can be either all of the even-numbered wires, or all of the odd-numbered wires in the cable.
There are a number of other attributes and characteristics of the 80-conductor cable, which I'm going to list in bullet form for easier absorption:
There are a couple of reasons why this coding was done. The main one is the issue mentioned in the discussion of the 40-conductor cable: it is not a good idea to connect a single drive to the middle connector on a ribbon cable, because the "stub" of left-over, unconnected cable causes signaling problems. With Ultra DMA this "stub" connection is not just "not recommended", it is illegal: a single device must be at the end of the cable. The other reason is that since these cables support cable select inherently, the position of each drive on the cable matters if cable select is being used. With these two needs combined, it just made sense to design the cable so that drive positioning was explicitly clear.
A standard 80-conductor Ultra DMA IDE/ATA interface
A comparison of the wires used in 80-conductor and
Aside from the above, the cable can be treated the same way as a 40-conductor cable. Since it is of higher quality, it can be used in place of a 40-conductor cable in older systems without any issues. However, it does not directly replace a 40-conductor cable select cable. Note also that the 18" length restriction associated with the original 40-conductor cable applies to this variation as well.