[Nich]

Hi.


I just wondered. Since there is a certain build-quality required in all

other cables, Digital and Analog alike, maybe the same holds true

for these cheap IDE cables.


Since I make my own cables out of 'very good' parts, I would hate

if my IDE's corrupted the data in the beginning of the chain.


Maybe some are better than others?


Thanks,


Nicholas

 

[Rop]

Well, personally I'm partial to silver cables. The bytes are just a little smoother. Of course, I put some solid brass weights on them just in case (those nasty resonances of drives, ruins the bit-transparency!).


Actually, if there's something wrong with an IDE cable you're going to find out in a hurry. I've used tons of IDE cables going back to the beginning of IDE itself and have never had a problem with any of them, cheap or expensive (including the ones that come 'free' with MBs and controllers). Not sure if there's error correction/detection at the level of the IDE interface (There is at the drive level), but with executables there's no room for error, a single faulty bit has more than a good chance to make the software crash.


-Rob-

 

[Nich]

"Well, personally I'm partial to silver cables. The bytes are just a little smoother. Of course, I put some solid brass weights on them just in case (those nasty resonances of drives, ruins the bit-transparency!). "







I was just thinking in terms of video-playback, especially HD materiel,

where it can't go back and get the same data once more since it's

needed on the screen. What do I know.


Nicholas

 

[Troubleshooter]

Hey if the cables cost more than the GDP of a small African country they GOTTA be good. That's how it works in the AV world right?!

Sorry Nich, honestly not making fun of you at all, just 'the industry ' As Rop said, nah an IDE cable is pretty much an IDE cable. You can get some nice round cables if you like to make the case look nicer inside and help airflow but its mostly just bling-bling.

-Trouble

 

[Nich]

Hmm...OK


Thanks. Thought so.


Nicholas

 

[Rop]

Talking about round cables: I got some nice ones from Indy PC . Cheap, fast service, no complaints (I have no connection with them, just a happy customer). Got the ATA133 ones for my HD and DVD, plus a round floppy cable, silver color (told you I was partial to silver ).


Main reason was to get a little better airflow since the ribbon IDE cable that came with my P4PE board pretty much blocked the inflow fans because that's where the HD is mounted. My HTPC case has a side window, so it looks a lot better now too.


System has been running flawlessly with those cables for about 3 months now, 24 hours a day, 7 days a week.


-Rob-

 

[Nich]

24/7 for three months??? Damn.


Nicholas

 

[RichCarling]

Some cables have the pull tabs at each connector making it a little easier to pull them off a drive, I also like ones that have the tab in the middle of the connector so that it is harder to cram them on upside down. Anything to minimize bending and breaking pins is a plus (I once trashed an IDE drive by forcing an IDE cable on, if your stubbon and tired enough, you can break almost anything).


The round cables are nice for the air flow as mentioned, and are neater in the box. Also, if you use a PCI IDE interface card for supporting large drives, the newer cables come in 24" and 36" lengths, I have found the 24" cables to work well with the PCI IDE interface cards and the PCI IDE interface allows you to have more IDE drives. I tie up two IDE connections with my DVD burner and separate DVD-ROM drive alone.

 

[daredevil23]

The difference is that older IDE cables will not work well with UDMA 66 or faster drives at the faster speeds.

 

[Rop]

Nich, the 3 months were not without reboots. The hardware is rock solid, the software (ie. XP Pro) is not.


I agree with RichCarling that one of the nice things of buying round cables is that you can decide on the length (guess you can do the same for ribbon cables, but most I've seen are just plain vanilla standard length with two connectors crimped on the end). I took a tape measure and figured out that the HD could do wth a 12" IDE cable, the DVD needed 18", the floppy just 12". Got all of them with a single connector at the end. Voila! No extraneous cables all over the place, no dangling connectors, just neat connections. A lot more tidy than the ribbon cables they're replacing.


Personally I don't like the pull tabs at the ends much, but that's mostly a matter of taste (looks tidier IMO without them). If you plan on connecting/disconnecting often they certainly help!


-Rob-

 

[jkaiser]

To answer the question, yes there is a difference.


OLD cables are 40 wires

New cables are 80 wires (a grounded wire between each of the 40 active wires to cut down on cross talk.


There are flat cables and round cables. Be very careful about the round cables and the manufacturer, there is a difference. Try to stay away from the 24" ones as they seldom work.

 

[coldie]

Cables make a big difference if you are running HW tests.


Given a new hard drive:


40 pin are the worst

80 pin generic

80 pin round

80 pin high quality are the best


If you're actually looking at the IDE bus, you'll see fewer errors with the better cable you use.


Does this matter? Not in the least. IDE is meant to be bit perfect, so if there's a transmission error, the packet is simply resent. The only effect is on overall throughput, and HW's are really damn fast anymore, so it doesn't matter.


Coldie

 

[Troubleshooter]

Can you even buy 40 pin ide cables anymore? The 80 pin UDMA cables work on non UDMA devices. I figured they didn't even bother to make the old ones anymore.

-Trouble

 

[Sailn]

The 80 wire cable is really just a 40 wire cable with a ground between each data line. As was stated before, the round cables do/may have some problems as they do deviate from the ide spec. 36 inches seems to be about all you can get an still stay within spec.


I have a friend who make them for me...that is to saw he will be making me some when I figure out what I really want.


While he does have the proper tool for putting the ends on, he claims the old wood working clamp (2 pc of wood with screw at each end to tighten it) woeks just as well.

 

[Rop]

Quote:

Originally posted by Sailn While he does have the proper tool for putting the ends on, he claims the old wood working clamp (2 pc of wood with screw at each end to tighten it) woeks just as well.

Yep, that works well. Made a bunch of cables using a vise mounted on a workbench. Just have to be careful not to squash the connector.


-Rob-

 

[BigguyZ]

The 40 pin cables are mostly used for optical drives, that don't require ATA100 or ATA133 standards. With those it's easy to "customize" the cables length so there's no slack between the connectors. Cleans it up a lot. The ATA 100 and 133 cables are the 80 pin cables you see.


I have read reviews on sheilded grounded cables where the difference in the HDD performance was significant. The cable was the Coolermaster ultra series I think....

 

[Biggs]

Quote:

Originally posted by Troubleshooter Hey if the cables cost more than the GDP of a small African country they GOTTA be good.

IOSS RD3XP Gladiator Super Shielded Round Cable


Hmmmm... interesting.

 

[Nich]

Thanks for all your replies.


How about when a IDE is connected to both. say, harddisk and DVD

drive? Is it better to only have it connected to one unit?


And final, is there anything to gain in cutting down on the length?

Lets say, from 30 to 20 cm?(none, I take it)


That should do it, thanks.


Nicholas

 

[Mastiff]

Oh, man, don't do that! You'll loose a bucket of performance from your harddisk since it will be forced to work at the same speed as the DVD drive, if I have understood this correctly. I always keep my harddisks on IDE1/RAID while my DVD/CD/RW drives are at IDE2. And you should have the DVD drive as master on IDE2 (I don't know why, but it is recommended by the best of sources).

 

[Ricardo Cruz]

As Biggs suggested, those shielded cables do keep the signal "cleaner". In the end, there are fewer errors transmitted so you may get an increased throughput (less packages re-sent).