Modified Rounded IDE Cables -Phase II (pics)

[jmroberts70]

My old thread is waaay long gone and it's finally time to start on my final wiring mod job involving parallel IDE cables. This all comes fairly late in the PC world as these cables are quickly being replaced with SATA drives and soon I'm sure I'll be running systems with absolutely no more PATA components. But in the mean time, I have quite an investment in PATA drives and plan on quite a bit more mods to this system before I'm done. So here goes...

To begin with, here's an example of some of the work I'm capable of and striving for:
http://www.unofficial3d.com/atx/cablingprototype02.html

I've still got to purchase a new power supply so I can gut it and start my power lines fresh from the circuit board to ensure the perfect cable lengths. I'm still debating weather to put a connector on the outside of the box or to just run the wires straight out of the PSU to the components to be more clean. I'm also yet to order more heat shrink and sleeving. So here are the pics...

Here's the "before" shot:

start with some basic rounded cables like so...

--and do this!!!

Here's some detail shots:

motherboard connection

Optical and removable drives

Additional IDE controller and drives

close-up of the hard drives

It may be hard to see but here's my temporary connection for the floppy drive. I've got to order some smaller sleeving for this little baby but in the mean time, it stays tucked away

here's the floppy cable connection on the mobo

another view of the lower end of the case.

I've still got a LOT of work to do on this baby. Like I said, I'm going to be ordering some additional materials and another power supply to butcher. I'm going to be working in all green this time (many people considered my original red materials to actually be pink and I can't be having any of that). My final case will resemble a lot of the original ideas I started in my prototype project page. I'll be sure to keep you all posted on the progress through the months!

 

[thewhiteguy]

Psssssst ... we've got SATA now

 

[ashmedai]

Hehehhehe

 

[jmroberts70]

Psssssst ... we've got SATA now

just in case you missed it earlier...in the post above yours...

This all comes fairly late in the PC world as these cables are quickly being replaced with SATA drives and soon I'm sure I'll be running systems with absolutely no more PATA components. But in the mean time, I have quite an investment in PATA drives

 

[Othersider]

just in case you missed it earlier...in the post above yours...

Why not take your wiring skills and make a really slick SATA with integrated power. It's already designed for it, with the connectors being in contact with each other, but no one's taken advantage of it with a true SATA data/power cable.

 

[zapperfish]

they look so nice, although i still like flat IDE cables

 

[jmroberts70]

Why not take your wiring skills and make a really slick SATA with integrated power. It's already designed for it, with the connectors being in contact with each other, but no one's taken advantage of it with a true SATA data/power cable.

That's true. Great idea. When I start using SATA drives, I'll give that a shot. At least it's not like there's nothing to mod in a SATA system.

 

[CrimandEvil]

I like, I've been meaning to cut the end cover things off of mine for a while now.

 

[Othersider]

That's true. Great idea. When I start using SATA drives, I'll give that a shot. At least it's not like there's nothing to mod in a SATA system.

Something like this needs a much neater implementation.

 

[jmroberts70]

Something like this needs a much neater implementation.

Well it's a step. The power wires are still not bound together -and $14 a cable? OMG that's expensive for some wire and a couple connectors. I'll wait for the price to drop before I start buying and playing with them. Thanks for the link though! It's good to know that some manufacturer is already on the right track.

 

[MeanieMan]

Isn't the custom twisting and bending of the wires bad? You end up putting the wrong channels next to each other and they can mess up each other's "pulses".

 

[ashmedai]

It can be if you screw it up...but then again that goes for alot of the things we do here, doesn't it?

 

[jmroberts70]

Isn't the custom twisting and bending of the wires bad? You end up putting the wrong channels next to each other and they can mess up each other's "pulses".

Well I thought that too as I know the whole idea of the ATA-133 standard with the 80-wire ribbon was to have every other wire be a ground to better isolate each signal wire but then I realized that rounded IDE cables in general go against that entire concept so I figured there would be no problem with what I was doing as it was only a further extension of the idea. In reality, I thought of doing benchmarks with and without flat 80-wire IDE cables to see if there actually is a performance hit with using the rounded cables but I never got around to it. In reality, I have been running this system for 2 weeks now with some rather disk-intensive operations as this is my main file server and have had no problems. I have also done several real-world implementations of these cable runs in the past with no noticible effect on disk performance or data loss.

I'm sure someone here will be able to better comment on the actualy physics of the cables and the IDE standard but for me, I've had no problems with them as they are.

 

[MeanieMan]

Well I thought that too as I know the whole idea of the ATA-133 standard with the 80-wire ribbon was to have every other wire be a ground to better isolate each signal wire but then I realized that rounded IDE cables in general go against that entire concept so I figured there would be no problem with what I was doing as it was only a further extension of the idea. In reality, I thought of doing benchmarks with and without flat 80-wire IDE cables to see if there actually is a performance hit with using the rounded cables but I never got around to it. In reality, I have been running this system for 2 weeks now with some rather disk-intensive operations as this is my main file server and have had no problems. I have also done several real-world implementations of these cable runs in the past with no noticible effect on disk performance or data loss.

I'm sure someone here will be able to better comment on the actualy physics of the cables and the IDE standard but for me, I've had no problems with them as they are.

I don't see why you get different benchmark scores. When they mess up its just going to cause something along the lines of a read or write error. Which could some serious headaches, especially if it happens doing defrag.

Not all rounded cables are the same. The good ones wrap the transfer wires in the grounding wires, then they are shielded from point to point. Your wires are bare naked and are folded and twisted into each other.

It only takes one incident to mess everything up. This is about the same idea as POS PSUs...its worked so far...

 

[ColinR]

I would have thought that there would be some interface-level parity checking going on, and maybe a CRC-style check going on just below file system level to protect against errors.

Surely someone's benched this for a comparison... 80 flat vs 80 round vs 40 round vs 40 flat. Rapid PC and rapid hard disks. There must be one or two people round here with the kit!

I'd do it, but I'm at work I'll do it later if I remember and have time (major chore to get at my hd connectors on my MSI).

 

[jmroberts70]

Well if no one beats me to it, I'm able to run the bench as I have all variety of cable and quality. But I won't be able to run the tests until this Wednesday afternoon. If I'm still the lab rat for the job, I'd love some suggestions of which HDD bench tests to use...