I need a hand - Rally the scientists

[parb]

yeah right....

im far more of a cynic than you aer. i am fairly convinced that mfgs who put beads on their cables are far more afraid of emissions from the cable than shielding. in fact i think this has far more to do with fcc part 15 compliance than signal integrity.... and im sure vendors wouldlove for you to pay for this compliqnce as a benefit...


also,you have some seriously impressive hard drives. generally its hard to squeeze much more than eight to ten megs per sec out of a consumer class diskdrive, even when doing full on sequential io's. you not only managed that, you got twentyseven mb reading and twentyseven writing! that is seriously impressive and borders on the impossible, unless you where bursting small amounts into disk buffers. either way its a bit of pointless test on usb since you entered a special usb signalling mode called high speed mode which relies on reduced signal amplitudes in exchange for higher signal to signal transmissions rates.

but your audio stream will rarely or should i say never enter highspeed signalling, you will sit in the isochronous protocol and need maybe four words worth of bandwidth at two hundred khz or so ~ maybe one megqbyte worth of datatransfer (24 bits of data sent at 192khz rate worst case)... hardly high speed signalling even if you quadrupled that bandwidth. isochronous mode will at best use up 60% of the bandwidth in the cable. but typical is a few percent, and here im assuming usb 2.0 is what you are using.

if youreallyvwoory about performance, just use shorter cables.... it will give you much better signal transfer characteristics and reduce retransmits from errors etc...

usb signalling isnt somewhat sophisticated and employs different transfer modes depending on usage and speeds requested, i think youll be fine with regular cables certified to the standard.

Originally Posted by danielwritesbac

In the efforts from most manufactures to make a substantial looking product for the least money, you'll find many somewhat defective but still marginally working USB cables. The exact same is true for internal drive cables like SATA and IDE.

Better USB cables are easy to spot. If the manufacturer actually cares about data integrity, the USB cable will have a ferrite. This clamps some of your signal but it clamps more noise than signal in theory. Its probably unnecessary, but I tested it.
Using a simple hard drive speed test to an external hard drive, I got 27 megabytes per second with the ferrite cable and somewhat less performance with other cables. The CD won't need that performance; however, I'm just guessing that you'd probably prefer a good cable. The difference is a few cents.

So, on my messages for this thread, I'm thinking "audiophile" and what will match with that, and what won't. In a system symmetry approach, you just want to try to do a little better. . . on every piece so that there's not a weak link.

The majority of this concern is manufacturing variance and muntzing. Muntzing - Wikipedia, the free encyclopedia

But its also a concern that CD discs are deprecated and supporting them as well as possible isn't a goal for the majority of manufacturers today, in 2010.
However, during the timeframe that the Sony 52x writer was manufactured, supporting CD writing was a primary goal. That is a major difference.

[danielwritesbac]

Originally Posted by parb

yeah right....

I'm far more of a cynic than you are. i am fairly convinced that mfgs who put beads on their cables are far more afraid of emissions from the cable than shielding. in fact i think this has far more to do with FCC part 15 compliance than signal integrity.... and I'm sure vendors would love for you to pay for this compliance as a benefit...

Well, it does seem odd that muffling some signal with a magnet choke would be beneficial to the speed; however, maybe it removes more unintentional noise than it does signal. One would hope.

The Rosewill R2 that was used in that 27 megabytes per second over USB example, is "touchy" about its USB cable. Recent production has begun shipping with ferrite choke cables, and that has decreased returns.

The power cord also has a ferrite, and that definitely does work for cleaner power. So does knots in the power cord, but that looks ridiculous. Yes, my 5-1/4" CD writer's enclosure has a few evenly spaced knots in the power cord. I wasn't sure if the bargain basement design of the enclosure or its power supply had any effective RF filtering, but there was no need to guess.

When all is said and done, it seems that digital needs a very definite signal in order to make zero or one, not 0.58943923847, which would be a noise (weak bit? weak byte? can't remember the term for that noise).

GRC SpinRite, its a program designed to remove the weak bit (weak byte?) noises from the hard drive, causing it to store only 0 and 1 after its maintenance cycle has run, usually overnight. Well used hard drives get this same problem. Likewise, CD writers, especially DVD writers can occasionally burn a too small hole, causing error correction to either temporarily muffle or emit a little "flip" sound during playback, that's barely noticeable. . . except on audiophile equipment designed to highlight differences, and then that's really an obnoxious standout difference in an otherwise near-perfect presentation.

Originally Posted by parb

also,you have some seriously impressive hard drives. generally its hard to squeeze much more than eight to ten megs per sec out of a consumer class diskdrive, even when doing full on sequential io's. you not only managed that, you got twentyseven mb reading and twentyseven writing! that is seriously impressive and borders on the impossible, unless you where bursting small amounts into disk buffers. . . .

Single drive speeds while directly connected to SATA may go up to 145 megabytes per second.

Tools:
Advanced Format Alignment Tool for Windows XP era operating systems
Photo showing backwards compatibility jumper for SATA 150 controllers (may not be necessary to use a jumper)

Examples:
Newegg.com - SAMSUNG Spinpoint F4 HD322GJ/U 320GB 7200 RPM 16MB Cache SATA 3.0Gb/s 3.5" Internal Hard Drive -Bare Drive
Newegg.com - SAMSUNG Spinpoint F3 HD103SJ 1TB 7200 RPM 32MB Cache SATA 3.0Gb/s 3.5" Internal Hard Drive -Bare Drive

Of course, at some point, you'll be wanting a decent quality cable. Because the data cables are bottom dollar components (no matter what you're charged for it), there's much variety, most of which is not optimal.

P.S.
Back on topic of getting your favorite old CD writer connected. . . since the 40 wire IDE standard is now long since incompatible, then a near optimal choice is an external enclosure capable of both USB and Firewire. Its not actually necessary to use the Firewire, but you'd like the controller better on that sort of enclosure, because its built for speed and thus you'd have a steady buffer when writing discs, surprisingly, even when using USB.
You'd still have to guess a bit when shopping, but maybe that tip will "improve the odds" much closer to probable. Actually, the whole point of using a dedicated CD-only writer would be to improve the odds. Just make sure its capable of the new "700" CD discs. see CDfreaks.com forum for more information and any necessary flash rom updates (that should be done before using an external enclosure). For reference, can't find drive, weird buffer run-out behavior (could make cd noise), and appearing, disappearing, reappearing in "My Computer" is probably a bad USB cable.

P.P.S. (the point)
Also reference CDfreaks.com forum for the common problem of finding a CD blank that is excellently compatible with both writer and player--a very common problem that's laborious to solve. But, solve that problem and you'll stop the playback noises (from error correction muffling, or lacking error correction brashness/flippy sound). That particular problem and the many wild explanations for it, is probably the cause of this thread. Its data integrity whereby half measures are noises, especially on digital.

[godfrey]

Originally Posted by danielwritesbac

Well, it does seem odd that muffling some signal with a magnet choke would be beneficial to the speed; however, maybe it removes more unintentional noise than it does signal. One would hope.

Ferrite around the cable only affects common mode signals, so common mode noise and interference is attenuated, but the desired signal should be unaffected.

Originally Posted by danielwritesbac

The power cord also has a ferrite, and that definitely does work for cleaner power. So does knots in the power cord, but that looks ridiculous. Yes, my 5-1/4" CD writer's enclosure has a few evenly spaced knots in the power cord. I wasn't sure if the bargain basement design of the enclosure or its power supply had any effective RF filtering, but there was no need to guess.

Looping the cable a couple of times through a ferrite ring should be good too.

[danielwritesbac]

Thank you very much. That's good information for to "better the odds" and it seems to work.