Hardware Canucks

Hardware Canucks (http://www.hardwarecanucks.com/forum/)
-   Storage (http://www.hardwarecanucks.com/forum/storage/)
-   -   Different quality of USB cables? (http://www.hardwarecanucks.com/forum/storage/29513-different-quality-usb-cables.html)

frontier204 March 4, 2010 09:55 PM

Different quality of USB cables?
I just purchased one of the newer Kingston SSDs, the SNV425-S2/128GB, notebook kit. I plan to create a "Member Review" for the SSD KIT, so I decided to start with the included 2.5" enclosure, and see how the SSD performs as an oversized USB flash drive. I wanted to compare the performance of the ENCLOSURE with a "Tsunami" cheapo drive enclosure that I had lying around. In performing an HDTune on the SSD + USB enclosure kit, I found terribly inconsistent results, which I narrowed down to the USB 2.0 A - to Mini - B cable that came with the Tsunami enclosure.

Cable A - Cable from Tsunami Enclosure (Length: ~115 cm)

Cable B - Another USB A - to Mini B cable I have lying around; I think it came with a camera or embedded systems development kit (Length: ~100 cm)

There's a ~7 MB/s difference in read speed, just by using a different cable! Anyway, the Tsunami cable is going in the trash, as I think this bench discrepancy and I/O issues I found while doing DBAN through the same cable are related. To give the benefit of the doubt, I was running BOINC for the good cable while benching, but not for the bad cable.

All other factors were held constant, EXCEPT the cable, and the BOINC ran intentionally while benching with the good cable:
System: Dell Inspiron 1420 / Intel Santa Rosa Chipset
CPU: Intel C2D T7500 2.2GHz
RAM: 2x2GB Corsair
OS: Windows 7 Pro 64-bit with all default drivers installed
USB Enclosure: Kingston SNA-DC/U (included with the SSD)

I just wanted to share the discrepancy I found between the two cables, and as a little reminder that the cable is an important factor. As a side question, have you seen any higher than ~30 MB/s from a USB 2.0 enclosure? I've run HDTune on other drives that easily have higher sustained read speeds than USB 2.0 can provide (e.g. Seagate Barracuda 7200.11, Samsung F1), and they average ~28 MB/s.

lowfat March 5, 2010 12:58 AM

That is quite baffling.

AkG March 5, 2010 05:20 AM

Either something else was robbing CPU cycles the USB enclosure needed....or that first cable had a intermittent short causing retransmission delays. My guess is that it was "almost" working but some of the transmissions were getting scrambled and thus had to be resent. That would easily explain the lower numbers as a working cable is a working cable.

frontier204 March 5, 2010 03:52 PM

I agree with you about "almost" working... I expected Windows to at least warn me about things like this - heck I copied hundreds of GB back and forth from that hard drive using that same cable for months. The benchie and an I/O error from Linux tipped me off to this occurrence.
I've experienced a very poor cable (actually, it was a hack where my research group tried non-shielded cable for USB), and the computer just reverted back to USB 1.1 speeds. I didn't know there was something BETWEEN fully-working USB 2.0 and junk cable that computer reverts to USB 1.1 (or flat-out nonworking cable).
That's new for me; I'll add that to my list of miscellaneous hardware troubleshooting encounters :thumb:

All times are GMT -7. The time now is 10:22 AM.