Previously Mark F said:
On Mon, 08 Sep 2008 13:57:34 -0700, Charliec
I don't think you can speed it up, although there is a chance
that using a different USB interface would speed things up.
(I assume you are using USB 2.0 - I am referring to performance
differences with different USB 2.0 interfaces, either on the
same machine or another machine.)
I had some film scanned to disk and it came back on a
Western Digital My Book Essential 1TB
UPC 7 18037 12270 0 R/N: ADC
Order
WDH1U10000N
Model
WD10000H1U-00
The drive inside (I open up just about everything AFTER I test it -
now that we know it works, lets see why.)
seems to be
WD P/N: WD10EAVS-00D780
MDL: WD10EAVS-00D7B0
DATE: 24 JUN 2008
(I didn't take things apart far enough to see if the
disk is a SATA test connected to an independent
SATA-to-USB converter, or if the electronics is
"melded together".)
(Assuming that I copied the model number correctly after
prying the My Book apart, the drive is a Western Digital
Caviar GP WD10EACS 1TB 5400 to 7200 RPM
16MB Cache SATA 3.0Gb/s Hard Drive - OEM
Perhaps what appears to be variable speed affects the performance.)
So it is a fairly recent drive and should have reasonable performance.
However, I found that it performed poorly for a SATA drive connected
to USB 2.0. (I also know from experience that USB 2.0 is typically
1/2 the speed of IEEE-1394a, and typically around 1/4 the speed
of an eSATA connection to a SATA disk.)
Here are the details:
When I copied from the My Book Essential it only
ran at about 15MB/second. (I typically see 22-25MB/second for
USB 2.0 to SATA [as was the case here], and about 40-45MB/second
for input from IEEE-1394a, and 60 MB/second for input from eSATA.)
In other words, the drive performed poorly compared to other
drives with USB 2.0/SATA interfaces, and USB 2.0 only runs
at 1/2 of IEEE-1393a, 1/3 of what I can see for SATA copies.
(My SATA adapter is on a PCI card, so my SATA-to-SATA copy operations
are limited by the PCI bus speed.)
HD Tach and HD Tune also showed low speeds compared to other
USB 2.0/SATA disk external boxes. The numbers were about
15MB/second for HD Tune 2.53 and 17MB/second for
HD Tach RW version 3.0.0. (I think the performance difference
is due to HD Tune using a smaller read block size than HD Tach.)
I used a (six year old) Gateway 700S 2.4 MHz, 512MB, Windows
XP Professional with Service Pack 2 system for all of the
tests and for the copy of the data files from the
WD My Book Essential.
(In my case, I suggested to the company that did the film scanning
that they try a Western Digital My Book Home Edition, which has
USB 2.0, IEEE-1394a, and eSATA if they couldn't get Studio
Edition's formatted for Windows instead of Mac. However the
company would have to weigh evaluating another disk type for their
use as compared what probably would only allow overnight
copies completing faster.)
Incidentially, it is not the fault of the drive. I have one of
these too, also opened, because I run SMART selftests (long) on
every disk periodically. Beginninf ogf the drive, it gives
abouy 88MB/s linear performance and about 45MB/s at the end.
However, since the enclosure is basically "free", as it costs
about the same as the bare drive (forget about that extra disk
memory, it costs WD nearly nothing and has almost no influence
on disk performance, since modern OSes use far larger
buffer/caches). I think for that the performance is reasonable.
The disk can be used without problem on its own (at least IO
had none), and the enclosure accepts other disks. I tried
an old Seagate 80GB ATA with a promise ATA-SATA adapter.
No problem at all.
Arno
