Installation

I had a good candidate drive lying around, a Maxtor 80GB 7200 RPM model that is representative of what you can find in a local Office Depot or CompUSA.  

I disassembled the enclosure by removing the four screws in the front panel.  This allowed me to slide the top aluminum plate off the enclosure.  The individual parts are well-manufactured and I had no quality-control issues.

Here is a close-up of the drive interface electronics. You can see the 80-pin ribbon cable stub has been split to allow some wiggle room.  The standard 4-pin Molex connector is provided for power.

Here are the parts ready for assembly.  The drive secures into the enclosure with four screws (included) tapped into the bottom of the enclosure.

Here is the (almost) completed assembly. The drive is secure, but not cramped.  I've been running the drive for a few days now, and have not noticed any heat or EMI-related issues.

Testing

By this point, I was eager to see if the drive could muster the performance to satisfy my micro-short patience threshold.  I first plugged in the AC adapter.  Next I switched the unit on, and lastly I plugged the USB cord into an available port in a test machine I am working on. The test machine I first used has older Version 1.1 USB ports, so I was not surprised to see the following message appear:

I'll address the performance of the unit later, but know that even though the USB port used was older, the drive still functioned fine.  If you aren't in a hurry. Knowing that this wouldn't be a fair test of the unit, I unplugged the unit from this machine and plugged it into the Bling Box (a project box from my Website) which has USB 2.0 ports (VIA 8237 Southbridge). Windows acknowledged the connection without issue, and all seemed to be fine.  The drive was not formatted or partitioned, so I had to take a small detour to the Computer Management screen to take care of business (START-> RUN -> compmgmt.msc).  Once there, I initialized, partitioned and formatted the drive with an NTFS partition.  Windows assigned a drive letter to the drive and then it appeared in My Computer along with my other drives.  My Computer did not differentiate or even hint that it was a removable unit.  It did not request or require the installation of any drivers or utilities to begin using the drive, so I tested it in this configuration.

In order to test the performance of the drive, I set it up to disable write-caching because I wanted to avoid data loss in the event that the power switch on the enclosure was suddenly turned off or in the event of a cord removal.  This is optional, but prudent, IMO.  I don't claim to be an expert on benchmarking hard drives, but I figured that the best way to see if the interface was holding the drive performance back was to test it first inside the enclosure, and then again with the same drive attached to the mainboard IDE port.  I used SiSoft Sandra's file system benchmark module to test the drive, and then tested it with a real-world file transfer.  Below is the output of the two configurations:

USB_Maxtor Test 1,  USB 2.0 Interface	USB_Maxtor Test 2, IDE interface
Benchmark Results	Benchmark Results
Drive Index : 23MB/s	Drive Index : 28MB/s
Performance Test Status	Performance Test Status
Run ID : DEN on Monday, August 30, 2004 at 4:31:32 PM	Run ID : DEN on Monday, August 30, 2004 at 4:56:00 PM
SMP Test : No	SMP Test : No
Total Test Threads : 1	Total Test Threads : 1
SMT Test : No	SMT Test : No
Dynamic MP/MT Load Balance : No	Dynamic MP/MT Load Balance : No
Processor Affinity : No	Processor Affinity : No
Operating System Disk Cache Used : No	Operating System Disk Cache Used : No
Use Overlapped I/O : Yes	Use Overlapped I/O : Yes
IO Queue Depth : 4 request(s)	IO Queue Depth : 4 request(s)
Test File Size : 1023MB	Test File Size : 1023MB
File Server Optimised : No	File Server Optimised : No
Benchmark Breakdown	Benchmark Breakdown
Buffered Read : 25 MB/s	Buffered Read : 29 MB/s
Sequential Read : 26 MB/s	Sequential Read : 33 MB/s
Random Read : 18 MB/s	Random Read : 24 MB/s
Buffered Write : 26 MB/s	Buffered Write : 25 MB/s
Sequential Write : 26 MB/s	Sequential Write : 31 MB/s
Random Write : 24 MB/s	Random Write : 16 MB/s
Average Access Time : 17 ms (estimated)	Average Access Time : 11 ms (estimated)
Drive	Drive
Drive Type : Hard Disk	Drive Type : Hard Disk
Total Size : 31GB	Total Size : 31GB
Free Space : 31GB, 100%	Free Space : 31GB, 100%

Here is a summary of my experience (your mileage may vary). I timed these transfers using a stopwatch, so I think two significant digits is all the accuracy you can glean from these informal tests. I tested the transfers in both directions.

313MB File Copy

(write)SATA to USB: 16 seconds (approx 19.6 MB/s)

(read)USB to SATA:  7.0 seconds (approx 44MB/s)

For reference, I also timed the drives when they were installed in the case via the onboard IDE/SATA ports. Same files.

(write)SATA to IDE: 15 seconds

(read)IDE to SATA: 12 seconds

I came away from this informal testing with the following conclusion:  In my configuration, with my drives, I did not find the USB enclosure to be in any measurable way slower than the internal interface, which basically floored me. Transferring 313MB from an external drive to my main box in less than ten seconds is stellar.  Some of you with more advanced boxes might not be impressed, but this is by far the fastest external storage system I have used. 

I have tried various solutions in the past, like ZIP, JAZ and Magneto-Optical drives and none of them have come close to this level of performance. Since the documentation lists the USB transfer rate at 480Mbps maximum (60MB/s), it stands to reason that the drive itself would be the bottleneck.  This is great news for folks that need internal-drive performance in a removable package. 

With multi-gigabytes of fast storage on hand, I'm sure that I don't need to spell out the possibilities. You may be wondering what kind of performance the drive attains in an older USB 1.1 equipped box.  Well, I tried that out, too.  Let me just say that I had full intentions of testing the same suite with the older interface, but when I started transferring the same folder over to my other box for testing, it was glaringly obvious that it was a waste of time.  A lot of time. 

To transfer the 313MB folder across the USB 1.1 interface, the copy files progress window estimated six minutes.  I let the stopwatch go for a few seconds to see if it was going to speed up, but no dice.  I stopped the stopwatch and went to fix a sandwich.  About six minutes later, while I was finishing off the sandwich, it finished off the file transfer.  That was the only benchmark I needed to run (and hopefully I won't be chastised too hard for not laboring through the rest of the process). 

Suffice it to say that I had visions of stacks of floppies and multi-part ZIP archives going through my head.  Painful.  But it worked without issue, and the machine recognized the drive right away. A burned CD-ROM would be a better option on this platform, albeit with a lower capacity.

Conclusion

All in all, I am very pleased with the performance of the unit when using it with the USB 2.0 interface.  The unit was simple to assemble, uses standard drives and looks very good with the LED's and aluminum construction.  It used no drivers, and Windows identified and configured the drive in less than 10 seconds from initial hook-up.  I give the KingWin Giga-Plus Hard Drive Enclosure (USB 2.0 version) high marks for functionality, appearance and ease-of-use.

Pros

• Large capacity, fast data access in a portable package

• No drivers to load (WinXP)

• Good Looks

• Complete Package

• Complete Documentation

Cons

• Short Power Cord

• Requires you to furnish the storage media

• Requires the OS to be loaded prior to use

• Limited OS compatibility