I don’t tend to run the latest and greatest when it comes to tech. Like, almost never. As a result, I’ve been using USB 2.0 devices constantly, despite the much better, MUCH faster USB 3.0 spec being finalized in 2008 and devices beginning to hit the consumer market by late 2009. It wasn’t until I upgraded my work laptop and added a dedicated USB 3.0 controller to my desktop workstation that I discovered what I had been missing all these years.

That’s a screenshot of images copying from a CF card to my local machine via a USB 3.0 multi-card reader/USB hub combo that I picked up for $24 and works like a charm. I also have been using an SATA “docking station” that accepts both 2.5″ and 3.5″ drives, to backup data every few weeks for storage in a fireproof safe. The model I bought was $23 and is unbelievably fast for file copying. Here’s a screenshot of a large file transfer from a 2.5″ SATA drive inserted in the docking station to my local RAID volume:

Out of sheer excitement to have this new (well, new to me) technology at my fingertips, I purchased two 32GB USB 3.0 thumb drives to play with. I always find myself shuffling files around on thumb drives, so I figured it certainly won’t hurt to have a couple more on hand. Both models I bought cost under $30, and both have fantastic reviews on Amazon. So I pit them against each other. My test case is copying large video files to and from each device and noting the speed reported by my operating system. Here are the results:

Now, to be fair, I ran numerous tests with the same file (a 1.4GB video), and observed results that were all over the place. The M70 didn’t do so well on my Windows machine, freezing up for up to 20 seconds between bursts of absurdly fast write speeds, but was smooth as butter on reading, sometimes reaching 210 MBps. On my linux laptop the M70 didn’t freeze up at all, but wasn’t all that much faster on writes as the B30.

The B30 was always smooth on writes, on both operating systems, though it was slower on Windows than on linux. Because I could more accurately measure transfer speeds on the linux machine (using rsync, which I realize isn’t exactly a scientific speed testing tool), the results recorded above were taken from my tests on the linux machine. I suspect Windows is trying to do some kind of buffering, especially on writes, and the M70 didn’t handle it well. Thus the poor experience copying files to the device.

So, for any of you geeks out there who are still using USB 2.0 devices, let me tell you, now that prices have come down on all sorts of USB 3.0 gear, it’s definitely time to upgrade. And even if you’re running an older desktop that doesn’t have a 3.0 controller, this one is working beautifully on my workstation.