The transfer speed here is mostly affected by the on-board flash memory bandwidth and, to a lesser extent, its read and write latencies (which of course differ a lot). Why is speed so important here? After all, it's just a pocket mass storage device...
Well, imagine yourself in a situation where you got to urgently copy something from a PC somewhere, before someone discovers you? Obviously, the extra speed would make a difference, especially if copying, say, a CD-ROM ISO file.
Here I had a look at the brand new Corsair Flash Voyager 1 GB stick, comparing it with my year-old workhorse, TransCend JetFlash 512 MB. I wanted to see how much faster these device turned in a year, and also if any other improvements are there.
Well, first of all, the Corsair stick dispenses with the usual silver-metallic accessory look present on the Transcend unit, in favour of black-blue rubbery, bouncy device. The Flash Voyager is supposedly water and shock-resistant, and it does look well-enough sealed for the first feature.
For the second one, I threw it against both floor (solid granite) and wall (not brittle American wood planks that any wind takes off, but solid heavy-duty Singapore concrete - here we build (nearly) all homes like fortresses, just in case), and it was fine.
I also pulled it hard from bottom front USB port of my test-torture PC, under a sharp angle - it tilted and bent the USB connector on the memory, but gentle push back put it in place, and all worked without any noticeable after effects or a break in the water proof sealing.
So it was as sturdy as advertised - and no, I didn't try it on the JetFlash, because I still want to keep that unit working, haha...
How about performance? Well, I did few tests here... one is generic Sandra 2005 flash memory benchmark (see results in the attached image), but the other one was practical copying and reading of files under my beloved (surely not?) Windows XP Professional SP1.
Two kinds of files were copied - one were 4 large files for a total of 380 MB, to test the contiguous large-file throughput with little directory overhead, while the other one was a ~390 MB set of three directories (each with many subdirectories and few thousand files altogether) and three root medium-size files. At the end, we also read the 4 large files from each stick. Here are the numbers:
Test LargeFile write LargeFile read Directory write
Corsair 0:27 s 0:22 s 1:47 s
Transcend 1:12 s 0:14 s 8:26 s
Hmm, what a difference! Hey, we even used the same USB port in each case... to make sure things were right. It's interesting, the Corsair part absolutety trounces the JetFlash in writing, especially once there are many files to handle, but it is somewhat slower in reading the large files.
In Sandra synthetic test, on the other hand, both parts seem to show similar performance, with a bit of advantage on the Corsair side. So, aside from the surprising read time showing (which I repeated twice on both units just in case), JetFlash looks good. µ
Sign up for INQbot – a weekly roundup of the best from the INQ