Speedier than most of your USB sticks: Corsair Flash Voyager 16GB USB 3.0 review

James Trevaskis
24 September 2011, 6:00 AM


Send to a friend Print

Corsair keeps the quality coming with an update to its Flash Voyager series.


For as long as USB storage devices have been in the market, Corsair has been pushing the technology to the limits. A few years ago it introduced its Flash Voyager series; with a flexible rubber casing and high transfer speeds the range has always exuded quality. Now with every motherboard on the market featuring USB 3.0, Corsair has updated its range to include this technology.



The USB 3.0 specification has a theoretical maximum transfer speed of 572MB/s, which is a 10 fold increase over the 57MB/s offered by USB 2.0. The Corsair Flash Voyager 16GB has a maximum read speed of 70MB/s and write speed of 19MB/s, these speeds seem very slow when you consider the maximum speed of USB 3.0 but there is more to consider than just the interface speed. The bottleneck now lies not in the interface, but in the speed of the flash memory on the USB stick itself.

Theoretical talk aside, this stick put up some very interesting numbers when we pit it against the Mach Xtreme FX 16GB USB 3.0 stick. Both units showed excellent sequential read speeds, with a small 1% difference between them. When comparing sequential write speeds Corsair came in with 20.24MB/s while Mach Xtreme dominated with 35.59MB/s. The tables were turned slightly when we compared random read speeds with Corsair taking a convincing 5% lead in these tests.

The picture becomes slightly clearer when we compare the current retail prices of $35 for the Corsair Flash Voyager and $75 for the Mach Xtreme FX, in fact Corsair offer a 32GB variant for as low as $75 and it boasts a 39MB/s write speed. Considering the performance, enclosure quality and price, the Corsair Flash Voyager 16GB is amazing value if you’re looking for a new USB stick.

Available from Corsair, retailing for $35.
APC rating: 8/10 (Highly Recommended)

Post your comment



Comments

 RSS feed Email alert

Michael J (User):

megabytes and megabits you retards. they are written differently.

24 September 2011, 6:31 PM (1 year ago)report abuse Send to a friend reply

Tony Sarno (APC staff):

Where exactly are we wrong? That's how Corsair itself also measures it.

25 September 2011, 5:03 AM (1 year ago)report abuse Send to a friend reply

apcuser (New user):

One must be extremely careful about displaying one's erudition, to wit: I can find absolutely nothing wrong with your article as regards statements of memory size. You use "MB" when you mean "megabytes". I cannot even find a reference to "megabits", which, I'm certain, you would have written as "mb".

Great article: explanatory, expository, concise--just the device I've been looking for; thanks.

ps--

I would like to point out a little-known fact which usb-memory manufacturers never mention (I'll leave it to you to determine the possible reason):

the usb standard specifies only the serial bit rate of the transmission, and the protocol. The protocol is the same for all usb vesions (1.x, 2.0, 3.0): four "housekeeping" bits for every six data bits in a 10-bit "package".
So, depending upon how you run the numbers, this is only about a 60% actual DATA rate relative to the USB device's speed. In the case of USB 3.0, with a transmission rate of 4.8 Gb/s, and considering that there are eight bits required to make up one byte, USB 3.0, with its 4.8 Gb/s speed, is capable of "only" delivering a DATA transmission rate of around 343 MB/s.

Just thought you'd like to know...

25 September 2011, 2:36 AM (1 year ago)report abuse Send to a friend reply

apcuser (New user):


"A little knowledge is dangerous thing; so is a lot." --A. Einstein
"It is better to remain silent and be thought a fool, than to open your mouth and remove all doubt." -- George Eliot

Prof. Einstein did not include the one other important circumstance, which oversight I'm certain he would have corrected had he had the privileege of reading one of the comments to this article: "...and also no knowledge."

I remain continually amazed at the individuals who will say (write) anything simply in order to see their masterpiece in print, facts be damned.
The curious thing is that most posts of this type almost always include personal attacks; obviously, the reasoning is that this (ad-hominum attacks) enhances the commenter's credibility, when just the opposite is true.

I have often wondered about the possibility that publications use the simple criterion of material (comments) containing any form of ad-hominum attack--either directed at the author or other commenters--be an automatic reason for rejection.

Warmest regards...

25 September 2011, 11:13 AM (1 year ago)report abuse Send to a friend reply

Fornax (User):

FYI not sure about the 16gb model but the 32gb model was physically much larger than most normal usb sticks. Also our department purchased 5 of the 32gb USB 3.0 and have had 2 of the 5 fail so far within 10 weeks of use. They are bing used in an IT department but my old Corsair 16gb lasted almost 2 years before failure. They are incredibly fast sticks even on usb 2 ports.

26 September 2011, 12:38 PM (1 year ago)report abuse Send to a friend reply

ss-rotel (User):

the downside with the older USB2 voyager stuff coming into OZ recently is they are being classed as "Refurbs". if that's so corsair can get around a tax or import thing, or that they really ARE refurbs, makes me not wanna buy a corsair pendrive...

but that's just me

26 September 2011, 4:17 PM (1 year ago)report abuse Send to a friend reply

ss-rotel (User):

quote apcuser - four "housekeeping" bits for every six data bits in a 10-bit "package".

I dont think that's correct. USUALLY bytes are 8 bit, with maybe a EC bit depending on the protocol, so to me what you said didn't make sense. so i looked it up

the site i found discribes data transmition over USB simular, but not the same, as the protocols used in TCP/IP. It discribes the USB transfer as having a token packet, data packet, acknoweldgement packet, and a Start of Frame packet.

without going into it in great details, it states that data packets can be up to 1024bytes for high speed, or USB2 devices, and discibes 4 different data types that can be sent identified with a 4bit PID.

Also, the above rates are all in megabytes per second, as the article states 57MBps transfer rate, were USB2 is rated USUALLY as 480megabits per sec.

480/8 = 60, which is close to 57MBps right? Add overhead and packet ACK, and you'll get 57MBps.


The article also states that there is a bulk transfer protocol, which is why you get faster transfer of small number of big files, or a large number of small files, which is basically an anologue of TCP/IP

so far, it's the mechanical side of the storage device that's the bottlekneck, not the USB interface or protocol. the FLASH memory or the HDD that you're using is not fast enough to except the data being moved over this high speed bus anyway, so it doesn't matter anyway

26 September 2011, 4:13 PM (1 year ago)report abuse Send to a friend reply

Not signed in
anonymous user Anonymous user