
The results for our simulated system backup in a single container file benchmark are similar to those seen in the multimedia benchmarks. Here, too, the transfer rate increase is barely noticeable. Only RAID 6 shows a notable result: there is a transfer rate increase of 8.1 MB/s.

As with the HD video playback benchmark, a noticeable increase in speed occurs in RAID 6 here, with a bump of 15.2 MB/s. The difference between the TS-559 Pro and the TS-559 Pro+ is slightly higher in RAID 0, providing 16.5 MB/s better throughput.

Significant differences between the Qnap TS-559 Pro and the TS-559 Pro+ are not to be found with regard to office files. The influence of the Intel Atom D525 processor in Qnap TS-559 Pro+ mode is limited to increased performance of only 0.2 to 0.7 MB/s. When seen through the lens of a margin of error, the difference between the two devices in this discipline is essentially zero.
More benchmark results can be found in our image gallery.
C'mon! I'm a knuckle-dragging FPS-player: I don't know what "SMB/CIFS protocols" stands for, let alone good for! Isn't there at lest a related article?
Ease of use. Very few users have the time, will and knowledge to build a NAS.
First off, it isn't better than a $500 Linux box. Linux requires Linux knowledge and you have to provide the software you need yourself - that costs small businesses money. This also offers failover and load balancing with its dual NIC card that you wouldn't have in a $500 Linux box.
The main thing is ease of installation. You don't need a highly technical person to get this box up and running and quickly backing up your companies data, whereas a Linux machine will require additional staff that a small business normally would not have on hand and have to pay to come onsite.
For $1000 I'd like one just to play around with myself, though it clearly is not targeted for home users.
There are ways to go with ATX cases, but that is not really comparable.
A NAS is a computer. Heck you can even build a PC put Windows 7/XP Home edition on it and turn it into a NAS all for ~$500 (and thats even with 2tb storage in raid 1, heck that is what I have done and it works great and I am even using a low power AMD CPU that is powerful enough to actually be useful rather than a pathetic atom cpu).
There is no ease of use factor or amount of time on earth that is worth $500+ dollars.
Agreed - I have been running a NAS based on an Atom processor for about two years now (but it's also a RADIUS server, SVN server, etc). However not everyone is technically capable and they are sometimes willing to pay for something off the shelf rather than putting it together themselves.
It's IMO a choice between fire and forget (if one of the thecus boxes fails the plant can replace it themselves) and an ongoing support need. For me that's worth the incremental cost.
Additionally, you'll be going with software raid if you roll your own (hardware raid cards would blow any cost advantage entirely) and then you're going to have to be really conscious of processing power required for solid performance. That means time spent tweaking, optimizing, and in component selection that will more than overtake the hardware cost differential if you cost your time at a reasonable rate.
This can be solved by purchasing a $20 NIC and adding it to the Linux system.
The important thing to note is that if you've got the time and inclination to learn to setup and administer the system, a Linux box will give you more bang for the buck compared to systems like the QNap or Thecus. the N7700's work great as a single host box (or for archive use), but I've not seen the kind of performance that could support even a 2 host cluster. You could get that out of an Openfiler box with some tweaking (though an Openfiler box with that performance would probably cost a grand or so with required peripherals like a dual nic and hot swap SATA chassis).
You may find them under $300!