Viewing page 5 of 7 pages. Previous 1 2 3 4 5 6 7 Next
The ATIV's A6-1450 proved to be efficient, managing to keep ahead of the older G-T56N (similar to the E-450) in all of my tests despite having a much lower clock speed. However, the only substantial gains were in my multi-threaded video encoding tests having twice as many cores really helped here. It absolutely crushed the older Atom N2600 across the board but it couldn't compete with an 18W dual core Ivy Bridge chip like the original NUC's Core i3-3217U. One of Intel's new Bay Trail Atoms perhaps would be a more appropriate match.
Navigating through Windows 8 felt fairly snappy but there were a few laggy
moments here and there. The Task Manager in particular would always seem to
take about four seconds to pop-up the first time and while looking at the CPU
utilization, I would find the Task Manager itself eating up 10% of the available
CPU cycles. While the benchmarks show modest single-threaded performance, this
weakness is obscured during general operation by the extra CPU cores and the
SSD. The entires system takes only 8 seconds to boot up from start to finish.
When I hear APU, "better than integrated graphics" comes to mind but this hasn't really been true since Intel upped their game with their last few generations of HD graphics. The A6-1450's HD 8250 graphics are an incremental upgrade over the G-T56N's HD 6320 graphics processor, but neither is truly capable. This 8W APU just doesn't have enough horses to run modern triple-A titles well, even at the native resolution of 1366x768, but older less demanding titles run just fine. I tried out World of Warcraft for half an hour on maximum quality settings and there was no lag at all.
HD Video playback was flawless, though it did require noticeably more resources than typical desktop hardware. CPU usage was about 20% (including the Task Manager) when playing back x264-encoded MKVs, and HD YouTube videos drove it up to ~70% for 720p and ~80% for 1080p content, and switching from Flash to HTML5 was even more taxing. The GPU handles some of the decoding but it really only lends a helping hand rather than performing any heavy lifting. More intensive Flash-based Facebook games were also somewhat choppy.
For the WiFi performance test, I sent a large file transfer (700MB) over 802.11n to and from a machine connected via gigabit ethernet and timed the operation to calculate the average transfer rate. I also checked signal strength to the various wireless networks in our area by using the "netsh" tool from the MS-DOS command line.
It should be noted that the 802.11n router servicing our lab is not the greatest, an Actiontec combination router/gateway from our ADSL provider. It is located in a central location, about a few feet away with only one wall between it and our test systems so it should produce ideal results.
Compared to the wireless adapters on the last few WiFi-equipped motherboards I've reviewed, the ATIV's NIC is unusual in that it produced a higher upload rather than download transfer rate. Not only is this opposite of typical, it's also not altogether useful for most users.
In our lab, the ATIV picked up the usual networks in our vicinity and the reported relative signal strength was about average. However it should be noted that when I brought the machine home, it detected a lot of SSIDs (my area has a greater WiFi saturation), a few of which I had never seen before.
|Help support this site, buy from one of our affiliate retailers!|