Intel NUC5i5RYK Review: A Broadwell-U UCFF PC for Enthusiasts
by Ganesh T S on February 20, 2015 8:00 AM ESTPerformance Metrics - II
In this section, we mainly look at benchmark modes in programs used on a day-to-day basis, i.e, application performance and not synthetic workloads.
x264 Benchmark
First off, we have some video encoding benchmarks courtesy of x264 HD Benchmark v5.0. This is simply a test of CPU performance. As expected, the configurations based on the Core i5-5250U are not as powerful as the Core i7-based ones. The effect of faster memory is almost non-existent. The slightly higher burst frequency of the Core i5-5250U (2.7 GHz) compared to the Core i5-4250U (2.6 GHz) help the NUC5i5RYK take a slight lead over the corresponding Haswell NUC.
7-Zip
7-Zip is a very effective and efficient compression program, often beating out OpenCL accelerated commercial programs in benchmarks even while using just the CPU power. 7-Zip has a benchmarking program that provides tons of details regarding the underlying CPU's efficiency. In this subsection, we are interested in the compression and decompression MIPS ratings when utilizing all the available threads.
Results in the compression benchmark are similar to what we observed for 3D rendering - Broadwell-U doesn't provide too much benefit over Haswell-U except when burst frequencies play a major role. This is understandable - 14nm helps sustain higher clock rates for the same power consumption.
TrueCrypt
As businesses (and even home consumers) become more security conscious, the importance of encryption can't be overstated. CPUs supporting the AES-NI instruction for accelerating the encryption and decryption processes have, till now, been the higher end SKUs. However, with Bay Trail, even the lowly Atom series has gained support for AES-NI. The Core i5-4200H in the BXi5G-760 does have AES-NI support. TrueCrypt, a popular open-source disk encryption program can take advantage of the AES-NI capabilities. The TrueCrypt internal benchmark provides some interesting cryptography-related numbers to ponder. In the graph below, we can get an idea of how fast a TrueCrypt volume would behave in the Intel NUC5i5RYK and how it would compare with other select PCs. This is a purely CPU feature / clock speed based test.
Agisoft Photoscan
Agisoft PhotoScan is a commercial program that converts 2D images into 3D point maps, meshes and textures. The program designers sent us a command line version in order to evaluate the efficiency of various systems that go under our review scanner. The command line version has two benchmark modes, one using the CPU and the other using both the CPU and GPU (via OpenCL). The benchmark takes around 50 photographs and does four stages of computation:
- Stage 1: Align Photographs
- Stage 2: Build Point Cloud (capable of OpenCL acceleration)
- Stage 3: Build Mesh
- Stage 4: Build Textures
We record the time taken for each stage. Since various elements of the software are single threaded, others multithreaded, and some use GPUs, it is interesting to record the effects of CPU generations, speeds, number of cores, DRAM parameters and the GPU using this software.
The results show that the faster DRAM helps despite increased latency. GPU acceleration proves beneficial in cutting down the time taken for the second stage considerably.
Dolphin Emulator
Wrapping up our application benchmark numbers is the Dolphin Emulator benchmark mode results. This is again a test of the CPU capabilities, and the Core i7-based models obviously trump the Core i5-based ones.
Facebook
Twitter
RSS
83 Comments
View All Comments
smegforbrain - Saturday, February 21, 2015 - link
Agreed. We initially bought a D34010WYKH1 (i3) at work as a test device last summer to see how well they worked. A couple of our employees have been using them for more than 6 months now.I ended up buying a D54250WYKH1 (i5) for myself to use at home as a media PC and other uses. I'm very happy with it, but I am slightly disappointed that they didn't go with full HDMI for the Broadwell refresh.
mavere - Friday, February 20, 2015 - link
VP9 hybrid acceleration is enabled in recent Intel Broadwell drivers, and apparently Chrome 41+ (currently in Beta) has the ability to hook into Intel's VP9 decoder.In future Broadwell tests, can you check Youtube power consumption with Chrome 41+? Chrome defaults to VP9 on so most Youtube videos nowadays, so this is an important and unexplored area for mobile power usage.
TrackSmart - Friday, February 20, 2015 - link
I don't much of a performance difference between the "mainstream" and the "enthusiast" versions used in this article (aside for artificial SSD benchmarks). Of course, the price difference between the configurations is only $66.Really the biggest performance difference among all of the candidates is between the one machine that has a spinning hard disk and all of the rest. Worth knowing none-the-less.
Essence_of_War - Friday, February 20, 2015 - link
Should I be getting the message that intel not want us to use thunderbolt?Jaybus - Saturday, February 21, 2015 - link
No. It is a matter of power. Thunderbolt standard specifies that every port supply 10 W of DC power. Also, the controller itself is very high bandwidth and requires > 2 W. It adds up on a system striving to be low power and fanless.Samus - Friday, February 20, 2015 - link
Wow Intel has the cleanest uefi bios interface I've ever seen. Shame they don't make motherboards anymore :(A4i - Saturday, February 21, 2015 - link
I got the old i5-4200U , mounted on industrial PC chassis (half inch thick aluminum side panel/radiator). It gets extremely hot, when operating at max CPU load (and ~0 GPU load). Don't hold you breath for power efficiency and productivity. Also that platform lacks any overclocking settings in BIOS or OS tools and is frustratingly painful to undervolt.KAlmquist - Saturday, February 21, 2015 - link
A benchmark question:The "enthusiast" version has 10-10-10-32 1866Mhz memory, while the "mainstream" version has 9-9-9-27 1600Mhz memory. So the difference in memory could account for at most a 17% difference in performance between the two versions (1866Mhz/1600Mhz). But the Cinebench R15 3D rendering multi-threaded benchmark shows the "enthusiast" version to be 34% faster, and the 7-Zip LZMA Decompression benchmark makes the "enthusiast" version 30% faster.
The only other difference between the two versions is the choice of SSD. This could in principle explain the differences listed above, but that would mean that the two benchmarks cited are largely disk I/O bound since we see the effect even with relatively fast disk drives (SSD rather than mechanical) and a relatively slow CPU (1.6 Ghz dual core). I thought those benchmarks were intended to measure computational throughput, and thus should not be affected by the storage subsystem at all. Am I wrong?
Kidster3001 - Friday, February 27, 2015 - link
If you're looking for a pur CPU benchmark that doesn't rely on other parts of teh system then you'll end up with synthetics that everyone complains about because they don't reflect the real world.Yes, Archive applications rely very heavily on I/O. If you want to use them for CPU benchmarks you should set up RAM Disks for the storage.
On a similar note, one of my pet peeves is using browser benchmarks to compare CPUs. Browser benchmarks measure the performance of the browser and it''s engine(s). They do very little good as a CPU comparison tool. Sunspider for instance can score wildly different on the exact same system just by using a different browser.
gfieldew - Saturday, February 21, 2015 - link
Could the FHD model drive a 4K TV/Monitor adequately? I'm not expecting HDMI 2.0 but 60Hz would be good. Thank-you!