Networking and Storage Performance

Networking and storage are two major aspects which influence our experience with any computing system. This section presents results from our evaluation of these aspects in the Intel NUC6i7KYK (Skull Canyon). On the storage side, one option would be repetition of our strenuous SSD review tests on the drive(s) in the PC. Fortunately, to avoid that overkill, PCMark 8 has a storage bench where certain common workloads such as loading games and document processing are replayed on the target drive. Results are presented in two forms, one being a benchmark number and the other, a bandwidth figure. We ran the PCMark 8 storage bench on selected PCs and the results are presented below.

Futuremark PCMark 8 Storage Bench - Score

Futuremark PCMark 8 Storage Bench - Bandwidth

The storage score (primary result) shows that there is not much to gain by going from the SM951 in the NUC6i5SYK to the 950 PRO in the NUC6i7KYK. It shows that workloads are more user-input and CPU-bound, rather than storage-bound. On the other hand, the storage bandwidth number (secondary result) shows a significant jump. Readers can refer to our explanation of how these numbers are calculated by PCMark 8. The secondary result is the total amount of data transferred (both reads and writes) divided by the storage I/O busy time (i.e, time duration during which the number of pending I/O operations was at least 1). The secondary result is a very important metric when idle time compression is involved, but it doesn't matter as much as the primary result when it comes to application responsiveness (as the workload might be CPU-bound, rather than storage-bound). In any case, the above result shows that a powerful CPU can drive up the secondary result very high.

On the networking side, we restricted ourselves to the evaluation of the WLAN component. Our standard test router is the Netgear R7000 Nighthawk configured with both 2.4 GHz and 5 GHz networks. The router is placed approximately 20 ft. away, separated by a drywall (as in a typical US building). A wired client (Zotac ID89-Plus) is connected to the R7000 and serves as one endpoint for iperf evaluation. The PC under test is made to connect to either the 5 GHz (preferred) or 2.4 GHz SSID and iperf tests are conducted for both TCP and UDP transfers. It is ensured that the PC under test is the only wireless client for the Netgear R7000. We evaluate total throughput for up to 32 simultaneous TCP connections using iperf and present the highest number in the graph below.

Wi-Fi TCP Throughput

In the UDP case, we try to transfer data at the highest rate possible for which we get less than 1% packet loss.

Wi-Fi UDP Throughput (< 1% Packet Loss)

The antenna placement and the system design ensure that the Intel 802.11ac AC8260 WLAN subsystem performs exceptionally well in our Wi-Fi testing and comes out at the top of the charts in both TCP and UDP tests.

Gaming Benchmarks HTPC Credentials
Comments Locked

133 Comments

View All Comments

  • ragenalien - Monday, May 23, 2016 - link

    Could we get a comparison between this and the iris pro 6200? Seems like there isn't much difference performance wise, but there should be.
  • defaultluser - Monday, May 23, 2016 - link

    Skylake gets much better GPU performance/watt than Broadwell did, as evidenced by the NUC with 48 EU 64MB eDRAM being fed by just 23w continuous. That's a huge improvement form the 45w this beast used to take!

    I think the only surprise for me was just 40% performance improvement over the 4770r. I always assumed the 4770r was bandwidth-limited, but I guess the eDRAM cache was enough to keep things fed.

    But yeah, pointless product continues to be pointless. Intel charges a premium for these things because they take up more die space and require dedicated eDRAM cache to feed them...just like discrete GPUs take up more die space, and require dedicated DRAM to feed them. Where is the efficiency gain in this crap?
  • defaultluser - Monday, May 23, 2016 - link

    Oh, I just noticed the review uses 2133 DDR4, which would account for the 40% performance increase we saw. I thought for sure a "premium gaming" platform like this would ship with z170, so I didn't give the test setup a second glance.

    I guess Intel cheaping-out with H170 has forever doomed this machine to mediocrity. Too bad, dropping ten bucks more on the Z170 would have allowed some much more interesting memory configurations. With DDR4 2133 we're probably castrating performance.
  • tipoo - Monday, May 23, 2016 - link

    "Our only concern is that the cooling solution keeps the temperature of the cores too close to the junction temperature during periods of heavy CPU load."

    My rMBP 15", Iris Pro only model, routinely hovers at the tJunction max at load. Is this a real concern? Or is it designed to do this?
  • BrokenCrayons - Monday, May 23, 2016 - link

    There's probably a little wiggle room built into the processor's design by its engineers, but according to Intel's site here: http://www.intel.com/content/www/us/en/support/pro...

    "Tjunction Max is the maximum temperature the cores can reach before thermal throttling is activated. Thermal throttling happens when the processor exceeds the maximum temperature. The processor shuts itself off in order to prevent permanent damage. Tjunction Max (Tj Max) is also referred to as TCC Activation Temperature in certain processor datasheets."

    Basically, reaching the Tjunction means the CPU is close to shutting itself off to prevent damage. That might mean there are longevity implications related to brushing up against that upper ceiling on a regular basis, but I haven't seen any statistical data regarding a meaningful sample of processors put under such conditions failing more often during their few years of useful life due to CPUs going bad just because the OEM decided to implement a cooling solution that allows the processor to wander up to the Tjunction temp when it's working hard.

    I think a bigger concern might be looking into whether or not the rMPB in general will approach Tjunction under load or if that's abnormal. Abnormalities might point to some sort of problem with your specific laptop. I don't know what's status quo for your hardware so its hard to say if that's something you should worry about.
  • tipoo - Monday, May 23, 2016 - link

    Yeah, that's something I looked into, Anandtechs own retina macbook pro 15" review only pegged them at going up to 76 ish celsius if memory serves, but that was the older dGPU model with the 650M. From the threads I'm seeing, the Iris Pro model does regularly hover at 99-101C, I'm guessing since the GPU grunt is right beside the CPU on a single die so heat isn't spread wider like the dedicated GPU model.

    I don't see any reports of this model failing though, so I'd hope they tested extensively at 100 degrees and found it was fine, and so allowed the processor to keep its boost long enough to get there.

    I do wish they could have just added another few mm so that the cooling was better and the CPU and GPU could stay at boost longer, and with that room they could have added some mm to the keyboard too (which I consider the absolute minimum in key travel now).
  • tipoo - Monday, May 23, 2016 - link

    Plus they also only tested Half Life 2, which probably allowed the CPU and GPU not to be at max all the time as it's so old.
  • 8steve8 - Monday, May 23, 2016 - link

    give us a 65W CPU with iris pro , add a couple inches in height... and use the stock retail CPU cooler. Add USB-c in the front. Use USB power delivery usb-c for power.

    Done, the perfect little pc.
  • TheinsanegamerN - Monday, May 23, 2016 - link

    Except the niche NUCs fill dont want a NUC the size of a MINI-ITX case. Not to mention intels stock cooler is not the quietest nor the best cooler in existence.
  • 8steve8 - Monday, May 23, 2016 - link

    well a couple inches of height would not put it near the average mini-itx case size.
    It's impressive that these NUCs are small, but they goes a bit extreme when they use laptop chips and cooler designs meant for laptops. We can get a very small desktop without sacrificing CPU performance and acoustics/thermals.

    Use 65W+ chips w/iris pro, full size intel retail heatsink, usb-c power delivery... no wasted space with expansion slots. 1 m.2 should be the only internal slot.

Log in

Don't have an account? Sign up now