Dell XPS 13 2016 (i7, 256 GB, QHD+) Notebook Review

3200x1800 pixels on 13 inches - that is quite imposing seeing that we are looking at 276 PPI here. This fine resolution is no longer perceptible as pixels to the eye (from about 30 cm/~1 ft), and we thus used our microscope. One pixel is now 0.092 millimeters. Like in the predecessor, the high resolution is combined with a 10-finger multi-touchscreen, namely the SHP1421 LQ133Y1 by Sharp. Unlike the former model (XPS 13-9343, also SHP1421) this is an IGZO panel. Sharp introduced this technology to the market, where amorphous silica (aSi TFTs) replaces indium gallium zinc oxide. That increases the electron mobility many times over, and leads to faster response speeds. That might be theory because the 30 or 40 milliseconds specified under Response Times are not what gamers would expect.

We measure the TFT's brightness using disabled adaptive brightness in a dark room by default. We did not observe visible fluctuation in the base brightness as is common for sensors (ambient light) after disabling this option. However, the sensor in our XPS remained active, which we recorded via different rates depending on whether we covered the sensor beside the webcam (center: 285 cd/m²), left it uncovered (308 cd/m²) or illuminated with a small lamp (323 cd/m²). The brightness thus still has a bit of potential. The screen's brightness no longer corresponds with the good rates of the XPS 13-9343 touch/non-touch, anyway. The reason might be Dell's Content Adaptive Backlight Control (CABC), which is a Firmware-based control of the screen's brightness according to the content. A CABC off option is not in the BIOS.

Upon inquiry, Dell replied that the Content Adaptive Brightness Control (CABC) functioned properly, but there were some inquiries from customers who find that annoying. Dell has a tool for QHD+ XPS 13 systems for disabling CABC for these users - or also for certain application fields. However, Dell had to admit that the trick does not work on XPS 13 FHD devices. Dell could not provide us with it at date of publication, but we will try it when it is available.

The screen from Sharp only covers 93% of the sRGB color space, and AdobeRGB should not even be considered with 59%. Calibrating does not change that, but it improves the color and grayscale accuracy. ColorChecker's DeltaE drops from 2.5 to 2.4, and Grayscale is reduced by 2.0 to 2.9. We use the non-calibrated DeltaE rates for the rating. Owners of the XPS 13 QHD+ can install the ICC profile linked in the box above in color management to achieve a similar improvement. Seen in total, the differences are low so that the naked eye usually will not perceive a difference. Thus, our QHD+ review sample comes close to the outstanding color reproduction of Microsoft's Surface Book when calibrated.

As usual, we look at the clock behavior during multiple Cinebench tests. The dual-core performs the multi-core test at a stable 2.9 GHz, which is clearly within the Turbo range. 3.0 GHz are even sometimes observed at initiation, but only for 20 to 40 seconds depending on casing and ambient temperature. That is an impressive and, above all, stable Turbo utilization at its very best. The SoC even maintains its 2.9 GHz after one hour of looping Cinebench R15. Kudos for such a slim casing.

The Core i5-6200U in the sister model performed these tests at 2.7 GHz (max. Turbo: 2.8 GHz), i.e. also with the best possible multi-core performance. There is nothing to complain about here, which the charts confirm. The 6200U only lags behind the 6500U installed here by 5/9% and 7/12% (R15/R11.5 both multi/single). The advantage in processing performance is very marginal, and thus buying the more expensive Core i7 model is only worthwhile for specific high-performance application fields.

The 2015 XPS 13 with the Broadwell Core i5 (-28/34 multi) is certainly outclassed significantly. The machines from Microsoft with an i5-6300U (2.4 - 3.0 GHz) are on par despite a 100 MHz lower clock (measuring tolerance). Consequently, computing power will hardly be the decisive point for buying.

System Noise

The noise production is restrained during undemanding use; the XPS is only noiseless in permanent idle phases. The fan rarely spins in high performance mode, and strongly depends on the background applications of the operating system. We measured a maximum idling noise of 34.8 dB(A), which is the same as under load. Although the fan has intermediate levels and does not speed up sporadically, we either did not hear anything or just the maximum noise in practice.

We installed "Dell Command Power Manager" since it was not preloaded on our review sample. In addition to battery specifications, the user will also find a temperature manager. The fan was permanently disabled when we set the "Quiet" mode here and "Passive" under Windows' system cooling policy, but only in balanced / high performance.

The differences to the XPS 13-9350  are due to another measuring environment determined by different base noise (silent = 28.1). 31.8 dB(A) stands for the inactive fan in this test. The maximum increase under load of 2-3 dB(A) is roughly the same in both XPS models.

The chart's brief statement: Dell likely has one of the quietest devices in its lineup with the XPS 13. The "noisy" Surface Pro 4 can take an example from that, not to mention Apple's MB Air 13 2015-03 that roars with 46 dB(A) in the stress test. The screenshot illustrates the minor overhead in the comparison of minimum average idling noise (green) and load noise (avg / max; red / yellow).

The XPS 13 i7 can get quite warm; we measured as much as 57 degrees Celsius (134.6 Fahrenheit) in some areas on the upper side and underside. We prompted these temperatures with our stress test that demands everything from the components. This test shows what the system does to control the waste heat in the worst case. It cannot be compared with normal use load, for example via video rendering.

The XPS 13 i7 manages this situation like its i5 sister model: The clock is clearly reduced to below the nominal clock of 2.5 GHz - i7 @ 1.5; i5 @ 1.4 GHz. The temperatures on the SoC climbed to just below 90 degrees (194 Fahrenheit) in both devices, and settled to around 80 degrees (176 Fahrenheit) after throttling stopped.

The iGPU only reduced its clock marginally under stress, so to 900 MHz. That is 50 to 100 MHz less than under simple GPU load via a Cinebench OpenGL test. The HD 520 in the XPS 13-9350 with Core i5 performed the stress test at 950 MHz. Consequently, the GPUs in both XPS models remain behind their potential of 1050 MHz in this (rare) application scenario.

The chart illustrates: Apparently, the XPS 13 i5 manages its waste heat a bit better; the rates are lower especially since we assessed the i7 model at a cooler room temperature. The Surface Pro 4 with its active cooling in the tablet unit is roughly on the same increased waste heat level. Some other rivals and the predecessor remain much cooler.

Power Consumption

The power consumption of the i7 review sample was measured with a new instrument (MetraHIT Energy), making the rates in the chart below only conditionally comparable. We did not perform power measurements with the XPS 13-9350 at the time. The chart shows the typically low power consumption only in the load range. The QHD+ model consumes considerably more power when idle than Apple's MB Air 13 and the Surface Pro 4 (MetraHIT measurement), both with similarly high resolutions. The devices' minimum brightness levels vary. The brightness alone cannot, however, explain the minimum idle noise differences.

i7 owners clearly are at a disadvantage in terms of battery life; the Wi-Fi test stopped 35% earlier than with the i5 model (~ 118 minutes). We used the balanced profile at 150 cd/m² without Dell's Command Power Manager features (was not preloaded) for the test. The XPS 13 QHD+ is consequently not on the high level of an Apple MB Air 13 2015 that lasts almost twice as long in the Wi-Fi test. The other tests (idle, load) are proportionate to this test. We performed the idle test a second time using Dell's energy-saving features (Power Manager) and the power-saving mode: A runtime of 9:18 hours is a bit shorter, but is still within the range of measuring tolerances.

Update 12/03/15: Dell frantically unleashes the 1.1.7 BIOS version. Its improvements are to lead to drastically longer battery runtimes. The manufacturer will officially introduce this BIOS update within the next few days. The testers received an unofficial download that we are not permitted to provide here. According to Dell, the present BIOS prevents the new NVMe-based configurations to utilize the Low-Power C-States from Skylake. We took the opportunity the second time around to also update the existing audio, Wi-Fi and VGA drivers. And in fact, the runtimes have improved considerably. The idle runtime increases from previously 9:37 (577 min) to 13:04 hours. Wi-Fi even improves its lead of 5:33 (333 minutes) to decent 7:12 hours. The XPS 13 can thus keep up quite well with its rivals.

Dell's Command Power Manager enables intervening in the automatic battery charge as well as in the power consumption. Clever users can even prompt battery mode although the device is connected to a power outlet. Of course, it is also possible to disable battery charging at pre-defined times. Dell has apparently designed its device for the decentralized power supply (Smart Power Grid) of the future. Whether the XPS user is at all interested in studying the power consumption to that extent is a personal decision. However, this feature can be seen as useful with a certain background knowledge.

The Battery Extender stores settings that ensure lower power consumption, such as reduced CPU performance or the brightness of the keyboard backlight. The system utilizes the settings made here when the plug is pulled from the outlet.

There are plenty of devices to choose from: Should I take the stylish XPS 13 Ultrabook with the razor-thin bezel, or the Surface Pro 4 tablet? Or plainly: XPS 13 FHD vs. XPS 13 QHD? The decision is not dependent on the computing power; both the XPS and Surface compute on the same high level without throttling. The performance difference to the lower priced Core i5 version can be neglected in our opinion.

The shorter runtime of the XPS 13 i7/QHD version compared with the i5/FHD sister model and the comparison to the market speak against it. 5:30 hours are simply too short for a premium-range Ultrabook. Details like ambient brightness control also used by PWM, the poor maintenance options and the poor webcam are certainly worthy of criticism. We will keep our readers updated about insights concerning Content Adaptive Brightness Control (CABC) here.

Many arguments speak for the XPS 13 - including the i5 version - as listed under Pros. The high-quality chassis with the thin bezel is mentioned there, just like the relatively good load performance, the still low noise development, and the high-contrast panel with a high resolution. The potential buyer will have to decide whether or not that is a must. The handling via touch also functions as impeccably as ever, and the same is true for the keys and ClickPad.

For 1549 Euros (~$1653, gross, free of shipping costs, and/or shops), the customer gets a sophisticated subnotebook, which may be one of the best. The former model from 2015 with an i7-5500U and 256 GB SSD is still going strong on the market, and starts at 1500 Euros (~$1601). Although the new version does not sport an extremely faster processor/graphics card, it has Thunderbolt 3 in USB Type C combined with USB 3.1 Gen 2. Thus, not much speaks for buying an older model from 2015; the user is more future-ready with the upcoming docking options via Thunderbolt 3.