In the course of our coverage of mini-PCs, we have seen offerings from vendors such as ASRock, GIGABYTE and Zotac targeting the gaming market. Usually, 'mini' doesn't fit the requirements of consumers in this space, but the appearance of power-efficient high performance GPUs have made the offerings in the gaming mini-PC space quite interesting. Zotac has been creating mini-PCs with a gaming focus by tying a mobile NVIDIA GPU with a Core U-series Intel CPU for a couple of generations now. Today, we will be taking a look at the Zotac ZBOX MAGNUS EN970 - a system combining a Broadwell-U CPU with a NVIDIA Maxwell GM204 mobile GPU.

Introduction and Setup Impressions

The ZBOX E-series targets the gaming market. In the previous years, the E-series adopted a tried and tested industrial design (for example, the chassis of the ZBOX EI750 was very similar to that of the PCs in the ZBOX ID series). The MAGNUS EN970 adopts a radically different industrial design. The unit is not as small as the NUCs, even though the height is similar. The area of the top side is around the same as that of the ASRock Vision series. However, the absence of an optical drive slot enables a chassis with considerably lower thickness.

The specifications of our Zotac ZBOX MAGNUS EN970 review configuration are summarized in the table below.

Zotac ZBOX MAGNUS EN970 Specifications
Processor	Intel Core I5-5200U Broadwell-U, 2C/4T, 2.7GHz, 14nm, 3MB L2, 15W
Memory	2 x 8 GB DDR3L @ 1600MHz
Graphics	NVIDIA GTX 960 (as per drivers), 'a rebadged GTX 970M (hardware-wise)'
Disk Drive(s)	128 GB OCZ Vector
Networking	2x 1GbE Realtek RTL8168 + 1x1 Intel Wireless-AC 3160 802.11ac
Audio	Capable of 5.1/7.1 digital output with HD audio bitstreaming (HDMI)
Operating System	Retail unit is barebones, but we installed Windows 8.1 Pro x64
Pricing (As configured)	USD $978, Barebones is $800
Full Specifications	Zotac ZBOX MAGNUS EN970 Specifications

The Zotac ZBOX MAGNUS EN970 kit doesn't come with any pre-installed OS, but does come with a CD and a read-only USB key containing the drivers. In any case, we ended up installing the latest drivers from Zotac's product support page. In addition to the main unit, the other components of the package include a 180 W (19.5V @ 9.23A) adapter, a US power cord, a single 2.4 GHz / 5 GHz antenna for the Wi-Fi feature, a driver CD / read-only USB key, user's manual and a quick-start guide.

The gallery below takes us around the hardware in the unit.

In the table below, we have an overview of the various systems that we are comparing the Zotac ZBOX MAGNUS EN970 against. Note that they may not belong to the same market segment. The relevant configuration details of the machines are provided so that readers have an understanding of why some benchmark numbers are skewed for or against the Zotac ZBOX MAGNUS EN970 when we come to those sections.

Comparative PC Configurations
Aspect	Zotac ZBOX MAGNUS EN970	Zotac ZBOX MAGNUS EN970ASRock VisionX 420DASRock VisionX 471DGIGABYTE GB-BXi5G-760GIGABYTE GB-BXi7-4770R (1)Intel NUC5i7RYH
CPU	Intel Core i5-5200U	Intel Core i5-5200U
GPU	NVIDIA GTX 960 (3GB) [GTX 970M]	NVIDIA GTX 960 (3GB) [GTX 970M]
RAM	Panram Intl PSD3L1600C118G2VS 11-11-11-28 @ 1600 MHz 2x8 GB	Panram Intl PSD3L1600C118G2VS 11-11-11-28 @ 1600 MHz 2x8 GB
Storage	OCZ Vector (128 GB; SATA 6Gb/s; 25nm; MLC)	OCZ Vector (128 GB; SATA 6Gb/s; 25nm; MLC)
Wi-Fi	Intel Dual Band Wireless-AC 3160 (1x1 802.11ac - 433 Mbps)	Intel Dual Band Wireless-AC 3160 (1x1 802.11ac - 433 Mbps)
Price (in USD, when built)	$978 Barebones is $800	$978

Performance Metrics - I

The Zotac ZBOX MAGNUS EN970 was evaluated using our standard test suite for mini / industrial PCs. Not all benchmarks were processed on all the machines due to updates in our testing procedures. Therefore, the list of PCs in each graph might not be the same.

Futuremark PCMark 8

PCMark 8 provides various usage scenarios (home, creative and work) and offers ways to benchmark both baseline (CPU-only) as well as OpenCL accelerated (CPU + GPU) performance. We benchmarked select PCs for the OpenCL accelerated performance in all three usage scenarios. These scores are heavily influenced by the CPU in the system. The issue for the MAGNUS EN970 is the fact that it comes with a U-series CPU, while the gaming mini-PCs from the competitors use the non-U CPUs. For example, the extra thickness of the ASRock VisionX series units allows them to integrate Core i7-4712MQ CPUs in the VisionX 471D and 420D systems. The GIGABYTE GB-BXi5G-760 uses a Core i5-5200H.

Miscellaneous Futuremark Benchmarks

3D Rendering - CINEBENCH R15

We have moved on from R11.5 to R15 for 3D rendering evaluation. CINEBENCH R15 provides three benchmark modes - OpenGL, single threaded and multi-threaded. Evaluation of select PCs in all three modes provided us the following results. The surprising aspect is the OpenGL results - the previous generation GTX 760 seems to perform better than the GTX 960. As we shall see further on in the review, this issue persists in a few other benchmarks as well. The other results can easily be explained by the low power CPU (Core i5-5200U) in the system.

Performance 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 U-series based system comes in at the bottom of the pile compared to the CPUs with much higher TDPs.

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.

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 i5-5200U in the ZBOX MAGNUS EN970 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 Zotac ZBOX MAGNUS EN970 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.

Dolphin Emulator

Wrapping up our application benchmark numbers is the Dolphin Emulator benchmark mode results. This is again a test of the CPU capabilities.

Gaming Benchmarks

The gaming credentials of the GTX 960 are quite impressive. Even though the tag indicates a desktop GPU, the internal hardware is still a mobile GPU - the GTX 970M. It is based on Maxwell GM204. The performance, however, should be similar to that of the desktop GTX 960.

For the purpose of benchmarking, we chose four different games (Sleeping Dogs, Tomb Raider, Bioshock Infinite and DiRT Showdown) at three different quality levels. In addition, starting with this review, we are also bringing three more benchmarks involving The Talos Principle and GRID Autosport. As someone focusing on HTPCs and multimedia aspects, I rarely get to process gaming benchmarks, even while evaluating GPUs. One of the aspects that I feared was spending lot of time in installing the same games again and again on different PCs under the review scanner. The solution was to go the Steam route. Unfortunately, Steam also likes to keep the game files updated. A quick online search revealed that Steam could make use of an external drive for storing the game executables and downloadable content. With the Steam drive on-the-go use-case being read-heavy, the Corsair Flash Voyager GS USB 3.0 128GB Flash Drive (with read speeds of up to 275 MBps) was ideal for use as a portable Steam drive.

Sleeping Dogs

Tomb Raider

Bioshock Infinite

DiRT Showdown

The Talos Principle

GRID Autosport

In the gaming benchmarks, the NVIDIA GTX 960 (GTX 970M) simply smokes the competition, particularly at the 1080p quality settings. The ZBOX MAGNUS EN970 more than makes up for the weak CPU with impressive gaming performance.

Networking and Storage Performance

We have recently started devoting a separate section to analyze the storage and networking credentials of the units under review. 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.

The OCZ Vector drive we used in our setup has been EOL-ed by OCZ, but there are plenty of similarly performing drives with similar costs in the market.

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.

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

The numbers are similar to what we obtained for other systems with the Intel AC-3160 WiFi 802.11ac PCIe WLAN card. The external antenna helps in provideing better performance.

HTPC Credentials

Despite being actively cooled, I was pleasantly surprised by the noise profile of the system. The fans have variable rotational speeds, and in general, for most media related activities, the fan noise is minimal. I would go as far to say that the noise is probably the lowest amongst all the gaming mini-PCs that we have evaluated.

Refresh Rate Accurancy

Starting with Haswell, Intel, AMD and NVIDIA have been on par with respect to display refresh rate accuracy. The most important refresh rate for videophiles is obviously 23.976 Hz (the 23 Hz setting). The Zotac ZBOX MAGNUS EN970 does not have the same out-of-the-box accuracy possessed by Intel and AMD. However, NVIDIA also allows for setting custom resolutions and refresh rates.

The gallery below presents some of the other refresh rates that we tested out. The first statistic in madVR's OSD indicates the display refresh rate.

Network Streaming Efficiency

Evaluation of OTT playback efficiency was done by playing back our standard YouTube test stream and five minutes from our standard Netflix test title. Using HTML5, the YouTube stream plays back a 720p encoding. Since YouTube now defaults to HTML5 for video playback, we have stopped evaluating Adobe Flash acceleration. Note that only NVIDIA exposes GPU and VPU loads separately. Both Intel and AMD bundle the decoder load along with the GPU load. The following two graphs show the power consumption at the wall for playback of the HTML5 stream in Mozilla Firefox (v 41.0).

GPU load and VPU load were around 16.01% and 21.42% for the YouTube HTML5 stream.GPU load in the steady state for the Netflix streaming case was 5.88% and the VPU load was 4.36%.

Netflix streaming evaluation was done using the Windows 8.1 Netflix app. Manual stream selection is available (Ctrl-Alt-Shift-S) and debug information / statistics can also be viewed (Ctrl-Alt-Shift-D). Statistics collected for the YouTube streaming experiment were also collected here.

Decoding and Rendering Benchmarks

In order to evaluate local file playback, we concentrate on EVR-CP and madVR. We already know that EVR works quite well even with the Intel IGP for our test streams.

In our earlier reviews, we focused on presenting the GPU loading and power consumption at the wall in a table (with problematic streams in bold). Starting with the Broadwell NUC review, we decided to represent the GPU load and power consumption in a graph with dual Y-axes. Nine different test streams of 90 seconds each were played back with a gap of 30 seconds between each of them. The characteristics of each stream are annotated at the bottom of the graph. Note that the GPU usage is graphed in red and needs to be considered against the left axis, while the at-wall power consumption is graphed in green and needs to be considered against the right axis.

Frame drops are evident whenever the GPU load consistently stays above the 85 - 90% mark. An important aspect to note here for NVIDIA GPUs is that the load reported doesn't take into account the GPU clock. For example, the 4Kp30 clip in the EVR-CP mode: Initially, the GPU and VPU operate at higher clocks, resulting in the reported load being 7 - 8% and the power consumption being around 34W. However, later on towards the end of the stream, the load jumps up to around 12%, but the power consumption is now only around 25 W. Essentially, the power consumption at the wall is the real metric of efficiency for decode and playback. As long as the GPU load stays below the 85 - 90% mark, frame drops don't occur.

EVR-CP and madVR Defaults don't present any challenges to the GTX 960

Since the GTX 960 is a powerful GPU, we evaluated madVR with both default and stress settings. The 'madVR stress settings' was configured with the upscaling algorithms set to Jinc 3-tap with anti-ringing activated and the downscaling algorithm set to Lanczos 3-tap, again with anti-ringing activated.

The ZBOX MAGNUS EN970 was able to handle the madVR stress configuration without any issues.

Moving on to codec support, the GTX 970M / GM204 is a known quantity with regards to the scope of supported hardware accelerated codecs.

There is no HEVC_VLD_Main10 support. This is a bit disappointing since the real desktop GTX 960 supports Main10 decoding in hardware. Other than that aspect, the ZBOX MAGNUS EN970 actually makes for a reasonably power-efficient madVR-capable HTPC.

Power Consumption and Thermal Performance

The power consumption at the wall was measured with a 1080p display being driven through the HDMI port. In the graphs below, we compare the idle and load power of the Zotac ZBOX MAGNUS EN970 with other low power PCs evaluated before. For load power consumption, we ran Furmark 1.12.0 and Prime95 v27.9 together. Despite consuming close to 19W at idle, the ZBOX MAGNUS EN970 actually happens to be the PC with the lowest idle power amongst all the discrete GPU-equipped machines that we have evaluated so far.

The load power is also amongst the highest in the set of numbers that we have seen till now. However, the big separation between the idle and load powers indicate that the sytem can operate efficiently over a wide range of loading conditions.

Our thermal stress routine starts with the system at idle, followed by 30 minutes of pure CPU loading. This is followed by another 30 minutes of both CPU and GPU being loaded simultaneously. After this, the CPU load gets removed, allowing the GPU to be loaded alone for another 30 minutes. The various clocks in the system as well as the temperatures within the unit are presented below.

According to Intel's official specifications, the junction temperature of the Core i5-5200U is 105C. We find that pure CPU loading keeps the clock frequency half-way between the base frequency (2.2 GHz) and the maximum burst frequency (2.7 GHz). However, the temperature remains well below the junction temperature (around 82C). Getting the GPU into the equation ramps up the motherboard temperature as well as that of the GPU and GPU. However, the CPU remains below the junction temperature despite going up to as high as 102C. The GPU stabilizes around 81C.

Final Words

The Zotac ZBOX MAGNUS EN970 provided us with the opportunity to take a look at what mobile Maxwell coupled with a Broadwell-U CPU could bring to the gaming mini-PC market. Simply put, this is the most powerful mini-PC we have evaluated so far for gaming workloads.

There are certain things that prospective consumers need to keep in mind. The M.2 slot in the unit supports only PCIe x2 devices and not x4 ones such as the Samsung SM951. The discrete GPU drives up the idle power, meaning that one can't have the cake and eat it too. It would be nice to have something similar to what is done in the notebook space to be implemented here - i.e, being able to use the integrated GPU as well when not gaming, or when attempting to use Quick Sync. Finally, we would like to see a higher performance CPU in the system - there are other CPUs at this TDP worth considering.

One of the aspects for which Zotac can't be blamed directly is the advertising of the NVIDIA GTX 970M as GTX 960 when they are completely different for HTPC uses. NVIDIA's rebranding scheme makes very little sense, at least in this situation. The 'NVIDIA GeForce GTX 960' in the ZOTAC ZBOX MAGNUS EN970 is a MXM card, and as such, can't be purchased standalone. As an OEM exclusive, it is only available in pre-built systems like the MAGNUS EN970. Despite being placed in a SFF system, the GTX 970M can provide roughly the same performance as that of a bonafide desktop GTX 960 - just not for HTPC workloads.

Excellent thermal design and a robust cooling mechanism make the MAGNUS EN970 one of the quietest gaming mini-PCs that we evaluated. To provide some rough numbers, a HTC One M7 running the Android Sound Meter Pro app recorded just 50 dB when placed directly outside the outlet vent during the benchmarking of GRID Autosport. With the GB-BXi5G-760, we recorded 65 dB under similar circumstances. During gameplay, or even 5 - 6 ft. away, the noise profile of the unit is simply not a concern at all.

The barebones version of the ZBOX MAGNUS EN970 is available for $800 on Amazon. For $100 extra, the PLUS version adds a 120GB SATA SSD and an 8GB SO-DIMM. The pricing is quite reasonable when the costs of the competing products are considered. On the whole, the plus points of the ZOTAC ZBOX MAGNUS EN970 far outweigh the issues. There is still scope for improvement that we have pointed out in this review, but it still makes an interesting mini-PC purchase.