The ML100 series of passively cooled PCs from Logic Supply utilize the NUC / UCFF (ultra-compact form factor) motherboards. Currently, that lineup includes Bay Trail, Broadwell, and Skylake-based units. The ML100G-50 that we are looking at today is the Skylake vPro version.

Introduction and Product Impressions

The ML100G-30 released in early 2015 came with a Broadwell vPro processor and used one of the Intel Broadwell NUC motherboards (NUC5i5MYBE). Logic Supply's expertise in passively cooled systems enabled the replacement of the fan in the kit with an effective fanless thermal solution. The ML100G-50, however, is equipped with an ASRock board that is similar in many ways to the one found in the ASRock Beebox-S 6200U. The main differences between the internals of the ML100G-50 and the Beebox-S 6200U are the processor (vPro-capable Core i5-6300U vs. non-vPro Core i5-6200U), the Wi-Fi card (Intel AC7260 in our review sample vs. Intel AC3160), and the Ethernet controller (I219-LM vs. I219-V, the former being necessary for vPro features). The internal layout (not the motherboard itself) is also slightly different, with the full-metal solution of the ML100G-50 making it necessary to have a couple of RP-SMA connectors in the rear panel. In terms of physical dimensions, the ML100G-50 is the same as the ML100G-30. Due to the thermal solution and the need for RP-SMA Wi-Fi antenna connectors, the dimensions are slightly larger than the ASRock Beebox-S 6200U (142mm x 107mm x 62mm vs. 119mm x 110mm x 46mm)

The full specifications of our Logic Supply ML100G-50 review configuration are summarized in the table below.

Logic Supply ML100G-50 Specifications
Processor	Intel Core i5-6300U (2C/4T @ 2.4GHz (Turbo 3.0 GHz), 14nm, 3 MB L2, 15W)
Memory	Transcend TS1GSH64V1H DDR4 15-15-15-36 @ 2133 MHz 2x8 GB
Graphics	Intel HD Graphics 520
Disk Drive(s)	Transcend MTS600 TS128GMTS600 (128 GB; M.2 Type 2260 SATA III; Micron 20nm; MLC)
Networking	Intel Dual Band Wireless-AC 7260 (2x2 802.11ac - 867 Mbps) Intel Gigabit Ethernet Connection I219-LM
Audio	3.5mm Headphone Jack Capable of 5.1/7.1 digital output with HD audio bitstreaming (HDMI)
Display	2x HDMI (1x 2.0a, 1x 1.4b) 1x Display Port 1.2
Miscellaneous I/O Ports	1x USB 3.1 Gen 2 Type-C 3x USB 3.0
Operating System	Retail unit is barebones, but we installed Windows 10 Pro x64
Pricing	$871 (barebones) / $926 (default configuration) $1141 (as configured)
Full Specifications	Logic Supply ML100G-50 Specifications

Logic Supply allows the ML100G-50 to be customized prior to shipping. By default, the unit comes with 4GB of DDR4-2133 RAM and a 32GB M.2 2260 SATA SSD. There is no WLAN support in the default configuration. Based on the purchase options, Logic Supply can pre-install up to two SO-DIMMs of 8GB each and a Transcend M.2 2260 SSD up to 512GB in capacity. Customers can opt to install their own SO-DIMMs (in which case, they can go up to 32GB of RAM) and/or M.2 2260 SSDs (Note that only M.2 2260 SSDs are officially supported - unlike the Beebox-S which allows M.2 2280 SSDs using an add-on plastic riser tab). Both SATA and PCIe drives are supported, though M.2 2260 PCIe SSDs are relatively rare in the market.

Other customization options include the WLAN card (the Intel AC8260 is available on the Logic Supply purchase page, though our review unit came with the AC7260). wall / VESA and DIN rail mounting kits, port and dust blocking kits, and choice of OS (Ubuntu 14.04 LTS, Windows 8.1 Embedded, Windows 10 Home / Pro / IoT Enterprise LTSB). While the standard warranty is 2 years, $99 extends that to 3.

We installed Windows 10 Pro x64 for the review process, and almost all drivers were available via the regular Windows Update process. The CIR driver (for the IR receiver) and the thermal framework drivers (Intel DPTF) had to be downloaded from ASRock's website for the Beebox-S 6200U

In addition to the main unit, the other components of the package include a 65 W (19V @ 3.42A) adapter, a US power cord, two 2.4 GHz / 5 GHz antennae for the Wi-Fi feature, adhesive rubber feet for the unit's base and a cable management tie along with some screws (probably for the wall mount / VESA mount - which is supported, but not included in our review package). The gallery below takes us around the chassis and also a view of the customer-accessible portion of the internals that allows the DRAM and SSD to be changed.

Note that there is a thick thermal pad right above the M.2 SSD. This should definitely help the drive keep its cool when subject to disk-intensive workloads. The all-metal construction and the finned structure of the chassis sides and top also help in drawing the heat away from the internal components. As usual, we will quantitatively evaluate the effectiveness of the thermal solution in a later section.

In the table below, we have an overview of the various systems that we are comparing the Logic Supply ML100G-50 against. Note that they may not belong to the same market segment. Of particular interest are the Logic Supply ML100G-30, the Logic Supply Core ML320, the Zotac ZBOX CI523 nano and the Zotac ZBOX CI540 nano - all of which are passively cooled U-series UCFF PCs. 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 Logic Supply ML100G-50 when we come to those sections.

Comparative PC Configurations
Aspect	Logic Supply ML100G-50	Logic Supply ML100G-50ECS LIVA CoreLogic Supply ML100G-30Logic Supply Core-ML320Zotac ZBOX CI540 nanoZotac ZBOX CI523 nanoSYSmark 2014 SE Calibration SystemZotac ZBOX CI320 nano
CPU	Intel Core i5-6300U	Intel Core i5-6300U
GPU	Intel HD Graphics 520	Intel HD Graphics 520
RAM	Transcend TS1GSH64V1H DDR4 15-15-15-36 @ 2133 MHz 2x8 GB	Transcend TS1GSH64V1H DDR4 15-15-15-36 @ 2133 MHz 2x8 GB
Storage	Transcend MTS600 TS128GMTS600 (128 GB; M.2 Type 2260 SATA III; Micron 20nm; MLC)	Transcend MTS600 TS128GMTS600 (128 GB; M.2 Type 2260 SATA III; Micron 20nm; MLC)
Wi-Fi	Intel Dual Band Wireless-AC 7260 (2x2 802.11ac - 867 Mbps)	Intel Dual Band Wireless-AC 7260 (2x2 802.11ac - 867 Mbps)
Price (in USD, when built)	$871 (barebones) $1141 (as configured)	$871 (barebones) $1141 (as configured)

Performance Metrics - I

The Logic Supply ML100G-50 was evaluated using our standard test suite for low power desktops / 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. In the first section, we will be looking at SYSmark 2014 SE, as well as some of the Futuremark benchmarks.

BAPCo SYSmark 2014 SE

BAPCo's SYSmark 2014 SE is an application-based benchmark that uses real-world applications to replay usage patterns of business users in the areas of office productivity, media creation and data/financial analysis. In addition, it also addresses the responsiveness aspect which deals with user experience as related to application and file launches, multi-tasking etc. Scores are meant to be compared against a reference desktop (the SYSmark 2014 SE calibration system in the graphs below). While the SYSmark 2014 benchmark used a Haswell-based desktop configuration, the SYSmark 2014 SE makes the move to a Lenovo ThinkCenter M800 (Intel Core i3-6100, 4GB RAM and a 256GB SATA SSD). The calibration system scores 1000 in each of the scenarios. A score of, say, 2000, would imply that the system under test is twice as fast as the reference system. Since the Logic Supply ML100G-50 is one of the first passively cooled PCs to be subject to the SYSmark 2014 SE benchmark, we only have the reference system to compare against in this subsection.

SYSmark 2014 SE also adds energy measurement to the mix. A high score in the SYSmark benchmarks might be nice to have, but, potential customers also need to determine the balance between power consumption and the efficiency of the system. For example, in the average office scenario, it might not be worth purchasing a noisy and power-hungry PC just because it ends up with a 2000 score in the SYSmark 2014 SE benchmarks. In order to provide a balanced perspective, SYSmark 2014 SE also allows vendors and decision makers to track the energy consumption during each workload. In the graphs below, we find the total energy consumed by the PC under test for a single iteration of each SYSmark 2014 SE workload and how it compares against the calibration systems.

The U-series Core i5-6300U is not going to be competitive against a full-blown 65W TDP desktop processor, but, in terms of energy consumption, the ML100G-50 is a lot better compared to the reference desktop.

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 Core i5-6300U enables the ML100G-50 to come out on top in all of the Futuremark benchmarks when compared with fanless systems evaluated earlier using the same methodology. The closest competitor is the Zotac ZBOX CI523 nano based on a Core i3-6100U.

Miscellaneous Futuremark Benchmarks

The Intel HD Graphics 520 is also one of the most powerful GPUs we have seen in fanless computing systems in this form factor, as shown by the 3DMark benchmarks in the above graphs.

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.

In the single-threaded version, the ML100G-50 performs similar to the ML100G-30 despite the 100 MHz bump in the core clock. However, the multi-threaded version gives it a clear lead. In the OpenGL version, the Zotac ZBOX CI53 nano using the Core i3-6100U with the same graphics configuration (HD Graphics 520 operating at 300 MHz / 1 GHz) has a slight lead.

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 Core i5-6300U helps the ML100G-50 come out on top. It is followed by the Broadwell-based ML100G-30 and the Skylake i3-based Zotac ZBOX CI523 nano.

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. The pattern of results is similar to that seen in the x264 benchmark.

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 have hardware acceleration for the encryption and decryption processes. The Core i5-6300U in the Logic Supply ML100G-50 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 Logic Supply ML100G-50 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 ML100G-50 comfortably comes out on top in all the stages, as expected.

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 i5-6300U's specifications ensure that the ML100G-50 comes out on top again.

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 Logic Supply ML100G-50. 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 Transcend MTS600 SATA SSD used in the Logic Supply ML100G-50 is not a high-end one in terms of performance. It does use Micron's 20nm MLC flash (which should be good in terms of endurance), but the Silicon Motion SM2246EN controller and the capacity point (128GB) makes it come below the higher capacity SATA SSDs used in the Zotac ZBOX series reviews.

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 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 2x2 802.11ac solution used in the ML100G-50 is the same as the one used in the ML100G-30. The benchmark numbers are expected to be the same, but, we did see some slight differences that can be attributed to antenna orientation and other environmental factors. In any case, the 2x2 solution delivers a big jump in performance over the 1x1 solution used in other fanless PCs.

HTPC Credentials

The fanless nature of the ML100G-50. coupled with its compact form factor, ensures that it is a good HTPC candidate. However, Skylake doesn't support 4K Netflix, and the Intel HD Graphics 520 doesn't have full hardware decode acceleration for HEVC Main10. All of these are addressed in Kaby Lake. Given that industrial PCs are usually a generation behind the fanless consumer PCs, we would not recommend the ML100G-50 solely for HTPC duties. That said, it is interesting to look at certain aspects that are specific to the system - namely, the refresh rate accuracy and OTT streaming efficiency.

Refresh Rate Accuracy

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). As expected, the Logic Supply ML100G-50 has no trouble with refreshing the display appropriately in this setting.

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 1080p H.264 video. 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 51.0.1).

GPU load was around 16.58% for the YouTube HTML5 stream and 0.0205% for the steady state 6 Mbps Netflix streaming case.

Netflix streaming evaluation was done using the Windows 10 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.

The ML100G-50 is not particularly power efficient for OTT streaming workloads. Video decoding is now hardware accelerated on almost all systems (including Atom-based PCs), and systems such as the ECS LIVA Core (Core M with a 7W TDP CPU) and the ZBOX CI320 nano (Bay Trail-M 4.5W SDP CPU) are more power efficient compared to the ML100G-50 (which is equipped with a 15W TDP CPU).

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 Logic Supply ML100G-50 with other low power PCs evaluated before. For load power consumption, we ran the AIDA64 System Stability Test with various stress components, as well as our custom Prime95 / Furmark stress routine, and noted the maximum sustained power consumption at the wall.

The power consumption numbers are typical of a fanless system sporting a 15W TDP processor. The addition of extra board components such as the USB 3.1 Gen 2 bridge chip (and the overall migration from a standard Intel NUC board to the ASRock Beebox one) makes the ML100G-50 idle at higher power levels compared to the ML100G-30. On the full loading side, it is interesting to note that the ML100G-50 is able to sustain a higher power draw compared to the ML100G-30.

Our thermal stress routine starts with the system at idle, followed by four stages of different system loading profiles using the AIDA64 System Stability Test (each of 30 minutes duration). In the first stage, we stress the CPU, caches and RAM. In the second stage, we add the GPU to the above list. In the third stage, we stress the GPU standalone. In the final stage, we stress all the system components (including the disks). Beyond this, we leave the unit idle in order to determine how quickly the various temperatures in the system can come back to normal idling range. The various clocks, temperatures and power consumption numbers for the system during the above routine are presented in the graphs below.

The AIDA64 system stability test uses real-world workloads to stress the system components. However, power virus tests such as the Prime 95 torture test and Furmark stability test can subject the system to greater stress. We repeated our thermal stress routine with 30 minutes of Prime 95 (v28.10), followed by 30 minutes of Prime 95 and Furmark (1.18.2). The Prime 95 load was then removed, and the GPU stressing Furmark test was allowed too run for another 30 minutes. The various clocks, temperatures and power consumption numbers for the system during the above routine are presented in the graphs below.

The two thermal stress routines reveal that the Core i5-6300U starts off with a 20W allowance for the package power consumption. Based on the workload, the thermal design appears to be able to handle it for around 15 - 20 minutes before the power throttling kicks in. The throttling kicks in when the package temperature reaches around 80C. The AVX instructions in Prime 95 tend to stress one part of the die, causing a more rapid rise in temperature, while the AIDA system stability test is more generic. Despite consuming 20W in both cases, the temperature reaches 80C faster with Prime 95. This causes the system to shift down to the 15W level earlier with the custom stress workload.

Another important aspect to keep note of while evaluating fanless PCs is the chassis temperature. Using the Android version of the FLIR One thermal imager, we observed the chassis temperature after the CPU package temperature reached the steady state value in the above graph. Similar to the ML100G-30, the ML100G-50 also ends up with a chassis temperature of around 67C after being subject to our thermal stress routines.

We have additional thermal images in the gallery below.

The thermal design of the ML100G-50 is very similar to that of the ML100G-30. Therefore, it is no surprise that the thermal profile for both systems under stress are almost the same. All in all, Logic Suply's chassis design ensures that the ML100G-50 is able to operate the CPU package at higher than the rated power numbers for an extended duration before falling back to the expected 15W levels.

Miscellaneous Aspects and Concluding Remarks

The primary selling point of the ML100G-50 is its vPro-capable Core i5-6300U. The only other Skylake industrial fanless PC that we have seen in this form factor is the Aleutia R50, but, that sports a Core i5-6260U and is not vPro-capable. In fact, the last vPro-capable fanless PC that we had evaluated was the Logic Supply ML100G-30. The Intel BIOS in it was very user-friendly, and had plenty of easy to understand options for configuring vPro and AMT. Unfortunately, the BIOS in the ML100G-50 is not as fancy as that of the Intel NUC or the ASRock Beebox, as shown in the gallery below.

Once the proper BIOS options under the Advanced > AMT Configuration were set up, Ctrl-P could be used to get into the ME BIOS. With the ML100G-30, we were able to configure the ME BIOS and set up remote access for administration. However, our ML100G-50 came with a locked ME BIOS password that we were unable to change.

We didn't take the trouble to get into the ME BIOS further, but, Intel's ME BIOS extensions are likely no different from what we saw in the ML100G-30 review.

In summary, Logic Supply has been able to deliver effectively on the promise of a fanless Skylake industrial PC. The chassis has been designed to adapt to the requirements of industrial PCs. The move from an Intel board to ASRock's Beebox NUC board delivers the best value for money (in terms of features such as multiple native display outputs, a USB 3.1 Gen 2 Type-C port etc.). The online ordering page provides plenty of options for customization.

Despite the excellent acoustics that come with being completely fanless and reasonable power consumption numbers, we do not recommend the unit as a passive HTPC. Simply put, the platform commands a premium for its vPro capabilities, and something like the Zotac ZBOX CI523 nano can deliver better value for money as a HTPC platform with the same capabilities.

In terms of scope for improvement, we have a spare SATA port on the board. The ability of the chassis to accommodate a 2.5" drive would be welcome, given that we have already sacrificed on the chassis height. We would also like Logic Supply to explore tweaks to the thermal design that can allow the CPU package to operate at the 20W package power level for a longer duration. Option to configure the unit with an integrated serial I/O port (something that is present in the ML100G-30) would also be welcome. Other than these minor quibbles, Logic Supply's ML100G-50 is a solid step up from the Broadwell-based ML100G-30.