08/13/2010 | 02:59 PM
In the spring of this year we completed our series of reviews of mainboards based on the Intel H55 and H57 Express chipsets. We might have expected the lack of excitement on our readers’ part because such mainboards are somewhat inferior to their counterparts based on the more popular Intel P55 Express. Moreover, if you don’t use a CPU-integrated graphics core, the video interfaces at the back panel of such an H55/H57-based mainboard are useless and take over the space that could be occupied by more demanded connectors. That said, the review of the tiny Zotac H55-ITX WiFi turned out extremely popular. The mini-ITX mainboard, even though not without some downsides, turned to be just as functional as its larger counterparts and even superior to them to some extent. So, as this topic seems to be interesting to our readers, we want to carry on with our reviews of mini-ITX products. Today, it will be the 17x17-centimeter Intel DH57JG mainboard based on the Intel H57 Express chipset.
If you’ve ever seen a mini-ITX mainboard, you won’t find anything surprising about the next mini-ITX product you come to know. The total available space of 289 cm2 provides but a limited place for experiments, so mini-ITX mainboards are all basically the same. However, Intel mainboards are always different from others in one way or another, and we will see that as we compare the DH57JG with Zotac H55-ITX WiFi that we reviewed earlier.
Like the Zotac board, Intel DH57JG is equipped with one graphics slot that works in full-speed PCI Express 2.0 x16 mode but does not have an additional mini PCI Express slot. It also has two DDR3 slots and thus supports no more than 8 gigabytes of system memory. The distance between the CPU socket and memory slots is some 5 millimeters longer than on the Zotac product, which makes it easier to install a CPU cooler (and the cooler itself may be somewhat larger). The memory slots are placed very close to each other, so you should take modules with slim heat-spreaders or without any heat-spreaders at all.
Then, as the memory slots are located along the right edge of the PCB, the 24-pin power connector has moved upwards. There should be no problems connecting it to the power supply, as opposed to the front-panel indicators and buttons. The appropriate pins are in the top left corner of the PCB rather than in the bottom right, as usual. They occupy the place of PS/2 connectors the Intel mainboard lacks. Some system cases may have too short cables that won’t be able to reach that far.
The mainboard uses a dual-phase CPU voltage regulator and has an ATX12V connector for additional power supply. Five out of the six Serial ATA ports supported by the chipset are implemented. You can connect four internal SATA drives to this mainboard and there is also an eSATA connector at its back panel. 12 out of the chipset’s 14 USB ports are available: six at the back panel and six more can be connected to the three onboard pin-connectors. Besides, the mainboard’s back panel features a LAN connector (the network interface is based on Intel 82578DC Gigabit Ethernet controller), HDMI and DVI connectors (these work only if you’ve got a Clarkdale processor with integrated graphics core), five analog audio connectors and an optical S/PDIF (the audio interfaces are provided by an eight-channel Realtek ALC889codec).
The mainboard also has two fan connectors, a chassis integrity sensor, and a piezoelectric speaker that cannot be turned off. One legacy interface is supported: a COM port.
The capabilities and features of this mainboard are summed up in the following table:
So, Intel DH57JG is functional enough, yet far inferior to Zotac H55-ITX WiFi in this respect. Notwithstanding the same form-factor and similar chipsets, these mainboards represent two different engineering approaches. The Zotac board can impress anybody with its specs, being far superior even to many full-size mainboards out there. Despite its small form-factor, it not only makes full use of the chipset’s features but complements them with a number of extra controllers. As opposed to that, the Intel mainboard has a minimum number of additional controllers and doesn’t even use all the capabilities of the chipset. We must acknowledge that this utilitarian and pragmatic approach is acceptable, too. Intel didn’t pack lots of controllers and interfaces into this product since most mini-ITX users won’t ever need them. The DH57JG is not meant for quad-core CPUs, so two phases in the CPU voltage regulator should suffice just fine. The four SATA (+ one eSATA) and twelve USB ports are also more than enough for such a small mainboard. We can’t really find any fault with this engineering approach which is rational and thus justifiable.
The lack of USB 3.0 and FireWire might be viewed as a downside, but these interfaces (and a mini PCI Express slot, too) are not really necessary. And even the obvious drawback – the improper location of the pins to connect the buttons and indicators of the system case – is not so serious. This drawback will only become a real problem if you install this mainboard into an ATX or microATX system case. There should be no problems if you take a small specialized system case instead.
Thus, for all its drawbacks, Intel DH57JG does not have any really serious defects. The difference is that the Zotac board is an engineering masterpiece whereas the Intel one is a practical product with no pretensions. Everyone is impressed with elite flagship mainboards, yet most users prefer to buy ordinary ones, like Intel DH57JG.
All performance tests were run on the following test platform:
We used Microsoft Windows 7 Ultimate 64 bit (Microsoft Windows, Version 6.1, Build 7600) operating system, Intel Chipset Software Installation Utility version 220.127.116.115, ATI Catalyst 10.2 graphics card driver.
We didn’t have any problems assembling our testbed. Intel DH57JG worked normally in its nominal mode except for temperature. The problem was that the CPU cooler fan rotated at 300 RPM and was not willing to speed up. As a result, the CPU was always at least 70°C hot and even higher under load. Although this fan management setup is ineffective, it is deliberate. Everyone knows that boxed Intel coolers are noisy at high speeds. In order to keep the noise within reasonable limits, the speed of the CPU cooler fan is restricted by the mainboard to low levels despite the increased temperature. The mainboard’s BIOS offers but primitive fan control options. You can just specify that you use a more or a less effective cooler compared to the standard boxed one. Our Scythe Samurai ZZ is much better than the boxed cooler, but we selected the more aggressive control method as if it were weaker. However, this didn’t bring any benefits to us. The fan speed would increase but for a short period of time, producing a paradoxical situation when the CPU was cooler under load than in idle mode. And then the fan would slow down again to 300-600 RPM, the CPU temperature rising back to 70°C and higher.
We also found a BIOS option that allows controlling the speed of the CPU fan through appropriate software. We literally found it since it had been hidden in the “Maintenance” section which is not visible when you enter the BIOS normally. You need to reset a special jumper to see it. However, we could not change the speed of the fan with the popular SpeedFan utility even after we enabled that BIOS option. The exclusive Intel Desktop Utilities software didn’t offer any fan control settings at all. Here is an illustration: the CPU is as hot as 75°C but the fan speed is still about 350 RPM.
By the way, this warning was issued because we had changed the temperature thresholds for it. By default, the software will only warn you if the CPU temperature rises above 97°C.
There is no pleasure in feeling waves of heat from a computer with an Intel LGA1366 or a six-core AMD processor inside, but you can put up with that because the increased heat dissipation is due to high performance. It is much harder to condone that a mainboard with a rather economical Intel Core i3-540 scorches your fingers. Intel DH57JG is “optimized” for inefficient and noisy boxed coolers from Intel although its user might install a better cooler. As a result, this mainboard is going to keep your CPU in rather harsh thermal conditions.
Oddly enough, the small Intel DH57JG is superior to its larger cousin, DH55TC, in terms of setup and overclocking functionality. Although these options are not as rich as on overclocker-friendly mainboards (for example, you cannot change the CPU voltage or adjust memory timings other than the main ones), they are quite sufficient for some overclocking. The following table lists the key features this mainboard offers:
Using these options, we managed to overclock our CPU to 3.4 GHz, which is a mere 100 MHz lower than what we had with other mainboards without increasing the CPU voltage. This difference can be explained by the lack of an option that would counteract a CPU voltage drop under load as well as by the harsher temperature conditions of the CPU on the Intel mainboard.
We also want to note something we encountered during our tests. The memory modules we used (OCZ DDR3 PC3-12800 Blade Series Low Voltage OCZ3B1600LV2GK) have a rated frequency of 1600 MHz with 6-6-6-24 timings and 1.65 V voltage. We could increase the voltage and change the timings but the mainboard would not start up at a CAS Latency value of 6. This is a drawback and we had to use higher timings which had been selected by the mainboard by default. On the other hand, we must acknowledge that the mainboard delivered somewhat higher performance than its opponents, and this seems to have been due to some memory optimizations. Perhaps this is why the timings cannot be set too low, yet the mainboard works fast. You will see exactly how fast it is in the next section of the review.
As usual, we are going to compare the mainboards speeds in two different modes: in nominal mode and during CPU and memory overclocking. The first mode is interesting because it shows how well the mainboards work with their default settings. It is a known fact that most users do not fine-tune their systems, they simple choose the optimal BIOS settings and do nothing else. That is why we also run a round of tests without interfering in any way with the default mainboard settings. We deliberately ran the tests in the same conditions as our Intel H55/H57 Express based mainboards reviewed previously, that s why we could use those results for comparison purposes today. The mainboards are listed on the diagrams according to their performance (from high to low). The results of Intel DH57JG mainboard are marked with a darker color for your convenience.
We started using the recently released Cinebench 11.5 program version. All tests were run five times and the average result of the five runs was taken for the performance charts.
We have been using Fritz Chess Benchmark utility for a long time already and it proved very illustrative. It generated repeated results, the performance in it is scales perfectly depending on the number of involved computational threads.
A small video in x264 HD Benchmark 3.0 is encoded in two passes and then the entire process is repeated four times. The average results of the second pass are displayed on the following diagram:
In the archiving test a 1 GB file is compressed using LZMA2 algorithms, while other compression settings remain at defaults.
Like in the data compression test, the faster 16 million of Pi digits are calculated, the better. This is the only benchmark where the number of processor cores doesn’t really matter, because it creates single-threaded load.
There are good and bad things about complex performance tests. However, 3D Mark Vantage has become extremely popular. The diagram below shows the results after three test runs:
Since we do not overclock graphics in our mainboard reviews, the next diagram shows only CPU test from the 3D Mark Vantage suite.
We use FC2 Benchmark Tool to go over Ranch Small map ten times in 1280x1024 resolution with medium and high image quality settings in DirectX 10.
Resident Evil 5 game also has a built-in performance test. Its peculiarity is that it can really take advantage of multi-core processor architecture. The tests were run in DirectX 10 in 1280x1024 resolution with medium image quality settings. The average of five test runs was taken for further analysis:
As you can see, Intel DH57JG takes a leading position in each of our tests. In some of them it is only inferior to the Biostar board which works at higher frequencies than most other mainboards. Thus, Intel DH57JG would be the fastest if we had the same test conditions for each mainboard. Moreover, Intel DH57JG won the 7-Zip compression test despite the Biostar’s advantageous settings. This test is influenced not only by the CPU frequency but also by the memory subsystem parameters. The CPU frequency was the same on each mainboard, save for the Biostar, and all of them fell behind the Biostar with the exception of Intel DH57JG. This indicates that the latter has some memory optimizations. Thus, the Intel mainboard has a small but clear advantage in performance, yet it may turn to be a downside: we could not set low memory timings on this mainboard and increase its performance during overclocking.
We also performed the same tests during processor and memory overclocking. Almost on all tested mainboards the CPU was overclocked to 4.5 GHz and the memory worked at 1564-1570 MHz frequency with 6-6-6-18-1T timings. The only exception were the EVGA H55 and Zotac H55-ITX WiFi mainboards, where we could only push the CPU to 3.9 GHz, while the memory worked at 1360 MHz with 6-6-6-16-1T timings, and our today’s hero that stopped at 3.4 GHz CPU frequency. I would like to remind you that Intel DH55TC mainboard wasn’t included into this test run, because it doesn’t overclock processors at all.
Here, the mainboards fall into three categories depending on the CPU frequency achieved. The mainboards with low CPU frequency cannot match the others even in the gaming tests which focus on the graphics subsystem performance. Anyway, overclockability is not a decisive factor for a small mini-ITX mainboard. It is just a nice bonus. Therefore, we are not going to blame Intel DH57JG for its lagging behind in this test.
We measured power consumption using Extech Power Analyzer 380803 device. This device was connected before the system PSU, i.e. it measured the power consumption of the entire system without the monitor, including the power losses that occur in the PSU itself. When we took the power readings in idle mode, the system was completely idle: there were even no requests sent to the hard drive at that time. We used LinX program to load the CPU. For more illustrative picture we created graphs showing the power consumption growth depending on the increase in CPU utilization as the number of active computational threads in LinX changed in nominal mode as well as during overclocking. The boards are sorted out in alphabetical order on the diagrams below.
The full-size ASUS P7H57D-V EVO consumes much more power than the rest of the mainboards, but it is a unique, maximum-functionality LGA1156 mainboard and this might be expected from it. ASUS P7H55D-M EVO and EVGA H55 are not economical when idle both in nominal and overclocked modes (the EVGA board seems to consume little power when overclocked only because it is not overclocked as much as the other mainboards). The rest of the mainboards are close to each other in terms of power consumption. We can only note that Intel DH55TC and MSI H57M-ED65 are somewhat more economical than their opponents in the nominal mode.
Zotac H55-ITX WiFi that we have compared Intel DH55JG with throughout this review is very energy-efficient when idle and under low load, both in its nominal and overclocked modes. However, it starts to consume a lot of power at high CPU loads because it sets the CPU voltage too high. We have not spotted Intel DH57JG do so, so this mainboard is energy-efficient both in the nominal and overclocked mode (but it has the lowest overclocking results because we could not increase the CPU voltage on it).
Intel DH57JG is neither a sensation nor a disaster. Its design is not flawless and its BIOS options are rather scanty, even though this can partially be explained by its small form-factor. Its specs are not gorgeous, either. Particularly, it doesn’t support USB 3.0 and FireWire. It couldn’t overclock our CPU to its maximum and did not allow setting low memory timings. Besides, its fan control system keeps the CPU under harsh thermal conditions. Combined with its rather high price, this makes us wonder who is going to buy it at all.
Well, Intel DH57JG does have an advantage, namely Intel H57 Express chipset. By the way, we were at first perplexed as to why it was not based on the H55 Express which would have been quite sufficient. The H57 Express supports more USB ports, but the mainboard doesn’t use them. It supports more PCI Express lanes, but the mainboard needs only 16 lanes for its single graphics slot. The single notable difference is that the H57 supports RAID arrays but does a small mini-ITX mainboard need this feature? Well, it does! For example, you can build a compact and high-performance file server!
There are many manufacturers that produce ready-made Network Attached Storage devices that accommodate from one to four hard disk drives and run an integrated, usually Linux-based, OS. Such NASes provide reliable file storage for a home or small office but all of them share common problems such as a low processor frequency and a small amount of system memory, which limit their performance. As a result, such a NAS may not provide a data-transfer rate higher than 100 Mbps even when connected via Gigabit Ethernet. As opposed to them, Intel HD57JG allows building a compact server which will be as fast as big desktop computers. But are there any mini-ITX enclosures that would accommodate four HDDs if even microATX system cases not always support that many hard drives? Well, it turns out that such PC enclosures do exist. For example, there is the cute Chenbro ES34169 that supports four HDDs with hot swap feature.
Or there is the Lian Li PC-Q08 series which can accommodate as many as six 3.5-inch HDDs and comes in several colors so that you could choose what suits your home interior best.
So, if you want a small but high-performance file server, Intel DH57JG is going to be a perfect choice for it because the rest of mini-ITX mainboards we know of are based on Intel H55 Express chipset which does not support RAID and cannot provide fault-tolerant data storage. But if you just want a small mini-ITX mainboard for other purposes, there are a lot of more interesting alternatives on the market.