Intel chipset with integrated graphics g31. Intel begins shipping budget P31 and G31 chipsets. Intel G31 Express Chipset

This review is a kind of continuation of the article published last year “Choosing a chipset for Core 2 Duo. The era before Intel 3 Series" and one of the last parts of the material dedicated to chipsets for the Intel LGA775 platform. This time we will consider system logic sets of the Intel 3 Series family, the products of which, although not in large quantities, are still present on the market.

The company released the Intel 3 Series line of chipsets in 2007 to support the expected 45-nm processors of the Penryn family with a 1333-MHz bus, which could not function on the then popular motherboards based on the Intel x965 and 945x system logic sets. In total, there were eight models in the line: the entry-level mass P31, P35 for mid-level systems, integrated G31, G33 and G35, the flagship X38, and solutions for the corporate market - Q35 and Q33. Of course, the new products were not limited to supporting future processors - with their appearance, the era of DDR3 memory and the PCI Express 2.0 bus began (the latter is true only for the high-end X38 chipset). In addition, the south bridge was updated, which became known as ICH9: the number of USB 2.0 ports increased to twelve, and support for Intel Turbo Memory technology appeared, which is based on the installation of an additional module with flash memory on the board, which allows for faster loading of applications in the operating system Windows Vista system.

Intel P31/G31 Express

The P31 and G31 chipsets replaced the Intel 945x/946x series with the only difference being support for the FSB bus with a frequency of up to 1333 MHz and DDR2-800 memory with a maximum capacity of 4 GB. The integrated solution has an updated GMA 3100 graphics core instead of the GMA 950 used previously. The rest of the functionality is limited to the south bridges ICH7 or ICH7R, which support only eight USB 2.0 ports, four PCI-E 1.1 lanes (six for ICH7R), four SATA II channels (the ability to organize RAID arrays of levels 0, 1, 5 and 10 for ICH7R) and Fast Ethernet.

Gigabyte GA-P31-ES3G

As a representative of a product based on the entry-level Intel P31 Express chipset, we used the Gigabyte GA-P31-ES3G board, which is now one of the most affordable solutions for Penryn family processors.

The motherboard comes in a green and black box that highlights the board's features, including Easy Energy Saver technology.

• instructions for installing AMD/Intel processors;
• disk with drivers and software;
• two SATA cables;
• one IDE cable;
• one FDD cable;
• rear I/O strip;
• sticker on the system unit.

The processor power subsystem is built according to a 3-phase circuit using solid-state capacitors and chokes in armored cores. The remaining circuits have conventional capacitances and simpler chokes. ATX12V standard processor power connector.

The functionality of the GA-P31-ES3G is minimal, due to the limited capabilities of the ICH7 south bridge. The board has four SATA II channels, eight USB 2.0 ports, four of which are located on the rear panel, one IDE channel, three PCI-E x1 connectors and three regular PCI. There are also COM and LPT ports.

BIOS

Additional software

The Gigabyte GA-P31-ES3G motherboard allows you to install the Easy Energy Saver utility, which has common features with Dynamic Energy Saver Advanced, but unlike the latter, it does not control the dynamic shutdown of processor power phases when the system is idle, but simply calculates the saved electricity when the voltage is reduced power supply and activation of C1E and EIST technologies for processors of the Core 2 family.

Overclocking

To determine the overclocking potential of the board, the following configuration was assembled:

• Video card: ASUS EN8800GS TOP 384M;
• Power supply: Silver Power SP-S850 (850 W).

As a rule, when overclocking, the PCI Express bus frequency must be locked at 100-110 MHz, but with these settings the system was only able to boot at the FSB frequency of 350 MHz. If PCI-E is left in Auto mode, then the ceiling is practically limited by the capabilities of the chipset itself.

The Intel P35/G33 Express system logic sets, which replaced the Intel P965/G965, were the first that Intel presented when announcing a new line of chipsets at Computex 2007. The rest had the status of a “paper” announcement and appeared a little later. For the first time, new products supported DDR3-1066/800 memory with a maximum capacity of 8 GB, in addition to DDR2-800/667, and motherboard manufacturers could produce products designed for any of these types, or even combined solutions. The ICH9 south bridge expanded the functionality of the system with twelve USB 2.0 ports; its modifications ICH9R and ICH9DO supported six SATA II channels with the ability to organize RAID arrays of levels 0, 1, 5 and 10. In addition, the new south bridges added eSATA support and the ability disable unused high-speed serial interface channels. The built-in graphics core of the G33 is similar to that of the G31, but with one difference, which is support for Clear Video video enhancement technology. The CrossFire operating mode for the P35 was stated to be “x16 + x4”, which, as in the case of the P965, limited the performance of the tandem. Intel Turbo Memory technology never became widespread, although high hopes were placed on it.

Intel Q35/Q33 Express

For the corporate market, Q35 and Q33 chipsets were released, which contained a GMA 3100 video core, supported DDR2-800/667 memory and were equipped with ICH9, ICH9R and ICH9DO south bridges with the ability to implement Intel Turbo Memory (Q35 only). These solutions differed in their support for Intel Trusted Execution and Intel Virtualization for Directed I/O (Q35) technologies.

Representatives of the business segment did not come to our testing, so we will consider a product based on the once mass-produced mid-level Intel P35 Express chipset. The ASUS P5K Deluxe/WiFi-AP motherboard is packaged in a black box with a handle and a hinged lid, which lists all the product features and supported technologies.

• instructions for the motherboard;
• instructions for the WiFi access point;
• disk with drivers and software;
• four SATA cables;
• one IDE cable;
• one FDD cable;
• rear I/O strip;
• WiFi antenna;
• set of branded Q-Connector pads.

The functionality of the board is good even by modern standards: three PCI slots, two PCI-E x1, one PCI-E x4 (PCI-E x16 slot) and one PCI-E x16, as well as ten USB 2.0 ports, six of which are on the rear panel, and six SATA II channels with the ability to organize RAID arrays 0, 1, 5 and 10 due to the ICH9R south bridge.

Two gigabit controllers are responsible for the network on the board, Realtek RTL8110SC and Marvell 88E8056, the first of which is located on the PCI bus, and the second on the PCI-E bus.

As you have already noticed, two USB ports are not available to the user - they are used by the WiFi module near the rear panel of the board.

BIOS

The board's BIOS, based on AMI microcode, allows users to fine-tune the system and, compared to modern ASUS boards, is less overloaded with various parameters.

Additional software

The board comes with the ASUS AI Suite program, which is responsible for monitoring, system overclocking, fan control and energy saving functions.

Overclocking

The board was overclocked with the same configuration as with the Gigabyte GA-P31-ES3G. The BIOS version was 0902, dated June 19, 2008 - more recent firmware is in beta status, so it was not used. With a voltage on the north bridge of 1.4 V and a FSB bus of 1.4 V, the maximum stable frequency was 550 MHz, and with an increase in NB Voltage to 1.55 V, the system was able to pass the stress test at a frequency of 566 MHz. For greater stability, the board was blown by a 120 mm fan.

After the announcement of the new line of chipsets, the “old man” i975X temporarily passed the baton of the company’s flagship to the Intel P35 Express system logic set, despite the limitations in the construction of CrossFire systems. Some manufacturers even produced P35-based boards with switch chips for the PCI Express bus, which allowed video cards to “communicate” with each other and the motherboard via eight PCI-E 1.1 lanes on each side. Of course, this affected the final cost of the finished solution, but high-level products have never had an affordable price.

In the fall of 2007, the high-end Intel X38 Express chipset was already introduced, which differed from the P35 by supporting 32 PCI Express 2.0 lanes, which had twice the bandwidth of PCI-E 1.1, and allowed two video cards to operate in “x16 + x16” mode . In addition, it became possible to work with DDR3-1333 memory, and support for XMP (eXtreme Memory Profiles) technology was introduced, which is an analogue of EPP (Enhanced Performance Profiles) from NVIDIA, but for DDR3 modules. Another interesting feature of the chipset was the presence of a heat distribution cover on the north bridge, which made it possible to evenly dissipate heat and avoid chipping the chip crystal.

Since products based on Intel X38 Express are already a certain rarity at the moment, the ASUS Rampage Formula board was chosen as a representative of solutions based on the top X48 chipset. Over the year, the cost of high-end motherboards has decreased by one and a half times and now you can buy a good product for $150-200, but given the imminent release of the P55 chipset and processors for Socket LGA1156, the feasibility of such a purchase is questionable. Although, it is not known how much the newest boards will cost (they don’t promise much) and it is not a fact that it will be possible to immediately switch to the next platform. So if you have a quad-core processor, but there is no desire to change it, and boards based on Intel P45 are not suitable for one reason or another, then a solution based on X48 will be the only candidate for purchase.

The board in question belongs to the Republic of Gamers series and comes in a proprietary box with a hinged lid and a carrying handle.

• instructions for the motherboard;
• disk with drivers and software
• disc with the game S.T.A.L.K.E.R.: Shadow of Chernobyl;
• six SATA cables;
• external sound card SupremeFX II;
• power adapter for SATA devices;
• one IDE cable;
• one FDD cable;
• briquette with two USB connectors and one mini-FireWire;
• rear I/O strip;
• turbine for radiator on power elements;
• set of branded Q-Connector pads
• set of couplers;
• remote LCD indicator LCD-Poster;
• sticker on the system unit with a logo.

The full-size ATX board is made on black PCB, like all expensive solutions from ASUS. The arrangement of the elements is more or less thought out, there is practically nothing to complain about: two PCI-E x16 slots are separated by a sufficient distance, memory can be replaced without removing the video card, the SATA and IDE connectors are rotated 90° and after installing the accelerators, each channel will be available for connection drives. Regarding memory - as noted above, the X48 chipset can work with the DDR2 standard without problems, and Rampage Formula is no exception - the board supports modules with frequencies above 1200 MHz, but finding such ones is now problematic. The maximum memory capacity can reach 8 GB and in this regard there are also no differences from X38-based solutions. But for new products from the real fourth series, the bar has already been pushed back to 16 gigabytes, although eight are now rarely used. Of the processors, you can install all modern models with a bus frequency from 800 to 1600 MHz.

The functionality of the product is at a high level, and the user can install two PCI expansion cards, two PCI-E x1, another slot (black) reserved for the SupremeFX II sound card, two video cards with PCI Express 2.0 interface, combining them in CrossFire mode.

The buttons are convenient to use on an open bench or with one video card in the case, but Clear CMOS may not even be needed, because during the entire testing period I did not have to press it, except for checking the function itself. The board is also equipped with three connectors for connecting external thermal sensors and a number of LEDs near the processor, memory, north and south bridges, which are responsible for the level of voltage supplied to the listed components.

BIOS

Additional software

In terms of software, the board is equipped with a standard set of utilities: for flashing the BIOS and changing the screensaver at system startup, the PC Probe II monitoring program and a more advanced tool for configuring Ai Suite, which we partially discussed above. This program allows you to configure the FSB frequency, processor multiplier, memory operating mode and its supply voltage, PCI-E frequency. It is also possible to configure the fan operating mode.

Turbo

Crazy

Overclocking

Well, now the real acceleration. The configuration is the same, without any changes. The board was additionally blown by a 120 mm fan. BIOS version is 0902.

Board specifications

Test configuration

Testing was carried out on the following configuration:

• Processor: Intel Core 2 Duo E8500 (3.16 GHz);
• RAM: G.Skill F2-8800CL5D-4GBPI (2x2048 MB, DDR2-1100, 5-5-5-15-2T, dual channel);
• Video card: XFX GF GTX295 576M 1792MB DDR3;
• Hard drive: Samsung HD252HJ (250 GB, SATA2);
• Power supply: Seasonic SS-600HM (600 W);

• Operating system: Microsoft Windows Vista Ultimate x86 SP1;
• Driver for motherboards: Intel Chipset Software Installation Utility 9.0.0.1008
• Driver for video card: NVIDIA GeForce 182.50

Test results

Memory subsystem

Conclusions

The first thing I would like to say to users who have motherboards based on Intel 3 Series system logic sets is that if there are no special reasons for switching to new products, then you can wait with the upgrade. All that you will lose with older solutions is about 5% in modern games when using the highest quality picture, a Radeon HD 4890-level card and more than 4 GHz Core 2 Duo, as you can see in one of our upcoming materials. But when building a system from scratch or when moving from a platform of venerable age, naturally, there is no point in buying motherboards on already outdated chipsets, and in this case it is better to immediately turn your attention to newer solutions. Or even wait a little and immediately switch to the LGA1156 platform, since the prices for finished products are promised to be not so high, in contrast to the beginning of expansion into the Nehalem microarchitecture market.

As for the reviewed products, the GA-P31-ES3G board based on the entry-level Intel P31 Express system logic set demonstrated good performance with fewer settings, and in some tests it was even able to show better results than more expensive solutions. The scope of delivery is minimal, but you can hardly expect anything more from a board that costs less than $60 - this is an ordinary workhorse that is installed into a system and is often forgotten about what it is.

ASUS P5K Deluxe/WiFi-AP is fully equipped and is one of the best representatives of motherboard models based on the P35 chipset. Two network cards, a WiFi module, excellent overclocking potential - perhaps the owners of these boards should only consider a replacement when switching to another platform.

The former flagship Intel chipset X48, also known as X38 without support for the 1600 MHz FSB bus and DDR3-1600 memory, will for a long time be the basis of motherboards for high-performance systems based on quad-core processors and a combination of a pair of Radeon HD video cards. The ASUS Rampage Formula reviewed is an example of what an overclocking product should be. Maximum settings for overclocking, concentrated in one section, thoughtful design and cooling system, additional functionality in the form of power and system reboot buttons, as well as for resetting BIOS Setup and external LCD-Postera settings - take it and overclock. The relationship of this motherboard to the gaming series is a little unclear - players have no use for all this. But it’s just right for overclockers, if the price suits them, of course. But top products are only rising in price these days...

We thank the following companies for providing test equipment:

• 1-Incom for G.Skill F2-8800CL5D-4GBPI memory kit;
• Quasar Micro for Gigabyte GA-P31-ES3G motherboard;
• Master Group for ASUS P5K Deluxe/WiFi-AP and ASUS Rampage Formula motherboards, as well as for the ASUS EN8800GS TOP 384MB video card;
• Max Point for Silver Power SP-S850 power supply;
  
• Noctua for the Noctua NH-U12P cooler and Noctua NT-H1 thermal paste;
• Syntex for the Seasonic SS-600HM power supply;
• XFX for the XFX GF GTX295 576M 1792MB DDR3 video card.

The diversity of Intel's chipset portfolio is difficult to understand, as some features overlap from model to model. Choosing a higher model number will not necessarily give you a more advanced chipset. So, for example, the G35 offers more advanced graphics, but doesn't support DDR3 memory or work with the newer ICH9 southbridge like the G33 does. However, the Intel G31 chipset is actually an entry-level chipset and consumes minimal power.

The G31 falls into the category of mass-market desktop chipsets designed for "core computing." This means that this set of system logic is completely unsuitable for high-end systems and does not support any advanced functions. The G31 chipset was designed to be a cost-effective option for the average user. Therefore, it is aimed at processors such as Core 2, Pentium Dual Core or similar Celeron processors based on the Core 2 microarchitecture.

The G31 chipset is limited to 4GB of memory, while the G33 and G35 support up to 8GB. The low-end chipset only supports dual-channel DDR2-800 memory (although this is not a disadvantage compared to DDR3) and works with an ICH7 southbridge instead of an ICH8, ICH9 or ICH10. As a result, the G31 supports only four SATA/300 ports, but provides two more UltraATA/100 channels, while newer chipsets support either one legacy ATA channel or none at all. The G31 with the ICH7 southbridge provides eight USB 2.0 ports, HD audio, traditional PCI slots, and just a 100Mbps network controller. If you need a faster Ethernet connection, then look for a motherboard that comes with a PCIe network controller to provide Gigabit Ethernet. Both motherboards presented in our review are exactly like this. And finally, although the G31 chipset has one x16 PCI Express slot for upgrade, it is not compatible with PCI Express 2.0.

Although FSB1066 was initially specified, all current G31 motherboards we've reviewed also support FSB1333 processors. Both of our motherboards support all Core 2 processors, including Core 2 Quad processors. Incredibly, even the three-phase voltage regulator on the Foxconn G31 motherboard is quite enough to run the Core 2 Quad Q9550, if you decide to abandon the dual-core processor in favor of a quad-core one.

GMA3100 graphics core

Integrated graphics solutions will not provide the same performance and features as decent 3D graphics cards such as Radeon HD4850/4870 and Nvidia GeForce 260/280. Even "massive" video cards on GPUs GeForce 9600GT vastly superior to integrated solutions. However, we decided to stick with integrated graphics to keep power consumption as low as possible.

The G31 chipset contains an integrated Intel GMA3100 graphics core, based on the GMA3000 core. This is the same graphics core that Intel first used in the 965 line of chipsets (called GMA X3000), and although the G965 supports pixel shaders 3.0, the G31, G33, Q35 chipsets are limited to the SM 2.0 model, which means support for DirectX 9.0c. However, this is sufficient for all the functions of the Aero interface in the Windows Vista operating system. Motherboards with the G31 chipset typically have one analogue D-SUB15 display output, and sometimes a digital DVI output. Since the GMA3100 is not suitable for HTPC (Home Theater), motherboards are not equipped with HDMI outputs; You shouldn't expect two digital outputs from such boards either.

CPU Intel Core 2 Duo came out in late summer 2006 and created a sensation. Surprisingly, AMD's Athlon 64 X2 processors were surpassed by a product that provided better performance while consuming less power. Launched with a clock speed of 2.66 GHz (2.93 GHz for the expensive Extreme version), the Core 2 Duo maintained its excellent status for more than two years.

The first update was to increase the processor bus frequency from FSB1066 to FSB1333 in the summer of 2007 with the release of the P35 platform and slightly modified processors. The second modification is the FSB1600 frequency in high-end models and a change in the technological process from 65 nm to 45 nm. All Core 2 Duo E7000, E8000, and Core 2 Quad Q8000 and Q9000 processors are based on 45nm Wolfdale cores. Quad-core Yorkfield processors use two Wolfdale crystals.

Wolfdale processors have improved over time; The most recent modification was the introduction of M0 stepping, which reduced the processor's idle power consumption compared to L steppings. We looked for a suitable processor that would use the latest stepping and found a Core 2 Duo E7200 with only 3 MB L2 cache instead of 6 MB . Of course, the reduced capacity of the L2 cache is another opportunity to reduce power consumption, which we did not fail to take advantage of.

Currently, the E7000 line consists of only two models: the E7300 with a clock frequency of 2.66 GHz and the E7200 with a frequency of 2.53 GHz. Both models are based on the Wolfdale core, but have only 3 MB of L2 cache and a reduced frequency of the FSB1066 system bus (for comparison: all Core 2 Duo E8000 processors have FSB1333 and 6 MB of L2 cache). The E7000 line does not support either Virtualization Technology (VT) or Trusted Execution Technology (TXT); For our purposes, this is not a problem, since both technologies are irrelevant for users of "mainstream" desktops. The reduced L2 cache capacity combined with lower clock speeds makes this processor an excellent choice for a highly power efficient computer, as you'll see in our testing. Despite the declared thermal package of 65 W, the Core 2 Duo E7200 consumes energy much less than this ceiling.

In recent years, the motherboard market has seen a steady trend towards increasing popularity of solutions with an integrated graphics core. This circumstance is quite understandable. A personal computer is gradually moving from the category of luxury items to the category of such ordinary, but practically vital items in the household, such as a television. If ten years ago buying a PC in connection with entering university was considered simply crazy happiness, today the practice of purchasing a “digital friend” for a primary school student is almost the norm. And not for fun, but because it is necessary. Naturally, not every parent/grandparent can buy the most powerful car for their child, and not everyone needs it. For studying, mastering the basics of programming, music, films, simple toys and getting to know the world of digital technology in general, the performance of the built-in graphics core of modern chipsets is quite enough. At the same time, the absence of the need to purchase a discrete video card reduces the cost of the system unit by at least 70, or even $100, that is, by about 25-30%. According to demand, supply also increases. If earlier motherboards based on a chipset with an integrated graphics core were rare, today in some stores they make up up to half of the entire assortment. However, this circumstance is also due to global trends. As soon as the hype about ATI's purchase by the second largest processor manufacturer, AMD, had died down, it was presented with its own chipset, and only in a version with built-in video. The world's first processor giant Intel was not long in coming, presenting, along with a new line of chipsets for Core 2 Duo processors with a 1333 MHz bus, a solution with an integrated graphics core Intel GMA 3100 - Intel G31 Express. We already looked at the features of this chipset when we got acquainted with the line as a whole, and today we have the opportunity to get acquainted with the first motherboard based on it. Meet FOXCONN G31MX-K.

FOXCONNG31 MX- K

The key feature of the board, in addition to the increased performance of the graphics core, is compatibility with the new Core 2 Duo processors based on the Penryn core with a system bus frequency of 1333 MHz. Despite the fact that the official set of frequencies supported by the Intel G31 Express chipset ends at 1066 MHz, the G31MX-K specification also includes a value of 1333 MHz. Otherwise, the characteristics of the FOXCONN G31MX-K look quite standard. Network capabilities are due to the use of a Realtek RTL8111 gigabit controller, which directly interacts with the chipset via the PCI Express bus. The sound is based on the six-channel Realtek ALC662 codec, a simplified modification of the popular eight-channel Realtek ALC888 chip.

Since the Intel G31 Express chipset uses the old ICH7 southbridge, the maximum DDR2 memory frequency supported by the board is officially limited to 800 MHz. However, there is nothing wrong here. Today, DDR2-800 modules with a capacity of 1 GB are the most profitable purchase in terms of price/performance ratio, but the cost of a gigabyte DDR2-1066 module can exceed the cost of the board itself. One of the positive aspects of the ICH7 chip is the presence of a built-in single-channel IDE device controller, which eliminates the need for the manufacturer to use additional chips. In conditions of a strictly limited budget, this significantly affects the cost.

Packaging and accessories

The FOXCONN G31MX-K motherboard is packaged in a box corresponding to its size, the design of which is dominated by dark tones.

The equipment is standard. There is everything you need to assemble a basic level system, and nothing more. Specifically, the box with the FOXCONN G31MX-K contained the following accessories:

80-wire IDE cable;

cable for FDD;

Serial ATA cable;

adapter from Molex power plug to two SATA;

plug for I/O panel on the rear wall of the case;

CD with drivers and software for Windows XP and Vista;

instruction manual;

quick assembly guide poster.

As you can see, there are no excesses.

Design and Layout

The FOXCONN G31MX-K motherboard is very tiny. Made in the Micro ATX form factor, the width of the board does not reach the maximum size specified by the standard, as much as 36 mm. However, the layout of the elements was practically unaffected by this.

One could complain about the block for the floppy drive that has “slipped” into the central part of the lower edge of the PCB, but we have already talked about its practical uselessness in our time more than once.

The latches of the DIMM slots will certainly be blocked by even the shortest video card, but it is quite possible that the board will never “see” this device. So this circumstance should not be considered a disadvantage. But as for the number of DIMM slots for DDR2 RAM, I would like to see four of them instead of two, since having bought two 1 GB modules today, expand the memory capacity to 2 GB by purchasing two more “slats” tomorrow, alas. it won't work. You will have to get rid of the old ones and buy new, more capacious modules.

The set of slots for additional devices looks quite logical. In addition to the PCI Express x16 connector for an external video card, one PCI Express x1 and two PCI slots are soldered onto the PCB.

The layout of the remaining elements is well thought out, and no problems are expected during assembly. Everything is quite convenient and compact.

The processor's switching power converter is made according to a standard three-phase circuit. As practice shows, with high-quality implementation, this is more than enough to ensure stable power supply for modern processors both in standard modes and with moderate overclocking, even if we are talking about a quad-core Core 2 Quad. In normal modes, the power MOSFET transistors, which perform the main work of voltage conversion, do not heat up by more than 44 "C. During overclocking, their temperature at times reached 50" C. Since for this type of element the maximum permissible operating temperature is 105"C, we can say that the MOSFETs of the board in question have a double margin.

We see that when creating the G31MX-K, FOXCONN engineers completely abandoned the use of newfangled capacitive elements based on solid polymer. However, the electrolytic capacitors soldered on the board are produced only by trusted manufacturers, so there is no doubt about their quality. Practice shows that if you take care of the ventilation of the system unit, high-quality electrolytes can serve until the very moment the PC is sent for scrap as obsolete.

The chipset cooling system is completely passive. The Northbridge is cooled by a large aluminum radiator.

What is installed on the southern one is more like a heat sink, but as a series of measurements using a MASTECH MS650 digital thermometer showed, this is more than enough.

The temperature of the south bridge radiator (at room temperature 26"C) did not exceed 36"C, and the north bridge - 39"C. So, the main sources of air heating for a PC system unit built on the FOXCONN G31MX-K will most likely be the processor and hard drive The board itself does not require any additional cooling or airflow.

The back panel of the FOXCONN G31MX-K looks quite traditional. The following ports and connectors can be found here:

two PS/2 for connecting a mouse and keyboard;

four USB 2.0;

one D-SUB (VGA);

• one network RJ-45;

  three three-pin inputs and outputs of the sound card (mini Jack).

Perhaps it wouldn’t hurt to add a digital output to a DVI monitor to this set, since today this connector has begun to appear even on budget models of LCD monitors. But regarding its absence, claims should be made not to the board manufacturer, but to the chipmaker, since this interface is not supported by the Intel G31 Express chipset.

Having finished studying the elements and their layout on the board, we assemble a test bench and begin studying the BIOS.

Test bench configuration,BIOSand overclocking

To immediately clarify the issue regarding the equipment used, let’s take a look at the configuration of the test bench:

processor: Intel Core 2 Duo E6400, 2133 GHz (8x266), 2 MB L2;

cooler: Scythe Ninja Plus with 120 mm fan at 1500 rpm;

RAM: 2 modules of 1024 MB each, Apacer DDR2-800, 4-4-4-15 400 MHz;

hard drive: Seagate ST3160811AS, 160 GB, 3 Gb/s SATA, 8 MB Cache, 7200 rpm;

power supply: FLOSTON 560 W (LXPW560W).

The appearance and structure of the utility for configuring the basic input/output system FOXCONN G31MX-K, built on the basis of the AWARD Phoenix BIOS 6.00PG microcode, is quite standard.

All settings necessary to configure the system are presented in full. Among a number of submenus with familiar names, one stands out, called Fox Central Control Unit. We've already seen a submenu like this on a number of mid-range FOXCONN motherboards, but this is the first time we've seen it on a budget chipset-based board with integrated graphics. Here, BIOS developers have collected a number of settings used to overclock the system. For novice overclockers, there are three system acceleration modes: Step 1, 2 and 3, where all overclocking parameters are predefined by the developers. In particular, in Step 3 mode, the FSB frequency increases to 287 MHz, the voltage on the processor - by 0.1500 V, on memory modules - by 0.3 V.

At the same time, the test model Core 2 Duo E6400 will be overclocked to 2300 MHz, and RAM to 860 MHz. Considering the overclocking potential of the E6400, let's face it: the acceleration is minimal. Moreover, with such a weak overclock, increasing the processor supply voltage by 0.15 V is completely pointless. In any case, in manual mode the results can be much more impressive. Fortunately, such an opportunity is provided.

By switching to manual mode, the FSB frequency can be changed in the range from 266 to 600 MHz in 1 MHz steps.

The processor supply voltage can be either lowered or increased in the range from -0.0500 to + 0.2500 V in steps of 0.0125 V.

The voltage on DIMM slots can be increased by 0.1, 0.2 and 0.3 V.

The user also has access to settings for RAM addressing delays and its frequency. The four main timings can be changed either together or individually, entrusting the selection of the rest to the system algorithm. This feature can be useful for beginning overclockers.

There are only two coefficients used to calculate the frequency of RAM. With standard FSB 1066 MHz clocking, they will comply with DDR2-667 and DDR2-800 standards.

Practical testing of the board's overclocking capabilities began at 333 MHz, which in terms of Quad Pumped Bus format is 1333 MHz, that is, it corresponds to the bus frequency of the new Core 2 Duo processors based on the Penryn core. The FOXCONN G31MX-K coped with this task easily.

However, further testing ended quite quickly - at around 340 MHz. When this threshold was exceeded, although the system did not freeze, it also “refused” to operate at the specified frequency, and did this in a rather unique way: the board simply reset the FSB clocking to the nominal 266 MHz. Therefore, conclusions regarding the overclocking capabilities of the G31MX-K are not entirely clear-cut. On the one hand, the overclocking potential of the board is low and is not suitable for serious overclocking. On the other hand, FOXCONN G31MX-K is a product intended for inexpensive home and office systems, and the real opportunity to get from one of the most affordable Core 2 Duo models, the older E6700, and even operating at the frequency of the newly minted Penryn bus, looks extremely tempting. So, overall, considering the price and specifics of the product, the overclocking capabilities of the board look quite decent.

To complete our introduction to the features of the FOXCONN G31MX-K BIOS Setup, let’s take a look at the PC Health Status submenu.

Here we see quite rich system monitoring capabilities, allowing us to control the voltage values ​​of key power circuits, processor temperature, air temperature in the system unit and fan speed. In addition, the user has access to control of Smart Fan technology, the task of which is to automatically adjust the rotation speed of the processor fan impeller depending on the temperature values ​​​​received from sensors built into the CPU cores.

Having become acquainted with the main features and capabilities of the board in question, we move on to the test part.

Testing

In this case, the main objective of the testing is to find out how the performance of the new Intel G31 Express chipset, and especially its GMA 3100 graphics core, differs from the previous Intel G965 Express with the GMA 3000 accelerator. The G965 chipset is represented by the ASUS P5G-V motherboard.

Memory subsystem

Obviously, the memory controller of the Intel G31 Express chipset is better optimized. As a result, the FOXCONN G31MX-K board is immediately ahead.

Complex tests

The most serious and “advanced” package for comprehensive system testing today is the updated SYSMark 2007 package from BAPCO. The main feature of SYSMark 2007 is the fact that it uses only real-life and widely used applications to test the system. The ones that run on their PCs every day and are used by millions of people around the world for work. The package consists of four scenarios, each of which includes a number of operations performed by a specific set of applications specific to a specific area of ​​​​PC use.

In the overall standings, when tested with the SYSMark 2007 package, the FOXCONN G31MX-K motherboard is noticeably faster.

The advance in speed is consistently observed when examining each scenario separately in detail.

Next comes the popular PCMark 2005 test suite. Unlike SYSMark, it only simulates the operation of real applications, but, nevertheless, at the moment it is capable of giving a completely adequate and comprehensive assessment of system performance.

Here the differences between the performance of the tested boards are more noticeable. The memory controller, graphics and disk subsystems work faster. The result is a landslide victory for the Intel G31 Express.

In the next test package CINEBENCH 9.5, based on professional software for creating 3D scenes - CINEMA 4D, the built-in GMA 3100 graphics core is significantly superior to the previous one.

Mathematical and scientific calculations

The ScienceMark 2.0 program emulates scientific calculations performed on a computer, such as determining the kinetic and potential energy of molecules in a metal crystal lattice at different temperatures, calculating nuclear and electron charges, and other complex mathematical calculations.

This test does not see any difference between the boards (chipsets). Only when calculating the potential energy of a silver molecule does the ASUS P5G-V lose to its rival by one second.

The only task performed by the Super Pi program is to determine the value of Pi (3.14) with high accuracy. That is, this is a mathematical problem in its purest form. In our case, the calculation was performed with an accuracy of 1 and 8 million decimal places.

The system calculates the number Pi with an accuracy of 1 million decimal places equally quickly, but the G31 coped with eight million decimal places two seconds faster.

Coding of video and audio data

The next set of applications, which includes the tasks of encoding DVD video with the most popular codecs - DivX and XviD, as well as converting it into the 3gp format that is understandable to the vast majority of mobile phones, loads the processor and memory subsystem, so here, thanks to a better optimized controller, FOXCONN G31MX- K works consistently faster.

The same applies to the task of compressing an audio stream using the Lame 3.97 codec into the popular MP3 format.

Image processing

Adobe Photoshop is the most popular and functional raster editor. To measure the performance of systems in this task, a script was used to process five-megapixel photographs in uncompressed TIF format (about 15 MB each) with more than 30 filters.

The Panorama Factory program is designed for stitching panoramic images from separately shot frames. The application is characterized by very high stitching accuracy, but, as a result, high resource consumption. Eight five-megapixel photographs were processed.

The G31 represented by the FOXCONN G31MX-K handles raster graphics significantly faster.

Data archiving

The WinRAR archiver is one of the most popular and effective data compression programs.

Here the unconditional advantage is again on the side of the G31.

Semi-synthetic gaming tests

In semi-synthetic gaming tests, the GMA 3100's integrated video is definitely faster, but the advantage is negligible. Perhaps more attention should be paid to the overall performance advantage here.

In games, the situation looks a little better, but it is still not possible to seriously say that the GMA 3100 is suitable for games 2-3 years old.

Interface throughput and power consumption

The speed characteristics of USB and SATA controllers are almost identical.

The power consumption level of the systems was measured using the power indicator of the FLOSTON LXPW560W power supply.

In terms of energy efficiency, the G31 is definitely better. The amount of energy consumed by the FOXCONN G31MX-K is noticeably lower.

Audio quality

The audio path based on the Realtek ALC662 HDA codec was tested using RightMark Audio Analyzer 5.5 in 16-bit, 44 kHz mode using a Creative Sound Blaster Audigy 4 SE sound card.

The final rating of “very good” allows us to say that the ALC662 audio codec installed on the board is inferior to the popular ALC888 only in the number of channels, but certainly not in sound quality.

Conclusions

As testing has shown, there is definitely progress as a result of updating the Intel GMA 3000 graphics core to version 3100. However, in quantitative terms, it is too small to justify the need for a new chipset to appear on the market. The situation could be corrected by adding new functionality, but the Intel G31 Express cannot boast of this either. The chipset uses the old ICH7 south bridge; the north bridge does not officially support the 1333 MHz bus frequency, which means the new Core 2 Duo processors based on the Penryn core. Naturally, support for DDR3 is also out of the question here. On the other hand, the G31 is still suitable for the role of a slightly more productive replacement for the Intel G960 Express chipset, because the price tag for Penryn and DDR3 memory modules are far from the budget. Moreover, the advantage of the G31 over its predecessor lies not so much in the graphics subsystem, but in subtle optimizations of the memory controller and various buses, which significantly increased the performance of the chipset in all applications. For them, Intel has prepared two more products with integrated video - G33 and G35.

In turn, the FOXCONN G31MX-K motherboard is a good, solid implementation of the Intel G31 chipset. Having qualitatively implemented all the capabilities of the chipset, the developers went further, adding the ability to increase the system bus frequency to 1333 MHz. Of course, it’s still pointless to buy Penryn and put it on the board because of this, it’s too expensive for it, but the G31MX-K definitely has room for overclocking regular Core 2 Duo. The FOXCONN G31MX-K also has a number of other advantages listed above. Of these, first of all, I would like to highlight the low level of heat generation. The board is really “cold”, so it is well suited for miniature, or cramped, cases. This is also facilitated by a well-thought-out design.

Photos were taken in the TECHLABS studio, photographer Dmitry Filatov

Thank you companyFOXCONN for the fee provided for testing.

So, Intel has paused for almost three years since the release of the revolutionary i9xx series of chipsets. Let us remember that at that time desktop systems were added at once: a new type of socket and a new power connector, DDR2 memory, PCI Express bus (including an option for connecting video accelerators) and High Definition Audio. Then, over the course of two generations of chipsets (i945/955/975 and i965), only increases in FSB and memory frequencies followed, as well as support for new families of processors (first dual-core, and then Core 2).

Now we are meeting a new generation of chipsets, which, along with a radical change in numbering, offers an update of such important architectural characteristics of the system as the general purpose bus and memory type.

Intel X38 Express

It’s logical to start reviewing the new family of chipsets with its top representative, which, however, has not yet been released to the market and will appear only in the third quarter, like the entire second wave of new chipsets. Note that previously the model number of a top product was specified by an increased numerical index (i915 - i925), but now a top product can be easily distinguished by the X prefix, which at Intel is responsible for any general improvements (not only for chipsets, but also for processors and video accelerators) . The block diagram below lists the key characteristics of the X38:

• support for “new” processors of the Celeron and Pentium families, as well as all processors of the Core 2 (Duo/Quad/Extreme) family with a system bus frequency of 800/1066 MHz, including future models with a system bus frequency of 1333 MHz;
• dual-channel memory controller DDR2-533/667/800 or DDR3-800/1066/1333 with support for up to 4 DIMM modules with a total capacity of up to 8 GB (with ECC) and Fast Memory Access and Flex Memory technologies;
• 2 PCI Express 2.0 x16 graphic interfaces;
• DMI bus (with ~2 GB/s bandwidth) to the new south bridge ICH9/R/DH/DO.

It is clearly noticeable that all the key characteristics of the chipset have been changed. Let's look at the innovations point by point.

Processor support. Here it should be noted right away that officially all 3x series chipsets do not support processors of the Celeron D, Pentium 4 and Pentium D families (as well as their Extreme Edition versions). The lack of support is not due to the changed characteristics of the processor bus, but to the new standard for creating FMB motherboards (specifically, the VRM processor power supply module), which provides support for future processors created according to the 45-nanometer process technology, instead of the old ones produced on the basis of 90 (or more) - nanometer technology. Of course, there is no direct connection between the chipset used and the power subsystem on the motherboard, but manufacturers, in the vast majority of cases, follow Intel design standards, so it seems extremely unlikely that we will see a significant number of Intel 3x models supporting processors from the pre-Core era 2". Not to mention boards that simultaneously support Prescott and Penryn.

As for Core 2 support, things can’t be better for the X38: all current and future Core 2 Duo, Core 2 Quad and Core 2 Extreme models (including quad-core versions) will officially work with this chipset, and for all of them there will be 1333 MHz bus is supported. Of the younger families of new processors (Celeron 400 and Pentium E2000), everyone will be able to work on the X38, although for marketing reasons, Celeron 400 support for the top chipset is not announced.

Memory support. The DDR2 controller capabilities of all new chipsets have not changed (in fact, no development in this area is expected, everything in the specification has already been implemented), but boards based on Intel 3x will be able to work with DDR3 memory. The features and theoretical performance of the new type of memory have already been discussed in a separate article on our website, but here we will limit ourselves to considering practical aspects. The first question that usually arises is whether it is possible to support DDR2 and DDR3 simultaneously? Here the situation is no different from the transition from DDR to DDR2: Intel does not officially test such combinations and does not check them for compatibility, but no one is stopping motherboard manufacturers from doing this on their own. Our readers who regularly view the news are undoubtedly already familiar with several models of combined motherboards, and we carried out today's testing on one of these (however, we are unlikely to see combined models on the X38). Note that simultaneous Job DDR2 and DDR3 memory is of course impossible: at startup the board will initiate work with memory of one or another type.

In the context of assembling systems, DDR3 is good for everyone: less heat dissipation (the supply voltage is lowered, so even DDR3-1066 will emit less than DDR2-800), a different location of the key in the connector will not allow you to confuse the DDR2 and DDR3 slots on combined boards. As you already know, DDR3 is expected to operate at frequencies up to 800 (1600) MHz, and X38 will allow you to immediately use almost the fastest option - DDR3-1333. The situation with the availability and timings of available memory at the time of launch of Intel 3x is terrible. DDR3 modules are not yet widely available on the market, and in such conditions even “elite” manufacturers (like Corsair) allow themselves to sell modules with frankly mediocre characteristics at an insane price. We recommend that all our reasonable readers wait, since over time, of course, prices will fall and characteristics will increase. In the meantime, analysts predict that DDR3 will reach 50% of the market presence only in 2009, and by the end of 2007 this type of memory will hardly gain even 10%. And of course, in the practical part of the article we will look at what they are offering us to overpay for.

PCI Express 2.0. Here, Intel strikes a preemptive blow, not only by finally creating a chipset with support for two full-speed PCI Express x16 interfaces, something that the top products of its competitors have long boasted of (there is no real tangible benefit to speak of from such a configuration in the vast majority of cases, but the principles are more expensive), but also by implementing the host controller of the second version of the standard. In practical terms, the use of PCI Express 2.0 will not interfere with the use of older video cards, since the same connectors are used and compatibility is maintained in both directions. When applied to a graphical interface, the innovations of PCI Express 2.0 will most likely not be very interesting, with two exceptions. First, the performance of each PCI Express lane (lane) has been doubled, so that a connection with a single lane (PCIEx1) now has a throughput of 500 MB/s in each direction at the same time, and for a 16-lane PCIEx16 interface the total throughput will be 16 GB /With. We emphasize that in the foreseeable future the systems will not receive any practical benefit from this.

Secondly, the power supplied via the bus has been increased by the same 2 times: the PCIEx16 slot of the first version of the standard provided up to 75 W, but now the video card can receive 150 W. (The question immediately arises of how these “extra” watts will get to the bus - will there be a special additional power connector on X38 boards?) However, immediately after the launch of the i915/925, video cards for PCI Express began to appear, but with their own power connector on board (the bus 75 W was not enough), and now top video accelerators only look condescendingly towards power from the PCIEx16 connector, offering, at best, to abandon one of the two on-board power connectors. However, here, of course, the “merit” of SLI/CrossFire is great: it is the top video cards that are primarily designed for pairing, and if one theoretically can still have enough power on the bus, then the second video accelerator, thoughtlessly deprived of its own power connector, simply cannot to start in such conditions. As for the possibility of combining a pair of Intel X38 video cards, everything is the same: CrossFire is officially supported, SLI is not officially supported and will not be in the foreseeable future.

Also paired with the X38 will be a new south bridge from the ICH9 family; we will consider the functionality of this family in detail below.

Intel P35 Express

Let us briefly list the main functional characteristics of the northbridge of this chipset:

There are fewer innovations here, with only DDR3 being the most significant. Support for processors is limited to the same models based on the 65nm and future 45nm process technology, but for the reasons described above (for P35 boards a simplified FMB design is assumed), Core 2 Extreme models (especially quad-core) will not work in P35 boards. The chipset also lacks support for DDR3-1333 memory (in fact, it lacks a divider for setting such a memory frequency). Instead of PCI Express 2.0, the now standard PCI Express x16 (version 1) graphics interface is used, and like the P965 and earlier chipsets, the P35 does not allow flexible configuration of this interface to support CrossFire. However, as before, this fact does not stop motherboard manufacturers - they create solutions for CrossFire based on the P35, connecting the second slot to the south bridge (where PCIEx1 peripheral interfaces go to it). The southbridge for this chipset is also one of the ICH9 family.

Intel G33 Express

The main integrated chipset of the new family bears the somewhat illogical name G33, while in terms of functionality it is on par with the P35. The reason is that in the third quarter Intel will release another integrated chipset (now G35), with an improved graphics core, and it was necessary that the newcomer not be equal in number to the top-end X38. So, the G33, which is a variant of the P35 with an integrated graphics core, has the following architecture:

Let us briefly list the main functional characteristics of the northbridge of this chipset:

• support for “new” processors of the Celeron and Pentium families, as well as Core 2 Duo/Quad processors with a system bus frequency of 800/1066 MHz, including future models with a system bus frequency of 1333 MHz;
• dual-channel memory controller DDR2-533/667/800 or DDR3-800/1067 with support for up to 4 DIMM modules with a total capacity of up to 8 GB (without ECC) and Fast Memory Access and Flex Memory technologies;
• PCI Express x16 graphic interface;
• integrated graphics core GMA X3100 with support for Clear Video technology;
• DMI bus (with ~2 GB/s bandwidth) to the new south bridge ICH9/R/DH.

Let us repeat, this chipset differs from the P35 only in the presence of integrated graphics.

Integrated GMA X3100 Graphics. Let's hope that video drivers for the X3100 will be ready quickly, and we will finally see everything that was promised to us since the days of the X3000 (G965). In fact, the new video core has not undergone major changes compared to the GMA 950 (i945G), so it is significantly inferior in architecture to the X3000; We will analyze the differences when/if we can fully carry out all the tests. For now, let us remind you that Clear Video technology is designed to accelerate and improve the quality (deinterlacing + color correction) of video playback (including HD) in hardware, as well as provide digital video interfaces (including HDMI) for image output. Of course, full support for the Aero interface in Windows Vista is promised. The GMA X3100 also claims to support playback of HD DVD and Blu-ray discs; we will look at the details after testing the boards on the G33.

Intel G31, G35, Q35 Express

Let's say a few words about the rest of the chipsets in the new line. All of them will be released to the market in the third quarter of 2007.

G31 is an entry-level integrated chipset; it would be a stretch to call it new. In fact, its functionality is at the level of the 945G chipsets, which it is intended to replace. Even the south bridge of this chipset is the same old ICH7/R - thus, at the same time, the leapfrog with PATA support, which is not at all desirable in the corporate sector, which has been going on since the days of ICH8, has been resolved. The G31 supports Core 2 Duo (but with an FSB frequency of no higher than 1066 MHz) and memory up to DDR2-800.

The G35 is an interesting integrated chipset with a redesigned graphics engine, which Intel promises will be the first [integrated] solution to support DirectX 10. In due course, we will, of course, talk about the G35 (and its GMA X3500) in more detail. Otherwise, the G35 promises to be very similar to the G965 (note that this also applies to integrated video), and architecturally it will be similar to Intel 3x chipsets only by support for 45-nanometer Wolfdale and Yorkfield and the new Core 2 Duo with a FSB frequency of 1333 MHz (memory DDR3 is also not supported). The old ICH8/R/DH will be used as the south bridge for the G35.

Q35 (and its simplified version Q33) is the basis for Intel vPro business systems, an integrated chipset with gaming capabilities disabled. The most interesting will be the combination of the Q35 with the ICH9DO (Digital Office) southbridge, which will provide support for technologies such as AMT (Active Management Technology) 3.0, Trusted Execution Technology and Virtualization Technology. The Q35 also does not support DDR3 memory.

Intel ICH9 South Bridges

New chipsets get updated south bridges. Quite comparable to its northern brothers, ICH9 has a number of evolutionary improvements compared to ICH8, and also supports (only ICH9R) one technology that can be considered revolutionary. Let us briefly list the main functional characteristics of the new family of south bridges:

• up to 6 PCIEx1 ports;
• up to 4 PCI slots;
• 4/6 (4 for ICH9, 6 for ICH9R) Serial ATA II ports for 4/6 SATA300 devices (SATA-II, second generation of the standard), with support for AHCI mode and functions like NCQ (for ICH9 this mode is guaranteed to work only under Windows Vista), with the ability to individually disable, with support for eSATA and port splitters;
• the ability to organize a RAID array (only for ICH9R) levels 0, 1, 0+1 (10) and 5 with the Matrix RAID function (one set of disks can be used in several RAID modes at once - for example, RAID 0 and RAID can be organized on two disks 1, each array will have its own part of the disk allocated);
• 12 USB 2.0 devices (on two EHCI host controllers) with the ability to individually disable;
• Gigabit Ethernet MAC controller and a special interface (LCI/GLCI) for connecting a PHY controller (i82566 for Gigabit Ethernet implementation, i82562 for Fast Ethernet implementation);
• Intel Turbo Memory support;
• High Definition Audio (7.1);
• harness for low-speed and outdated peripherals, etc.

ICH9R traditionally differs from ICH9 in the presence of support for RAID arrays, as well as two extra SATA ports. Special versions of the southbridge ICH9DO (Digital Office) and ICH9DH (Digital Home) are based on the ICH9R, but the first of them offers additional functions of Active Management Technology 3.0, Trusted Execution Technology and Virtualization Technology, and the second - Viiv Technology (the positioning of both of these variations is obvious) .

Minor evolutionary changes include the increased number of USB 2.0 ports to 12, the implementation of the eSATA function and port splitters (which is relevant specifically for external eSATA connectors) for chipset SATA ports, and SATA connectors are now (like USB, starting with ICH8) subject to individual customization shutdown. An alternative to creating RAID arrays for data safety can be the new Intel Rapid Recover Technology, which allows you to create a disk image on another hard drive, quickly update it without touching unchanged files, and quickly restore data if the first hard drive is damaged. The southbridge still integrates a Gigabit Ethernet MAC controller, but we have not seen its use in any i965-based board - apparently, for ordinary desktop systems, a network controller from Marvell, Broadcom, Realtek and others like them, connected via PCI Express bus turns out to be cheaper. At the same time, users of corporate vPro systems will certainly appreciate the functions of the proprietary Intel controller. It would be strange to expect the return of PATA support after abandoning it in ICH8, and it really did not happen - Intel considers this issue closed despite the abundance of problems with “replacements” for chipset PATA.

The most intriguing thing about the new series of south bridges is the support for Intel Turbo Memory technology (under development known as Robson Technology). Its essence is to install a module with a certain amount of NAND flash memory on the board (to begin with, it is planned to produce variants with 512 MB and 1 GB). Basically, apparently, the module will be installed in the PCIEx1 slot, although in principle other connection options are possible (for example, to the contacts for an external USB port). Windows Vista users will benefit from Turbo Memory, and unlike, say, USB keys with flash memory, the module integrated on the board can be used by the new Microsoft OS for both ReadyDrive and ReadyBoost.

Briefly, in the first case, we get the opportunity to use a flash drive as cache memory for a hard drive - for linear read-write operations there cannot be much gain here (flash memory is slower than a hard drive), so the benefits of ReadyDrive will be observed with regular operations of exchanging small pieces of data, which are typical for reading and updating the paging file (the access time of flash memory is noticeably less than that of a hard drive). An additional advantage is the reduction in the number of accesses to the hard drive (data is merged onto the disk in batches during idle moments, and reading is not performed at all if the necessary data is available in the Turbo Memory cache), which saves energy - of course, this is a real benefit only for mobile devices devices.

ReadyBoost expands the available amount of memory for preliminary reading and caching of data (from the hard drive), and although flash drives cannot compete with RAM in speed, they still read not from the hard drive, but from flash memory with its low random access time allows you to significantly speed up loading applications and opening files (numbers are called up to 2 times). The disadvantage of Turbo Memory is the potential fragility of flash drives, the best of which are characterized by a number of rewrite cycles on the order of a million (possibly several million), which, even taking into account some capacity reserve, can lead to loss of drive capacity long before the end of the PC’s service life, in which it is installed.

Heat dissipation. The heat dissipation of the new chipsets deserves special mention. Despite being manufactured on the same 90nm process and with more complex logic, the 3x series chipsets consume noticeably less than their predecessors. Thus, the TDP for the P35 is 16 W (for the P965 - 19 W), and this despite the fact that the TDP of the new chipset is calculated based on increased FSB frequencies (1333 MHz) and memory (1066 MHz DDR3), that is, in equal conditions the difference is much more than 3 W in favor of the P35. Likewise, the new chipsets have noticeably lower maximum idle heat dissipation (5.9 W for the P35 and 10 W for the P965), although here a small concession is allowed for newcomers: idle measurements are carried out for the case of 2 DIMMs, and not 4, as before . The G33 is, in principle, characterized by the same consumption values, but since this chipset can be used without an external video card, we will give its heat dissipation for this case as a reference: idle - 5.75 W (versus 13 W for the G965), and TDP is 14.5 W (G965 has a record 28 W).

compare the reference heatsink for the new chipsets and the one used by MSI

As a result, the difference is so noticeable that it can be easily determined even by touch, when touching the chipset radiators. By the way, the reduced heat generation, of course, entailed a redesign of the standard cooling system, and Intel documentation provides a recommended version of the chipset cooler, with a significantly lower weight and surface area. Fortunately, those P35-based boards that we saw (including models from Intel itself) retained the same type of radiators (used for the i945/965 chipsets), and top products from all manufacturers, of course, will continue to be equipped with powerful designs using heat pipes - the situation is obligatory, although now this will only become relevant in case of serious overclocking. As a result, we have a turning point in an extremely unpleasant trend, when after the hot i965 and scorching nForce 600i it seemed that we would soon have to develop new standards for chipset cooling devices.

Performance Research

Test stand:

• Processor: Intel Core 2 Duo E6600 (2.4 GHz)
• Motherboards:
  • MSI P35 Neo Combo (BIOS version V1.0B16 from 04/20/2007) on the Intel P35 chipset
  • Gigabyte 965P-DQ6 (BIOS version D25) on Intel P965 chipset
  • EVGA nForce 680i LT SLI (BIOS version P03) on the NVIDIA nForce 680i LT SLI chipset
• Memory:
  • 2 modules of 1 GB Corsair XMS3-1066C7 (DDR3-1066)
  • 2 modules of 1 GB Corsair CM2X1024-9136C5D (DDR2-1142)
• Video card: ATI Radeon X1900 XTX, 512 MB
• Hard drive: Seagate Barracuda 7200.7 (SATA), 7200 rpm

Software:

• OS and drivers:
  • Windows XP Professional SP2
  • DirectX 9.0c
  • Intel Chipset Drivers 8.2.0.1014
  • NVIDIA Chipset Drivers 9.53
  • ATI Catalyst 6.8
• Test applications:
  • RightMark Memory Analyzer 3.72
  • 7-Zip 4.10b
  • WinRAR 3.41
  • codec XviD 1.0.2 (08/29/2004)
  • SPECviewperf 8.01
  • Doom 3 (v1.0.1282)
  • FarCry (v1.1.3.1337)

Test platform

Thanks to the fact that we received several MSI motherboards based on the P35 chipset for testing, including one with simultaneous support for DDR2 and DDR3, as well as a set of DDR3 memory modules from Corsair, today's testing will help answer two questions at once. Firstly, we will find out how the speed of DDR2 and DDR3 compare on one platform (P35), and secondly, we will compare both versions of this platform with other chipsets on the market today. As the latter, it is logical to take the P965 (which is being replaced by the P35) and the top chipset of the latest NVIDIA series - nForce 680i LT SLI (we have already found out that there is no difference between the nForce 680i LT SLI and nForce 680i SLI either in speed or functionality, and we had at our disposal a board based on nForce 680i LT SLI).

Comparing the two types of memory turned out to be more difficult, since the pre-sale BIOS versions of MSI boards turned out to be practically unprepared for DDR3: the BIOS of the P35 Neo Combo model did not provide the ability to set the normal (for DDR3) supply voltage (1.5 V) and timings (they were limited to the standard DDR2 circuit, so it was impossible to set values ​​greater than 6 for the main timings). At the same time, the Corsair modules we had in DDR3-1066 mode did not agree to work with timings lower than 7-7-7, so the board had to be launched in the mode of setting timings using SPD. Additional problems were created by the novelty of the platform, which did not allow checking the accuracy of timings (and other memory parameters) using numerous Windows utilities. Fortunately, the latest version of CPU-Z already understands both the P35 chipset and DDR3, so, with discounts for everything said above, we were able to bring some clarity.

In DDR3-1066 mode (SPD timings), according to CPU-Z, the timing scheme turned out to be as follows: 7-7-7-20. Since the board did not allow us to set the main timings above 6, we ran DDR2 memory at 1066 MHz with timings of 6-6-6-18 to get the results as close as possible. At the same time, at 800 MHz our DDR3 modules unexpectedly easily agreed to work even with timings of 4-4-4-12, which made it possible to compare this configuration with the P965 and nForce 680i LT SLI in their standard mode with DDR2-800@4- 4-4-12. Since we didn’t have a P965 board at hand that would allow us to run our memory in DDR2-1066 mode, only NVIDIA products represent previous generations of chipsets in this mode (remember, in tests it is extremely close to i965).

Now, before moving on to presenting the test results, let's consider the issue theoretically. Under equal conditions (at the same frequency with the same timings), DDR3 cannot be noticeably faster than DDR2, and the main hopes for acceleration from using a new type of memory can only relate to modes with reduced timings at high frequencies. Indeed, in terms of absolute timing values, the DDR2-800@4-4-4-12 and DDR3-1600@8-8-8-24 modes are equal, so if memory manufacturers manage to produce low-latency modules, DDR3 may be more efficient even in " normal" conditions.

The second possible advantage of DDR3 is increased bandwidth, since this memory can run on a O higher frequencies. Unfortunately, this gain can only appear on future processors, since at a FSB frequency of 1066 MHz the throughput of this bus is only ~8.5 GB/s, which corresponds to the throughput of a dual-channel DDR2-533! As practice shows, in such cases, usually increasing the memory frequency “by one step” can still bring a small gain, but in reality even DDR2-800 will be sufficient with a margin even for future processors with a 1333 MHz bus, while current processors are neither DDR3-1066 nor Moreover, DDR3-1600 is not needed.

Test results

As usual, let's start with a low-level study of memory potential using a test developed by our programmers.

Let us remind you that despite the similarity of speed indicators in real applications, the NVIDIA and i965 chipsets look very different in the synthetic RMMA test, so we will not focus on this difference.

The P35, which is noticeably inferior to both competitors in terms of data read speed, demonstrates a very interesting effect: when operating memory (both DDR2 and DDR3) at 1066 MHz, its performance is higher than in DDR2-800 mode, although the nForce 680i LT SLI has decrease slightly. Let us leave this fact, which is poorly consistent with our theoretical reflections, for now, and turn to other relationships. Actually, it remains for us to note that DDR3 looks noticeably worse than DDR2 even with equal timings. We deliberately do not indicate the exact magnitude of the differences here, since it would be hasty to estimate percentage differences before moving on to actual tests.

When testing write speed, we are not interested in the maximum performance achieved when using the direct data storage method, since they will be the same on a processor of the same architecture. In terms of the actually achievable speed of writing to memory, the picture roughly corresponds to that of reading: the new chipset is noticeably slower than competitors, DDR3 is slower than DDR2 (especially at 800 MHz), and still the transition to a memory frequency of 1066 MHz speeds up the P35 with both types of memory, but slows down the NVIDIA chipset.

Finally, a test of memory latency, and here the first surprise is the implementation in the P35 memory controller of a technology similar to NVIDIA's DASP - when during pseudo-random reading from memory (without going beyond one page), the latency is reduced radically, by several times. Obviously, we are dealing with the same caching buffer with prefetching. Nevertheless, even in such a successful test of pseudo-random reading from memory, the P35 is significantly inferior to its competitors (in this case, the nForce 680i LT SLI). In a comparison of DDR2 and DDR3 on the P35, the old type of memory again wins, this difference is especially noticeable in the DDR2/3-1066 mode, where DDR3 has higher timings.

It’s curious, but here too the transition to a memory operating frequency of 1066 MHz leads to acceleration, although the ratio of absolute timing values ​​should have led to the opposite: taking into account the clock cycle, CL4 for DDR2/3-800 corresponds to 10 ns, and CL6 for DDR2-1066 - 11.25 ns (not to mention CL7 for DDR3-1066 - 13.13 ns). Why is this so? Two possible explanations come to mind. Firstly, the correspondence of the bus frequency between the Core 2 Duo E6600 and the DDR2/3-1066 memory is noteworthy: perhaps such a synchronous operating mode provides some advantage. However, the absence of such an effect in the NVIDIA chipset suggests that some internal optimizations of the memory controller also have an effect, just like in the i965, which allows you to get a small gain from running the memory at any higher frequency.

Well, now let’s move on from considering theoretical aspects to real tests, and here, with numbers in hand, we’ll evaluate the advantages of certain configurations.

So, based on the real results, we can already draw the first conclusions. On the one hand, all the ratios we noted earlier have been preserved: the P35 is slightly (now we can say specifically - up to 7%) inferior to the P965 and nForce 680i LT SLI, DDR2-800 on the P35 is faster than DDR3-800 at equal timings (by 3%) , and DDR2/3-1066 on the P35 is faster than the same type of memory at a frequency of 800 MHz (it is impossible to give an exact estimate here, since the timings of DDR2 and DDR3 are different), and without a discount for a significantly lower O longer timings. On the other hand, it is worth noting that a 7% difference is observed in only one test, and working with DDR2-800 is obviously not the P35's strong point. Concealing the differences even further is the fact that DDR2-800@4-4-4 is memory with almost extremely low latency, while DDR3-1066@7-7-7 is the standard option, which Corsair and company are sure to use very soon they will offer an alternative with significantly lower timings.

But let’s not rush to a conclusion, let’s look at the results of other tests.

There are no surprises to be expected from testing for video encoding speed (measured according to our open method); here, as usual, all competitors look the same, since the limiting factor is processor performance.

In the SPECviewperf package of professional 3D applications, only NVIDIA chipsets manage to shine, which is quite likely due to their optimized graphics bus controller, since different memory operating modes (and even different types of memory) affect the speed only nominally.

We don’t see anything new in games either; the only thing worth noting is that in one of the Doom 3 modes (for the first and last time during today’s testing), the P35 emerges as the absolute winner (and of course, with memory running at 1066 MHz). However, the difference between the chipsets in Doom 3 is generally small, no more than 3%, and the losses due to the use of DDR3 instead of DDR2 on the P35 are even less - about 2%. In FarCry, the spread of results is a little more significant, up to 4%, but all three patterns we noted today remain valid.

Conclusions

It is difficult to evaluate chipsets that introduce several revolutionary innovations at once. In this case, the announcement turned out to be smoothed out, because PCI Express 2.0 will appear only in the third quarter, with the release of X38, and compatibility problems are not expected due to the transition to a new version of the standard. The second new product, DDR3 memory, did not make much of an impression on us with its speed characteristics, but, fortunately, at least in the first generation of chipsets the choice between DDR2 and DDR3 will be available, so you can safely wait for the prices to drop and the characteristics of the new type of memory to improve. Support for new processors is probably the main trump card of the Intel 3x series. True, by the time these new processors become available, it may turn out that other chipsets also support them, including products from competitors, of which at least the nForce 600i series claims support for FSB 1333 MHz, but no one can yet declare actual support for 45-nanometer models. The new southbridge is moderately progressive, adding a little here and there, and its main intriguing feature, Intel Turbo Memory, needs hands-on testing before reaching a verdict.

Before moving on to performance evaluation, I would like to note that we are waiting, firstly, for confirmation of the speed level shown by MSI boards. Indeed, all three boards we received demonstrated absolutely the same level of performance when working with DDR2 (two of them support only this type of memory), however, just before the end of the tests, we received new firmware for the P35 Platinum, which slightly (by several percent) increased the speed this model. In addition, although we cannot say that combined solutions are inferior in speed to “dedicated” ones, general concerns of this kind remain, so it is too early to put an end to the issue of DDR3 performance. If we take into account the release of [inexpensive, that is, mass-produced] processors with a 1333 MHz FSB bus, the picture may change further. However, having performed a significant amount of testing, it would be foolish not to draw any conclusions based on it. Our conclusions are as follows: taking into account all the listed and implied reservations, the new series chipsets still look a little slower than the old ones (both i965 and NVIDIA nForce 600i), DDR3 memory under equal conditions can lead to a loss of 2-3% of performance, and for the P35 it is better Memory operating at 1066 MHz is suitable, regardless of the timings.

Speaking globally about the fate of new chipsets on the market, X38 will undoubtedly find its, albeit few, fans of top-end solutions, being one of the best chipsets on the market in terms of functionality. P35, after exiting the early start streak, should show a decent level of performance, and its high-quality functionality, low heat dissipation, support for promising processors and memory types allow us to recommend buying boards based on the new chipset today instead of solutions of a similar class from competitors and old chipsets itself O th company. Turbo Memory technology, with a clear demonstration of all the promised properties, can become another very powerful argument for Intel 3x. We promise to talk about integrated options separately later.

Used from older versions to resolve conflicts regarding PATA support.

Specifications

The system bus of the G31 chipset, called QPB 800, operates at a frequency of 1066 megahertz. This device is capable of supporting dual-channel operational data stream mode with a frequency of up to 800 megahertz. The maximum amount of RAM is four gigabytes. The G31 chipset type is not designed to work with servers, so it is called pseudo-synchronous.

As for the integrated graphics adapter, the GMA 3100 provides good image quality and supports DirectX version nine.

The data transfer rate between bridges is two gigabytes per second, that is, one gigabyte per second goes out in one direction.

The chipset is capable of supporting four SerialATA channels, which, according to the characteristics of the G31 chipset, means the ability to connect four hard drives that will operate in SATA 300 mode. The latter designation demonstrates the data transfer speed within the system, that is, the maximum speed will reach 300 megabytes per second.

As far as power consumption is concerned, there is nothing to worry about since motherboards based on the G31 chipset are budget ones and do not have advanced features.

Supported processors

Motherboards with the G31 chipset support processors based on the Core 2 Duo microarchitecture with a processor bus frequency of no more than 1066 megahertz. working with this chipset is about 50 watts. The chipset also works with Pentium and Celeron processors, but only with support for socket 775.

Overclocking potential

For the example of overclocking the G31 chipset, we took it as a basis. To use the board's memory settings, you need to go to the overclocking section, which is called Fox Central Control Unit. After this, you need to select the optimal frequency, that is, the highest. The higher the operating frequency, the higher the productivity. Having selected the highest value, you need to look at the system monitoring section. The temperature of the current state of the entire system will be displayed there.

Now you can go directly to overclocking, and to do this you need to go to the Fox Central Control Unit section. By choosing the maximum value, you can see the performance gain. The G31 chipset is capable of overclocking the processor from 333 to a stable 600 megahertz.

Example of a motherboard with this chipset

As an example of a motherboard, a microATX format model from Asus is presented. This device is capable of working with both dual-core processors and quad-core representatives of the Intel Core 2 and Quad Core series. The motherboard has a P5KPL-AM 775 socket, which means that only 45-nanometer processors will fit.

The universal system bus operating at frequencies of 800, 1066, 1333 and 1600 megahertz can support DDR2 RAM up to 1066 megahertz in operation.

To unlock the full potential of the processors running on this motherboard, you can overclock the system bus to 1600 megahertz.

As you know, to make working with applications that use 3D graphics more productive, the motherboard has two slots for RAM. The motherboard can support dual-channel data flow with a frequency of up to 1066 megahertz, which will increase the speed of demanding applications.

To implement and play modern computer games at that time, the PCI architecture was improved. Now this bus is called PCI Express. With four times the bandwidth, you can enjoy every second of 3D gaming.

Also, the motherboard running on the G31 chipset has support for high-quality sound, additional settings for the BIOS from the manufacturer, and a controller for system cooling that optimizes its operation.

Conclusion

The choice of a motherboard running on a specific chipset determines the choice of all other equipment. Thanks to the chipset, the capabilities of the entire system change: the number of supported processors, component frequencies, parameters of the integrated graphics processor, power consumption and much more.

Depending on the chipset, you can guess how powerful the system will be and what its overclocking potential will be. The choice is yours.

Motherboards based on the G31 chipset are budget and are not designed for server manipulation and other advanced functions. This version of the motherboard is ideal for an average user, that is, for working with simple applications, surfing websites and running not too demanding games.

Different manufacturers add many additional functions to their motherboards, for example, such as monitoring power consumption or restoring BIOS program settings.