windwithme

Performance test for Intel HD Graphics HD3000 The preset value of GPU at BIOS is 30 which is equal to 1500 MHz after conversion. 3DMark Vantage => P2500 StreetFighter IV Benchmark 1920 X 1080 => 30.58 FPS In the Sandy Bridge, two GPUs, i.e., HD2000/HD3000 are built in by Intel, thus the performance is much advanced than the previous edition. The detailed test data are shared in several related articles shared by me recently (windwithme). The HD3000 has about 6-7 times 3D performance of the previous platform built-in GPU of Intel, thus it is believed that the applicable range is wider. The following part is about sharing of the overclocking for 2700K. New-style UEFI graphical interfaces are introduced into the BIOS items by BIOSTAR. O.N.E debugging page By changing 2700K at Turbo Mode to 50, the clock rate, i.e., 100 X 50 => 5GHz, will be executed. The Internal PLL Voltage Override is a special Intel overclock option, which effectively strengthens the stability of the CPU after substantially overclocking. It is suggested that the C1E should be powered off when overclocking, thereby facilitating control of clock rate and voltage of the CPU. When setting clock rate and parameters for DRAM, the smaller the numerical values are, the higher the performance is. The following example is DDR3 1866 CL8 10-9-27 1T, and the set values for the other detailed options are adjusted for optimization. User can perform miniature adjustment on the clock rate or the parameters according to the condition of the DDR3 on hand, so as to increase the performance of the DRAM. The CPU voltage adjustment is one of the key points for overclocking. It can be adjusted according to the condition of CPU and the heat-dissipation system of the platform; and the CPU VCoce LoadLine is Enabled. The DRAM voltage is another option to be adjusted for overclocking. Substantially, the overclocking at the Sandy Bridge platform is much simpler than that of the previous platforms, so the overclocking can be realized by grasping CPU frequency multiplication, DDR3 parameters, and voltage adjustment for both. PC Health Status It’s still felt that the temperature test result for CPU is on the much higher side; actually, the wind temperature of the heat sinks or the temperature of the software for measuring temperature are much lower. The above are personally stable set values for 2700K OC 5GHz, which may be readjusted for overclocking according to different hardware or difference between conditions individually. However, the general direction is at least right, which can be considered as reference for setting BIOS for the overclocking users. Overclocking performance CPU 100 X 50 => 4998.9MHz, at full speed is 1.416V. DDR3 1866.4 CL8 10-9-27 1T 1.600V Hyper PI 32M X 4 => 11m 52.875s CPUMARK 99 => 771 Nuclearus Multi Core => 32119 Fritz Chess Benchmark => 38.95/18694

Nuclearus Multi Core => 23908 Fritz Chess Benchmark => 27.86/13374 CrystalMark 2004R3 => 264699 CINEBENCH R11.5 CPU => 6.93 pts CPU(Single Core) => 1.57 pts Experience index of Windows- CPU 7.6 PCMark Vantage => 19386 The frequency multiplication of the 2700K in the mode of Turbo Boost 2.0 is 39, 38, 37 and 36 respectively; and the basic clock rate is 3.5 GHz. The above specification is 100 MHz higher than that of the 2600 K, the performance of this part is presented in the test, but the improvement is not much. Practically, both the 2600K and the 2700K are the highest-level CPUs in the framework of the Sandy Bridge, and the data obtained without overclocking is fairly high already. For a general use environment, most of the software requirements can be met by using the high performance preset in 2700K. DRAM bandwidth test DDR3 1599.8 CL7 8-7-22 1T ADIA64 Memory Read - 19618 MB/s Sandra Memory Bandwidth - 21335 MB/s MaXXMEM Memory-Copy - 21168 MB/s The DDR3bandwidth is much higher than that of our own platform or a competitor’s platform, which is the advantage of the improved LGA 1155. The high transmission performance of nearly 20000 MB/s can be achieved only by using DDR3 1600, and the voltage is lower than that for the previous platforms. Temperature performance (about 26 DEG C at room temperature) System standby - 23~33 CPU full speed - 40~46 Intel Burn Test v2.4，Stress Level Maximum The temperature performance when system standby is good by enabling the energy-saving function of C1E. With the latest Hydro Series H80 of CORSAIR and the high thermal diffusivity of the hydro-cooling framework, the temperature is only 46 DEG C at full speed. Only by working with excellent air-cooling or hydro-cooling equipment, the state of the preset value on 2700K can still be lowered to a low temperature. Power consumption test No software running on the OS desktop - 45W CPU full speed- 103W Due to introduction of 32nm, the power consumption is much lower than that of the previous generation Core i7-870 with the process of 45nm. The power consumption of the clock rate at the environment of C1E is only 58W higher than that of the preset maximum clock rate. Based on several articles about AMD APU shared by me recently, in a test with the same 32 nm and 4Cores, the power consumption of the APU platform is nearly twice that of AMD. It is thus clear that the performance and the power consumption under the Intel’s framework have good advantages under the same process.

Intel released the new generation framework of Sandy Bridge at early January, 2011; meanwhile, LGA 1155 is provided for support. Compared with the specifications of several platforms in the past, the LGA 1155 is provided with a lot of modifications in CPU overclocking (OC) at this time. Firstly, when a CPU with the external frequency of 100 MHz is overclocking, the other surrounding clock rates will be increased accordingly, thus the upper limit will fall around 110 MHz. Although the new 32nm process is introduced into the new generation 4 cores CPUs, overclocking can be only achieved by using frequency multiplication. The model names of i5-2500K and i7-2600K that are two kinds of CPUs without lock frequency multiplication have been released for a long time, which are considered as the representatives of the overclocking line. However, the overclocking can be only done by working with the chipset of overclocking frequency multiplication, i.e., P67 or Z68. The overclocking environment for LGA 1155 is simply declared above, hoping the users can aware of the hardware for the equipment clearly. The middle-level CPU featured by overclocking is Core i5-2500K, while the advanced-level is Core i7-2600K. The main difference between them is that the 2600K is provided with Hyper-Threading (HT) technology and an 8MB L3 cache, while the 2500K is only provided 6 MB L3 cache without HT. This time I pursue Core i7-2700K which will appear on the market very soon and replace the advanced 2600 K. The total clock rate of 2700K is 3.5 GHz, which supports the new generation Turbo Boost 2.0 and can reach the maximum performance of 3.9 GHz. The real 4 Cores are incorporated into the Hyper-Threading technology, so that eight threading can be executed, 4C/8T for short. With the 32 nm process, the power consumption of 95W and the 8-MB L3 Cache, the 2700K, surpassing the 2600K 100MHz, becomes the most advanced CPU of LGA 1155. The rear side of the 2700K is still D2 Revision, so that there is little difference in appearance between the 2700K and the 2600K at this part. The MB is the latest Z68 high-level chipset of LGA 1155, i.e., the latest version TZ68K+ released by BIOSTAR. The appearance and the specification are the same as that of the TZ68A+ shared before and the primary difference is supply voltage of CPU. Although the TZ68K+ is the new edition of BIOSTAR Z68, the price is still about US$99, thus it can challenge the low-price market of ATX Z68. By using the well-marked red-black matching, it has the most specifications as that of the Z68 series with the built-in Power/Reset button and a simple debugging LED. With respect to IO, there are three types among which the two blue USBs 3.0 can provide high-speed transmission. If provided two more USBs 2.0, it would be more perfect with respect to the scalability. The previous TZ68A+ is designed with four-phase power supply only, while the new edition TZ68K+ is provided with eight-phase power supply, which is the primary difference. However, as mentioned before, one-phase power supply can provide about 30-40 W, so , the four-phase power supply is already quite enough for overclocking of power consumption of Intel CPU. The heat sinks above the Z68 chipset look good because of being cut by a special manner. Two native SATA3 units of Z68 are positioned at the right side, which are used at a general environment or in a Raid0 state. Test platform CPU: Intel Core i7-2700K MB: BIOSTAR TSERIES TZ68K+ DRAM: CORSAIR CMZ8GX3M2A1866C9R VGA: Intel HD Graphics 3000 HD: Intel 510 Series 120GB POWER: Thermaltake TR2 450W Cooler: CORSAIR Hydro Series H80 OS: Windows7 Ultimate 64bit Firstly, a performance test is performed by using the preset value of CPU. Preset the performance CPU 100 X 35 => 3500 MHz (booting Turbo Boost, and then C1E) DDR3 1599.8 CL7 8-7-22 1T Hyper PI 32M X8 => 15m 21.977s CPUMARK 99 => 597

FINAL FANTASY XIV 1920 X 1080 => 4743 Due to Z68 PCI-E limited bandwidth, official spec cannot support 3Way graphics. GTX560 Ti performance should be satisfied by most gamers. If you need high resolution and effects, you can use 2-way SLi or higher grade VGA. GIGABYTE G1.Sniper2 Good 1.G1.Sniper2 is G1-Killer series which has top quality packing and components. 2.Built-in Creative CA20K2 audio chip with rare Japanese made high end caps and 128MB Memory cache. 3.Killer E2100 NPU chip with 1GB DDR2 cache provides better network traffic control. 4.High end Z68 with HDMI output supports Intel IGP exclusive Lucid Virtu technology. 5.Provides exclusive Front Access Control Panel to support USB 3.0 and quick OC. It also supports audio amplifier chip. 6.Special design military style heat sink has good quality and lookgin. Weak 1.CPU PWM should provide at least 12-phase to fit the product price. 2.No external chip to support 3Way VGA. 3.Can consider to make audio or LAN sub-card for more options. But it may cause users to have less PCI-E slots. 4.No GUI BIOS interface. Performance ★★★★★★★★★★ Components ★★★★★★★★★☆ Specification ★★★★★★★★★☆ Appearance ★★★★★★★★☆☆ C/P Value ★★★★★★★☆☆☆ For Intel Z68 exclusive Lucid Virtu Virtualization technology and SSD speed up Smart Response technology. I have shared in former several Z68 articles for detail introduction, using guide and performance. Currently, entry Z68 and high end one price difference is around 1.5~2c times. The main reason is components and spec. G1.Sniper2 positions at Gaming market. The audio and LAN chips are the top level in MB. Now the high end MB is for OC or Gaming. G1.Sniper2 integrates both advantages. If you buy same level audio and LAN cards, it will cost you much. G1.Sniper2 price is top. These two hardware components can enhance the product value. For looking for high end and OC Z68 gaming lovers, G1.Sniper2 can put into your wish list. Former, some readers suggest me to test more mid-end product, so this year I am trying to review more mid end or entry products. Like H61/H67, Pentium G620 and Core i5-2500K, if you are interested, you can search for that. This review takes me over 3 days, due to take pics, BIOS set and OC fine tune. Some readers may think it’s too long to read. If it’s possible, I am considering to separate into 2 parts like very popular Taiwanese movie, Seedig Bale. But after considering, I still decide to make in one article. If you are not familiar with OC or interested in buying in Z68, you can use it as your reference guide. This article is also post in my own blog WIND3C Welcome your visit.

DDR3 Test ADIA64 Memory Read - 24923 MB/s Sandra Memory Bandwidth - 27863 MB/s MaXXMEM Memory-Copy - 27807 MB/s Z68 set DDR3 clock at 2133, the ADIA64 and MaXXMEM performance is over X58 DDR3 2000. But, Z68 Sandra benchmark is lower than X58. After Sandy Bridge enhance the memory controller, the performance can compete with 3-channel. If you want to OC LGA 1155 DDR3 to 2200~2300, you have to pull CPU clock to over 100MHz first. Temperature Test (Room Temperature 28℃) Enter to OS Desktop - 36~42 CPU Full Speed - 70~80 Intel Burn Test v2.4，Stress Level Maximum Enter to OS desktop, the temperature is not big difference. 5GHz full speed max temperature is 80 which is also acceptable. There are 2 issues in OC, one is CPU voltage too low and the other is temperature rise due to CPU voltage too high. It takes your time to fine tune the best CPU voltage during OC. Using CORSAIR H80 water cooling system, the temperature is good. If it’s winter, the CPU temperature will be even lower. Power Supply is Thermaltake high end product, Toughpower Grand Series. Toughpower Grand Series is 80Plus Golf Certificate which is excellent performance for power saving. TPG-1200M is highest watt version in Toughpower Grand Series. Of course, all spec is the highest. My first impression is the shell design which is curve design. It’s dual 12V to support 40A and 85A. The max power output is 1200W. 14” dual ball baring flower shape fan is also special. The acoustic is very quiet. Modulized design is the must for high end Power Supply. It’s all Japanese made electrolytic and solid caps. The official warranty is 7 years. You can be more confident with Thermaltake. Power Consumption Test Enter to OS Desktop - 116W OCCT POWER SUPPLY Mode - 468W CPU voltage pull to 1.4V when OC to 5GHz, the power consumption only increase 16W when enter to OS desktop. Full speed is only 58W higher which is acceptable range. 2600K OC 5GHz can be run stable in safe temperature range. Some users think bigger watt Power Supply will consume more power. It’s not a right idea. Like the same test configuration, if I use 450W PSU, the power consumption will be 14W higher. 1200W means max output watt. The power consumption level is decided by 80Plus grades. It’s like a big pipe. The water flow amount is decided by tap. 1200W doesn’t mean it always work under max power output condition. Higher watt Power Supply is mainly for upgradability. My friend used 550W in the beginning and he upgraded to 750W when he started to use SLi. Recently, he upgraded to 850W due to buy more high-end VGA. So if you would like to reserve some upgradability for game or VGA card, I suggest you better to get higher watt PSU. VGA is nVIDIA GTX560 Ti 3DMark Vantage => P23771 StreetFighter IV Benchmark 1920 X 1080 => 305.18 FPS

Clicking "Quick Boost" will enable "Smart Quick Boost" and “@BIOS" will enable "@BIOS". These two are GIGABYTE exclusive applications. Touch BIOS integrates both for users easy to install. Killer E2100 LAN Chipset SOftware - Bigfoot Networks Killer Network Manager This network manager features are very rich. Users can know the network status through this software. Upper left, EZ Smart Response, enable system HDD can switch to RAID model and enable ISRT under AHCI/IDE mode automatically. Creative provides Entertainment, Game, Audio Creation modes. The Game mode panel and items as right hand side. It includes rich fine tune items. Entertainment mode panel interface is familiar to me as my X-FI audio card. Audio Creation mode provides different panel and items. It looks like Creative own EMU audio panel, but actually the items and features are different. Previously, I tested the audio effect in G1 X58. I would like to repeat some key points. X-FI audio is well benefit the video or gaming multimedia. You can feel very good sound filed. No matter bass, fighting or demolish, it’s really shocked. G1 builds in such high end audio chip and caps, it’s good C/P for gamers. You don’t need to buy an external sound card. For who owns 2500K or 2600K, many people want to use at 5GHz. However, for this high range OC, you need good quality CPU and high end cooler. If you don’t set well, the system will be unstable and CPU will be in high temperature. 5GHz is love and hate setting. Below is stable 5GHz OC Test CPU 100 X 50 => 5001.4MHz 1.400V DDR3 2134.2 CL7 10-7-27 1T 1.620V Hyper PI 32M X8 => 12m 09.442s CPUMARK 99 => 771 Nuclearus Multi Core => 31325 Fritz Chess Benchmark => 39.12/18776 CrystalMark 2004R3 => 400655 CINEBENCH R11.5 CPU => 9.84 pts CPU(Single Core) => 2.03 pts PCMark Vantage => 26169 From 4.5GHz to 5GHz, the CPU performance increase around 10~12%. CPU performance enhance range is not much. However, 5GHz is an excellent feeling for most users. 2600K single core performance and OC range is the best in the market. The multi-tasking performance is only behind 6C12T 980X and 990X. AMD will launch 8-core CPU in the future, called Bulldozer. Its multi-tasking performance may be higher than 2600K.

CrystalMark 2004R3 => 381995 CINEBENCH R11.5 CPU => 8.80 pts CPU(Single Core) => 1.83 pts PCMark Vantage => 24328 The CPU voltage will be various as different CPU quality. Actually, if the cooling is good, 2600K no need to add too much voltage from 1.245V to 4.5GHz. The above test is setting as 1.250V. It can pass most benchmark software and burning test stabile. Sandy Bridge as Intel 2nd generation Core I structure. The CPU single thread performance is excellent. 4.5GHz clock to test all benchmark, the score and performance are very good. Also, OC to 4.5GHz is a stage. You have to control voltage and temperature well. DDR3 Test ADIA64 Memory Read - 22282 MB/s Sandra Memory Bandwidth - 24726 MB/s MaXXMEM Memory-Copy - 24863 MB/s Sandy Bridge DDR3 bandwidth improves a lot. It’s even same as LGA 1366 3-channel. If DDR3 quality is good enough, even CPU clock is the same, you still can run at 2133. Using CORSAIR DOMINATOR-GT to test 1866 CL6, the bandwidth and performance are very good. Temperature Test (Room Temperature 28℃) Enter to OS Desktop - 33~38 CPU Full Speed - 56~64 Intel Burn Test v2.4，Stress Level Maximum The temperature result is good. It owes to H80 water cooling system as well. 2600K OC to 4.5GHz won’t be a problem if you use mid-high end coolers. This is also 32nm advantage. I am looking forward to Panther Point next year. Power Consumption Test Enter to OS Desktop - 100W OCCT Power Supply Mode - 410W For 2600K OC to 4.5GHz add GTX 560 Ti, the power consumption is very low. For full speed, CPU full speed is around 200W and above test OCCT CPU and VGA is 410W. This value just needs 550W PSU for whole system. GIGABYTE hasn’t implemented UEFI interface, but it provides alternative solution called Touch BIOS software. Users can adjust BIOS items in OS and also support touchscreen LCD. This software also integrates many GIGABYTE features. It also provide multilingual interface. The screen shot is Traditional Chinese. BIOS is not UEFI design, but it means user can use BIOS interface in OS? When I see items called "Miscellaneous Settings", I can imagine how rich the settings are...

Advanced DDR3 Parameters Setup Page The screen shows setting as DDR3 2133 CL7 10-7-27 1T Voltage Page Milti-Steps Load-Line Disabled/Level 1~Level10 CPU Vcore 0.750~1.700V QPI/Vtt Voltage 0.860~1.530V Systrm Agent Voltage 0.715~1.495V Graphics Core 0.850~1.705V CPU PLL 1.195~2.435V DRAM Voltage 0.890~2.135V Sniper2 built in one HDMI output. User can use for Intel built-in GPU. It bundles with Lucid Virtu GPU Virtualization technology for user to switch high end 3D graphic card and Intel Quick Sync Video technology. PC Health Status UEFI is a very common word recently. It means 3TB support plus GUI. GIGABYTE all Z68 BIOS support EFI 3TB HDD. For GUI, they still need to catch up. This review I use Intel Core i7-2600K. As 2600K default performance, windwithme has test end of 2010. Recently, the other Z68 OC guide can be reference. If you want to refer to the 2600K default performance, you can search my previous article. The BIOS pics above are 2600K OC 5GHz setup. You may need to fine tune your system voltage and clock due to CPU and DDR3 quality. Windwithme reviews are taking long time to fine tune for stable OC. The OC ratio is passing stress burning which is my own requirement. Maybe it’s just an easy performance pic, it takes me several hours for fine tune the most suitable voltage and CPU clock for stable running. With DDR3 OC and fine tune parameters, finally, the CPU and DDR3 are all stable at highest performance. Then I start to lower voltage to find the lowest value. It takes few hours to days. I think the every MB best OC setup is worthy and really useful for users who are interested in OC. I also hope the OC will bring the budgeted hardware into highest performance. System Configuration CPU: Intel Core i7-2600K MB: GIGABYTE G1.Sniper2 DRAM: CORSAIR DOMINATOR-GT CMT4GX3M2B2133C9 VGA: msi N560GTX-Ti Twin Frozr II HD: Intel 510 Series 120GB POWER: Thermaltake Toughpower Grand 1200W Cooler: CORSAIR Hydro Series H80 OS: Windows7 Ultimate 64bit Cooler is CORSAIR latest water cooling product. Hydro Series has 4 models, H50, H60, H80 and H100 from low to high. Below is H80 packing. It’s 5 years official warranty. The package is bigger than perviousH70. All 4 models support all CPU sockets include future LGA2011. Left hand is the all black radiator. The size is 120mm x 152mm x 27mm. The shell is also metal delicate which is better than H70. The both sides can install 12cm fans. The rotation rate can adjust by the button above cooler head. There are 1300, 2000 and 2500 RPM. The fan airflow is 46~92 CFM and acoustic is 22~39dBA. Hydro Series H80 advantage is can install in standard ATC chassis. This hydro cooling is very convenient for users. H80 new design improves the cooler head. The first one, the new mounting bracket is more stable and durable. The bottom still keeps the big dimension copper. Next is two 4-pin headers for 12cm fans. Near the cooler head, there are one small 4-pin and one big 4-pin headers. They are for MB or PSU to supply power to cooler motor. First of all, let’s test 2600K OC 4.5GHz performance. This setup is most ideal OC range for most suers. CPU 100.0 X 45 => 4501.5MHz 1.250V DDR3 1867.4 CL6 9-6-24 1T 1.610V Hyper PI 32M X8 => 13m 40.699s CPUMARK 99 => 693 Nuclearus Multi Core => 29030 Fritz Chess Benchmark => 35.31/16949

One of G1.Sniper2 key features is Killer E2100 LAN chip, called NPU. It bundles 1GB DDR2 for network traffic control. It can help to offload the network traffic from CPU. Another function is Game Networking DNA technology. It offloads gaming data directly to the NPU by bypassing the Windows Network Stack to save time. The other key feature is Creative Sound Blaster X-Fi audio processor, chipset code is 20K2. It built in 128MB Memory cache for 20K2. It’s only for high end Creative sound card. Nichicon MUSE ES(Green) and MW(Yellow) Japanese made audio caps with metal cover which is same as high end sound card design and components. Next is G1-Killer series special design heat-pipe. The design concept is shooting game, also called Locked and Loaded Heat-pipe. The part is a big heat sink. The center has 5 green LED which is on as booting up. The shape is same as magazine. The top is a copper bullet. The heat sink dimension and quality are impressive. VRM heat sink is similar to barrel. CPU is 8 phases Driver MOSFETs design for better power efficiency and temperature control. Booting Screen Main BIOS Tuning Menu, called M.I.T. It shows BIOS version, CPU/DRAM clock, DDR3 capacity, CPU temperature and CPU/DRAM voltage etc. You can adjust CPU ratio, clock and DDR3 multiplier. The screen is CPU 100.2Mhz/DDR3 2133 setup. CPU Detail Items 2600K default ratio is 38, 37, 36 and 35. Turbo Boost 2.0 technology has more critical CPU boost mode. Internal PLL Voltage Override items enhance CPU stability when OC to 5GHz. Z68 default is Enable. When OC CPU Clock Ratio to 50, I like to disable Intel Turbo Boost and C1E for better stability. Memory Setting Page Sandy Bridge multiplier has 800, 1066, 1333, 1600, 1866 and 2133.

GIGABYTE launched new series in early 2011 called G1-Killer with Intel chipset. First series use X58 chipset and there are 3 models, G1.Sniper, G1.Guerrilla and G1.Assassin. They are same as own high end model UD7, but have less power phases and PCI-E slots. G1 uses high end product components and focus on Gaming which enhance audio and network design. So far, Intel most popular product is Sandy Bridge. It’s been launching less than 1 year, however, there are Z68, P67, H67 and H61 chipset can supports SNB CPU. You can see Sandy Bridge can combine from entry to high end as users demand. From LGA 1155 CPU and chipset quantity, you can see the product life cycle should be longer than LGA 1156. This time, I got GIGABYTE 4th G1 product using Intel latest Z68 chipset with G1 series design concept. This model name is G1.Sniper2. You can see the color tone is black with green. It’s same as previous 3 G1 X58 series as focusing on war visual concept. Sniper2 is ATX and the size is 30.5 cm X 26.4 cm. It includes 5 Smart Fan controllable fan header which can adjust Z68 platform temperature more efficient. This time the magazine heat sink has no G1.Killer print. It’s just whole black design. Accessories Product manual, Software installation guide, Driver CD, IO panel, SATA cables and SLI bridge. Front Access Control Panel provides 2 USB 3.0 and 1 Power eSATA ports. Left side Quick Boost button can let user switch to OC mode to increase system performance easily. Bottom is G1-Killer poster and sticker. Lower-Left Corner 2 X PCI-E X16 supports AMD CrossFireX/nVIDIA SLI as X8 + X8. 2 X PCI-E X1 2 X PCI Very rare Bigfoot Killer E2100 LAN Chip. Built-in Creative CA20K2 audio chip supports Dolby Digital Live, DTS Connect, X-Fi Xtreme Fidelity and EAX Advanced HD 5.0 technology. It supports max 7.1 channels and High Definition Audio technology. Design in Taipei Lower-Right Corner 3 X Black SATA provided by Z68 supporting SATA2. 2 X White SATA provided by Z68 supporting SATA. You can build hybrid RAID 0, RAID 1, RAID 5 and RAID 10. The max performance depends on SATA devices. 2 X Grey SATA provided by Marvell 9182 supporting SATA3, RAID 0 and RAID 1. 2 X 32 Mbit flash Dual BIOS for dual protection. Upper-Right Corner 4 X DIMM DDR3 support 800/1066/1333/1600/1866/2133 and DDR3 max capacity is 32GB. It supports Extreme Memory Profile. DDR3 uses 2 phases PWM and next is 24-PIN power connector. Upper-Left Corner LGA 1155 CPU Socket. The cover is plating process for better quality. G1.Sipner2 uses 8+2+2 phases PWM. DRAM and Chipset have individual 2 phases. Behind the upper heat sink is an 8PIN ATX 12V power connector. IO 1 X PS2 KB/MS 1 X HDMI 1 X O.C. Button 7 X USB 2.0(Red/Black) 2 X USB 3.0(Blue) 1 X eSATA/USB 2.0 Combo(Black) 1 X RJ-45 LAN 1 X S/PDIF optical fiber output

BIOSTAR TSERIES TA75A+ PROS 1. The package and material are both above par. The product falls in the mid-price range. 2. The internal display has three kinds of output interfaces, and the 3D performance of the A8-3850 internal display comes close to some basic VGAs. 3. It includes a Japanese solid capacitor, built-in POWER/RESET buttons and a simple debug LED. 4. Its BIOS has been upgraded to UEFI interface, with rich options, a large voltage range, as well as an excellent overclocking capacity. 5. It provides two PCI-E X1/PCIs and integrates the native SATA3 and USB 3.0 technology. CONS 1. BIOSTAR has no current distribution channel in Taiwan. 2. The DDR3 compatibility could need further enhancement. Performance ratio ★★★★★★★★☆☆ Material ratio ★★★★★★★★☆☆ Specification ratio ★★★★★★★★★☆ Appearance ratio ★★★★★★★★☆☆ Performance vs. Price ratio ★★★★★★★★★☆ The BIOSTAR TA75A+ costs about US$99(NT$2875) in the U.S. Its BIOS has been upgraded to UEFI interface, with excellent overclocking capacity and debug LED design. It has also been integrated with the original new-era specification of the A75 chipset. It has high competitiveness among the A75 MBs of various brands. However, BIOSTAR still needs to exert effort in terms of increasing its market awareness. In addition, it should be noted that in CPU overclocking, as the A75 chipset increases the CPU 100MHz overclocking, the SATA frequency also increases accordingly. If the CPU overclocking is used for a long time, it is not certain whether it will affect the function of the HDD/SSD when operating with increased frequency. To safely increase the performance of the internal display, the DDR3 frequency should be increased first, and then adjusted to 1600/1866 to achieve stability according to the physical strength of the DDR3. The A75 MB with A8-3850/A6-3650 is an all-in-one platform for the mid-range market. It is highly attractive for users who require high specifications. The major difference with the A55 chipset is the absence of the USB 3.0 and SATA3. But the product is sold at a lower price and it is definitely a valuable option for entry-level computers. Specifically, an earlier model of the A6-3500 or a lower-level of the A4/E4 CPU sells for less the price. When the new-generation APU platform becomes popular in the future, it will be more affordable and valuable. This is good news to consumers who prefer to purchase low-configuration or entry-level computers. This article is also post in my own blog WIND3C Welcome your visit.

DDR3 1947 CL8 10-8-27 1T ADIA64 Memory Read - 10971 MB/s Sandra Memory Bandwidth - 18509 MB/s MaXXMEM Memory-Copy - 13163 MB/s Increasing the CPU frequency causes the new FM1 platform to increase the bandwidth for the DDR3 as well. However, the increased performance is only about 500 MB/s, which is exactly the required enhancement for AMD architecture. When the overclocking is close to DDR3 1947, it can offer nearly 11000 MB/s bandwidth, which is relatively good for the AMD platform. The overclocking range of the DDR3 is also increased in the FM1 platform which is normally between 1650-1800 on the AMD platform. Temperature performance (ambient temperature is about 32 Celsius. ) System standby - 13~16 CPU full speed - 55 Intel Burn Test v2.4，Stress Level Maximum Test with the AMD C&Q power-saving technology turned on. The CPU frequency is greatly reduced when the system is on stand-by therefore the temperature is much lower. When the CPU is at full speed while working with an advanced heat sink, the temperature is acceptable at around 55 Celsius . I hope that AMD can attain the 32nm process in its CPU series in the future, so as to help increase its competitiveness. Power consumption test System standby - 28W CPU and GPU at full speed - 178W Power Supply test in OCCT When the system is on stand-by, power consumption is quite satisfactory due to the power-saving feature. On the other hand, when at full speed, power consumption is still high. The 178W-power consumption at full speed is primarily caused by the CPU. This is not good since it adopts a 32nm process. Compare this with previous tests. The power consumption of Intel’s 32nm CPU at full speed is a dozen watts less than that of the A8-3850. Power consumption is another aspect that AMD needs to improve on in the future. It takes a long time to realize the 32nm process, and power consumption could not be reduced effectively. 3D test Internal display AMD Radeon HD6550D 3DMark Vantage => P5216 StreetFighter IV Benchmark 1280 X 720 => 109.58 FPS Unigine Heaven Benchmark => 15.3 FPS In my previous article, I shared the test data of CPU/DDR3 with 100/1866 ratio, and the 3DMark Vantage was about P4500. This score has reached the same level as the nVIDIA GT430, and it is 10% better than the Radeon HD5570 under the same brand. In this test, it works with CPU overclocking, so as to increase overall 3D performance by 10~20%. According to the Llano APU data posted on the internet, the 3DMark Vantage can achieve nearly P7000 at most, which has an amazing 3D internal display performance. Video Playback Testing Start up it to play the original copy of a Blu-ray video on hand –named Inception. Below is the publishing corporation’s copyright ownership. PowerDVD 10 software Starting menu Until now, the CPU occupancy rate is as low as 1-6%. Playing process The CPU occupancy rate is 2-10%. The A8-3850 internal display is represented by the Radeon HD6550D, which is capable of playing 1080P videos. Generally, the new generation Intel Sandy Bridge and AMD Llano APU have integrated the internal display function to usher in a new era of technology. Aside from multiplying 3D performance several times, the CPU occupancy rate is much lower than the old platform when playing high-definition videos. In the future, you need not worry about lagging problems when selecting the internal display platform for playing videos.

Performance test CPU 138.0 X 25 => 3450.3MHz DDR3 1840.2 CL8 10-8-27 1T Hyper PI 32M X 4 => 19m 30.146s CPUMARK 99 => 499 Nuclearus Multi Core => 14018 Fritz Chess Benchmark => 17.77/8530 CrystalMark 2004R3 => 199037 CINEBENCH R11.5 CPU => 4.11 pts CPU(Single Core) => 1.05 pts PCMark Vantage => 7420 Windows experiential index - CPU 7.4 Increasing frequency multiplication is a feature not available in the A8-3850. Also, the external frequency range is not very wide, so the AMD 32nm CPU overclocking limit has yet to be confirmed. When the black-box edition CPU is released, the overclocking limit of the CPU’s resultant frequency would have a better performance. According to the above test software, the CPU performance can be increased by 20% after the A8-3850 2.9GHz overclocks to 3.45GHz. DDR3 is part of the U.S. CORSAIR VENGEANCE series, with model number CMZ8GX3M2A1866C9R It is indicated on the package that the product supports Intel/AMD platforms, and the DDR3 capacity is 2 X 4GB The frequency is DDR3 1866, and the parameter is CL9 10-9-24 1.50V The AMD FM1’s preset DDR3 voltage is 1.50V. Given the preset settings, the DDR3 frequency can be increased to 1866. If you prefer to just rely on the high-frequency DDR3 and can do without debugging, the DDR3 1866 product is quite good and convenient. The appearance is also different from the earlier common Dominator design. VENGEANCE uses the new and large heat sink with elegant red color. DRAM bandwidth DDR3 1840.2 CL8 10-8-27 1T ADIA64 Memory Read - 10555 MB/s Sandra Memory Bandwidth - 17485 MB/s MaXXMEM Memory-Copy - 12701 MB/s DDR3 1866.8 CL8 10-8-27 1T ADIA64 Memory Read - 9933 MB/s Sandra Memory Bandwidth - 16127 MB/s MaXXMEM Memory-Copy - 10979 MB/s

The heat sink covers the AMD A75 chipset. Despite its small volume, the BIOSTAR heat sink appears to have better texture. The heat sink in the power supply uses eye-catching colors. The design makes heat dissipation conveniently larger. BIOS home screen O.N.E. debugging page The CPU Clock is one of the key points of overclocking. If the HDD/SSD on the SATA device is powerful enough, it could reach 140~155MHz. The IGD Clock Control adjusts GPU frequency, but increasing the frequency doesn’t help 3D performance. To start from the Core FID is recommended followed by a reduction of the CPU frequency multiplication and an increase in the external frequency of the CPU; these would result in a wide overclocking range. Voltage page APU-Core Over Voltage +0.050~1.450V APU-NB Over Voltage +0.050~0.200V DDR Memory Over Voltage -0.400~+0.450V(BIOS data 1.596V) APU DDR-PHY/PCI-E Over Voltage +0.010~0.450V FCH Over Voltage +0.010~0.450V DRAM parameter settings The most efficient way to improve the 3D performance of the Llano APU is to increase the frequency of the DDR3. To increase the external frequency of the CPU, it is recommended to first set the DDR3 at 1066/1333. More advanced DDR3 parameter options PC Health Status So far, the A75 overclocking has two internal display GPU methods. One is by taking the default CPU value and setting the DDR3 at 1600 or 1866 operation. The other is by reducing the CPU frequency multiplication and the DDR3 frequency, followed by increasing the CPU external frequency to achieve the main purpose. The former is a simple and stable overclocking method while the latter is for users who want to explore the CPU/DDR3/GPU performance limits. The BIOS setting mentioned above applies the latter’s overclocking method. It took me quite some time to achieve this setting. Test platform CPU: AMD A8-3850 MB: BIOSTAR TSERIES TA75A+ DRAM: CORSAIR CMZ8GX3M2A1866C9R VGA: AMD Radeon HD6550D HD: WD3200BPVT 320GB POWER: CORSAIR Builder Series CX430 Cooler: Thermaltake BigTyp 14Pro OS: Windows7 Ultimate 64bit SP1

In early July, AMD released a new platform with FM1 pin grid array, which supports the CPU and adopts an updated APU technology. It has updated its own architecture and integrated a GPU with higher performance into the CPU. For users who support the reasonably-priced AMD, this is definitely another option with new specifications. In 2011, AMD plans to release a CPU with higher performance called the Bulldozer which is set to strengthen its medium and high-end product lines. The Llano APU platform has two kinds of chipsets, the A75 and A55. The newly released A55 has low configurations. In the past, the quad-core CPU only had two options, the A8-3850 and A6-3650. However, recently the tri-core A6-3500 was released to the market. When the A55 gains a foothold in the market in the future, it can combine with the A6-3500 to present a valuable, new generation high performance internal display platform. I (nickname: windwithme) have discussed the A75 performance of the Micro ATX in my last article, and now I will talk about the A75 MB of the ATX specification. The greatest feature of the A75 is its capability to provide native support of USB 3.0 and SATA3, as well as a wider range of specifications. Major MB brands have many A75 models available in the market and hopefully, there would be an A75 product with ITX specification released in the future. I patronize the BIOSTAR TA75A+. The BIOSTAR brand is affordable and has overclocking features; it represents a relatively good C/P value. The package adopts a simple design with red color and words/patterns that describe the supported functions and technologies. Attached accessories User’s manual, IO board, SATA wires and software driver disc BIOSTAR TA75A+ appearance The advantage of the ATX is its good scalability. Its large size makes the computer mainframe larger. Choosing the A75 of the Micro ATX or ATX specification depends on the different requirements of users. The PCB is black in color while the other extended slots are either red or white. I believe that replacing white with black would bring better texture. Currently, many advanced MB products generally use black and red colors. This is because black appeals to a lot of consumers. The lower left section of the motherboard 2 X PCI-E 2.0 X16 supports 2Way CrossFireX technology, and the bandwidth is X16+X4 2 X PCI-E X1 2 X PCI Realtek RTL8111E network chip Realtek ALC892 audio chip supports 8-channel HD Audio The lower right section of the motherboard 4 X red SATAII connectors are provided by the A75 chip; these are of SATA3 specification and support RAID 0, RAID 1, and RAID 10. Blue is the front extended USB 3.0 slot, Power, Reset buttons, and a simple built-in Debug LED The upper right section of the motherboard 4 X DIMM DDR3 slots support 800/1066/1333/1600/1866/2000(OC), and the highest DDR3 capacity supported is 32GB. Specification of DDR3 2000 can be only achieved through CPU overclocking, along with a 24-pin ATX power connector. The upper left section of the motherboard TA75A+ requires 4+1 phase power supply, supports AMD A8/A6 CPU and A4/E4 series to be released in the future. The 8-pin power input is positioned in the upper left. In the past, the AM2+/AM3 heat sink were also installed on the FM1 pin grid array. IO 1 X PS2 keyboard VGA/DVI/HDMI 4 X USB 2.0(black) 2 X USB 3.0(blue) 1 X RJ-45network hole

StreetFighter IV Benchmark 1280 X 720 => 98.62 FPS FINAL FANTASY XIV 1280 X 720 => 1818 LOST PLANET 2 DX9C 1280 X 720 TEST B - 10.3 fps A8-3850 built-in Radeon HD6550D and A6-3650 built-in Radeon HD6530D. The spec and performance is a little different. The test is higher level HD 6550D. 3D performance is close to entry level VGA card. This is impressive part for Llano APU. Game Benchmark is 3~4 times more than pervious 890GX. You can see the advantage for GPU leading company here. As Llano APU 3D performance, it’s decided by DDR3 clock and CPU FSB. If you set DDR3 at default 1066, 3DVANTAGE is about P3000. Comparing to DDR3 1866, 3DVANTAGE increase about 50%. Of course, APU 3D limitation is not only this. If you OC CPU FSB and higher DDR3 clock, it can be over P6000. GIGABYTE A75M-D2H Good 1. 100% Japanese made solid caps, USB 3x power design and On/Off Charge practical features. 2. Rich BIOS items for complete CPU/GPU/DDR3 OC requirement. 3. APU latest IGP technology. Most high end HD 6550D 3D performance is leaping. 4. Native 6 SATA3 and 2 USB 3.0. Also support 2 front USB 3.0 devices. 5. AMD Dual Graphics and Dual-link DVI is good for Gaming or multimedia. Weakness 1. CPU PWM has no heat sink. 2. 4 DIMM DDR3 will be better 2. AMD 3850/3650 + A75 MB price is too high currently Performance ★★★★★★★★☆☆ Components ★★★★★★★☆☆☆ Specification ★★★★★★★☆☆☆ Outlook ★★★★★★★★☆☆ C/P Value ★★★★★★★★☆☆ A8-3850 is 32nm which has been waiting for long time by AMD fans. 3D IGP performance is leaping and A75 also supports USB 3.0 and SATA3. It enhances AMD advantages in market competition. However, for CPU performance, power consumption and temperature is same after moving to 32nm. Also DDR3 bandwidth hasn’t been improved for years. AMD should improve these points soon. Maybe, the future 8 Cores CPU, Bulldozer, performance will be better. Llano APU built-in 3D performance is very impressive. It’s may stimulated by Intel which focus on IGP performance improvement in these 2 years. AMD(ATI) is one of the two GPU leading manufacturer. However, this time they cannot care entry VGA market profit this time. After merging ATI, AMD must improve IGP performance to lead in 3D field. OF course, you won’t disappoint the AMD GPU. HD6550D win back the IGP leading position again. As previous AMD is for mid to low end market, Llano APU price is higher. If you think it’s X4 955 and entry level VGA, APU platform is cheaper actually. As AMD just launched Llano APU structure in early July, it’s only A75 chipset now. It will have A55 chipset later. CPU is only A8-3800 and A6-3600. If A6-3600 plus A55 MB, the cost performance will be enhanced. Intel Sandy Bridge fires the IGP war, and AMD Llano APU hold the leading positon. Even though they both launched only high end product in the beginning, as new structure be more popular, you will see more mid to entry products. According to 2011 IGP performance war, I believe users are the final winners. You can get better performance system with lower price. This article is also post in my own blog WIND3C Welcome your visit.

Temperature(It’s 32℃ in room. All temperature is showed by ℃ in the test below.) Enter to OS Desktop - 34 CPU Full Speed - 65 Intel Burn Test v2.4，Stress Level Maximum For temperature, ADIA64 score is much lower and Core Temp only can test single core. So I pick ADIA64 and Easy Tune6 for standard. Personally, I feel Easy Tune6 is close to real temperature. Comparing to Athlon II or Phenom II I used to test, the temperature didn’t drop much as 32nm. It might causes by CPU integrated GPU. I hope the future yield rate can improve this issue. Power Consumption Test Enter to OS Desktop - 32W CPU Full Speed - 146W Intel Burn Test v2.4，Stress Level Maximum CPU and GPU Full Speed - 171W OCCT Power Supply Test BIOS enable Cool&Quiet technology and just enter to OS desktop, the temperature is very good. CPU full speed, it increases to 113W, and CPU/GPU full speed up to 134W. You can see GPU effect 26W. The main power consumption of whole system is CPU. Built-in GPU is not high as Desktop system. A75 chipset native support SATA3. SATA transfer rate is another key for AMD platform. I use CORSAIR Performance 3 Series, supports SATA3, for higher SSD performance. Model is P3-256. The size is common 2.5”. Capacity is 256GB and weight is 80 grams. Official spec, max reading is 480 MB/s and writing is 320 MB/s. (Test by ATTO Disk Benchmark) Silver shell and MTBF 150000 hours. 3 years warranty. ATTO DISK Benchmark over 32k test is very close to max performance. The reading is 482.5 Mb/s and writing is 328.2 MB/s. File Benchmark over 512k test the writing is over 300 MB/s. AS SSD Benchmark Seq Read - 438.20 MB/s Write - 288.70 MB/s CrystalDiskMark reading is 444.8 Mb/s and writing is 313.9 MB/s A75 SATA3 transfer rate is over standard. ATTO max Read/Write can achieve 438/328 MB/s. In other benchmark software, the scores are not far from official spec. You can see AMD chipset great leap in this part. It supports 6 SATA3 ports. It’s good for more and more SATA3 HDD or SSD products in the markret. AMD Radeon HD6550D Performance Test Default Clock 600MHz 3DMark Vantage => P4450

CPU is AMD latest A8-3850. There is another model, A6-3650, in the market. It’s also Llano APU structure. The core name is Llano, 32nm, TDP 100W, clock 2.90GHz and 4MB L2 Cache. In L2 cache, it’s higher than Phenom II 2MB, but it’s no L3 cache. Performance Test CPU 100.0 X 29 => 2900.1MHz DDR3 1866.8 CL8 10-8-27 1T GPU 600MHz Hyper PI 32M X 4 => 22m 43.084s CPUMARK 99 => 453 Nuclearus Multi Core => 12559 Fritz Chess Benchmark => 15.08/7240 CrystalMark 2004R3 => 204026 CINEBENCH R11.5 CPU => 3.46 pts CPU(Single Core) => 0.88 pts PCMark Vantage => 12276 Windows Experience Index - CPU 7.3 A8-3850 double L2 cache and no L3 cache compare to Phenom II. The performance is similar. Even though phase in 32nm is exciting for AMD fans. However, they might be upset for same performance. I do hope Bulldozer 8Core CPU can show users the leap improvement. DRAM Bandwidth DDR3 1333.4 CL6 6-6-18 1T ADIA64 Memory Read - 8088 MB/s Sandra Memory Bandwidth - 12640 MB/s MaXXMEM Memory-Copy -9455 MB/s DDR3 1600.2 CL7 8-7-24 1T ADIA64 Memory Read - 9229 MB/s Sandra Memory Bandwidth - 15266 MB/s MaXXMEM Memory-Copy - 10655 MB/s DDR3 1866.8 CL8 10-8-27 1T ADIA64 Memory Read - 9556 MB/s Sandra Memory Bandwidth - 16192 MB/s MaXXMEM Memory-Copy - 10979 MB/s FM1 DDR3 spec is 1.50V. CPU in default clock, memory can reach DDR3 1866 which is improved. You can see the performance improve most from DDR3 1333 to 1600. From 1600 to 1866, the improvement is much less. For DRAM bandwidth, it’s weak point for AMD for many years. They should try to solve this issue soon…

Even though AMD A75 chipset is mid end range, if MOSFET has heat sink will be better. A75 chipset heat sink is new design this year and the matte handle quality is good. A75M-D2H use Realtek ALC889 audio chip and GIGABYTE emphasize it’s 108dB SNR, Signal-to-Noise Ratio. If I get time, I will test it with other MBs ALC889 by high end speakers or headset to compare the difference. BIOS M.I.T. Menu Voltage Range DDR3 Voltage Control 1.025~1800V CPU VDDP Voltage Control 0.900~1.480V CPU Vcore 0.750~1.795V CPU NB VID Control -0.600~+0.300V CPU Voltage Control -0.600~+0.400V Memory Clock AM3 platform can support up to DDR3 1600. FM1 socket can pull clock to DDR3 1866. IGX Setup Items You can set share memory as 256, 512 or 1024MB. Dual-link DVI enable item can support highest resolution to 2560 x 1600. VGA Core Clock can adjust GPU clock. However, APU 3D performance is not pulling by this item. Advanced DDR3 Parameters Setup CORSAIR XMS 8GB DDR3 1600 CL9 OC to 1866 CL8 CPU related items, you can enter IGX setup page from here. PC Health Status BIOS items are similar with AM2+/AM3. If you used to use previous AMD platform, it won’t be problem to you. A75M-D2H provides plenty items for users to fine tune. Test Configuration CPU: AMD A8-3850 MB: GIGABYTE A75M-D2H DRAM: CORSAIR CMX8GX3M2A1600C9 VGA: AMD Radeon HD6550D HD: CORSAIR Performance 3 Series P3-256 SSD POWER: Thermaltake TR2 450W Cooler: Thermaltake Jing OS: Windows7 Ultimate 64bit SP1

Few months ago, AMD launched new product line called APU. It’s AMD dual core E-350 CPU which integrate GPU and the chipset is Hudson-M1 FCH. This combo main competitor is Intel Atom platform, and not popular in consumer market. E-350 CPU performance is similar to Atom, however, its GPU performance is much better than Atom which even better than AMD own IGP performance. It breaks the trend of IGP performance slow improvement. In the past, built-in graphic is integrating in Chipset. Since last year, Intel LGA 1155 platform integrated GPU into CPU. It makes Intel IGP performance catch up AMD platform. This is the first time Intel lead the IGP performance. AMD IGP is always the leader in the market. However, each generation 3D performance improvement is small; it causes the UMA market no breakthrough products. 2011, Intel LGA 1156 changed the trend. Early this year, Intel launched LGA1155 Sandy Bridge, the performance improves 20~30% more. Intel HD2000/3000 performance is leaping. Finally, AMD launched new Llano APU high performance IGP structure which is the key role this review. Llano APU chipset has A75 and A55. So far you only can find A75 products in Taiwan. First of all, let’s see GIGABYTE A75M-D2H package. It is white background and feature and technology logos. GIGABYTE recently promotes SUPER 4 in entry products. At the first sight, I thought it’s new Japanese Super Sentai series... GIGABYTE A75M-D2H SUPER 4 are Super Safe, Super Speed, Super Savings, and Super Sound. If you are interested in, you can search internet to compare with other brands A75 MB. This year, GIGABYTE mid to high end MB is black PCB design. Personally, I prefer water blue or sky blue color. A75M-D2H color tone is pretty good. It’s Micro ATX and size is 24.4 cmx 22.5 cm. Bundle Accessories Product manual, quick installation guide, I/O panel, SATA cables and Driver CD. Lower-Left Corner 2 X PCI-E X16, first one is X16 and second one is X4 which support AMD CrossFireX technology 1 X PCI-E X1 1 X PCI Realtek ALC889 audio chip supports 7.1 channel and High Definition Audio technology. Realtek RTL8111E LAN chip and TPM module connector in black. Lower-Right Corner 6 X Blue SATA provided by A75. It’s SATA3 and supports RAID 0, RAID 1, RAID 10, JBOD. Left side has USB 2.0/3.0 front panel connector and right side is DualBIOS. Upper-Right Corner 2 X DIMM DDR3 supports 1066/1333/1600/1866/2400(OC) and DDR3 max capacity is 32GB. Next is 24-PIN power connector. Upper-Left Corner AMD new Socket FM1 latch is different with AM2+/AM3, but the cooler is common in use. It’s 4+1 phase PWM. One phase can supply 30~40W, it’s really enough design. Upper-Left is 4-pin power input. IO 1 X PS2 KB/MS D-SUB/DVI/HDMI Output 1 X S/PDIF Fiber Output 4 X USB 2.0(Red) 2 X USB 3.0(Blue) 1 X RJ-45 LAN

G850 Enter to Desktop - 30W G620 OCCT POWER SUPPLY Mode - 64W G850 OCCT POWER SUPPLY Mode - 75W Comparing previous H55 plus G6950 platform, it’s 32nm CPU structure as well, however, G620/G850 Enter to Desktop is lower 8W and Full Speed is lower 30W. Sandy Bridge platform show it’s advantage in power consumption. These 2 CPU is testing under OC GPU, so the power consumption is higher than standard. Intel HD Graphics Performance Test 3DMark Vantage => P1292 3DMark Vantage => P1455 StreetFighter IV Benchmark 1280 X 720 => 30.98 FPS StreetFighter IV Benchmark 1280 X 720 => 34.22 FPS FINAL FANTASY XIV 1280 X 720 => 427 FINAL FANTASY XIV 1280 X 720 => 479 You can know Sandy Bridge built-in GPUs are HD Graphics 2000 and HD Graphics 3000. Pentium G series GPU named HD Graphics and follow 2000/3000. However, the clock is same, 850/1100, so 3D performance is same as HD2000. For GPU OC, G850 clock can be higher to achieve 1433MHz and pass 3D tset. Comparing to previous LGA 1156 GPU, 3D performance is much higher. This is the other new platform strength. Intel launched 3 Pentium CPU this time and price is lower than Core i3-2100. MB you also can have Micro ATX H61 or H67. It’s good small PC or office PC. ITX boards are getting more and more in the market. Mainly, the chipset is H61 or H67. Usually, HTPC hardware price is higher. More and more makers sell ITX MB in these 2 years. HTPC users can lower the HTPC budget to normal ATX or Micro ATX PC. H67N-USB3-B3 features and expansion are quite good. The rich BIOS features can OC GPU, fine tune DDR3 parameters, also can lower CPU voltage to make system be good performance and green. I also saw some H61N-USB3 news in website. I think the C/P value will be even higher. The article is sharing Intel latest entry level CPU mainly. You can see the some improvement of CPU performance and power consumption. However, the DDR3 bandwidth and 3D performance improvement is bigger. Recently, H61 price is lower and many MB makers start to kill price. No matter ITX or Micro ATX platform, you can take it as your reference. I wish everyone can use lower budget to build higher performance next generation PC. This article is also in my blog. WIND3C Welcome your visit.

Windows Experience Index - CPU 6.4 Windows Experience Index - CPU 6.7 Windows Experience Index is simple and fast benchmark software. User can understand system performance easily. G620 and G850 are both 2 Cores CPU, but get good score. I hope this software can revise to be more detail test. As my experience, i3 3.3GHz is 7.1分, i5 2.8GHz s 7.4, i5 3.3GHz is 7.7 and i5 4.8GHz is 7.7 I think the full score should be 7.9. If the hardware performance is good enough the index over 7 is difficult to make a distinction. DDR3 Test G620 DDR3 1064.2 CL5 6-5-18 1T ADIA64 Memory Read - 13035 MB/s Sandra Memory Bandwidth - 13849 MB/s MaXXMEM Memory-Copy - 14196 MB/s G850 DDR3 1330.4 CL6 7-6-18 1T ADIA64 Memory Read - 15553 MB/s Sandra Memory Bandwidth - 17250 MB/s MaXXMEM Memory-Copy - 16723 MB/s LGA 1156 CPU, built-in GPU, causes low DDR3 performance. LGA 1155 Sandy Bridge platform has corrected this issue. The bandwidth is higher 10~20% than previous platform. BIOS setup is AUTO. G620 is DDR3 1066 and G850 is DDR3 1333, which are the official spec. Actually, if you use high end chipset, the DDR3 clock can be even higher. G620 DDR3 1872.2 CL9 9-8-24 1T ADIA64 Memory Read - 16638 MB/s Sandra Memory Bandwidth - 21978 MB/s MaXXMEM Memory-Copy - 16605 MB/s P67/Z68 OC series chipset, if your DDR3 quality is good enough, you can set as 1866 or 2133. Due to G series is entry model for higher C/P or office PC, most users won’t consider to use P67/Z68. I just would like to show entry CPU has higher DDR3 performance, so changing to Z68 here to share G620 high DDR3 bandwidth. Temperature G620 Enter to Desktop - 31~32(Air condition 27℃) G850 Enter to Desktop - 42(Room temperature 33℃) G620 Full Speed - 49~52(Air condition 27℃) Intel Burn Test v2.4，Stress Level Maximum G850 Full Speed - 53~56(Room temperature 33℃) Intel Burn Test v2.4，Stress Level Maximum As these 2 CPUs clock are similar, so I make different environment to share the temperature test. In hot summer, the highest temperature in open case is 56 and G850 in chassis is around 60. If you change to air conditioning room, G850 full speed is 50~52 which is higher 2~3 than G620. The temperature is key advantage for 32nm CPU. Power Consumption G620 Enter to Desktop - 30W

Performance Test G620 CPU 99.8 X 26 => 2594.1MHz DDR3 1064.2 CL5 6-5-18 1T GPU 1100 OC 1350MHz G850 CPU 99.8 X 29 => 2893.4MHz DDR3 1330.4 CL6 7-6-18 1T GPU 1100 OC 1433MHz Below is using same benchmark software. The first is G620 and second is G850. Hyper PI 32M X 2 => 13m 31.794s CPUMARK 99 => 400 Hyper PI 32M X 2 => 12m 07.461s CPUMARK 99 => 446 CPUMARK is simple benchmark for testing CPU single core performance. Hyper PI can test system stability and multi-core software. Currently, Intel structure is faster. G620/G850 clock difference is 300MHz, so the performance is 10% gap. Nuclearus Multi Core => 9059 Fritz Chess Benchmark => 8.08/3876 Nuclearus Multi Core => 10055 Fritz Chess Benchmark => 9.01/4324 You can see the performance under Fritz Chess Benchmark is around 10%. CrystalMark 2004R3 => 148458 CrystalMark 2004R3 => 162889 CINEBENCH R11.5 CPU => 2.01 pts CPU(Single Core) => 1.04 pts CINEBENCH R11.5 CPU => 2.25 pts CPU(Single Core) => 1.15 pts PCMark Vantage => 8598 PCMark Vantage => 10325 It’s the H67 platform and I use CORSAIR F40 SSD to set RAID0. You can see the advantage. Comparing to last generation Intel Pentium Dual Core G6950, both are 32nm, G series performance is improved. Even though CPU performance is not big improved, there are many parity price H61/H67 MB in the market. For new generation office PC, C/P is better. Also, Sandy Bridge has more features.

Boot Screen with some feature logos. BIOS Main Page MB Intelligent Tweaker, called M.I.T. Clock Setup Page CPUFSB/Ratio, DDR3 Clock, Integrated Graphics clock… CPU Related Items You can adjust CPU cores and C1E is power saving feature. DDR3 Clock Setup Advanced DDR3 Parameters Setup Voltage Page CPU Vcore 0.750~1.795V QPI/Vtt Voltage 0.900~1.500V Graphics Core 0.850~+1.705V PCH Core 0.900~1.500V DRAM 1.100~2.400V PC Health Status H67 is setting as IGP chipset, so it cannot OC as 2500K/2600K. GIGABYE H67N-USB3-B3 BIOS provides rich features for end users to adjust. For non-K i3, i5 or Pentium CPU, you can fine tune system by CPU voltage, DRAM parameters and GPU clock OC. Test Configuration CPU: Intel Pentium Processor G620/G850 MB: GIGABYTE H67N-USB3-B3 DRAM: CORSAIR CMX8GX3M2A1600C9 VGA: Integrated Graphics HD: CORSAIR CSSD-F40GB2 Raid0 POWER: Thermaltake TR2 450W Cooler: Intel Cooler OS: Windows7 Ultimate 64bit

Jan 2011, Intel launched new structure LGA 1155 socket Sandy Bridge platform. In the first stage, Intel focus on mid to high Core i5 and Core i7. The chipset is P67 and H67 which are also positioning at mid end. For the entry product, Intel launched H61 and Core i3 in March. In the market, Core i3 2100 or 2120 price is still high for entry CPU segment. For the users, lower than 100USD Sandy Bridge CPU is more suitable for office or worksheet users. LGA 1156 has Pentium G6950 which is cheaper and enough performance. However, in Taiwan, G6950 is the only model for 1156 platform. Sandy Bridge launched 3 Pentium CPU, G620/G840/G850, in May. It’s better structure and richer options for Intel platform CPU. First of all, let’s check Intel Pentium Processor G620, the clock is 2.6GHz. Physical 2 Cores don’t support Hyper-Threading. It can support 2 threads which called 2C/2T. 32nm and Max TDP is 65W. L3 Cache is 3MB. It’s most entry 2 Cores CPU in LGA 1155. Lower-Left is G620 back. Upper-Right is bottom of original cooler for 2 Cores CPU. Intel uses non0-sooper design. Another more high end G series, Intel Pentium Processor G850, clock is 2.9GHz. The other specs are same as G620. The main difference of G850 and i3 -2100 are Hyper-Threading and Intel GPU version. For advanced technology, i3 can support more. You can find the detail in official spec documents. G850 back is same as G620. Besides the clock, G850 DDR3 also support to 1333. It owns Embedded Options Available. These are key difference between G620 and G850. MB is H67 chipset, GIGABYTE H67N-USB3-B3. The price is not high as ITX and H67. Also the components and features are good quality. C/P is acceptable. CPU is 4+1 phases PWM. 2 DIMM DDR3 support max 16GB. It built-in 2 SATA3 and 1 PCI-E X16 which is quite good spec for ITX IO 1 X S/PDIF fiber and coaxial output 1 X D-SUB/2 X HDMI 4 X USB 2.0(Black) 2 X USB 3.0(Vlue) 1 X RJ-45 LAN 1 X eSATA(Yellow) USB 3.0 chip uses NEC D720200AF1, made in Japan. If GIGABYTE can add Bluetooth and mSATA wireless slot will be more practical for HTPC. AMD normally has more advantage in ITX MB. Some brands have many product lines but not easy to find in Taiwan. Since Intel IGP performance is much improved, H55 and H67/H61 all have many ITX MB. The price is cheaper than AMD ITX MB. Users can have more options within budget. H67N-USB3-B3 price is higher than entry H61 boards. Normally, USB 3.0 H61/H67 is around 10~15 USD higher. If you don’t need RAID or higher spec, you can choose H61 to have better C/P.