The Motherboard Memory Lane series returns today with a look at the Intel X79 platform and the era of Sandy Bridge-E and Ivy Bridge-E architecture processors. The X79 platform was in fact the first update Intel’s High-Performance Desktop (HEDT) segment since the launch of the aging Intel X58 platform. As usual we will examine the platform itself, the most popular motherboards and CPUs of that particular era, and the record scores that were made.
The Intel HEDT platform became wholly refreshed with the arrival of the X79 platform. HEDT describes a high-end, high-priced offering for enthusiasts that simply want the most powerful system that money can buy. The platform included the promise of hexa-core computing with enough PCIe lanes to support maxed out multi-GPU configurations, plus quad-channel memory.
We continue our Motherboard Memory Lanes series today with a look at the Intel Z87 platform, a launch that coincided with a revised socket design and a brand new Haswell architecture processor lineup. We’ll focus in on the new technologies that the platform included, the motherboards and CPUs that were popular with overclockers at that time on HWBOT and of course, the records scores that were made in this particular era.
Officially launched in June 2013, the Intel Z87 platform continued the big cat codenames that were used with previous Cougar Point and Panther Point platforms. The new Lynx Point platform arrived with similar enterprise and budget offerings that included Intel H81, B85, Q85, Q87, H87 and Z87 PCH parts. As with previous Sandy Bridge and Ivy Bridge designs, integrated graphics, video outputs and memory controllers were all integrated into the CPU itself. As with Ivy Bridge, Haswell CPUs offered support for up to three displays (digital outputs direct from the CPU, VGA from the PCH itself). In terms of memory, the new platform supported dual channel DDR3 at default speeds of up to 1,600MHz and also supported low power DDR3L. Most enthusiast systems were capable of DDR3-2400 and above straight out of the box.
We return today with our Motherboard Memory Lane series here on HWBOT, this time with our sights clearly set on the Intel Z77 platform and the launch of the Intel Ivy Bridge series processor lineup. Let’s take a look at the new technologies that the platform brought to the world, the motherboards and CPUs that were most popular with overclockers at that time and the records that were broken in this particular era.
The Intel 7-series platform was launched in April of 2012, replacing the 6-series family of chipsets that had arrived almost a year earlier. In terms of platform codenames, Panther Point replaced Cougar Point. In reality however from the perspective of the two PCH chips they were actually pretty similar. Although Panther Point arrived with a new 2nd generation Ivy Bridge Core architecture processor line up, all 7-series boards used the same LGA 1155 socket and supported previous generation Sandy Bridge processors. Likewise, customers looking to try the Ivy Bridge silicon could stick with their old Z68 or even P67 board with just a simple BIOS update.
Today we continue with our Motherboard Memory Lane series, taking a look at older Intel chipsets and processor platforms, the motherboards and processors that were popular and the benchmark records broken in that era. Today we turn our attention to a chipset that could well be described as the strange uncle of the Intel chipset series. The Intel Z68 platform was a slightly unusual platform launch in that it didn't actually coincide with a new processor series launch. Let’s take a look in a little more detail:
The Intel Z68 Express Platform Hub Controller, to give it its full title, was launched on May 11th of 2011, just four months after its predecessor the Intel P67. It used the same LGA 1155 socket and supported the same 2nd Generation Intel Core Sandy Bridge architecture processors. To better understand the Z68 platform, let’s first examine its predecessor, the Intel P67.
If we look again at the Intel P67 and its more affordable alternative, the Intel H67, we can see that the P67 supported CPU overclocking, while the H67 did not. The only other difference is the fact the P67 could also split its PCIe lanes in two 8x lanes for more effective multi-GPU configurations. One area however where the H67 excelled however, was the fact that it also offered support for Intel’s integrated HD Graphics. No P67 motherboards featured video outputs on the back panel, a fact that denied enthusiast customers the option of accessing a GPU that was present on all Sandy Bridge processors. Intel’s logic was that P67 customers that are attracted to the idea of overclocking multi-GPU configurations, would not require integrated graphics.
Today we roll out the eighth edition in our Mother Memory Lane series, this time focusing on the Intel P67 platform. Dubbed the 2nd Generation Intel Core processor platform, the P67 chipset arrived alongside the new and shiny Sandy Bridge architecture CPUs, probably the biggest game change in processor design that Intel had experienced since the arrival of Conroe several years earlier. In terms of Overclocking, the P67 platform saw Intel offer ‘unlocked’ K-SKU processors for the first time, another major shift. Today we’re going to take a look at the P67 platform itself, the most popular motherboards and processors of that particular generation and the record scores that were made around that time.
The Intel P67 chipset was launched on January 2011 and was codenamed Cougar Point. Like its predecessor, the P55 chipset, it was a single chip solution technically referred to as a PCH, or Platform Controller Hub. Cougar Point included several PCH options of which the P67 variant was deemed the ‘Premium’ offering. Other 6-series PCH chips in the Cougar Point family included H61, B65, Q65 and H67. Being the premium PCH offering targeting enthusiasts, the P67 PCH was in fact the only variant that offered full CPU overclocking, provided you also had a K-SKU Core i7 or i5 processor. Indeed this was the first time that overclocking was embraced by Intel as an enthusiast feature, a feature used for the first time in both platform and CPU-level marketing.
In Week 17 of 2017, we received 3701 benchmark results from 1016 registered overclockers around the world. The majority of the submissions is coming from Rookie overclockers representing 60% of the active community. They were responsible for 40% of the submissions. We had a peek at the most valuable submissions in a breakdown per league.
During Week 17 of this year we have a total of 5 overclockers with a golden cup in the Most Valuable Submission leaderboard. First in line is Splave from the United States with a Global First Place in the XTU 2xCPU benchmark. The American pushed his Core i3 7350K to 6700 MHz and scored 1216 marks. That is two more than runner-up Rsannino. Next up is the inevitable Dancop from Germany. The Number One overclocker in the world pushed the NVIDIA Titan Xp to 2265/1624 Mhz and combined with a 7 GHz Core i7 7700K this resulted in a new 3DMark03 World Record. Dancop scored353660 marks, which is more than 500 points higher than Smoke. Then we come to the Hardware First Place results. Oldscarface from the Netherlands grabbed gold with the Pentium E2160 in the CPU Frequency benchmark. His chip reached 568.43 MHz FSB resulting in an overall CPU frequency of 5115.89 MHz. Next in line with a HFP is Scannick from Italy, who scored a global cup in the Radeon HD 4870 3DMark01 ranking. Combining a 6.8 GHz Core i7 7700K and a graphics card overclocked to 1150/1050 MHz results in a score of 156998 marks. Last but not least we also celebrate Samsarulz from Chile who scores a golden cup in the Radeon HD 7950 3DMark11 Performance benchmkark. The graphics card is overclocked to 1300/1800 MHz and is combined with an Intel Core i7 5960X Haswell-E processor. Congratulations to everyone making the leaderboard!
The overclocking results submitted during Week 17 generated in total 245 World Record Points, 7774.6 Global Points, and 8570.4 Hardware Points. The distribution per League is as follows: 22% for Elite, 38% for Extreme, 14% for Apprentice, 14% for Enthusiast, 5% for Novice, and 21% for Rookie. The representation of the active community is as follows: 2% Elite, 8% Extreme, 4% Apprentice, 18% Enthusiast, 8% Novice, and 60% Rookie.
The MSI Dragon Squad is an exclusive group of social media personalities that MSI describes as brand ambassadors. As a member of the MSI Dragon squad you get exclusive access to the latest MSI products and a chance to create some exclusive content. Dragon Squad member Blunty was fortunate to get the chance to visit AMD Headquarters in Austin Texas. One section of the trip involved a presentation from AMD executives, including a briefing with Robert Hallock, AMD’s Evangelist for Processor Technologies. The briefing involved a really interesting look at overclocking memory on the new AMD Ryzen platform.
Robert kicks off my explaining what he describes as some ‘absurdly named BIOS options’ that may well seem unintelligible to even the most ardent enthusiasts. These include ProODT, which stands for Processor On-Die Termination –it’s a value that determines whether or not, or indeed when the electrical signal between memory and the CPU should be terminated. Robert then recommends somewhere between 40-60ohms as ideal ProODT settings. The setting is there to aid stability with certain memory DIMMs.
One other area that Robert explores is the Ryzen architecture in terms of voltages and the Uncore settings (also known as SOC) which concerns things like the PCIe bus, memory controller, USB etc. Altering the Vcore settings adjust only the voltage for the Zen cores themselves, not the other aspects of the chip that reside in the SOC voltage. He also explains that raising the SOC voltage to 1.1v can dramatically improve high speed memory stability.
When it comes to choosing a good quality thermal paste there are certainly plenty of options with tons of different price points and specialist features. Some are designed with serious subzero overclocking mind, while some are more suited to everyday systems that are water or air cooled. For most Overclockers and enthusiasts it can be case of simply going with the one that your peers recommend, or the one that suits your budget. Gavin Bonshor (known to HWBOT members as gavbon) writing for Play3r.net decided to separate myth from reality and embarked on a mission to find out the truth – the result is an exhaustive study of 26 different thermal paste products. Over to Gavin:
What is the best thermal paste to use? - It really depends on the type of cooling you are using. My recommendations are different based on if you’re using conventional cooling methods such as air, water and even sub-zero; liquid nitrogen and dry ice. What I will say though is for air and water cooling, most of the ‘decent’ aftermarket thermal pastes will have a couple of degrees Celsius between them…aside from those with high thermal conductivity ratings such as liquid metal.
Different types of thermal paste consist of different materials; some are better than others in terms of thermal conductivity. A good example is liquid metal against ceramic based pastes. The liquid metal not only plugs the gaps better between the IHS and the CPU cooler, but it also features better conductivity meaning more heat can be transferred per application than other types used. The drawback to liquid metal paste however is the application can be messy, the paste is more expensive and overall, some people just prefer to go with the easiest route; I don’t blame them in all honesty!
The Poitiers 2017 leg of the HWBOT World Tour took place a few weeks ago, bringing overclocking workshops and contests to Gamers Assembly in France. Arguably the highlight of the event was the Overclocking World Championship Poitiers 2017 Qualifier which invited Europe’s most feared and respected extreme overclockers to compete for a place in the OCWC Final at the end of the year. The great news today is that you can now relive the grand final between Germany’s Dancop and France’s Niuulh thanks to OverClocking-TV who as always, were at ground zero hosting a live stream.
The OCWC17 Poitiers Qualifier got underway with a qualification segment where twenty extreme overlockers competed on three stages; SuperPi 32M, Cinebench R15 and 3DMark11 Physics. The top four qualified for the semi-finals and were invited back the next day – Wizerty, Dancop orion24 and Niuulh made the cut. Semi-final 1v1 matches were held, each lasting 30 minutes using randomly drawn Core i7 7700K CPUs. Benchmarks were also drawn at random with each Overclocker possessing one veto to avoid an unfavorable selection. Semi-Final matches went ahead and saw French No.1 lose out to outsider Niuulh who is ranked 3rd in his native France and 42nd globally. Meanwhile Dancop showed his class in live overclocking contests by beating orion24. The grand final was setup.
You can watch a replay of the Final between Dacncop and Niuulh here on the OverClocking-TV YouTube channel. As well as the final itself the video also includes interviews with both overclockers plus some great commentary from host Trouff and Buildzoid who offers technical summaries of all the action. Enjoy!
Once we again we revisit a news post from the past that evokes happy Overclocking memories. This week we take a peek behind the curtain of time to recall a meetup organized by Team Finland, a story we covered in April 2010. The meetup was essentially an Extreme Overclocking workshop / try out which included overclockers SF3D, Asmola, Junksu, hOLIC, SeMbEr and Koneksi. Here’s what we wrote back on April 9th 2010:
“It’s definitely not easy getting started with extreme overclocking, or overclocking for benchmark scores for that matter. In the old days, when men were still men and most of us still playing Counter Strike getting started with extreme overclocking meant having the guts to order a couple kilos of Dry Ice, killing a couple of boards because of the lack of insulation and, most of all, spending hours and hours reading about extreme overclocking. Nowadays, experienced overclockers such as SF3D appear on local team sessions to meet up with the upcoming guarde to show how LN2 works and let them get familiar with the ‘stuff’.”
[Press Release] GIGABYTE TECHNOLOGY Co. Ltd., a leading manufacturer of motherboards and graphics cards, is excited to announce the start of Summer Spectacular 2017, the third of four contests in GIGABYTE’s 2017 overclocking season. A total prize pool of over $2,500 USD in enthusiast-grade hardware is available for the top three contestants in both Ambient and Extreme cooling categories along with six “Lucky Draw” winners.
With this contest restricted to only Intel® Core i3-7350K CPUs and NVIDIA® GT730 graphics cards, the playing fields are evened as participants look to push budget hardware to their limits! For those just learning the ins and outs of overclocking—a lucky draw is also part of this competition, where prizes including two GIGABYTE Z270X-UD3 are up for grabs for participants who submit scores in all stages of the competition. Prize info and event rules for Summer Spectacular 2017 are listed below:
Summer Spectacular 2017: May 1st, 2017 – May 31th, 2017
AMD Ryzen platform has been with us for a few months now. Although motherboard BIOSes continue to be updated and improved while enthusiasts and overclockers also learn more about the subtle nuances involved with pushing Ryzen CPUs, there’s still plenty to learn. In fact when it comes to different CPU models and different motherboards from different vendors AMD Ryzen can be a challenge – all of which is the reason why Shawn Jennings from Overclockers.com (known to many of us on HWBOT as Johan45) has put together a comprehensive Ryzen Overclocking guide.
The guide contains of all the findings and data garnered from weeks and testing and reviewing, making it a really solid resource for anyone who really wants to improve their game with AMD Ryzen processors. Here’s a sample of the introduction from Shawn in which he outlines a few of the challenges that Ryzen overclocking can involve:
Overclocking AMD Ryzen in General
I’d like to start this by saying overclocking is quite unique to the system/CPU. Different motherboards, memory, and CPUs will yield different results. There really are no “plug and play” settings for all systems. This is a process that needs to be done slowly and methodically with a lot of testing along the way. This is really the only way to assure you have decent stability and won’t corrupt your operating system over time.
One observation I have made is the majority of Ryzen 7 CPUs will run at 3.8 GHz with 1.35 V or less and the CPU “binning” appears to be quite parallel to the model. Meaning the best chance of the high Core clocks would be with the 1800x then the 1700x and finally the 1700. Now that doesn’t mean there aren’t Ryzen 7 1700 CPUs that will easily do 4.0 GHz at 1.35 V because there are but the odds are not in your favor. Realistically, you should expect to see 3.8-3.9 GHz for 24/7 operation on a Ryzen 7 1700.
Just a few weeks ago we witnessed the conclusion Round 1 of the Road to Pro Challenger Divisions here on OC-ESPORTS. Round 1 was as competitive as ever and drew good participation numbers across the Divisions as a whole. Be sure to check out our roundup articles with all the winners, scores and more here on the OC-ESPORTS newsfeed. Today we look forward to Round 2 which gets underway in just a matter of days.
Road to Pro Challenger Series Round 2: May 1st to June 30th 2017
The Road to Pro Challenger series is arguably the most comprehensive Overclocking contest ever. It spans a total of eight Divisions and includes challenges on a broad range of hardware including Intel, AMD and Nvidia plus classic hardware platforms from the past. The idea behind the series is very simple - it offers something for everyone. A place where we can all find our niche in terms of a preferred platform and available resources. Once you know what your focus is, you can then compete head to head against the best there is.
Every season the Challenger Series consists of three rounds, each lasting two months. Round 2 of the 2017 series runs from May 1st and ends on June 30th. Let’s take a look at the stages and benchmarks involved with each of the Divisions in Round 2.
In the last week or so there have been reports in the media that it is possible to give your old RX 480 card a boost in clock speeds by flashing your card’s BIOS with a new one designed to be used with the newly launched RX 580 cards. This is not a new phenomenon of course – overclockers have been unlocking cores and upping clock speeds and more since television was called books. But as outlined by this article from ExtremeTech’s Joel Hruska using a non-verified BIOS can be a source of agony as well as joy, if things go wrong. The good news for anyone considering flashing his RX 4-series card for a new 5-series card is that Buildzoid has plenty of thoughts on the matter, all which he has just shared with on his YouTube channel.
The video kicks off with advice about flashing a RX 470 card with a new RX 570 BIOS. It will not work. However it is possible to flash a RX 480 card with a RX 570 BIOS. The problem is that you will inadvertently have lost some shaders, an outcome that again is most unwelcome. So as a general rule, -70 cards should only ever be flashed with -70 BIOSes. The same goes for -80 cards.
The second question raised is - Which BIOSes work on which cards from which vendor? The answer is very hard to establish. One thing that does make things easier is having a dual BIOS card. Of course this allows you to experiment and and find out which one works without actually bricking your card. The key rule to follow is that you should always flash just one of the two BIOSes – seems pretty self explanatory and obvious, but yes. Point well made.
A few weeks ago South Africa’s No.1 ranked overclocker DrWeez got his hands on an ASUS ROG Maximus IX Apex motherboard, the perfect platform for him to get more familiar with the latest Intel Kaby Lake architecture CPUs. Last week he returned to YouTube screens with Overclocking Session #52 and an effort to push past the limitations he encountered in the previous session. In Session #51 Andrew managed to push his Core i7 7700k to 6.7GHz, a limitation that hoped to push past in session #52.
To kick off the session Andrew loads up a profile created by Russian overclocker Slamms, with a few alterations to the CPU core voltage settings – just to rein it in a touch. Next he turns off HyperThreading and also turns off two of the processors cores. In terms of benchmarks he aims to run a series of single threaded 2D benchmarks including SuperPi, PiFast and others with two cores and two threads and a target clock of around 6.9GHz – 200MHz above the highest clocks achieved in the previous session.