Intel "Sky Lake" - Core i5/i7 6XXX K series

[Jorge-Vieira]

Overclockers Crank World-Record Speeds out of Skylake Core i7-6700K

Two renowned overclockers have pushed Intel’s Core i7-6700K CPU to its absolute limit, breaking the world record in the process. The German pair, Der8auer and Dancop, overclocked the retail Skylake processor to hit 7007.85MHz, using an ASUS ROG Maximus VIII Gene motherboard and 2x4GB G.Skill DDR4 RAM, TechFrag reports.
The validated data, submitted to CPU-Z, shows that the Germans raised the multiplier to 69x, using a base clock frequency of 101.56MHz, while the overclocking was achieved using only one of the CPU’s four cores enabled and an uncore frequency of 2438MHz.

The overclocking yielded 4.8GHz using air cooling, but, when they implemented -190^oC cooling, managed to raise speeds up to 7007.85MHz.

The overclocking is a new world record, breaking the previous score of 6998.88MHz achieved by Chi-Kui Lam.

Processor (CPU)

CPU Name: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
Threading: 1 CPU – 1 Core – 1 Thread
Frequency: 7007.85 MHz (69 * 101.56 MHz) – Uncore: 2437.5 MHz
Multiplier: Current: 69 / Min: 8 / Max: 50
Architecture: Skylake / R0-Step (14 nm)
Cpuid / Ext.: 6.E.3 / 6.5E
IA Extensions: MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, EM64T, VT-x, AES, AVX, AVX2, FMA3, TSX
Caches: L1D : 32 KB / L2 : 256 KB / L3 : 8192 KB
Caches Assoc.: L1D : 8-way / L2 : 4-way / L3 : 16-way
Microcode: Rev. 0x0000023
TDP / Vcore: 95 Watts / 2.064 Volts
Type: Retail (Original Frequency : 4000 MHz)

Noticia:
http://www.eteknix.com/overclockers-...core-i7-6700k/

[Jorge-Vieira]

Intel asks for non-K Skylake overclocking capabilities to be pulled

Stop it now, it is too damn useful
There have been rumours that Chipzilla was leaning on motherboard vendors to pull overclocking of non-K Skylake processor capabilities from their feature lists. Now it is starting to look like it is true.

Hexus has reported that ASRock's has removed its SKY OC feature in a BIOS update that appeared yesterday.
ASRock, MSI, Biostar, ASUS and Gigabyte motherboards started putting the feature in their cards in December last year. ASRock and Biostar promoted it. Biostar even called its non-K overclocking HyperOC.
Other vendors seemed pretty certain that they would get an angry call from Chipzilla eventually and did not mention the new tech.
The performance uplift possible via this BCLK overclocking method was impressive. They could make a Core i3-6100 perform faster in Cinebench than the Core i5-4430, even when all four cores were flat out. In other tests, including gaming, these two chips were pretty much the same.
Of course that was never going to sit well with Intel as it charges rather a lot more for its i5 chips than the i3.
ASRock has updated its Intel Z170 motherboards with two tweaks; 1. Update CPU microcode to 0x76, and 2. Remove SKY OC function. It did not say why.
If you want to keep the tech there should be no pressure for you to upgrade. An earlier BIOS update from ASRock, just a fortnight ago, included a microcode to fix the Skylake complex workloads bug and otherwise there is little need for an upgrade.
You can even see the older BIOS versions on the ASRock website.
Our guess is that there will be other motherboard vendors who will be following ASRock soon.

Noticia:
http://www.fudzilla.com/news/motherb...s-to-be-pulled


Parece que a festa vai mesmo acabar, resta ver quem são os proximos a retirar as BIOS antigas.

[Enzo]

Certamente já deve de haver uma data de cópias delas a circular por ai pela Web. Foi bom enquanto durou

[Jorge-Vieira]

Intel i3 6100 Overclocking? Worth it?

Introduction


A couple of months ago we reported on ASRock releasing their SkyOC bios that allowed for overclocking non K sku Skylake cpu’s by BCLK overclocking. This set the overclocking community into a bit of a frenzy as it’s the been the first time since K chips were introduced that large gains in clock frequency have been possible off of BCLK tweaks. Yes, with Sandy Bridge through Haswell small bumps were possible off the 100mhz BCLK but only a small amount, nothing like what we’ve seen come out of a Skylake cpu on the Z170 platform.
Unfortunately Intel doesn’t seem so keen on this venture and has began the process of having Vendors, once again starting with ASRock, to pull the UEFI’s and release updates that lock out the ability to push the BCLK to extremes. Citing concerns such as the loss of temperature monitoring as one of their reasons to ‘protect’ the end user. However, there’s nothing forcing the consumer to replace this BETA overclocking friendly UEFI with their new locked down version so it’ll be up to the individual to do so for now we’re beginning to see repositories spring up in various forums to ensure these BETA UEFI’s aren’t lost to the void.

Advertisements


I personally love overclocking for fun, nothing extreme but I like to see how far I can take something on traditional cooling and find what benefits it yields. So once I saw this becoming a thing I picked up a Z170 board off of a forum sale and decided to order up an i3-6100 to toy around with myself. I considered an i5-6400, but in my personal opinion having an unlocked i5 available it simply seemed worth it spring for that and not have to fight any of the shortcomings that these BETA UEFI’s bring along, most notable being temperature monitoring. And, with no unlocked i3 being available it felt right to me. A low cost ($129) cpu with an already high clock speed of 3.7ghz would require very little work on the BCLK side of things. So I installed the MSI BETA UEFI on my Z170 XPower Gaming motherboard and found out if it was really worth the hassle and trouble.

Review:
http://wccftech.com/review/intel-i3-...locking-worth/

[Jorge-Vieira]

Intel Confirms: Overclocking Non-K Skylake Processors Not Recommended – Motherboard Vendors To Voluntarily Issue BIOS Update

Last week, we reported that Intel might disable overclocking support on Non-K Skylake processors. The guys over at PCWorld seem to have an official update on the matter directly from Intel. Several motherboard AIBs have started rolling out latest firmware which disables overclocking support which was previously enabled to offer overclocking support to users running Non-K processors and Intel has now officially stated why they have restricted overclocking on such processors because they aren’t designed to be overclocked.

Intel Responds On Skylake Non-K Overclocking Support: “Company does not warranty the operation of the processor beyond its specifications”

When we heard rumors that Intel was disabling overclocking support on Non-K processors, it reminded us of a similar case that happened with Haswell processors when motherboard vendors pushed out new BIOS updates that allowed overclocking of unlocked Haswell processors across all motherboards. Intel later released updates that disabled overclocking support and limited it to their Z87 chipset based boards. This feature was enabled once again when Intel launched Haswell Refresh generation and several 9-Series motherboards allowed over-clock for Devil’s Canyon processors.
Back in 2015, Intel faced a similar situation where motherboard vendors allowed overclocking support on Non-K Skylake chips soon after they launched in the market. This allowed users to overclock their Celeron, Pentium, Core i3 and Core i5 chips on a support Z170 motherboards. ASRock, one of Intel’s partners, went one step ahead by offering their SkyOC feature that allowed not only Non-K overclocking support but also supported overclocking across their entire 100-series motherboard family (Non-Z OC).

Folks running low-cost and budget oriented Skylake chips were really enjoying overclocking support for these processors as they allowed better performance at the fraction of a cost compared to a high-end K-Series Core i7 or Core i5 processor. A review we posted just yesterday showed the significant gains a Core i3-6100 CPU had delivered once it was overclocked.
Advertisements

However, after months of overclocking support, ASRock was the first to disable their SkyOC feature after the reports emerged that Intel could forcefully try to ban the unofficial overclocking support. Now, PCWorld, in an exclusive interview with Intel has received an update on the matter:

”Intel regularly issues updates for our processors which our partners voluntarily incorporate into their BIOS,” an Intel spokesman said. “The latest update provided to partners includes, among other things, code that aligns with the position that we do not recommend overclocking processors that have not been designed to do so. Additionally, Intel does not warranty the operation of the processor beyond its specifications.” via PCWorld

It seems like the update isn’t being forced down to motherboard vendors but is voluntarily. This means that vendors have the choice to either release the latest firmware to disable overclocking support as of whole or let the audience overclock their chips but in that case, Intel won’t accept any warranty as they have officially stated that they don’t warranty any processor that is running beyond its specification unless or unit it is intended to do so. The K-Series processors from Intel are the only series which come with official overclock support. Anything aside from these wouldn’t apply for warranty from Intel and I think this much a user already knows.





ASRock is the first vendor to disable overclocking support and it remains to be seen just how many more vendors voluntarily accept Intel’s update. There are still a couple of board vendors in the market that allow overclocking support for Non-K processors including MSI and EVGA. In any case, if you have the older BIOS which allows overclocking and you want that support, then simply don’t update to the latest firmware. But do note that your processor won’t be applicable for warranty if it is affected during the overclocking procedure.
An Intel Core i3-6100 CPU Overclocked to 4.5 GHz on an MSI Z170A Motherboard:

Noticia:
http://wccftech.com/intel-confirms-o...s-bios-update/


A Intel a fazer valer a sus posição dominante!!!

[Enzo]

Mesmo...parece que não lhes compensa ter mais i3 à venda que i5/7-k

[Jorge-Vieira]

Intel closes base clock loophole in latest microcode update

Intel has confirmed that it has removed the accidentally-introduced ability to modify the base clock of non-K series processors in its latest microcode update.









Intel is officially closing the loophole that led to the ability to modify the base clock on non-K series processors, issuing a microcode update that will once again lock the base clock of all affected chips.

In recent years, Intel has restricted support for overclocking its chips to selected models aimed at the enthusiast market and indicated by the presence of a K suffix. In December of last year, however, motherboard manufacturers began releasing BIOS updates which allowed the base clock to be modified even on non-K chips - allowing buyers on a budget to pick up cheaper chips and squeeze extra performance from them. For the motherboard manufacturers and customers, the news was good; for Intel, a terrible threat to the high profit margins of its enthusiast products.

It should come as no surprise, then, that the party is coming to a close: Intel is to issue a microcode update which will disable base clock modification on all affected non-K processors. Speaking to PC World, an Intel spokesperson stated: 'The latest update provided to partners includes, among other things, code that aligns with the position that we do not recommend overclocking processors that have not been designed to do so. Additionally, Intel does not warranty the operation of the processor beyond its specifications'

While confirming that inclusion of the microcode update in future BIOS revisions is entirely voluntary on the part of the motherboard manufacturers, the company's updates include other functionality - including fixes for system-hanging errata - which are unlikely to be easily abstracted away from the feature-killing code.

Noticia:
http://www.bit-tech.net/news/hardwar...ock-loophole/1


É o fim da festa
É nestas alturas que se sente a falta de uma AMD ao estilo dos bons velhos tempos!!!!!

[Jorge-Vieira]

Intel Skylake i5-6500, i5-6400 & i3-6100 Review

When Intel first launched their Skylake processors, buyers’ choices were essentially limited to a pair of options: the i7 6700K and later the i5 6600K. Both boasted features enthusiasts looked for like unlocked multipliers and fast out-of-box performance but they were also priced accordingly. These remained the only two possibilities for Skylake-bound buyers until late last year when Intel finally rolled out a top-to-bottom lineup.

The way Intel has approached Skylake’s cadence into lower price points has been drastically different from previous generations. Instead of offering a broad range of products from day one, additional i5, i3 and Pentium-class CPUs became available in fits and spurts, sometimes months after their initial “launch” dates. Much of this odd rollout seems to have been dictated by challenges brought about by Intel’s advanced 14nm manufacturing process, one which has effectively broken the ground rules set out by Moore’s Law. Simply put, Intel can’t produce Skylake processors fast enough to satisfy the market’s demand.

While this situation could eventually play into AMD’s hands provided their Zen microarchitecture meets expectations, for the time being this isn’t negatively impact Intel’s commanding market share. It has however contributed to prices remaining high which has pushed potential customers to lower price points. Many have asked themselves whether or not those unlocked multipliers that distinguish K-series processors are really worth the investment.

Enthusiasts who really want to push their systems will always gravitate towards overclocking. However, some of these more affordable alternatives could be an excellent option if a user believes the money saved versus something like the i7 6700K would be better invested in a motherboard with a huge feature set or a better graphics card.

This all brings us to the three processors in today’s review, namely the i5-6500, i5-6400 and i3-6100. All of these have been chosen for their price points above all else but there is a notable absence: the i3-6300. Unfortunately, since we purchased all of these processors it was the only one which wasn’t available but, on paper at least, it really doesn’t offer that much more than the i3-6100.

The $192USD i5-6500 is a processor many have considered to be a fitting replacement for the significantly higher priced i5-6600K. It offers Base and Boost frequencies that are within a mere 10% of Intel’s lowest-priced unlocked Skylake CPU while still incorporating four native processor cores, a 6MB cache and a 26W lower TDP. Unfortunately the 3.6GHz max Turbo speed is a far cry away from the 8-threaded i7-6700K’s 4.2GHz but the i5-6500 costs nearly $150 less.

Judging from the response on various online forums, the i5-6400 is considered by many to be the black sheep of Intel’s lineup. While it still includes four native processor cores and a quartet of threads, its base clock (typically what the processor will run at under heavily multithreaded workloads) and graphics clock are much lower than the 6500’s despite a slim $15 cost savings. TDP also remains identical to its more expensive sibling at 65W.

The benefits of a native quad core processor in today’s workloads can’t be overlooked. Hence, regardless of how efficient Intel’s Hyperthreading technology is, a dual core CPU like the i3-6100 which can process up to four threads will typically fall behind its more expensive quad core siblings. Nonetheless, this inexpensive sub-$120 processor can still tackle multi-threaded workloads with aplomb and its “virtual” threads will become transparent to certain applications. There’s also the potential to see performance pushed well ahead of previous-generation Haswell processors since Skylake’s HyperThreading algorithms have been improved.

Past its obvious efficiency advantage there’s a lot to differentiate the i3-6100 from the other processors in this review. It goes about its business without Turbo Boost functionality so instead of dynamically increasing clock speeds when the chip’s TDP allows, 3.7GHz is the constant running frequency. That’s actually quite impressive given the i5-6500 tops out at 3.6GHz. However, while clock speeds seem to be well aligned with higher end alternatives, the i3-6100’s meager 3MB of unified cache will likely hold back performance in some applications.

Added to our i7-6700K, we now have the potential to get a good snapshot of the Skylake lineup at several different price points. While every buyer loves to see how enthusiast level products handle themselves, the reality is that very few actually end up buying those expensive leading-edge wares. Our intent is to see how these less popular family members line up. Many will end up looking towards the i5 and i3 product stacks for their upgrade or new system needs; at the high end (the i5-6600K) they offer unlocked multipliers while something like the i3-6100 could allow someone to maximize the money invested in other components. Honestly, few require overclocking and even fewer games make use of anything more than four threads so any one of these three processors we’re about to cover could be an excellent choice.

Typically our introduction would end here and the benchmarks would start rolling but overclocking does need to be mentioned since it’s recently become something of a hot potato issue. While the locked nature of non-K processors effectively eliminates straightforward overclocking by simply increasing the multiplier, motherboard manufacturers had begun rolling out BIOSes which allowed Base Clock (BCLK) increases. This was significant since the CPU’s running frequency is derived from the base clock (typically 100MHz) and multiplier so with the latter being locked down, higher clock speeds could be realized by manually increasing the BCLK.

With Skylake’s BCLK infinitely more tractable than previous generations, budget-minded overclockers could suddenly buy a less expensive processor that could be overclocked to a point where it competed against much more expensive K-series chips. For example, we saw i5-6400’s easily reaching 4.1GHz with a Base Clock of 125MHz. Actually achieving those overclocks was more challenging than multiplier-based overclocking but it was still possible…..until last week.

We won’t delve into all the speculation surrounding this but for whatever reason, the non-K overclocking BIOSes which motherboard manufacturers were so quick to market suddenly started disappearing. All of them are still available on archive sites but key microcode and platform fixes which were rolled out in the last few weeks aren’t included on those earlier files. They’re essentially a dead end and no longer even supported by the manufacturers who originally released them.

Taking this into account, we’ve actually cut out overclocking from this review but the lack of any possible performance boost through higher than reference frequencies shouldn't dissuade you in any way. All of these CPUs offer some pretty impressive results straight out of the box.

Toda a review:
http://www.hardwarecanucks.com/forum...00-review.html

[Jorge-Vieira]

Intel Skylake GT4e Core Photo Shows Very Long Chip!

Congatec has released images of the conga-TS170 module that has been designed for graphics intensive applications. The board uses 6th Generation Intel Core processor with the flagship CPU being the Intel Core i7-6820EQ (4 x 2.8 / 3.5 GHz, 8MB L2 cache, 45 W). That means this module sports Intel Gen9 HD Graphics (GT4e) with 72 execution units, 128 MB of eDRAM and a maximum clock rate of 1150MHz. This graphics solution supports OpenCL 2.0, OpenGL 4.3 DirectX12 and HEVC (H.265) as well as hardware-accelerated encode and decode of HEVC, VP8, VP9 and VDENC video. With all this power it also means that three independent 4K (3840 x 2160) displays can be connected to the module via HDMI 1.4 and DisplayPort 1.2. Congatec’s new SoC module looks like it will appeal to those looking to build a media server or IoT edge node server where video transcoding or other parallel computing tasks will be run.
This is the first time that we can recall seeing the Intel Skylake GT4e core and it’s fairly large and very long!
image: http://www.legitreviews.com/wp-conte...70-645x476.jpg

The die appears to be between 210mm^2 to 220mm^2 when the above image is measured off the known 77mm^2 eDRAM die size. Over 65% of the die is said to be for the integrated graphics (IGP) and it appears that Intel is continuing to take graphics very seriously.

Noticia:
http://www.legitreviews.com/intel-sk...bMuFs6Ls0So.99

[Jorge-Vieira]

Rockit Cool launches de-lidding tool Kickstarter

Rockit Cool has launched a crowd-funding campaign for its Rockit 88, a self-contained de-lidding tool for removing the IHS from Intel's processors.

US-startup Rockit Cool has launched a crowd-funding campaign to produce an affordable de-lidding tool for Intel processors, to make removing the integrated heatspreader (IHS) quick and safe.

Intel has come under fire in recent years for changes to the way it bonds the integrated heatspreader (IHS), a plate of metal that serves to increase the surface area of the die for cooling purposes while also offering mechanical protection against crushing, to the die of its processors. Older processors would typically see the die soldered to the heatspreader, providing a high-quality thermal interface between the two; more recently, the company has begun using thermal grease instead. Removing the IHS and applying higher-quality thermal grease, or even leaving the ISH off entirely for direct-to-die cooling, is increasingly popular but fraught with risk: one slip of the vice or razor and your precious CPU is toast.

As a result, there has been increasing interest in tools to make de-lidding a CPU easier and safer. In November last year Roman 'der8auer' Hartung's Delid Die Mate launched in the UK, but its £62.99 purchase price meant it was a difficult purchase to justify for someone trying to shave a few degrees off the temperature of a single processor. James Holbrook of Rockit Cool believes he has a solution: a much cheaper but still high-quality delidding tool.

Dubbed the Rockit 88, the tool is entirely standalone: using bolts included with the kit, the lid is gently prized away from the package with no need to lock the chip into a vise or whip out a razor blade. Better still, it's considerably cheaper than the competition: the Rockit 88's crowd-funding campaign sees the device priced at $35 (around £24 excluding taxes), with limited numbers of early-bird units available at $28 (around £19). Even allowing for the somewhat steep $20 shipping fee to the UK (around £14), that makes the Rockit 88 £38 plus taxes - considerably lower than the Delid Die Mate's £62.99.

There are risks, of course: the device is Rockit Cool's first crowd-funding campaign, and while the company has produced prototypes that is no guarantee it will be able to produce at volume in the quality and quantity required. Those willing to take a punt, however, will find more details on the company's Kickstarter page.

Noticia:
http://www.bit-tech.net/news/hardwar...-kickstarter/1

[Jorge-Vieira]

Intel releases i7-6660U

Low powered Skylake-U

Chipzilla has quietly added a new mobile chip to its pricelist.

The i7-6660U is a Skylake-U processor which has low power consumption. It appears to have replaced the Core i7-6650U which has been installed in Laptops since September. The new chip can manage 200 MHz operating frequency at the same price level as the older version.
According to Chipzilla the chip is made using 14nm and has two CPU cores, clocked at 2.4 GHz, and 4 MB of last level cache. It can run 4 threads simultaneously. It has the bog standard Iris 540 graphics and the 15 Watt TDP.
The i7-6660U supports all the usual Skylake extensions and technologies, including AVX 2, AES, Turbo Boost, and VT-x / VT-d virtualisation.It is hitting the shops with a $415 it is not clear when it will be under the bonnet of any notebook.

Noticia:
http://www.fudzilla.com/news/process...eases-i7-6660u

[Jorge-Vieira]

Everything We Know About Intel's Skylake Platform

Intel's Skylake architecture and corresponding platform represent a huge evolution in connectivity, overclocking and, ultimately, system performance. This resource should help answer any questions you have about the company's current desktop PC design.

Since the release of Intel's Skylake-based processors, we have been working tirelessly to learn everything there is to know about the architecture and its associated platform. The information we have accumulated over a number of months, but it seems to be constantly changing, creating confusion among tech enthusiasts. And so we're aiming to clear up conflicting information and condense everything we know about the Skylake platform down into a single resource.

Intel released a total of six chipsets targeting its consumer and business customers. These platform controller hubs follow the same structure used for years to cover Intel's relevant markets. Three are consumer-oriented SKUs and three are designed for business.
Skylake Consumer Chipsets

Intel 100 Series Consumer Chipsets
Chipset	Z170	H170	H110
CPU PCIe 3.0 Config Support	1x16, 2x8 or 1x8 + 2x4	1x16	1x16
Embedded Options	No	No	Yes
TDP	6W	6W	6W
Recommended Customer Price	$47	$32	$26
Independent Display Support	3	3	2
Memory Channels/ DIMMs per Channel	2/2	2/2	2/1
CPU Overclocking Support	Yes	No	No
Intel Smart Sound Technology	Yes	Yes	No
Intel Small Business Advantage 4.0	No	Yes	No
Intel Small Business Basics	No	Yes	Yes
Intel RAID Support 0/1/5/10	Yes	Yes	No
Intel Smart Response Technology	Yes	Yes	No
Max Intel RST for PCIe	3	2	0
I/O Port Flexibility	Yes	Yes	No
Maximum HSIO Lanes	26	22	14
Chipset PCIe Support	20 PCIe 3.0 Lanes	16 PCIe 3.0 Lanes	6 PCIe 2.0 Lanes
USB Support (USB 3.0)	14 (10)	14 (8)	10 (4)
SATA 6Gb/s Ports	6	6	4
DMI	3.0	3.0	2.0

At the high end, we have the enthusiast Z170 chipset. In addition to featuring RAID 0/1/5/10, multi-GPU support and offering a wide selection of connection options, it is also the only PCH officially able to overclock Skylake-based processors. Z170 is followed up by the mid-range H170, which also offers a lot of I/O connectivity and RAID 0/1/5/10 functionality, but lacks overclocking and multi-GPU support via the CPU.

H170 is often advertised as supporting multiple graphics cards, but motherboards based on it can only run a multi-GPU setup in a x16/x4 configuration. This is because the PCH doesn't allow the CPU to divide its PCIe lanes between multiple devices. A second GPU can be connected over a four-lane PCIe 3.0 link through the PCH, which may be sufficient for some GPUs, but Nvidia doesn't let you enable SLI over a x4 connection.

At the bottom of the consumer chipset stack is H110, by far the most limited PCH. Instead of using a DMI 3.0 connection between the CPU and core logic, it employs the older DMI 2.0 technology. It's basically a budget-oriented approach to supporting Skylake-based processors.

Skylake Business Chipsets

Intel 100 Series Business Chipsets
Chipset	Q170	Q150	B150
CPU PCIe 3.0 Config Support	1x16, 2x8 or 1x8 + 2x4	1x16	1x16
Embedded Options	Yes	No	No
TDP	6W	6W	6W
Recommended Customer Price	$47	$43	$28
Independent Display Support	3	3	3
Memory Channels/ DIMMs per Channel	2/2	2/2	2/2
Intel SIPP Eligible	Yes	Yes	No
Intel vPro Technology Eligible	Yes	No	No
Intel Active Management Technology	Yes	No	No
Intel RAID Support 0/1/5/10	Yes	No	No
Intel Smart Response Technology	Yes	No	No
Max Intel RST for PCIe	3	0	0
Maximum HSIO Lanes	26	20	18
USB Support (USB 3.0)	14 (10)	14 (8)	12 (6)
SATA 6Gb/s Ports	6	6	6
Chipset PCIe Lanes	20 PCIe 3.0	10 PCIe 3.0	8 PCIe 3.0
DMI	3.0	3.0	3.0

The business-oriented chipsets are similar to the consumer offerings, though they include a few extra features that consumers don't necessarily need.
Q170 is nearly identical to Z170, but of course lacks overclocking support. Moreover, it features Intel SIPP, vPro Technology and Active Management Technology. Q150 is nearly the same as H170 as well, except it has fewer HSIO and PCIe lanes. Plus it lacks RAID and Intel’s Small Business Basics software package. It is eligible for Intel’s SIPP technology, though.
The low-end B150 business chipset, however, does not match up well with H110. It offers more connectivity, DMI 3.0 and PCIe 3.0. Because of its budget price and specs, B150 is often used by consumers as a more feature-rich alternative to H110.

MORE: Best CPUs
MORE: Intel & AMD Processor Hierarchy
MORE: All CPU Content

Skylake Workstation Chipsets

Starting with the introduction of its Allendale- and Conroe-based microprocessors, Intel also made single-socket Xeon processor support available on its consumer-oriented platforms. Over the years, these Xeon CPUs became popular alternatives to Intel’s more expensive Core i7s because they have similar specs but sometimes cost less or include other useful features.

To stop desktop enthusiasts from turning to Xeons rather than Core i7s, Intel doesn't facilitate Xeon support on its Skylake-based consumer- and business-oriented platforms.

Instead, the company created two workstation-oriented hubs for its LGA 1151 interface: C232 and C236. Intel hasn’t given us all of their specs, but a representative did convey the basics. What we're missing is how many HSIO lanes are available on C232 and C236, which is an important piece of information (more on that below). So, we reached out to the motherboard OEMs and were told by one product manager that C232 has the same connectivity support as B150, while C236 is identical to Z170.

Comparison Of C232 & C236 Chipsets
Chipset	C232	C236
CPU Support	Celeron, Pentium, Core, Xeon	Celeron, Pentium, Core, Xeon
Embedded Options	No	Yes
TDP	6W	6W
Recommended Customer Price	$34	$49
Max PCIe Lanes	8	20
CPU PCIe Configurations	1x16, 2x8, 1x8+2x4	1x16, 2x8, 1x8+2x4
USB 2.0	6	4
USB 3.0	Up To 6	Up To 10
SATA 6Gb/s	6	8
RAID	0/1/5/10	0/1/5/10
Intel Virtualization technology	No	Yes
Intel vPro	No	Yes
Intel RST	No	Yes
Intel RST Enterprise	Yes	Yes
Intel Node Manager	No	Yes
Intel Standard Manageability	No	Yes
Intel Smart Response Technology	No	Yes
Trusted Execution Technology	Yes	Yes
Intel HD Audio Technology	No	No
Intel SIPP	No	No
Intel Small Business Advantage	No	No
Intel HD Graphics Support	No	No

These chipsets are supposed to last longer and boast several advanced technologies that are not available on the desktop side. Other than that, their biggest advantages are multi-GPU and Xeon support.

Limiting LGA 1151-based Xeons to the C232 and C236 chipsets makes it more difficult for enthusiasts to build PCs using a Xeon processor, of course, as there are fewer motherboards sporting either PCH. Those that do exist are often pricier due to the validation work that goes into a more professional product.

Although a Xeon-based build can still be achieved for less money than a high-end Core i7, the delta is no longer as large. Most C236-based boards cost too much to make buying a Xeon economically viable. Falling back to C232 might save you some cash, but you'll also have far fewer features to use compared to Z170.

HSIO Lanes And Connectivity

Intel has been using a similar technology to HSIO (high-speed I/O) lanes (Flex /IO on Haswell-based chipsets) to interface between add-ins for years. Compared to its Haswell- and Broadwell-based platforms, however, Skylake is even more reliant on this technology, which sometimes makes understanding the connectivity options of each chipset more confusing.
Nearly every connection between the PCH and another device uses HSIO lanes. The only major connections that don’t are the USB 2.0 ports and the link between DMI link between the CPU and PCH. All USB 3.0 ports, SATA interfaces and PCIe slots consume at least one HSIO lane.

For example, Z170 exposes a total of 26 HSIO lanes, six of which are consumed by six permanently-enabled USB 3.0 ports. So, the chipset ends up with 20 configurable HSIO lanes that can be assigned to other devices. Each SATA port uses an HSIO lane as well, unless it's connected through a third-party controller (though that controller would need at least one lane to communicate with the PCH). As you can see in the diagram, GbE controllers and PCIe-based SSDs also consume available HSIO lanes.

The confusion happens when you hear what the chipset can support. Yes, you can do up to 10 USB 3.0 ports, eight SATA 6Gb/s ports, 20 PCIe 3.0 lanes and gigabit Ethernet. But the platform can only handle some of those I/O options simultaneously.

Skylake Chipsets (Real) PCIe Connectivity
	Max PCIe 3.0 Lanes	Max PCIe 3.0 Lanes If All USB, SATA & Single GbE In Use	DMI
Z170	20	9	3.0
H170	16	7	3.0
H110	6 (PCIe 2.0)	6 (PCIe 2.0)	2.0
Q170	20	9	3.0
Q150	10	5	3.0
B150	8	5	3.0
C236	20	7	3.0
C232	8	5	3.0

The biggest issue is that the maximum number of PCIe 3.0 lanes on each chipset will likely never be exposed. In order to have 20 lanes configured, which is technically possible on Z170, Q170 and C236, you would have to give up all SATA-based storage, native GbE and USB 3.0 ports beyond the six hard-wired ones.

Motherboard manufacturers make the situation more difficult to explain by launched products with more physical connections than the PCH can support at any one time. Engineers make all of the I/O functional by tying multiple devices to a single HSIO lane. Devices that share a lane cannot function simultaneously. So, often, connecting one piece of hardware disables other ports or features elsewhere. And it doesn't help that board vendors don't make this well-known. Most of the spec sheets we've seen do spell out which connections share HSIO lanes, but it's sometimes hidden in fine print somewhere at the bottom. As a result, even enthusiasts get caught buying motherboards based on their connectivity options without realizing they can't all be used together.

Memory And Bandwidth

DMI And Bandwidth

Another potential issue is the interface between the CPU and PCH. DMI 3.0 is essentially equivalent to a four-lane PCIe 3.0 link, offering roughly 4 GB/s of bi-directional bandwidth. All I/O from your USB-attached thumb drive, SATA-based SSD and gigabit Ethernet network goes through the PCH and across that interface before landing in system memory and eventually the CPU or GPU.


Using multiple devices simultaneously connected to the PCH forces them to compete for bandwidth. Intel claims that there shouldn't be as much contention now that the third-gen DMI doubles the previous generation's peak throughput, but it's still a plausible concern. That's one reason you probably wouldn't want to use a multi-GPU configuration on a chipset like H170 that won't divide the CPU's PCIe lanes between multiple graphics cards. It's also one of the reasons why Nvidia doesn't allow SLI across four-lane links.

Memory Support

Starting with Skylake, Intel added DDR4 support to its memory controller. But because the technology is still fairly new, the company retained support for DDR3 as well, easing adoption of its most modern platform.

Don't take that to mean any DDR3 module will work with Skylake, though. Only DDR3 operating at or below 1.35V is officially supported, and using DDR3 at higher voltage levels could damage the CPU's integrated memory controller.
Several board vendors list support for RAM operating at higher voltage levels, and you may not run into problems using RAM rated for 1.5 or 1.65V, but Intel doesn't recommend it. A lot of damage materializes slowly over time due to electromigration. As such, you'll want to carefully weigh the risks of dropping older modules into Skylake-based systems, providing you have a motherboard with DDR3 slots at all.

Overclocking

As it created previously, Intel has a couple of Skylake-based CPUs with unlocked clock multipliers (the K-series SKUs). These can be overclocked by simply increasing the ratio on a Z170 motherboard. The company also made changes to its PCH that again enable BCLK-based overclocking as well.

As a result, some motherboard vendors are trying to arm their platforms with the ability to overclock non-K-series CPUs, and to make some of Intel's other chipsets overclock as well. Although these capabilities would be immensely popular among enthusiasts, there are a few problems with trying to tune multiplier-locked Skylake-based CPUs.

For instance, once the BCLK of a non-K-series processor rises above its stock 100MHz, certain power-saving features, Turbo Boost and specific instructions no longer work. Thus far, this happens on all motherboards capable of overclocking non-K-series processors, regardless of chipset. The board vendors are forced to disable those features while overclocking to ensure stability. Their loss isn't debilitating, especially since power-saving features are often disabled as part of an overclock anyways. However, we know nobody likes restrictions imposed on them forcibly.
Supermicro

Currently, there are only two companies with motherboards capable of overclocking non-K-series CPUs. Supermicro was first on the scene, and it also has a platform that can overclock based on a chipset other than Z170. Its C7H170-M employs the H170 PCH and is built a lot like competing Z170 boards. As of this writing, the company doesn't have any other boards capable of overclocking non-K-series CPUs.

ASRock's Shifting Position

ASRock also has non-Z170 motherboards capable of overclocking non-K-series Skylake CPUs, but they've been hit by several setbacks. ASRock initially announced “Sky OC” both as a feature for Z170 boards and as a family of overclockable non-Z170 motherboards. Thus, any product with the Sky OC feature was able to overclock non-K processors, and the boards based on chipsets other than Z170 could overclock K-series CPUs using BCLK adjustments as well.

Not long after ASRock announced these boards, the company contacted Tom’s Hardware to let us know it abandoned their development under pressure from Intel. It was also forced to remove the Sky OC feature from its Z170 motherboards.

The quote went a little something like, “ASRock has decided to remove SKY OC technology from these motherboards’ feature lists since it is not compliant with Intel Skylake CPU specifications.”

We expected that to be the end of ASRock’s venture into overclocking non-K-series CPUs and overclockable non-Z170 chipsets, but then the company released its cancelled C232 Fatal1ty E3V5 Performance Gaming/OC board. It also announced its new “Hyper” family of non-Z170 motherboards able to overclock non-K-series processors using BCLK manipulation. Currently, the C232 Fatal1ty E3V5 Performance Gaming/OC is the only one of these boards available for sale online. The others may materialize in time, but it's also possible that Intel again applied pressure to stop their introduction.

Fallen Attempts At Non-K OC

Biostar tried to arm its motherboards with the ability to overclock non-K CPUs using a BIOS update for its Z170 boards. The approach was similar to the way ASRock enabled Sky OC. And like ASRock, Biostar removed the feature through another BIOS build that included new microcode from Intel.

According to a representative, “Intel regularly issues updates for our processors which our partners voluntarily incorporate into their BIOS. The latest update provided to partners includes, among other things, code that aligns with the position that we do not recommend overclocking processors that have not been designed to do so. Additionally, Intel does not warranty the operation of the processor beyond its specifications.”
Supermicro does not appear to have applied this update to its C7H170-M motherboard, but that could be just a matter of time. Technically, all of these boards are still able to overclock non-K CPUs, though to do so requires an outdated BIOS that might have other issues. This makes unsanctioned overclocking essentially dead.

Are You Looking At A Skylake Motherboard?


Intel's Skylake-based processor launch was one of the most hectic in recent history. The company's focus has changed, and the way it disseminates information isn't as thorough as it once was. Consequently, new information about the architecture and its supporting chipsets seems to trickle out slowly, whereas we were previously much better connected to the minds responsible for platform development. Several of the most recent disclosures were sponsored by motherboard OEMs exploring features and capabilities to give them an advantage over the competition.
Those board vendors may still have a few tricks up their sleeves. Overclocking non-K-series CPUs is still a hot topic, so it's possible that OEMs will continue looking for more graceful ways to achieve this. Meanwhile, we expect Intel will continue inhibiting the value-seeking tendencies of enthusiasts looking to buy something inexpensively and extract maximum performance from it. Should Intel surprise us and back off of its aggressive stance, other board vendors may try to create non-Z170 platforms with overclocking features as well. So, don't be surprised to see this story change over time as we work to keep an up-to-date reference on everything there is to know about Skylake.

Ver toda a noticia:
http://www.tomshardware.com/reviews/...-101,4498.html

[Jorge-Vieira]

Rockit 88 CPU delidder tool pops the top on recent Intel chips

Rockit Cool's Rockit 88 is a Kickstarter project that aims to make delidding—the practice of removing a CPU's integrated heat spreader, or IHS—easier. Delidding a CPU for extra cooling potential is a well-known but risky practice for extreme overclockers, since it usually involves pliers, a vise, and an X-Acto knife. A slip-up with any of these tools could harm the CPU. The Rockit 88 appears to work by holding Haswell and Skylake CPUs in a dedicated bracket and shearing off the IHS with a moving block. Rockit Cool claims that reducing the thickness of the thermal interface material between the processor die and the IHS on an unspecified, overclocked Devil's Canyon CPU resulted in a 10°C improvement in load temperatures.

Even with the safety and reliability of a delid tool, we're guessing that the risk of damage to the CPU isn't completely eliminated. Even with a dedicated tool, the thermal benefit of delidding is questionable for all but the most extreme overclockers. Going by testing on an older Intel CPU performed by an Anandtech forum member, the upper limit of benefits is about 10°C under load. Linus Tech Tips also tried delidding, and that anecdote suggests even more modest benefits. Still, the Rockit 88 might make the process safer, and it's simple-looking enough that its Kickstarter isn't as scary to back as more ambitious projects might be. Aspiring delidders can pledge $35 in exchange for a complete Rockit 88 kit. Rockit Cool expects to begin delivering Rockit 88s in May.

Noticia:
http://techreport.com/news/29968/roc...nt-intel-chips

[Jorge-Vieira]

Intel Pentium G4400 Processor Review – Budget Skylake

Intel Pentium G4400 Skylake Processor – A Low-Cost Gamers Skylake Processor?

When Intel releases a new processor family, we immediately start seeing reviews and opinions on the enthusiast grade processors such as the i5-6600K and the i7-6700K, both of which are great processors, however there is a downside; they are some of the more expensive processors. Not everybody wants to spend a lot of money on a new processor nor does everybody need the added features and power of these processors. Because of this, Intel releases a range of processors that start at $64.99 with free shipping, such as the Intel Pentium G4400, and quickly moves up in price depending on your need.
image: http://www.legitreviews.com/wp-conte...ox-645x645.jpg
Intel Pentium G4400 Retail Box

Why would somebody want to look at a processor such as the Pentium G4400? For a low cost processor, its specifications are pretty decent for an average home use type processor. It is a dual core processor that can handle 2 threads at a time at 3.3GHz with a TDP of 54W. Beyond that, it has many of the features of other Skylake processors such as support for up to 64GB of DDR4 or DDR3L memory in dual channel. Integrated graphics are provided by the Intel HD Graphics 510, that will utilize a maximum of 1.7GB of memory, in addition it provides for resolutions up to 4096×2304 @60Hz using Display Port in addition to supporting up to 3 displays.
Other more advanced technologies that is supported by the G4400 is Virtualization, AES N-I, and Secure Key. What it doesn’t support though is Turbo Boost, vPro, Hyper-Threading and a few other Intel Technologies that many home users will have little use for.
image: http://www.legitreviews.com/wp-conte...PU-645x578.jpg
Intel Pentium G4400 SR2DC CPU

Taking a look at the Intel Pentium G4400, the Integrated Head Spreader has the typical markings on it. Here we see the S-Spec of SR2DC and the clock speed of 3.3GHz. We then find the batch number which is X540C198.
image: http://www.legitreviews.com/wp-conte...ns-645x333.jpg
Intel Haswell versus Skylake

Taking a look at the bottom of the processor, it looks identical to other Skylake processors, which makes sense as it uses the same motherboard chipset. The Pentium G4400 is a LGA 1151 socket processor. Like other recent Intel processors, it has the capacitors in the middle, with the usual pins arranged around it. As we expect, with each new chipset, Intel changes the pin and key location to make sure you are installing the processor into the right socket.
image: http://www.legitreviews.com/wp-conte...er-645x614.jpg
Intel Pentium G4400 CPU Cooler

The Intel Pentium G4400 includes a basic CPU cooler that is designed to keep the processor running within it’s thermal tolerance. Those that are familiar with coolers know that the include CPU cooler doesn’t do a great job at cooling the processor, but it does work.
image: http://www.legitreviews.com/wp-conte...ed-645x388.jpg
Intel Pentium G4400 Installed in EVGA Z170 FTW

We will be using the EVGA Z170 FTW motherboard to benchmark and test the Intel Pentium G4400. Let’s take a quick look at the test system for the Intel Pentium G4400 before we get to doing some benchmarks.

Toda a review:
http://www.legitreviews.com/intel-pe...LqRe2PXtCof.99

[Jorge-Vieira]

Intel to launch next-gen Celeron, Pentium chips later this year

At its developer forum in China this week, Intel revealed plans to ship updated versions of its entry-level Celeron and Pentium processors later this year.
Codenamed Apollo Bay, the new chips are based on Intel’s 14-nanometer Goldmont CPU architecture that will succeed existing Celeron and Pentium processors based on the older Airmont architecture. Executives at IDF said Apollo Bay chips will offer better CPU and graphics performance in addition to improved efficiency, resulting in prolonged battery life on portable devices.
Intel’s cream of the crop, the Core series, is also expected to get an update in the second half of this year.
Originally, Intel planned to make the jump from today’s 14-nanometer Skylake architecture to its 10-nanometer Cannonlake chip sometime in 2016 following the familiar tick-tock development cycle it has relied on for the past decade.

In the wake of mounting evidence, Intel officially confirmed last month that it was ending the tick-tock era. That means the first 10-nanometer chips are being delayed until at least 2017 to make room for a third 14-nanometer product family.
The delay and introduction of a third 14-nanometer product family isn’t all that surprising as manufacturing on smaller processes has become increasingly difficult. When you consider the global slowdown in PC sales and the fact that AMD is no longer a serious threat, it makes perfect business sense for Intel to ride the wave as long as possible.

Noticia:
http://www.techspot.com/news/64454-i...ips-later.html