Intel 8th generation processors coffee lake. Why you shouldn't buy 8th Gen Coffee Lake Intel processors just yet. ⇡ Processors and Platforms: Specifications

A month after the announcement of the eighth generation Core processors for laptops, Intel Corporation officially announced new formation chips and for desktop computerscodenamed Coffee Lake. They are manufactured using an improved 14nm process technology and, as in the case of mobile Kaby Lake Refresh, contain more processing cores than their predecessors. HEDT-class solutions aside, this is the first increase in the number of cores in Intel desktop CPUs since 2006, when the Core 2 Extreme QX6700 was released.

The Core i7 and i5 have six cores, while the Core i3 has four. At the same time, the i7 series models are equipped with HyperThreading technology, thanks to which they execute 12 threads simultaneously. All six new products, a list of which is presented in the slide below, are equipped with an integrated GPU Intel HD Graphics 630 and can work with Intel Optane drives. Support for DDR4-2666 is also announced, the only exception is Core i3, compatible with DDR4-2400.

The rated clock speed of the most powerful member of the family - the Core i7-8700K - is 3.7 GHz, which is 500 MHz less than last year's Core i7-7700K. At the same time, under load, the chip develops 200 MHz more - 4.7 GHz. The difference between the "passport" frequency and the turbo mode reaches almost 27%, but the dynamic overclocking Turbo Boost Max 3.0 is not used here, we are talking only about the usual Turbo Boost 2.0. Obviously, Intel has resorted to the new frequency formula in order to achieve higher performance without a serious increase in heat dissipation requirements: the TDP of the Core i7-8700K is 95 W, which is only 4 W more than the i7-7700K.

Speaking of the speed of the new processors, the developers promise a 25% increase in the frame rate in modern games, 65% faster in such content creation applications as Adobe Photoshop, and 32% faster 4K video processing. Together with computing power prices have also increased: for example, the cost of the i7-8700K in batches of 1000+ is $ 359, which is 18% more expensive than the 7700K. The new items will go into retail on October 5 this year, deliveries to computer manufacturers will begin in the fourth quarter.

Along with CPU Coffee Lake, Intel announced a set supporting them system logic Z370. The press release states that motherboards based on the chipset meet the increased power requirements of the 8th Generation six-core Core processors and allow rAM DDR4-2666 standard. The first solutions based on the Z370 will also be announced on October 5, but some of them have already been online before the deadline.

08.21.2017, Mon, 09:36, Moscow time , Text: Vladimir Bakhur

Intel has announced the addition of eighth-generation Core chips to its U-series mobile processors. The next generation of Coffee Lake desktop processors will also arrive later this year.

The first four processors of the new eighth generation

Intel has unveiled four new mobile Core i5 and Core i7 processors in the U line. All new chips have four processing cores with support for hyper-Threading technology, which in total allows providing up to eight computational threads for each crystal.

Previous generations of Mobile Core processors came with two physical cores and supported four threads with Hyper-Threading Technology.

The official working name of the new mobile processors is Kaby Lake Refresh, that is, they are based on the improved seventh generation Kaby Lake architecture.

All the 8th generation Core processors presented today (Kaby Lake Refresh), like their predecessors, are manufactured in compliance with the 14nm technological process, but “with improved characteristics”, which led to the announcement of the new 8th generation. According to Intel, the transition to the 10 nm technical process will take place later in the fall, but within the framework of the same eighth generation.

The "real" next-generation architecture, tentatively named Coffee Lake, will be unveiled even later and will add to the list of 8th-generation Core chips. However, these chips will also be manufactured at 14 nm.

Processors Intel Core new 8th generation

The move to 10nm will be the next milestone and will debut with the Cannon Lake architecture. Thus, the list of eighth generation Core processors will include i7 / i5 / i3-8xxx chips of three different architectures: Kaby Lake Refresh, Coffee Lake and Cannon Lake. Earlier one core generation usually there were two kinds of architectures.

Details about architecture

The new eighth generation Core processors operate at relatively low main clock frequencies (no higher than 1.9 GHz for the older i7-8650U model), due to which all models fit into a thermal package (TDP) of up to 15 W with four processing cores.

8th generation Core processor appearance

At the same time, thanks to intel technology Turbo Boost Technology 2.0, chips are able to dynamically grow clock frequency more than doubled (up to 4.2 GHz for the older model i7-8650U), which allows you to significantly increase system performance on demand and stay in a "cold" state in standby mode.

Baseline Specs for First Four 8th Gen Core Processors

All new mobile intel processors Core 8 generations are equipped with built-in graphics core Intel UHD Graphics 620 with support for up to three independent displays, inherited with some changes from 7th generation processors (Kaby Lake, Intel HD Graphics 620). Integrated UHD Graphics 620 supports HEVC and VP9 codecs, and allows 4K video with 10-bit color depth.

Photo of the crystal of the new 8th generation Intel Core chip

The new 8th generation mobile processors received 8MB or 6MB of L3 cache, as well as a fast 2-channel memory controller with support for DDR4-2400 and LPDDR3-2133 modules.

About productivity and savings

The company's new eighth-generation mobile Core i7 and i5 chips deliver up to 40% performance gains over chips, according to internal company benchmarks. previous generation, and is twice as fast as chips from five years ago, for example, when comparing the new Core i5-8250U with the Core i5-3317U.

Intel's pipes are burning. So it seems at first glance at least. After a very successful line of Ryzen processors from AMD and a very warm welcome in the market, Intel is in a hurry to respond somehow. The company announced that Intel Coffee Lake desktop processors will begin shipping on October 5, just 10 months after the 7th Gen Kaby Lake unveiled.

The upcoming lineup includes the Core i3, Core i5 and Core i7 models, all with locked and unlocked multiplier options (overclocking an Intel Core i3 now). The most important change will be the increase in the number of cores in all announced models. The standard set for 4 cores will be supplemented with two more and now the mid and high end segment will be available with six cores. Also, the "budget" Core i3 will receive an increase in performance for the first time due to the transition to a quad-core model.

As it became known this summer, new Intel Core processors will require a new Z370 chipset from the motherboard. In other words, you cannot upgrade the processors in your computers without upgrading your motherboard. In addition, Intel has confirmed that the seventh and sixth generation processors will not work on the new Coffee Lake boards. In order not to panic, Intel claims that at the time of the start of sales, more than 50 different motherboard options will be available in stores for every taste, size and budget. Rumors are also quite amusing, according to which Intel will show new eight-core mainstream processors next year on another Z390 chipset.

The eighth generation of Core desktop processors are built on a process that Intel calls 14nm ++. However, the company does not name the size or number of transistors under the processor cover. However, according to Intel, gamers should see a performance boost of around 25% over Kaby Lake. As productivity rises, so does the price. The Core i7 will be $ 20 more expensive than its seventh generation equivalent, and the Core i5 will be $ 15 more expensive than its Kaby Lake sibling. Most likely, such a move is justified by the increase in cores and the desire to avoid the cannibalism of the seventh generation processors.

The family of the eighth generation Intel Core Coffee Lake processors is as follows:

• Intel Core i7-8700K. 3.7GHz base frequency, 4.7GHz single core overclocked, 6 cores / 12 threads, TDP 95W, 12MB cache, DDR4-2666 support, Intel Optane support. Unlocked multiplier. $ 359.
• Intel Core i7-8700. 3.2GHz base frequency, 4.6GHz single core overclocked, 6 cores / 12 threads, 655W TDP, 12MB cache, DDR4-2666 support, Intel Optane support. $ 303.
• Intel Core i7-8600K. 3.6GHz base frequency, 4.3GHz single core overclocked, 6 cores / 6 threads, TDP 95W, 9MB cache, DDR4-2666 support, Intel Optane support. Unlocked multiplier. $ 257.
• Intel Core i7-8400. 2.8GHz base frequency, 4GHz single core overclocked, 6 cores / 6 threads, 65W TDP, 9MB cache, DDR4-2666 support, Intel Optane support. $ 182.
• Intel Core i7-8350K. 4 GHz base frequency, 4 cores / 4 threads, TDP 91W, 6MB cache, DDR4-2400 support, Intel Optane support. Unlocked multiplier. $ 168.
• Intel Core i7-8100. 3.6GHz base frequency, 4 cores / 4 threads, 65W TDP, 6MB cache, DDR4-2400 support, Intel Optane support. $ 117.

Finally, the situation with intel chipsets 300-series for 8th Gen Coffee Lake processors. Although there is no official information on the latest chipset on Intel's site yet, the Z390, nevertheless, almost all its specifications are known. Next, I will briefly describe what it is, and in general, let's look at the chipsets for the 8th generation Intel Coffee Lake for socket 1151 v2, make a small comparison, and determine how they differ. This can help when choosing a motherboard for the latest generation CPU.

Chipset characteristics comparison table

Let's start by putting together the main characteristics of the chipsets:

Chipset	H310	B360	H370	Z370	Z390	Q370
PCI-Express version	2.0	3.0	3.0	3.0	3.0	3.0
Possible configurations of PCI Express processor lines	1x16	1x16	1x16	1x16 /	1x16 /	1x16 /
Max. number of PCI Express lanes	6	12	20	24	24	24
PCI Express Configurations	x1, x2, x4	x1, x2, x4	x1, x2, x4	x1, x2, x4	x1, x2, x4	x1, x2, x4
# Of memory channels	1	2	2	2	2	2
Max. number of DIMMs	2	4	4	4	4	4
Intel Optane support	-	+	+	+	+	+
Max. # of SATA 3.0	4	6	6	6	6	6
Max. qty x2 M.2 or x4 M.2	0	1	2	3	3	3
RAID configuration	-	-	PCIe 0.1.5 /	PCIe 0.1.5 /	PCIe 0.1.5 /	-
Wi-fi and Bluetooth support	Intel Wireless-AC MAC			-	Intel Wireless-AC MAC
Max. USB quantity	10	12	14	14	14	14
Max. USB 3.1 Gen 2 quantity	-	4	4	-	6	6
Max. USB 3.1 Gen 1 quantity	4	6	8	10	10	10
Max. number of USB 2.0	10	12	14	14	14	14
Overclocking support	-	-	-	+	+	-
# Of Displays Supported	2	3	3	3	3	3

It is immediately evident that Intel adheres to the already familiar hierarchy of its products, in which chipsets with the letter “Z” are endowed with the greatest number of functions. However, not quite so. There are moments that, it would seem, should be present in the same Z370, but they are not there. At the same time, the H370 and younger have them.

This is due to the fact that we got acquainted with the Z370 last year, but all the others began to appear relatively recently. In addition, there are no motherboards based on the latest chip, the Z390, at all. However, I will gradually tell you about everything.

Intel H310

Let's start with the youngest. As it was in the 1хх and 2хх series, this is the most "stripped down" variant with a minimum of possibilities. Probably a great choice for office computer: cheap, no "bells and whistles", but with all that is necessary for a less powerful PC.

The block diagram is "self-made", because I did not find the official one, and if it is somewhere, then in the comments leave a link to the picture so that I can insert the official image here.

The chipset does not support the PCI-Express bus of the latest version 3.0 to date, but manages the previous, 2nd version with a corresponding decrease in bandwidth - 8GB / s versus 5GB / s. By the way, the DMI bus used for communication with the processor is also version 2, although all other chipsets use DMI 3.0.

The memory controller is single-channel, with support for two DIMMs. Everything else is clear from the table.

B360

The chipset is one step higher than the previous one, but this step is quite large, because here we already see the use of PCI-Express and DMI (and, in fact, they are the same) version 3.0. The memory controller is already two-channel.

SATA drives can be connected 6, instead of 4 like in the H310. There was also the possibility of using the M.2 / U.2 connector. True, collect RAID array the chipset still cannot.

The 300 series of chipsets now have built-in support for USB 3.1 Gen 2, which allows you to do without the use of third-party controllers such as ASMedia. By the way, this USB version is only available on chipsets released this year.

In general, this is the first "full-fledged" chipset in the family, already suitable for using motherboards built on it to build high-performance computers.

H370

A very interesting chipset, which is in many ways similar to the Z370 described below, and in some cases even having more possibilitiesthan he is. Yes, overclocking processors by multiplier is still not available, and the number of PCI-Express lanes is somewhat less. Also, only two M.2 / U.2 connectors are supported against three in the Z370.

What's the advantage? First, there is USB 3.1 Gen 2. Second, there is support for wireless networks... Let's dwell on this in more detail.

This chipset, like the lower versions, has an integrated CNVi controller that supports Wi-fi 802.11ac and Bluetooth 5.0. Now to reap the benefits wireless, it is enough to install only the RF-module, which is small fee form factor M.2 2230 or 2216. The motherboard must have an M.2 connector with an E key for this.

Among the offered RF modules one can distinguish "budget" options - Wireless-AC 9461 and Wireless-AC 9462, which support Bluetooth 5.0 and Wi-fi 802.11ac with a maximum speed of up to 433 Mb / s, but which do not have what is in the Intel Wireless-AC 9560 module, namely the ability to work in a 2x2 scheme (including 802.11 Wave 2), as well as support for a 160 MHz channel, which gives a theoretical speed of up to 1.73 Gb / s.

As a result, we have a very advanced chipset, and if you are not interested in overclocking issues, but you need to assemble a fairly productive system, including a gaming one, then the H370 should be carefully considered.

Z370

As mentioned above, this first chipset of the 300 series is, on the one hand, one of the most sophisticated, and overclocking is available only in Z variants. On the other hand, it lacks some "chips" that appeared in chipsets only this year, which was also mentioned above when describing the H370.

This can cause some confusion when choosing the right motherboard. Still, the Z370 has a lot to offer to build a top-end system. This is, first of all, the configuration of video cards that can be used more than one. The second argument is the overclocking capabilities, because what a top system is without overclocking and squeezing all the megahertz out of the processor to a drop.

Another, maybe small, plus in the Z370's piggy bank - support for 3 M.2 / U.2 drives with the ability to combine PCIe SSD into a RAID array. The H370 also has such an option, but the chipset only supports 2 M.2.

Z390

A novelty (at the moment), which intrigued last year, when they published a roadmap of new generation chipsets. The Z390 settled down somewhere to the side, and it was not very clear what it was.

Now everything fell into place. Expectations that this chipset will be higher than the top-end Z370 were only partially justified. It does not offer anything special, and is, rather, what the Z370 really should have been, if it came out in due time, and not because there was no chipset for the new Coffee Lake processors released , and it was necessary to offer at least some.

So, the Z390 is very similar to the Z370 in storage configuration, pCI-Express configurations lines and so on. The changes are in the addition of something that was not in the Z370, but is present in later chipsets that are in the hierarchy at lower levels. So, appeared uSB support 3.1 Gen 2, and of course Intel Wireless-AC MAC.

Nothing more new, at least for the moment, is foreseen.

Q370

This chipset looks pretty attractive, at least in terms of features. In fact, this is the same Z370, but without overclocking wings and without RAID. On the plus side is the ability to use multiple video cards, which is still available only in Z-versions.

As the name suggests, the chipset is focused on corporate needs and, most likely, it will be bypassed by motherboard manufacturers, since the number of offered models based on Q-versions of chipsets is traditionally very scarce, and they are not widespread.

Conclusion. Chipsets for 8th Gen Intel - now everything is in place

The only thing left is to wait for the final specifications of the last 300 series chipset - the Z390, but it is already clear that it should replace the hastily released Z370.

All of these chipsets will also have to support future next-generation CPUs called Cannon Lake. For one "but" - according to available information, the "old" Z370 will not work with new CPUs. This can be seen as a kind of "setup" from Intel, since everyone who fell for the new "coffee" processors and, having no alternative, purchased a motherboard based on the Z370, when moving to the next generation of Intel processors, they will have to change and motherboard.

As for the rest - the usual hierarchy of chipset models, each of which will find its place when assembling a particular system, from budget to top-end ... well, to almost top-end, because only processors belonging to the HEDT level can be considered the real peak, for which it has already been announced corresponding X399 chipset. Yes, it is named the same as the processor chipset AMD Ryzen Threadripper. So now to the question "Which motherboard do you need on which chipset?" answering "X399" will not be enough. Need to clarify, Intel or AMD.

As you know, Intel is currently preparing to release a new, eighth, generation of Intel Core processors (Coffee Lake), which will compete with AMD Ryzen chips. Thanks to the leaks, we already know that Coffee Lake should bring six-core Core i7 and Core i5 models, as well as quad-core Core i3 models to the consumer desktop segment. And now the date of the official announcement of these processors has become known - August 21.

True, it is not yet clear whether this will be a full-fledged announcement or a purely paper premiere, and actual sales will begin much later. With the latest rumors pointing to Coffee Lake coming out at the end of the year, the first option seems more likely.

Intel Coffee Lake processors will be manufactured using 14nm technology and based on the same architecture used in the Kaby Lake and Skylake families, but with a number of improvements. But even in the absence of significant architectural differences, the new processors are likely to require new motherboards based on future 300-series chipsets to operate.

Intel previously promised that the performance advantage that Coffee Lake will give over current processors will be up to 30% (of course, not always and not in all applications), but all this increase can be easily attributed to the increased number of cores.

The event dedicated to the announcement of Intel Coffee Lake desktop processors will be streamed on Facebook Live, starting at 18:00 Kyiv time. The broadcast can also be viewed on the company's website.

Meanwhile, the characteristics of two Coffee Lake generation processors belonging to the younger Core i3 line have hit the network. We are talking about the Core i3-8100 and Core i3-8350K models. Whether the table with the characteristics of the processors found by the VideoCardz resource on the Chinese forums is reliable, you cannot say for sure, but there seems to be no reason to doubt the opposite. Moreover, this is not the first leak, pointing to the debut of quad-core Core i3 models in the new generation.

So, the table shows that both chips lack support for Hyper-Threading technology, which puts them on a par with the latest generation Core i5 processors, while the model with the K index in the name will be characterized by an unlocked multiplier, which can be considered a big plus given the lack of Supports Intel Turbo Boost Technology. The 4.0 GHz i3-8350K will have a TDP of 91W, while the 3.6GHz i3-8100 will have a TDP of 65W.