Thanks to the improvement of the technical process, it was possible to achieve a significant increase in performance, which will be more than 15% by test SYSMARK. Thus, this year the performance of Core i7 processors will grow more than in the past. This is shown on the slide from the presentation at the top under the heading "Promotion of the Moore Act by 14 Nm."

A new generation of processors on an improved platform of 14 nm is scheduled for the second half of 2017. They will be marked as a family of Core i7 / I5 / I3-8000 and replace the existing family of the 7th generation.

At the presentation for Intel investors, nothing spoke nothing about the plans for the release of the Cannonlake family (the former name of Skymont) - microprocessors on the 10-nm technological process. It is assumed that they should come out at the end of 2017, and the working model of Cannonlake on 10 nm recently showed at the CES exhibition. It was the CANNONLAKE family that was previously positioned as the 8th generation of processor architecture, which Skylake will change within the framework of the Tik-So strategy. Now another family has appeared that has nothing to do with Cannonlake. Perhaps this is an attempt to sell the old product in a new packaging.

Cancellation of the strategy "Tik-so"

Intel invariably adhered to the "Tik-So" strategy since 2006. Since then, every two years, she produced processors on a new process, significantly increasing the number of transistors on the crystal. Each transition to a new technical process was designated as "TIK", and the subsequent improvement in the microarchitecture with the same technical process is "so". Gigant semiconductor industry for ten years worked as a clock, issuing new architectures without failures.

It seems that in 2016 the "watches" of Intel jerked a bit for 14 nm, and the company announced.

In principle, nothing terrible in it. We repeat, this year the growth of the performance of chips (more than 15%) will even more than in the past (15%), said Intel. Maybe it is really better to squeeze the entire reserve from the existing process, optimizing it, and then move on. We cannot criticize Intel for a departure from the strategy that she most voluntarily installed.

One way or another, but now the strategy "Tik-so" was modified in a different appearance.

Instead of measured metronome, a new procedure is now implemented with a large focus on optimization. Perhaps the new architecture will not leave every two years, as it was before.

Why Intel does not force the transition to 10 nm? She doesn't need to do this, because she believes that he has been severely broken off in its technological superiority from competitors in the semiconductor industry (Samsung, TSMC and others). The company evaluates this gap around three years.

Such a stock allows you to feel quite confident.

New plant for 7 nm

The bright future of the Moore law should provide a new Intel Fab 42 plant, which will be able to provide production under the process of 7 nm.

Construction and equipment will take another three or four years and will require significant investment. The plant in Chandler (Arizona) will reduce the number of local unemployed for about 3,000 people (+ another 10,000 jobs will be added indirectly).

The construction of the plant in Chandler began in 2011. It should be the most advanced and innovative semiconductor enterprise in the world. The building itself was completed in 2013, but instead of installing equipment for 14 Nm at the beginning of 2014, Intel decided to postpone the launch of the conveyor. Currently, the plant is ready: air conditioning systems, heating and others - everything functions, it remains only to install and establish equipment. Intel does not plan to use this factory for the production of 10 Nm process inside, so in a few years here, it is likely to elapse the production according to the next rate of 7 nm.

According to Intel, the equipment will cost about $ 7 billion. Such is the cost of a modern industrial enterprise. It is not yet known which particular equipment will need. It is possible that Intel will begin to use photolithography in deep ultraviolet (EUV).

In the dawn, the two thousandth Intel hoped that by 2005 the frequency of processors would grow up to 10 GHz, and they will work under voltage below Volta. As we know, this did not happen. Approximately ten years ago, the law of scaling of Dennard ceased to work, which argued that with a decrease in the size of the transistors, it is possible to reduce the voltage supplied to the shutter and increase the switching speed. Since then, rarely, which processor receives a standard operating frequency above 4 GHz, but the nuclei has become more, the north bridge was moving on the crystal from the motherboard, other optimization and acceleration appeared. Now the Moore law slows down, empirical observation, which indicates a constant increase in the number of transistors on the crystal due to the decrease in their size.

A month after the announcement of the Core processors of the eight-generation for laptops, Intel officially introduced a new formation of chips and for desktop computers known as Coffee Lake codenamed. They are manufactured under an improved 14-nm technical process and, as in the case of mobile Kaby Lake Refresh, the number of computational nuclei compared to predecessors. If you do not take into account the HEDT class solutions, then this is the first increase in the number of cores in the "desktop" CPU Intel since 2006, when Core 2 Extreme QX6700 was released.

Core i7 and i5 nuclei have six, in Core i3 - four. At the same time, in the i7 series models, Hyperthreading technology is implemented, thanks to which they perform 12 threads at the same time. All six new products, the list of which is presented at the slide below, are equipped with an integrated GPU Intel HD Graphics 630 and can work with Intel Optane drives. Also supported DDR4-2666, the exception is only Core i3 compatible with DDR4-2400.

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

Speaking about the speed of new processors, the developers promise a risk of personnel frequency in modern games by 25%, 65% of the speed in such applications to create content as Adobe Photoshop, and 32% faster processing 4K video. Along with the computing power, prices have increased: for example, the cost of I7-8700K in part of 1000 pieces is $ 359, which is 18% more expensive than the 7700K model. On October 5th, the retail sale will arrive on October 5th of the current year, the supply of computers will begin in the fourth quarter.

Simultaneously with the CFU Coffee Lake, Intel has announced a support for the Z370 system logic kit. The press release states that the motherboards on the basis of the chipset meet the increased requirements for the power supply of the six-core processors of the eight-generation and allow you to set the RAM standard DDR4-2666. The first decisions on the Z370 base will also be announced on October 5, but some of them have already managed on the network until the term.

On June 2, Intel announced ten new 14 nanometer processors for desktop and mobile PCs of the Intel Core Fifth Generation Family (code name Broadwell-C) and five new 14-nanometer processors of the Intel Xeon E3-1200 V4 family.

Of the ten new Intel Core processors for the fifth generation (Broadwell-C) for desktop and mobile PCs, only two processors are focused on desktop PCs and have a LGA 1150 connector: these are quad-core models intel Core i5-5775c and Core i5-5675C. All other Intel Core processors of the fifth generation have BGA-execution and laptops are oriented. Brief specifications of new Broadwell-C processors are presented in the table.

	Connector	Number of nuclei / streams	Cache size L3, MB		TDP, W.	Graphic kernel
Core i7-5950hq	BGA.	4/8	6	2,9/3,7	47	Iris Pro Graphics 6200
Core i7-5850hq	BGA.	4/8	6	2,7/3,6	47	Iris Pro Graphics 6200
Core i7-5750hq	BGA.	4/8	6	2,5/3,4	47	Iris Pro Graphics 6200
Core i7-5700hq	BGA.	4/8	6	2,7/3,5	47	Intel HD Graphics 5600
Core i5-5350h	BGA.	2/4	4	3,1/3,5	47	Iris Pro Graphics 6200
Core i7-5775r.	BGA.	4/8	6	3,3/3,8	65	Iris Pro Graphics 6200
Core i5-5675r.	BGA.	4/4	4	3,1/3,6	65	Iris Pro Graphics 6200
Core i5-5575r.	BGA.	4/4	4	2,8/3,3	65	Iris Pro Graphics 6200
Core i7-5775c.	LGA 1150.	4/8	6	3,3/3,7	65	Iris Pro Graphics 6200
Core i5-5675c.	LGA 1150.	4/4	4	3,1/3,6	65	Iris Pro Graphics 6200

Of the five new processors of the Intel Xeon E3-1200 V4 family, only three models (Xeon E3-1285 V4, Xeon E3-1285L V4, Xeon E3-1265L V4) have a LGA 1150 connector, and two more models are made in BGA housing and are not intended for self-installing on the motherboard. Brief characteristics of new processors of the Intel Xeon E3-1200 V4 family are presented in the table.

	Connector	Number of nuclei / streams	Cache size L3, MB	Rated / Maximum Frequency, GHz	TDP, W.	Graphic kernel
Xeon E3-1285 V4.	LGA 1150.	4/8	6	3,5/3,8	95	Iris Pro Graphics P6300
Xeon E3-1285L V4.	LGA 1150.	4/8	6	3,4/3,8	65	Iris Pro Graphics P6300
Xeon E3-1265L V4.	LGA 1150.	4/8	6	2,3/3,3	35	Iris Pro Graphics P6300
Xeon E3-1278L V4.	BGA.	4/8	6	2,0/3,3	47	Iris Pro Graphics P6300
Xeon E3-1258L V4.	BGA.	2/4	6	1,8/3,2	47	Intel HD Graphics P5700

Thus, from 15 new Intel processors, only five models have a LGA 1150 connector and are focused on desktop systems. For users, the choice, of course, is small, especially if you consider that the processors of the Intel Xeon E3-1200 V4 family are oriented to servers, and not to custom PCs.

In the future, we will focus on the consideration of new 14 nanometer processors with the LGA 1150 connector.

So, the main features of the new Intel Core processors of the fifth generation and processors of the Intel Xeon E3-1200 V4 family is a new 14-nanometer microarchitecture of the codes name Broadwell. In principle, no fundamental difference between the processors of the Intel Xeon E3-1200 V4 family and the Intel Core processors of the fifth generation for desktop systems are not, therefore, in the future, all these processors will be denoted as Broadwell.

In general, it should be noted that the Broadwell microarchitecture is not just a Haswell in a 14-nanometer version. Rather, it is a bit of improved Haswell microarchitecture. However, Intel always does this: when moving to a new process, changes are made to the microarchitecture itself. In the case of Broadwell we are talking about cosmetic improvements. In particular, the volumes of internal buffers are increased, there are changes in the actuators of the processor kernel (changed a circuit of performing multiplication and division of floating semicolons).

We will not consider in detail all the features of the Broadwell microarchitecture. Of course, the transition to a new technical process allowed us to reduce the power consumption of processors (with an equal clock frequency), but no significant performance gains should be expected.

Perhaps the most significant difference between new Broadwell processors from Haswell is the fourth-level cache (L4-cache) Crystalwell. We will specify that such a cache L4 was present in the Haswell processors, but only in the top models of mobile processors, and in the Haswell processors for desktop PCs with the LGA 1150 connector it was not.

Recall that in some top models of Haswell's mobile processors, the IRIS Pro graphic core with additional EDRAM memory (Embedded DRAM) was implemented, which made it possible to solve the problem with the insufficient bandwidth of the memory used for GPU. EDRAM memory was a separate crystal that was located on one substrate with a processor crystal. This crystal received Crystalwell code name.

The EDRAM memory had a size of 128 MB and was manufactured using a 22-nanometer process. But the most important thing is that this EDRAM memory has been used not only for the needs of GPU, but also for the computational cores of the processor itself. That is, in fact, Crystalwell was an L4 cache separated between GPU and processor computational cores.

In all new Processors, Broadwell also has a separate EDRAM memory crystal with a size of 128 MB, which acts as a L4 cache and can be used by the graphics core and processor computing cores. Moreover, we note that the EDRAM memory in the 14-nanometer Processors Broadwell is exactly the same as in the top mobile processors of Haswell, that is, it is performed on a 22-nanometer process.

The following feature of the new Broadwell processors is a new graphic core with the code name Broadwell GT3E. In the version of the processors for desktop and mobile PCs (Intel Core i5 / i7) is the IRIS Pro Graphics 6200, and in the processors of the Intel Xeon E3-1200 V4 family - this is the IRIS Pro Graphics P6300 (except for the Xeon E3-1258L V4 model). We will not deliver in particular the architecture of Broadwell GT3E graphic cores (this is a topic for a separate article) and only briefly consider its main features.

Recall that the graphic core of Iris Pro was present only in Haswell mobile processors (Iris Pro Graphics 5100 and 5200). Moreover, in the IRIS Pro Graphics 5100 and 5200 graphics nuclei, there are 40 actuators (EU). New Iris Pro Graphics 6200 and Iris Pro Graphics 6200 and IRIS Pro Graphics P6300 have already endowed 48 EUs, and the EU organization system has changed. Each individual graphics processor unit contains 8 EUs, and the graphical module combines three graphic blocks. That is, in one graphical module contains 24 EU, and in the IRIS Pro Graphics 6200 or Iris Pro Graphics 6200 or IRIS Pro Graphics P6300, two modules are combined, that is, in the amount we get 48 EU.

As for the difference between the IRIS Pro Graphics 6200 graphics cores and Iris Pro Graphics P6300, then at the Iron level, this is the same (Broadwell GT3E), but they have different drivers. In the IRIS Pro Graphics P6300 driver, the driver is optimized for tasks specific to servers and graphics stations.

Before switching to a detailed consideration of the results of Broadwell testing, we will tell us more about several features of new processors.

First of all, new Broadwell processors (including Xeon E3-1200 V4) are compatible with motherboards based on Intel 9-series chipsets. We cannot argue that any Intel 9-Series chipset database will support these new Broadwell processors, but most of their boards support them. True, for this you will have to update the BIOS on the board, and the BIOS must support new processors. For example, for testing, we used the ASRock Z97 OC Formula board and without updating the BIOS, the system worked only if there is a discrete video card, and the image output through the graphics core of Broadwell processors was impossible.

The following feature of the new Broadwell processors are that the Core i7-5775C and Core i5-5675C models have a unlocked multiplication ratio, that is, accelerated. In the HASWELL processor family, such processors with a unlocked multiplication coefficient accounted for a K-series, and in the Broadwell family, instead of the letter "K", the letter "C" is used. But Xeon E3-1200 V4 processors are not supported (they cannot be increased multiplication ratio).

Now let's get acquainted with those processors that came to us for testing. This is a model, and. In fact, out of five new models with the LGA 1150 connector lacks only the Xeon E3-1285L V4 processor, which differs from the Xeon E3-1285 V4 model only with lowest power consumption (65 W instead of 95 W) and the fact that the nominal clock Just below (3.4 GHz instead of 3.5 GHz). In addition, for comparison, we also added Intel Core i7-4790K, which is the top processor in the Haswell family.

Characteristics of all tested processors are presented in the table:

	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i7-5775c.	Core i5-5675s	Core i7-4790k.
TEHPROTSESS, NM	14	14	14	14	22
Connector	LGA 1150.	LGA 1150.	LGA 1150.	LGA 1150.	LGA 1150.
Number of Cores	4	4	4	4	4
Number of streams	8	8	8	4	8
Cache L3, MB	6	6	6	4	8
Cash L4 (EDRAM), MB	128	128	128	128	N / A.
Nominal frequency, GHz	3,5	2,3	3,3	3,1	4,0
Maximum frequency, GHz	3,8	3,3	3,7	3,6	4,4
TDP, W.	95	35	65	65	88
Memory type	DDR3-1333 / 1600/1866		DDR3 -1333/1600.
Graphic kernel	Iris Pro Graphics P6300	Iris Pro Graphics P6300	Iris Pro Graphics 6200	Iris Pro Graphics 6200	HD Graphics 4600.
Number of GPU executive blocks	48 (Broadwell GT3E)	48 (Broadwell GT3E)	48 (Broadwell GT3E)	48 (Broadwell GT3E)	20 (Haswell GT2)
Rated graphics processor frequency, MHz	300	300	300	300	350
Maximum graphics processor frequency, GHz	1,15	1,05	1,15	1,1	1,25
VPRO technology	+	+	−	−	−
VT-X technology	+	+	+	+	+
VT-D technology	+	+	+	+	+
Cost, $	556	417	366	276	339

And now, after our express review of new Broadwell processors, we proceed directly to testing new products.

Test stand

For testing processors, we used the stand configuration stand:

Testing technique

Testing processors was carried out using our scripted benchmarks, and. More precisely, we took the methodology for testing workstations, but expanded it by adding tests from the IXBT Application Benchmark 2015 package and IXBT Game Benchmark 2015 game tests.

Thus, the following applications and benchmarks were used to test processors:

• Mediacoder x64 0.8.33.5680
• SVPmark 3.0
• Adobe Premiere Pro CC 2014.1 (Build 8.1.0)
• Adobe After Effects CC 2014.1.1 (Version 13.1.1.3)
• PhotoDex ProShow Producer 6.0.3410
• Adobe Photoshop CC 2014.2.1
• ACDSEE Pro 8.
• Adobe Illustrator CC 2014.1.1
• Adobe Audition CC 2014.2
• ABBYY FineReader 12.
• WinRar 5.11
• Dassault SolidWorks 2014 SP3 (Flow Simulation Package)
• Specapc for 3DS MAX 2015
• Specapc for maya 2012
• POV-RAY 3.7
• Maxon Cinebench R15
• SpecViewPerf v.12.0.2.
• SpecWPC 1.2.

In addition, games and gaming benchmarks were used for testing from the IXBT Game Benchmark package 2015. Testing in games was performed at the resolution of 1920x1080.

Additionally, we measured the power consumption of processors in idle mode and stress loading. To do this, a specialized software and hardware complex was used, connected to the rupture of power circuit circuits, that is, between the power supply and the motherboard.

To create a stress processor loading, we used the AIDA64 utility (Stress FPU and Stress GPU tests).

Test results

Power consumption of processors

So, let's start with the results of testing processors to power consumption. Test results are presented in the diagram.

The most voracious energy consumption, as expected, turned out to be the Intel Core i7-4790K processor with the declared TDP 88 W. Its actual power consumption in the stress mode was 119 W. At the same time, the temperature of the processor cores was 95 ° C and the trottling was observed.

The next power consumption was the Intel Core i7-5775C processor with the declared TDP 65 W. For this processor, power consumption in the stress mode was 72.5 watts. The temperature of the processor core reached 90 ° C, but the trottling was not observed.

The third power consumption is the Intel Xeon E3-1285 V4 C TDP 95 W processor. Its power consumption in the stress mode was 71 watts, and the temperature of the processor cores was 78 ° C

And the most economical in terms of energy consumption was the Intel Xeon E3-1265L V4 C TDP 35 W processor. In the stress load mode, the power consumption of this processor did not exceed 39 W, and the temperature of the processor cores was only 56 ° C.

Well, if you need to focus on the power consumption of processors, then it is necessary to state that Broadwell has significantly lower power consumption in comparison with Haswell.

Tests from the IXBT Application Benchmark 2015 package

Let's start with the tests that are part of the Benchmarck IXBT Application Benchmark 2015. We note that the integral performance result we calculated as the average geometric results in the logical groups of tests (video converting and video processing, creating a video content, etc.). To calculate the results in logical test groups, the same reference system was used as in Benchmark IXBT Application Benchmark 2015.

Full results Testing are shown in the table. In addition, we present test results on logical groups of tests in charts in a normalized form. The result of the Core i7-4790K processor is accepted for the reference.

Logical group tests	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i5-5675c.	Core i7-5775c.	Core i7-4790k.
Video converting and video processing, scores	364,3	316,7	272,6	280,5	314,0
Mediacoder x64 0.8.33.5680, seconds	125,4	144,8	170,7	155,4	132,3
SVPMARK 3.0, points	3349,6	2924,6	2552,7	2462,2	2627,3
Creating a video content, scores	302,6	264,4	273,3	264,5	290,9
Adobe Premiere Pro CC 2014.1, Seconds	503,0	579,0	634,6	612,0	556,9
Adobe After Effects CC 2014.1.1 (Test # 1), Seconds	666,8	768,0	802,0	758,8	695,3
Adobe After Effects CC 2014.1.1 (Test # 2), Seconds	330,0	372,2	327,3	372,4	342,0
PhotoDex ProShow Producer 6.0.3410, seconds	436,2	500,4	435,1	477,7	426,7
Processing digital photos, points	295,2	258,5	254,1	288,1	287.0
Adobe Photoshop CC 2014.2.1, Seconds	677,5	770,9	789,4	695,4	765,0
ACDSEE Pro 8, Seconds	289,1	331,4	334,8	295,8	271,0
Vector graphics, scores	150,6	130,7	140,6	147,2	177,7
Adobe Illustrator CC 2014.1.1, Seconds	341,9	394,0	366,3	349,9	289,8
Audio treatment, points	231,3	203,7	202,3	228,2	260,9
Adobe Audition CC 2014.2, Seconds	452,6	514,0	517,6	458,8	401,3
Declamation of text, scores	302,4	263,6	205,8	269,9	310,6
ABBYY FineReader 12, seconds	181,4	208,1	266,6	203,3	176,6
Archiving and unzipping data, scores	228,4	203,0	178,6	220,7	228,9
WinRAR 5.11 Archiving, Seconds	105,6	120,7	154,8	112,6	110,5
WinRAR 5.11 Unzipping, Seconds	7,3	8,1	8,29	7,4	7,0
Integral performance result, scores	259,1	226,8	212,8	237,6	262,7

So, as can be seen from the results of testing, the integral performance of the Intel Xeon E3-1285 V4 processor is almost no different from the Intel Core i7-4790K processor. However, this is an integral result for the totality of all applications used in the benchmark.

However, there are a number of applications in which the advantage on the Intel Xeon E3-1285 V4 processor side. These are applications such as Mediacoder x64 0.8.33.5680 and SVPMARK 3.0 (video converting and video processing), Adobe Premiere Pro CC 2014.1 and Adobe After Effects CC 2014.1.1 (Creating a video content), Adobe Photoshop CC 2014.2.1 and ACDSEE Pro 8 (digital processing Photos). In these applications, the higher clock frequency of the Intel Core i7-4790K processor does not allow him advantage over the Intel Xeon E3-1285 V4 processor.

But in such applications as Adobe Illustrator CC 2014.1.1 (Vector Graphics), Adobe Audition CC 2014.2 (Audio Processing), ABBYY FineReader 12 (text recognition) The advantage is on the side of the higher-frequency processor Intel Xeon E3-1285 V4. It is interesting to note here, tests based on Adobe Illustrator CC 2014.1.1 and Adobe Audition CC 2014.2 and Adobe Audition CC 2014.2 to a lesser extent (in comparison with other applications), the processor kernels are loaded.

And of course, there are tests in which the Intel Xeon E3-1285 V4 and Intel Core i7-4790K processors demonstrate the same performance. For example, this is a test based on WinRAR application 5.11.

In general, it should be noted that the Intel Core i7-4790K processor demonstrates higher performance (in comparison with the Intel Xeon E3-1285 V4 processor) in those applications in which not all the processor kernels or the loading of the nuclei is not fully. At the same time, all the processor kernels are loaded in tests, leadership on the Intel Xeon E3-1285 V4 processor side.

Calculations in the Dassault SolidWorks 2014 SP3 (Flow Simulation)

Application based on the Dassault SolidWorks 2014 SP3 application with an additional Flow Simulation package, we have been made separately, since the test system does not use a reference system as in Benchmark Benchmark 2015 in the Benchmark tests of IXBT Application Benchmark 2015.

Recall that in this test we are talking about hydro / aerodynamic and thermal calculations. In total, six different models are calculated, and the results of each subtest is the calculation time in seconds.

Detailed test results are presented in the table.

Test	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i5-5675c.	Core i7-5775c.	Core i7-4790k.
conjugate Heat TRANSFER, seconds	353.7	402.0	382.3	328.7	415.7
textile Machine, Seconds	399.3	449.3	441.0	415.0	510.0
rotating Impeller, seconds	247.0	278.7	271.3	246.3	318.7
cPU Cooler, Seconds	710.3	795.3	784.7	678.7	814.3
halogen Floodlight, seconds	322.3	373.3	352.7	331.3	366.3
electronic Components, Seconds	510.0	583.7	559.3	448.7	602.0
Total settlement time, second	2542,7	2882,3	2791,3	2448,7	3027,0

In addition, we also provide the normalized result of the rate of calculation (the amount, inverse the total calculation time). The result of the Core i7-4790K processor is accepted for the reference.

As can be seen according to the test results, in these specific calculations, leadership on the side of the Broadwell processors. All four Broadwell processors demonstrate a higher calculation rate in comparison with the Core i7-4790K processor. Apparently, these specific calculations affect the improvements in the executive blocks that were implemented in the Broadwell microarchitecture.

Specapc for 3DS MAX 2015

Next, consider the test results Specapc for 3DS MAX 2015 for Autodesk 3DS Max 2015 SP1. Detailed results of this test are presented in the table, and normalized results for CPU Composite Score and GPU Composite Score - in diagrams. The result of the Core i7-4790K processor is accepted for the reference.

Test	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i5-5675c.	Core i7-5775c.	Core i7-4790k.
CPU Composite Score	4,52	3,97	4,09	4,51	4,54
GPU Composite Score	2,36	2,16	2,35	2,37	1,39
LARGE MODEL COMPOSITE SCORE	1,75	1,59	1,68	1,73	1,21
Large Model CPU.	2,62	2,32	2,50	2,56	2,79
Large Model GPU.	1,17	1,08	1,13	1,17	0,52
Interacive Graphics.	2,45	2,22	2,49	2,46	1,61
Advanced Visual Styles.	2,29	2,08	2,23	2,25	1,19
Modeling	1,96	1,80	1,94	1,98	1,12
CPU Computing	3,38	3,04	3,15	3,37	3,35
CPU Rendering	5,99	5,18	5,29	6,01	5,99
GPU Rendering	3,13	2,86	3,07	3,16	1,74

The Specapc 3DS for Max 2015 test leads the Broadwell processors. Moreover, if the CPU Composite SCORE (CPU Composite SCORE), CER I7-4790K and Xeon E3-1285 V4 and Xeon E3-1285 V4 are demonstrating equal performance, then in GPU Composite Score), all Broadwell processors are significantly Ahead of the Core i7-4790K processor.

Specapc for maya 2012

Now let's look at the result of another test of three-dimensional modeling - Specapc for Maya 2012. Recall that this benchmark launched a pair of Autodesk Maya 2015.

The results of this test are presented in the table, and the normalized results are in diagrams. The result of the Core i7-4790K processor is accepted for the reference.

Test	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i5-5675c.	Core i7-5775c.	Core i7-4790k.
GFX Score.	1,96	1,75	1,87	1,91	1,67
CPU Score	5,47	4,79	4,76	5,41	5,35

In this test, the Xeon E3-1285 V4 processor demonstrates a slightly higher performance in comparison with the Core i7-4790K processor, however, the difference is not so significant as in the SPECAPC 3DS for Max 2015 package.

POV-RAY 3.7

In the POV-Ray 3.7 test (rendering of the three-dimensional model), the Core i7-4790K processor is the leader. In this case, a higher clock frequency (with an equal number of kernels) gives the advantage of the processor.

Test	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i5-5675c.	Core i7-5775c.	Core i7-4790k.
Render Average, PPS	1568,18	1348,81	1396,3	1560.6	1754,48

Cinebench R15

In Benchmark Cinebench R15, the result was ambiguous. In the OpenGL test, all Broadwell processors significantly exceed the Core i7-4790K processor, which is natural, since they are integrated into more productive graphics core. But in the processor test, on the contrary, the Core i7-4790K processor is more productive.

Test	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i5-5675c.	Core i7-5775c.	Core i7-4790k.
OpenGL, FPS.	71,88	66,4	72,57	73	33,5
CPU, CB.	774	667	572	771	850

SpecViewPerf v.12.0.2.

In the tests of the SPECViewPerf Package V.12.0.2, the results are determined mainly by the performance of the graphics core of the processor and, moreover, the optimization of the video driver to those or other applications. Therefore, in these tests, the Core i7-4790K processor is significantly lagging behind Broadwell processors.

Test results are presented in the table, as well as normalized on diagrams. The result of the Core i7-4790K processor is accepted for the reference.

Test	Xeon E3-1285 V4.	Xeon E3-1265L V4.	Core i5-5675c.	Core i7-5775c.	Core i7-4790k.
catia-04.	20,55	18,94	20,10	20,91	12,75
creo-01.	16,56	15,52	15,33	15,55	9,53
eNERGY-01.	0,11	0,10	0,10	0,10	0,08
maya-04.	19,47	18,31	19,87	20,32	2,83
mEDICAL-01.	2,16	1,98	2,06	2,15	1,60
showcase-01.	10,46	9,96	10,17	10,39	5,64
sNX-02.	12,72	11,92	3,51	3,55	3,71
sW-03.	31,32	28,47	28,93	29,60	22,63

2,36 Blender2,43 2,11 1,82 2,38 2,59 Handbrake.2,33 2,01 1,87 2,22 2,56 LuxRender.2,63 2,24 1,97 2,62 2,86 Iometer.15,9 15,98 16,07 15,87 16,06 Maya.1,73 1,63 1,71 1,68 0,24 Product Development.3,08 2,73 2,6 2,44 2,49 Rodinia.3,2 2,8 2,54 1,86 2,41 Calculix.1,77 1,27 1,49 1,76 1,97 Wpccfg.2,15 2,01 1,98 1,63 1,72 Iometer.20,97 20,84 20,91 20,89 21,13 catia-04.1,31 1,21 1,28 1,32 0,81 showcase-01.1,02 0,97 0,99 1,00 0,55 sNX-02.0,69 0,65 0,19 0,19 0,2 sW-03.1,51 1,36 1,38 1,4 1,08 Life Sciences.2,73 2,49 2,39 2,61 2,44 Lammp2,52 2,31 2,08 2,54 2,29 namd.2,47 2,14 2,1 2,46 2,63 Rodinia.2,89 2,51 2,23 2,37 2,3 MEDICAL-01.0,73 0,67 0,69 0,72 0,54 Iometer.11,59 11,51 11,49 11,45 11,5 FINANCIAL SERVICES.2,42 2,08 1,95 2,42 2,59 Monte Carlo2,55 2,20 2,21 2,55 2,63 Black Scholes.2,57 2,21 1,62 2,56 2,68 Binomial2,12 1,83 1,97 2,12 2,44 Energy2,72 2,46 2,18 2,62 2,72 FFTW.1,8 1,72 1,52 1,83 2,0 CONVOLUTION2,97 2,56 1,35 2,98 3,5 ENERGY-01.0,81 0,77 0,78 0,81 0,6 sRMP.3,2 2,83 2,49 3,15 2,87 Kirchhoff Migration.3,58 3,07 3,12 3,54 3,54 Poisson.1,79 1,52 1,56 1,41 2,12 Iometer.12,26 12,24 12,22 12,27 12,25 General Operation3,85 3,6 3,53 3,83 4,27 7zip.2,48 2,18 1,96 2,46 2,58 Python1,58 1,59 1,48 1,64 2,06 OCTAVE1,51 1,31 1,44 1,44 1,68 Iometer.37,21 36,95 37,2 37,03 37,4

It is impossible to say that in this test everything is definitely. In some scenarios (Media and Entertaiment, Product Development, Life Sciences), a higher result demonstrate Broadwell processors. There are scenarios (Financial Services, Energy, General Operation), where the advantage on the Core i7-4790K processor side or the results are about the same.

Gaming tests

And in conclusion, consider the results of testing processors in gaming tests. Recall that for testing, we used the following games and gaming benchmarks:

• Aliens vs Predator.
• WORLD OF TANKS 0.9.5
• Grid 2.
• Metro: LL REDUX
• Metro: 2033 Redux
• Hitman: Absolution
• Thief.
• Tomb Raider.
• Sleeping Dogs.
• Sniper Elite V2.

Testing was performed when resolving the screen 1920 × 1080 and in two setup modes: for maximum and minimum quality. Test results are presented in diagrams. In this case, the results are not normalized.

In gaming tests, the results are: All Broadwell processors demonstrate very close results, which is natural, since they use the same graphic core of Broadwell GT3E. And the most important thing is that when settings for the minimum quality, the Broadwell processors allow you to comfortably play (at FPS more than 40) in most games (with a resolution of 1920 × 1080).

On the other hand, if a discrete graphic card is used in the system, then there is simply no special meaning in the new Broadwell processors. That is, it makes no sense to change Haswell on Broadwell. Yes, and the price of Broadwells is not so much attractive. For example, Intel Core i7-5775C is more expensive than Intel Core i7-4790K.

However, Intel seems to do not bet on Broadwell desktop processors. The range of models is extremely modest, and on the SKYLAKE processor approach, so the Intel Core i7-575C and Core i5-5675C processors are unlikely to use special demand.

Server processors of the Xeon E3-1200 V4 family - this is a separate market segment. For most ordinary home users, such processors do not represent interest, but in the corporate sector of the market, these processors may and will be in demand.

08/21/2017, Mon, 09:36, MSK , Text: Vladimir Bahur

Intel announced the replenishment of the line of their mobile processors U-series Core chips of the eighth generation. The new generation of Coffee Lake processors for desktop PCs also will also appear this year, but later.

Four-generation four-year-old processors

Intel introduced four new Mobile Core i5 and Core i7 mobile processors in the U Line. All new chips have four computing cores with support for Hyper-Threading technology, which in total allows you to provide up to eight computing streams on each crystal.

Previous generations of Core mobile processors were produced with two physical cores and with Hyper-Threading technology supported four computing streams.

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

All Core 8 generation processors presented today (Kaby Lake Refresh), as well as their predecessors are manufactured in compliance with the norms of 14 Nm of the technological process, but "with improved characteristics", which led the announcement of the new 8 generation. According to Intel, the transition to the rules of 10 nm of the technical process will be held in the fall, but within the same time, the eighth generation.

The "real" architecture of the new generation under the working title Coffee Lake will be presented even later and will replenish the list of Core 8 generation chips. However, these chips will also be made according to the standards of 14 nm.

Intel Core processors of the new 8 generation

The transition to the norms of 10 nm will be the next step and debuts with the CANNON LAKE architecture. Thus, the list of Core processors of the eighth generation will include chips i7 / i5 / i3-8xxx three different architectures: Kaby Lake Refresh, Coffee Lake and Cannon Lake. Previously, one generation of Core was usually two varieties of architectures.

Details about architecture

New eight-generation Core processors operate on relatively low main clock frequencies (not higher than 1.9 GHz in the older model i7-8650U), making all the models to fit into a thermobacket (TDP) to 15 W at four computing kernels.

External view of the Core 8 generation processor

At the same time, thanks to Intel Turbo Boost Technology 2.0 technology, the chips are capable of dynamically increase the clock frequency more than twice (up to 4.2 GHz at the older model i7-8650U), which allows you to significantly increase the system performance as needed and remain in "Cold" state in standby mode.

Basic characteristics of the first four Core 8 generation processors

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

Photography of the crystal of the new Intel Core 8 generation chip

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

About performance and savings

According to the internal tests of the company, the new Core i7 and I5 generation mobile chips provide a capacity gain of up to 40% compared to the chips of the previous generation, and twice overtake chips of five years ago, for example, when comparing the new Core i5-8250u with Core i5- 3317U.

Almost 3 times higher Speed: 802.11AX 2x2 160 MHz allows you to develop a maximum theoretical data transfer rate up to 2402 Mbps, almost 3 times (2.8 times) higher than 802.11ac 2x2 80 MHz (867 Mbps ), as documented in the IEEE 802.11 wireless standard specifications. It requires the use of 802.11AX wireless router with similar configuration.

Compared to other I / O technologies for PCs, including ESATA, USB, and IEEE 1394 FireWire *. Real performance values \u200b\u200bmay vary depending on the hardware and software used. Be sure to use the device with THUNDERBOLT ™ technology. Additional information can be found on the website.

The best-in-class technology Wi-Fi 6: Intel® Wi-Fi 6 (GIG +) adapters support additional channels of 160 MHz, which makes it possible to achieve the maximum possible theoretical velocity (2402 Mbps) for typical Wi-Fi adapters 2x2 802.11ax PC. Intel® Premium Adapters Wi-Fi 6 (GIG +) allow 2-4 times the maximum theoretical speed compared with standard Wi-Fi 802.11ax PC 2x2 adapters (1201 Mbps) or 1x1 (600 Mbps), which supports 80 MHz channels on the mandatory requirement.

According to the results of the comparative test of the AIXPRT workload, performed for the Intel® Core ™ I7-1065G7 pre-generation processor and the Intel® Core ™ i7-8565 8-generation processor (Int8 results). The results of performance tests are based on testing as of May 23, 2019 and may not reflect all public security updates. Detailed information is presented in the configuration description. No system can be completely protected.

Intel Corporation is a sponsor and participant in the Benchmark XPRT developer community, as well as the main developer of XPRT performance tests. Principled Technologies is a Publisher of the XPRT Performance Tests Family. It is necessary to contact other sources of information and performance tests to get a complete product assessment that you plan to buy.

Changing the clock frequency or voltage can damage or reduce the service life of the processor and other system components, and can also lead to a deterioration in the stability and performance of the system. In the event of a change in processor specifications, products may not be subject to warranty service. For more information, contact the manufacturers of the system and components.

Intel and Intel logo are trademarks of Intel or its divisions in the United States and / or other countries.

* Other names and trademarks are the property of their legal owners. (If third-party names and trademarks are used)