What are the differences between Core i3 and i5 or i7 processors. Intel Core i5 processors

In the process of assembling or buying a new computer, users are sure to have a question. In this article, we will look at Intel Core i3, i5 and i7 processors, and also tell you what is the difference between these chips and what is better to choose for your computer.

Difference No. 1. The number of cores and support for Hyper-threading.

Perhaps, the main difference between Intel Core i3, i5 and i7 processors is the number of physical cores and support for Hyper-threading technology, which creates two threads of computation for each real physical core. Creation of two computation threads for each core allows more efficient use of the processing power of the processor core. Therefore, processors with Hyper-threading support have some performance advantage.

The number of cores and Hyper-threading technology support for most Intel Core i3, i5, and i7 processors can be summarized in the following table.

	Number of physical cores	Support for Hyper-threading technology	Number of threads
Intel Core i3	2	Yes	4
Intel Core i5	4	Not	4
Intel Core i7	4	Yes	8

But there are exceptions to this table.. Firstly, these are the Intel Core i7 processors of their "Extreme" line. These processors can have 6 or 8 physical processing cores. At the same time, they, like all Core i7 processors, have support for Hyper-threading technology, which means that the number of threads is twice the number of cores. Second, some mobile processors (laptop processors) are exempt. So some Intel Core i5 mobile processors have only 2 physical cores, but at the same time they have Hyper-threading support.

It should also be noted that Intel has already planned to increase the number of cores in its processors. According to the latest news, the Intel Core i5 and i7 processors with Coffee Lake architecture, which are scheduled for release in 2018, will have 6 physical cores and 12 threads.

Therefore, you should not completely trust the above table. If you are interested in the number of cores in a particular Intel processor, then it is better to check the official information on the website.

Difference number 2. The amount of cache memory.

Also, Intel Core i3, i5 and i7 processors differ in the amount of cache memory. The higher the processor class, the more cache memory it gets. Intel Core i7 processors get the most cache memory, Intel Core i5 gets a little less, and Intel Core i3 gets even less. Specific values ​​​​should be viewed in the characteristics of the processors. But for example, you can compare several processors from the 6th generation.

	Level 1 cache	Level 2 cache	Level 3 cache
Intel Core i7-6700	4 x 32 KB	4 x 256 KB	8 MB
Intel Core i5-6500	4 x 32 KB	4 x 256 KB	6 MB
Intel Core i3-6100	2 x 32 KB	2 x 256 KB	3 MB

It must be understood that a decrease in the amount of cache memory is associated with a decrease in the number of cores and threads. But, nevertheless, there is such a difference.

Difference No. 3. Clock speeds.

Typically, higher-end processors come with higher clock speeds. But, not everything is so clear-cut here. Not infrequently, Intel Core i3 can have higher frequencies than Intel Core i7. For example, let's take 3 processors from the 6th generation line.

	Clock frequency
Intel Core i7-6700	3.4GHz
Intel Core i5-6500	3.2GHz
Intel Core i3-6100	3.7GHz

Thus, Intel is trying to maintain the performance of Intel Core i3 processors at the right level.

Difference No. 4. Heat dissipation.

Another important difference between the Intel Core i3, i5 and i7 processors is the level of heat dissipation. The characteristic known as TDP or thermal design power is responsible for this. This characteristic tells how much heat the processor cooling system should remove. For example, let's take the TDP of three 6th generation Intel processors. As can be seen from the table, the higher the class of the processor, the more heat it produces and the more powerful the cooling system is needed.

	TDP
Intel Core i7-6700	65 W
Intel Core i5-6500	65 W
Intel Core i3-6100	51 W

It should be noted that TDP tends to decrease. With each generation of processors, TDP is getting lower. For example, the TDP of the 2nd generation Intel Core i5 processor was 95W. Now, as we see, only 65 watts.

Which is better Intel Core i3, i5 or i7?

The answer to this question depends on what kind of performance you need. The difference in the number of cores, threads, cache memory and clock speeds creates a noticeable difference in performance between the Core i3, i5 and i7.

• The Intel Core i3 processor is a great option for an office or budget home computer. If you have a video card of the appropriate level, it is quite possible to play computer games on a computer with an Intel Core i3 processor.
• Intel Core i5 processor - suitable for a powerful work or gaming computer. A modern Intel Core i5 can handle any graphics card without any problems, so you can play any games on a computer with such a processor, even at maximum settings.
• An Intel Core i7 processor is an option for those who know exactly why it needs such performance. A computer with such a processor is suitable, for example, for editing video or conducting game streams.

Once, a great sage in captain's uniform said that a computer would not be able to work without a processor. Since then, everyone considers it his duty to find the very processor, thanks to which his system will fly like a fighter.

From this article you will learn:

Since we simply cannot cover all known science chips, we want to focus on one interesting family of the Intelovich family - Core i5. They have very interesting characteristics and good performance.

Why this particular series and not i3 or i7? It's simple: excellent potential without overpaying for unnecessary instructions that the seventh line sins with. Yes, and more cores than in Core i3. You will quite naturally start arguing about support and you will be partially right, but 4 physical cores can do much more than 2 + 2 virtual ones.

Series history

Today on the agenda we have a comparison of Intel Core i5 processors of different generations. Here I would like to touch on such pressing topics as heat pack and the presence of solder under the lid. And if there is a mood, then we will also push especially interesting stones together with our foreheads. So let's go.

I would like to start with the fact that only desktop processors will be considered, and not options for a laptop. There will be a comparison of mobile chips, but another time.

The output frequency table looks like this:

Generation	Year of issue	Architecture	Series	socket	Number of cores/threads	Level 3 cache
1	2009 (2010)	Hehalem (Westmere)	i5-7xx (i5-6xx)	LGA 1156	4/4 (2/4)	8 MB (4 MB)
2	2011	Sandy Bridge	i5-2xxx	LGA 1155	4/4	6 MB
3	2012	Ivy Bridge	i5-3xxx	LGA 1155	4/4	6 MB
4	2013	Haswell	i5-4xxx	LGA 1150	4/4	6 MB
5	2015	Broadwell	i5-5xxx	LGA 1150	4/4	4 MB
6	2015	skylake	i5-6xxx	LGA 1151	4/4	6 MB
7	2017	Kaby Lake	i5-7xxx	LGA 1151	4/4	6 MB
8	2018	coffee lake	i5-8xxx	LGA 1151v2	6/6	9 MB

2009

The first representatives of the series saw the light back in 2009. They were created on 2 different architectures: Nehalem (45nm) and Westmere (32nm). The brightest representatives of the line should be called i5-750 (4 × 2.8 GHz) and i5-655K (3.2 GHz). The latter additionally had an unlocked multiplier and the possibility of overclocking, which indicated its high performance in games and not only.

The differences between the architectures lie in the fact that Westmare is built according to the 32 nm process technology and has 2 generation gates. Yes, they use less energy.

2011

This year saw the light of the second generation of processors - Sandy Bridge. Their distinguishing feature was the presence of a built-in Intel HD 2000 video core.

Among the abundance of i5-2xxx models, I would especially like to single out a CPU with an index of 2500K. At one time, it made a splash among gamers and enthusiasts, combining a high frequency of 3.2 GHz with Turbo Boost support and low cost. And yes, there was solder under the cover, not thermal paste, which additionally contributed to the high-quality acceleration of the stone without consequences.

2012

The debut of Ivy Bridge brought a 22nm process, higher frequencies, new DDR3, DDR3L and PCI-E 3.0 controllers, and USB 3.0 support (but only for i7).

The integrated graphics have evolved to the Intel HD 4000.

The most interesting solution on this platform was the Core i5-3570K with an unlocked multiplier and a frequency of up to 3.8 GHz in boost.

2013

The Haswell generation brought nothing supernatural except for the new LGA 1150 socket, the AVX 2.0 instruction set and the new HD 4600 graphics. In fact, all the emphasis was on energy saving, which the company managed to achieve.

But as a fly in the ointment, there is a replacement of solder with a thermal interface, which greatly reduced the overclocking potential of the top i5-4670K (and its updated version 4690K from the Haswell Refresh line).

2015

In fact, this is the same Haswell, transferred to the 14 nm architecture.

2016

The sixth iteration, named Skylake, brought an updated LGA 1151 socket, support for DDR4 RAM, 9th generation IGP, AVX 3.2 instructions, and SATA Express.

Among the processors, it is worth highlighting the i5-6600K and 6400T. The first was loved for high frequencies and an unlocked multiplier, and the second for its low cost and extremely low heat dissipation of 35 W despite Turbo Boost support.

2017

The era of Kaby Lake is the most controversial, as it brought absolutely nothing new to the desktop processor segment other than native USB 3.1 support. also, these stones completely refuse to run on Windows 7, 8 and 8.1, not to mention older versions.

The socket remained the same - LGA 1151. And the set of interesting processors has not changed - 7600K and 7400T. The reasons for people's love are the same as for Skylake.

2018

Goffee Lake processors are fundamentally different from their predecessors. Four cores were replaced by 6, which previously only the top versions of the X-series i7 could afford. The size of the L3 cache was increased to 9 MB, and the heat pack in most cases does not exceed 65 watts.

Of the entire collection, the i5-8600K model is considered the most interesting for the possibility of overclocking up to 4.3 GHz (though only 1 core). However, the public prefers the i5-8400 as the cheapest entry ticket.

Instead of totals

If we were asked what we would offer the lion's share of gamers, we would say without hesitation that the i5-8400. The benefits are obvious:

• price below $190
• 6 full physical cores;
• frequency up to 4 GHz in Turbo Boost
• heat pack 65 W
• complete fan.

Additionally, you do not have to select a “certain” RAM, as for the Ryzen 1600 (the main competitor, by the way), and the cores themselves in Intel. You lose additional virtual threads, but practice shows that in games they only reduce FPS without introducing certain adjustments to the gameplay.

By the way, if you don’t know where to buy, I recommend paying attention to some very popular and serious online store- at the same time you can find out there prices for i5 8400 From time to time I buy different gadgets here.

In any case, it's up to you. Until we meet again, do not forget to subscribe to the blog.

And more news for those who follow (solid state drives) - this rarely happens.

Within the framework of this material, using the example of models of microprocessors with indices 8600K and 8700K, a comparison of the CPU will be carried out and an answer will be given to what technological features will also be considered and recommendations regarding their use will be given. In addition to this, reviews of PC owners are given based on a database of chip models and their real cost today.

Specialization. Purpose

Before we figure out how the i5 differs from the i7, let's find out the scope of the families of these microprocessors. Like i5 - 8600K, and i7 - 8700K are designed to build high-performance modern computing systems. These include gaming PCs, workstations or graphics stations, computers for transcoding multimedia content, and even entry-level servers. The performance level of the chips in question is sufficient and allows you to run any software without any problems, including the most demanding ones.

CPU Specifications

Now let's look at the technical characteristics of these two microprocessors and, based on them, find the key differences between i5 and i7. The i5-8600K and i7-8700K CPUs belong to the 8th generation of Intel chips based on the Core microarchitecture, codenamed CoffeeLake. They are manufactured according to the technological standards of 14 nm and are equipped with 6 computing units. The clock frequency of the first of them can vary from 3.6 to 4.3 GHz, and the second - from 3.7 to 4.7 GHz. They also have an unlocked multiplier, and due to this they can be dispersed.

An important difference between i5 - 8600K and i7 - 8700K is the organization of the cache. The first CPU has the third level of fast volatile memory is 9 MB, and the second - 12 MB. In terms of RAM, these two CPU models are identical. They can address 64 GB of memory and are designed to work in combination with DDR4 modules. The recommended frequency of the latter is 2666 MHz. Also, the chips have an identical thermal package, which is equal to 95 watts.

Technological features of microprocessors

Comparison of i5 and i7 indicates that the former does not support HT technology, while the latter can be activated by the CPU. Due to this, on the i7 - 8700K microprocessor, it is possible to process twelve logical streams of program code on six real cores. This allows you to get in some cases a performance increase of 10-12 percent. The only important nuance is software optimization. That is, it must use all twelve threads in the process. Such software today includes various packages for transcoding multimedia information.

Regarding the technology of automatic increase in the clock frequency of the TurboBoost microprocessor, it can be noted that both the i5-8600K and the i7-8700K support it. Only in the second case, the algorithm of its work is more “hard”. That is, the first model of the CPU can automatically increase its frequency to 4.3 GHz, and the second - to 4.7 GHz. Therefore, the i7-8700K already has higher performance from the start. Moreover, this difference in some applications can be a solid 15%.

Booth configuration

Comparing the speed of the i7-8700K and i5-8600K is most correct with the i7-7700K, i5-7600K and Ryzen 1700X CPUs. Their cost and performance are comparable.

The test bench configuration includes the following components:

1. Motherboards Z270 (for chips 7 (for 8th generation CPU) and B350 (for AMD solution).
2. Thermalright Archon proprietary cooler.
3. Graphics accelerator GTX 1080 8Gb GDDR5.
4. RAM standard DDR4 with a nominal frequency of 2666 MHz 2 modules of 8 GB.
5. 960 GB SSD drive.
6. Power supply for 1.2 kW.

synthetic tests. Analysis of the results

Synthetic tests of i5 and i7 show a slight advantage of the CPU of the second family. In this case, the difference can reach a very solid 30 percent.

In the CineBench R15 utility in multi-threaded mode, the following results were obtained in conditional scores:

1. 1700X - 1534.
2. 8700K - 1392.
3. 8600K - 1021.
4. 7700K - 969.
5. 7600K - 664.

The presence of 8 cores and 16 threads allows AMD's flagship to easily outperform competitors from Intel. The difference between the 8700K and 8600K is due to the HT technology, increased frequencies, and increased L3 cache size. For the same reason, the heroes of this review are inferior to the flagship processor devices of the previous generation. They have four physical cores and due to this, lower performance.,

But in the single-threaded CineBench R15 test, the situation changes as follows:

1. 8700K - 198.
2. 7700K - 194.
3. 8600K - 184.
4. 7600K - 173.
5. 1700X - 155.

In this case, only one computing core is used, and the higher its frequency, the higher the level of performance. Therefore, the leader is 8700K, which can operate at 4.7 GHz. Then comes the 7700K running at 4.5GHz and the 8600K running at 4.3GHz. Next up is the 7600K running at 4.2GHz. Outsiders in this test were 1700X from AMD. Its frequency is 4.2 GHz.

The advantage of i7 over i5 in multi-threaded software can be up to 30%, and in single-threaded software this gap is reduced to 15%.

Game applications. Analysis of results

The performance of the i5 and i7 differs even less in gaming applications than in synthetic benchmarks. As a rule, such programs are optimized for one computing core. In rare cases, they may already use two such blocks. So the difference becomes even smaller.

In the game Dirt Rally, we got the following number of fps at FullHD resolution and average output image quality:

1. 8700K - 161 - 245.
2. 8600K - 149 - 230.
3. 7700K - 136 - 215.
4. 7600K - 134 - 206.
5. 1700X - 106 - 164.

The advantage of i7 chips over i5 CPUs is 10-15 frames per second. In percentage terms, this difference will be equal to 6-8%. The flagship AMD in this case is again far behind. It has a low clock frequency and this is the main limiting factor that does not yet allow it to compete on equal terms with Intel microprocessors.

CPU cost

The current recommended price for an Intel Core i5 with an 8600K index is $257. The flagship is priced at $359 by the manufacturer. But in this case, the choice must be made, not based on the cost, but on the tasks that are planned to be implemented in the future on the computer system being assembled.

Introduction This summer, Intel did something strange: it managed to replace two generations of processors that are aimed at mainstream personal computers. First, Haswell was replaced by processors with the Broadwell microarchitecture, but then within just a couple of months they lost their novelty status and gave way to Skylake processors, which will remain the most progressive CPUs for at least another year and a half. This generational leapfrog occurred mainly due to Intel's problems with the introduction of a new 14nm process technology, which is used in the production of both Broadwell and Skylake. Broadwell microarchitecture performance carriers were greatly delayed on their way to desktop systems, and their successors came out on a predetermined schedule, which led to the crumpled announcement of the fifth generation Core processors and a serious reduction in their life cycle. As a result of all these perturbations, in the desktop segment, Broadwell has occupied a very narrow niche of economical processors with a powerful graphics core and is now content with only a small level of sales characteristic of highly specialized products. The attention of the advanced part of users switched to the followers of Broadwell - Skylake processors.

It should be noted that over the past few years, Intel has not pleased its fans at all with an increase in the performance of its products. Each new generation of processors adds only a few percent in specific performance, which ultimately leads to a lack of clear incentives for users to upgrade old systems. But the release of Skylake - the generation of CPUs, on the way to which Intel, in fact, jumped over the step - inspired certain hopes that we would get a really worthwhile update to the most common computing platform. However, nothing like this happened: Intel performed in its usual repertoire. Broadwell was introduced to the public as an offshoot of the mainstream desktop processor line, while Skylake proved marginally faster than Haswell in most applications.

Therefore, despite all expectations, the appearance of Skylake on sale caused many skepticism. After reviewing the results of real tests, many buyers simply did not see the real point in switching to sixth generation Core processors. And indeed, the main trump card of fresh CPUs is primarily a new platform with accelerated internal interfaces, but not a new processor microarchitecture. And this means that Skylake offers little real incentive to upgrade past-generation based systems.

However, we still would not dissuade all users without exception from switching Skylake. The fact is that even though Intel is increasing the performance of its processors at a very restrained pace, since the advent of Sandy Bridge, which are still working in many systems, four generations of microarchitecture have already changed. Each step along the path of progress contributed to the increase in performance, and to this day, Skylake is able to offer a fairly significant increase in performance compared to its earlier predecessors. Just to see this, you need to compare it not with Haswell, but with the earlier representatives of the Core family that appeared before it.

In fact, that's exactly what we're going to do today. With all that said, we decided to see how much the performance of Core i7 processors has grown since 2011, and collected older Core i7s from the Sandy Bridge, Ivy Bridge, Haswell, Broadwell and Skylake generations in a single test. Having received the results of such testing, we will try to understand which processor owners should start upgrading old systems, and which of them can wait until the next generations of CPUs appear. Along the way, we will also look at the performance level of the new Core i7-5775C and Core i7-6700K processors of the Broadwell and Skylake generations, which have not yet been tested in our laboratory.

Comparative characteristics of tested CPUs

From Sandy Bridge to Skylake: Specific Performance Comparison

In order to remember how the specific performance of Intel processors has changed over the past five years, we decided to start with a simple test in which we compared the speed of Sandy Bridge, Ivy Bridge, Haswell, Broadwell and Skylake, reduced to the same frequency 4 .0 GHz. In this comparison, we used the Core i7 processors, that is, quad-core processors with Hyper-Threading technology.

The SYSmark 2014 1.5 complex test was taken as the main test tool, which is good because it reproduces typical user activity in common office applications, when creating and processing multimedia content, and when solving computing problems. The following graphs show the results obtained. For ease of perception, they are normalized, the performance of Sandy Bridge is taken as 100 percent.

The integral indicator SYSmark 2014 1.5 allows us to make the following observations. The transition from Sandy Bridge to Ivy Bridge increased the specific productivity very slightly - by about 3-4 percent. The next move to Haswell was far more rewarding, resulting in a 12 percent improvement in performance. And this is the maximum increase that can be observed on the above graph. After all, Broadwell overtakes Haswell by only 7 percent, and the transition from Broadwell to Skylake increases the specific performance by only 1-2 percent. All the progress from Sandy Bridge to Skylake translates into a 26 percent increase in performance at a constant clock speed.

A more detailed interpretation of the obtained SYSmark 2014 1.5 indicators can be seen in the following three graphs, where the integral performance index is decomposed into components by application type.

Pay attention, most noticeably with the introduction of new versions of microarchitectures, multimedia applications are added to the speed of execution. In them, the Skylake microarchitecture outperforms Sandy Bridge by as much as 33 percent. But in counting problems, on the contrary, progress is manifested least of all. Moreover, with such a load, the step from Broadwell to Skylake even turns into a slight decrease in specific performance.

Now that we have an idea of ​​what happened to the specific performance of Intel processors over the past few years, let's try to figure out what the observed changes were due to.

From Sandy Bridge to Skylake: what has changed in Intel processors

We decided to make the reference point in the comparison of different Core i7 representatives of the Sandy Bridge generation for a reason. It was this design that laid a solid foundation for all further improvement of productive Intel processors up to today's Skylake. Thus, representatives of the Sandy Bridge family became the first highly integrated CPUs in which both computing and graphics cores, as well as a north bridge with an L3 cache and a memory controller, were assembled in one semiconductor chip. In addition, for the first time they began to use an internal ring bus, through which the problem of highly efficient interaction of all structural units that make up such a complex processor was solved. All subsequent generations of CPUs continue to follow these universal principles of construction laid down in the Sandy Bridge microarchitecture without any serious adjustments.

The internal microarchitecture of computing cores has undergone significant changes in Sandy Bridge. It not only implemented support for the new AES-NI and AVX instruction sets, but also found numerous major improvements in the depths of the execution pipeline. It was in Sandy Bridge that a separate zero-level cache was added for decoded instructions; a completely new command reordering block has appeared, based on the use of a physical register file; branch prediction algorithms have been significantly improved; and in addition, two of the three execution ports for working with data have become unified. Such heterogeneous reforms, carried out at once at all stages of the pipeline, made it possible to seriously increase the specific performance of Sandy Bridge, which immediately increased by almost 15 percent compared to the previous generation Nehalem processors. To this was added a 15% increase in nominal clock frequencies and excellent overclocking potential, as a result of which, in total, we got a family of processors, which Intel still sets as an example, as an exemplary embodiment of the "so" phase in the company's pendulum development concept.

Indeed, we have not seen improvements in the microarchitecture after Sandy Bridge that are similar in terms of mass and effectiveness. All subsequent generations of processor designs have made much smaller improvements to the cores. Perhaps this is a reflection of the lack of real competition in the processor market, perhaps the reason for the slowdown in progress lies in Intel's desire to focus on improving graphics cores, or maybe Sandy Bridge just turned out to be such a successful project that its further development requires too much effort.

The transition from Sandy Bridge to Ivy Bridge perfectly illustrates the decline in the intensity of innovation that has occurred. Despite the fact that the next generation of processors after Sandy Bridge was transferred to a new production technology with 22nm standards, its clock speeds did not increase at all. The improvements made in the design mainly affected the memory controller that became more flexible and the PCI Express bus controller, which received compatibility with the third version of this standard. As for the microarchitecture of the computing cores, some cosmetic changes made it possible to speed up the execution of division operations and slightly increase the efficiency of Hyper-Threading technology, and nothing more. As a result, the increase in specific productivity amounted to no more than 5 percent.

At the same time, the introduction of Ivy Bridge brought something that the millionth army of overclockers now bitterly regrets. Starting with processors of this generation, Intel abandoned the pairing of the CPU semiconductor chip and the cover covering it by means of flux-free soldering and switched to filling the space between them with a polymer thermal interface material with very dubious heat-conducting properties. This artificially worsened the frequency potential and made the Ivy Bridge processors, as well as all their followers, noticeably less overclockable compared to the "oldies" Sandy Bridge, which are very peppy in this regard.

However, Ivy Bridge is just a tick, and therefore no one promised any special breakthroughs in these processors. However, the next generation, Haswell, did not bring any inspiring performance growth, which, unlike Ivy Bridge, is already in the “so” phase. And this is actually a little strange, since there are a lot of various improvements in the Haswell microarchitecture, and they are dispersed in different parts of the execution pipeline, which in total could well increase the overall pace of command execution.

For example, in the input part of the pipeline, branch prediction performance has been improved, and the queue of decoded instructions has been dynamically shared between parallel threads coexisting within Hyper-Threading technology. Along the way, there was an increase in the window of out-of-order execution of commands, which in total should have increased the share of the code executed in parallel by the processor. Directly in the execution unit, two additional functional ports were added, aimed at processing integer commands, servicing branches and saving data. Thanks to this, Haswell was able to process up to eight micro-ops per clock - a third more than its predecessors. What's more, the new microarchitecture also doubled the throughput of the L1 and L2 caches.

Thus, improvements in the Haswell microarchitecture did not affect only the speed of the decoder, which seems to have become the bottleneck in modern Core processors at the moment. After all, despite the impressive list of improvements, the increase in specific performance in Haswell compared to Ivy Bridge was only about 5-10 percent. But for the sake of justice, it should be noted that the acceleration is noticeably much stronger on vector operations. And the greatest benefit can be seen in applications using the new AVX2 and FMA commands, support for which has also appeared in this microarchitecture.

Haswell processors, like Ivy Bridge, were also not particularly liked by enthusiasts at first. Especially when you consider the fact that in the original version they did not offer any increase in clock frequencies. However, a year after their debut, Haswell began to seem noticeably more attractive. First, there has been an increase in applications that capitalize on the strengths of this architecture and use vector instructions. Secondly, Intel was able to correct the situation with frequencies. Later versions of Haswell, which received their own code name Devil's Canyon, were able to increase the advantage over their predecessors by increasing the clock speed, which finally broke through the 4 GHz ceiling. In addition, following the lead of overclockers, Intel improved the polymer thermal interface under the processor cover, which made Devil's Canyon more suitable for overclocking. Of course, not as malleable as Sandy Bridge, but nonetheless.

And with such baggage, Intel approached Broadwell. Since the main key feature of these processors was to be a new manufacturing technology with 14-nm standards, no significant innovations were planned in their microarchitecture - it should have been almost the most banal “tick”. Everything necessary for the success of new products could well be provided by only one thin process technology with second-generation FinFET transistors, which in theory allows reducing power consumption and raising frequencies. However, the practical implementation of the new technology turned into a series of failures, as a result of which Broadwell got only economy, but not high frequencies. As a result, those processors of this generation that Intel introduced for desktop systems came out more like mobile CPUs than like followers of the Devil's Canyon business. Moreover, in addition to truncated thermal packages and rolled back frequencies, they differ from their predecessors in a smaller L3 cache, which, however, is somewhat offset by the appearance of a fourth-level cache located on a separate chip.

At the same frequency as Haswell, Broadwell processors show a roughly 7% advantage, provided by both the addition of an additional data caching layer and another improvement in the branch prediction algorithm along with an increase in the main internal buffers. In addition, Broadwell has new and faster execution schemes for multiply and divide instructions. However, all these small improvements are canceled out by the clock speed fiasco, which takes us back to the pre-Sandy Bridge era. So, for example, the older overclocker Core i7-5775C of the Broadwell generation is inferior in frequency to the Core i7-4790K by as much as 700 MHz. It is clear that it is pointless to expect some kind of increase in productivity against this background, if only there were no serious drop in it.

In many ways, it was precisely because of this that Broadwell turned out to be unattractive to the bulk of users. Yes, the processors of this family are highly economical and even fit into a thermal package with 65-watt frames, but who cares, by and large? The overclocking potential of the first generation 14nm CPU turned out to be quite restrained. We are not talking about any work at frequencies approaching the 5 GHz bar. The maximum that can be achieved from Broadwell using air cooling lies in the vicinity of 4.2 GHz. In other words, the fifth generation of Core came out at Intel, at least strange. Which, by the way, the microprocessor giant eventually regretted: Intel representatives note that the late release of Broadwell for desktop computers, its shortened life cycle and atypical characteristics negatively affected sales, and the company does not plan to embark on such experiments anymore.

Against this background, the newest Skylake is presented not so much as a further development of the Intel microarchitecture, but as a kind of work on bugs. Despite the fact that the production of this generation of CPUs uses the same 14nm process technology as in the case of Broadwell, Skylake has no problems with high frequencies. The nominal frequencies of the sixth generation Core processors returned to those indicators that were characteristic of their 22nm predecessors, and the overclocking potential even increased slightly. Overclockers played into the hands of the fact that in Skylake the processor power converter again migrated to the motherboard and thereby reduced the total heat dissipation of the CPU during overclocking. The only pity is that Intel never returned to using an effective thermal interface between the chip and the processor cover.

But as for the basic microarchitecture of computing cores, despite the fact that Skylake, like Haswell, is the embodiment of the “so” phase, there are very few innovations in it. Moreover, most of them are aimed at expanding the input part of the execution pipeline, while the rest of the pipeline remained without any significant changes. The changes relate to improving the performance of branch prediction and improving the efficiency of the prefetch block, and nothing more. At the same time, some of the optimizations serve not so much to improve performance as they are aimed at another increase in energy efficiency. Therefore, one should not be surprised that Skylake is almost the same as Broadwell in terms of its specific performance.

However, there are exceptions: in some cases, Skylake can outperform its predecessors in performance and more noticeably. The fact is that in this microarchitecture the memory subsystem has been improved. The in-processor ring bus became faster, and this ultimately increased the bandwidth of the L3 cache. Plus, the memory controller received support for DDR4 SDRAM memory operating at high frequencies.

But in the end, nevertheless, it turns out, no matter what Intel says about the progressiveness of Skylake, from the point of view of ordinary users, this is a rather weak update. The main improvements in Skylake are made in the graphics core and in energy efficiency, which opens the way for such CPUs into fanless tablet form factor systems. Desktop representatives of this generation differ from the same Haswell not too noticeably. Even if we close our eyes to the existence of the intermediate generation of Broadwell, and compare Skylake directly with Haswell, then the observed increase in specific productivity will be about 7-8 percent, which can hardly be called an impressive manifestation of technical progress.

Along the way, it should be noted that the improvement of technological production processes does not live up to expectations. On the way from Sandy Bridge to Skylake, Intel changed two semiconductor technologies and more than halved the thickness of transistor gates. However, the modern 14nm process technology, compared to the 32nm technology five years ago, did not allow increasing the operating frequencies of processors. All Core processors of the last five generations have very similar clock speeds, which, if they exceed the 4 GHz mark, are very insignificant.

For a visual illustration of this fact, you can look at the following graph, which displays the clock frequency of older overclocking Core i7 processors of different generations.

Moreover, the peak clock frequency is not even on Skylake. Haswell processors belonging to the Devil's Canyon subgroup can boast of the maximum frequency. Their nominal frequency is 4.0 GHz, but thanks to the turbo mode in real conditions they are able to accelerate to 4.4 GHz. For modern Skylake, the maximum frequency is only 4.2 GHz.

All this, of course, affects the final performance of real representatives of various CPU families. And then we propose to see how all this affects the performance of platforms built on the basis of the flagship processors of each of the Sandy Bridge, Ivy Bridge, Haswell, Broadwell and Skylake families.

How We Tested

The comparison involved five Core i7 processors of different generations: Core i7-2700K, Core i7-3770K, Core i7-4790K, Core i7-5775C and Core i7-6700K. Therefore, the list of components involved in testing turned out to be quite extensive:

Processors:

Intel Core i7-2600K (Sandy Bridge, 4 cores + HT, 3.4-3.8 GHz, 8 MB L3);
Intel Core i7-3770K (Ivy Bridge, 4 cores + HT, 3.5-3.9 GHz, 8 MB L3);
Intel Core i7-4790K (Haswell Refresh, 4 cores + HT, 4.0-4.4 GHz, 8 MB L3);
Intel Core i7-5775C (Broadwell, 4 cores, 3.3-3.7GHz, 6MB L3, 128MB L4).
Intel Core i7-6700K (Skylake, 4 cores, 4.0-4.2 GHz, 8 MB L3).

CPU cooler: Noctua NH-U14S.
Motherboards:

ASUS Z170 Pro Gaming (LGA 1151, Intel Z170);
ASUS Z97-Pro (LGA 1150, Intel Z97);
ASUS P8Z77-V Deluxe (LGA1155, Intel Z77).

Memory:

2x8 GB DDR3-2133 SDRAM, 9-11-11-31 (G.Skill F3-2133C9D-16GTX);
2x8 GB DDR4-2666 SDRAM, 15-15-15-35 (Corsair Vengeance LPX CMK16GX4M2A2666C16R).

Video Card: NVIDIA GeForce GTX 980 Ti (6 GB/384-bit GDDR5, 1000-1076/7010 MHz)
Disk subsystem: Kingston HyperX Savage 480 GB (SHSS37A/480G).
Power supply: Corsair RM850i ​​(80 Plus Gold, 850 W).

Testing was performed on the Microsoft Windows 10 Enterprise Build 10240 operating system using the following set of drivers:

Intel Chipset Driver 10.1.1.8;
Intel Management Engine Interface Driver 11.0.0.1157;
NVIDIA GeForce 358.50 Driver.

Performance

Overall Performance

To evaluate the performance of processors in common tasks, we traditionally use the Bapco SYSmark test package, which simulates the user's work in real common modern office programs and applications for creating and processing digital content. The idea of ​​the test is very simple: it produces a single metric that characterizes the average weighted speed of a computer during everyday use. After the release of the Windows 10 operating system, this benchmark has been updated once again, and now we are using the latest version - SYSmark 2014 1.5.

When comparing Core i7 of different generations, when they operate in their nominal modes, the results are not at all the same as when compared at a single clock frequency. Still, the real frequency and features of the turbo mode have a fairly significant impact on performance. For example, according to the data obtained, the Core i7-6700K is faster than the Core i7-5775C by as much as 11 percent, but its advantage over the Core i7-4790K is very small - it is only about 3 percent. At the same time, one cannot ignore the fact that the latest Skylake is significantly faster than the processors of the Sandy Bridge and Ivy Bridge generations. Its advantage over the Core i7-2700K and Core i7-3770K reaches 33 and 28 percent, respectively.

A deeper understanding of the SYSmark 2014 1.5 results can provide insight into the performance scores obtained in various system usage scenarios. The Office Productivity scenario models typical office work: word preparation, spreadsheet processing, e-mail, and Internet browsing. The script uses the following set of applications: Adobe Acrobat XI Pro, Google Chrome 32, Microsoft Excel 2013, Microsoft OneNote 2013, Microsoft Outlook 2013, Microsoft PowerPoint 2013, Microsoft Word 2013, WinZip Pro 17.5 Pro.

The Media Creation scenario simulates the creation of a commercial using pre-captured digital images and video. For this purpose, the popular Adobe Photoshop CS6 Extended, Adobe Premiere Pro CS6 and Trimble SketchUp Pro 2013 packages are used.

The Data/Financial Analysis scenario is dedicated to statistical analysis and investment forecasting based on a certain financial model. The scenario uses large amounts of numerical data and two applications Microsoft Excel 2013 and WinZip Pro 17.5 Pro.

The results obtained by us under various load scenarios qualitatively repeat the general indicators of SYSmark 2014 1.5. Only the fact that the Core i7-4790K processor does not look outdated at all attracts attention. It noticeably loses to the latest Core i7-6700K only in the Data/Financial Analysis calculation scenario, and in other cases it is either inferior to its successor by a very inconspicuous amount, or even turns out to be faster. For example, a member of the Haswell family is ahead of the new Skylake in office applications. But processors from older release years, the Core i7-2700K and Core i7-3770K, seem to be somewhat outdated offerings. They lose from 25 to 40 percent to the novelty in different types of tasks, and this, perhaps, is quite sufficient reason for the Core i7-6700K to be considered as a worthy replacement.

Gaming Performance

As you know, the performance of platforms equipped with high-performance processors in the vast majority of modern games is determined by the power of the graphics subsystem. That is why, when testing processors, we choose the most processor-intensive games, and measure the number of frames twice. The first pass tests are carried out without turning on anti-aliasing and setting far from the highest resolutions. Such settings allow you to evaluate how well processors perform with a gaming load in general, which means they allow you to speculate about how the tested computing platforms will behave in the future, when faster variants of graphics accelerators appear on the market. The second pass is performed with realistic settings - when choosing FullHD-resolution and the maximum level of full-screen anti-aliasing. In our opinion, these results are no less interesting, as they answer the frequently asked question about what level of gaming performance processors can provide right now - in modern conditions.

However, in this test, we have assembled a powerful graphics subsystem based on the flagship NVIDIA GeForce GTX 980 Ti graphics card. And as a result, in some games, the frame rate showed dependence on processor performance even in FullHD resolution.

Results in FullHD resolution with maximum quality settings

Typically, the impact of processors on gaming performance, especially when it comes to powerful representatives of the Core i7 series, is negligible. However, when comparing five Core i7 different generations, the results are not at all uniform. Even at the highest quality settings, the graphics of the Core i7-6700K and Core i7-5775C show the highest gaming performance, while the older Core i7 lag behind them. Thus, the frame rate obtained in a system with a Core i7-6700K exceeds the performance of a system based on a Core i7-4770K by an inconspicuous one percent, but the Core i7-2700K and Core i7-3770K processors already seem to be a significantly worse basis for a gaming system. Switching from a Core i7-2700K or Core i7-3770K to the latest Core i7-6700K results in a 5-7 percent increase in fps, which can have quite a noticeable impact on the quality of the gameplay.

You can see all this much more clearly if you look at the gaming performance of processors with reduced image quality, when the frame rate does not rest against the power of the graphics subsystem.

Results at reduced resolution

The latest Core i7-6700K again manages to show the highest performance among all the latest generations of Core i7. Its superiority over the Core i7-5775C is about 5 percent, and over the Core i7-4690K - about 10 percent. There is nothing strange in this: games are quite sensitive to the speed of the memory subsystem, and it is in this direction that Skylake has made serious improvements. But the superiority of the Core i7-6700K over the Core i7-2700K and Core i7-3770K is much more noticeable. The older Sandy Bridge lags behind the novelty by 30-35 percent, and Ivy Bridge loses to it in the region of 20-30 percent. In other words, no matter how Intel was scolded for too slow improvement of its own processors, the company was able to increase the speed of its CPUs by a third over the past five years, and this is a very tangible result.

Testing in real games is completed by the results of the popular synthetic benchmark Futuremark 3DMark.

They echo the gaming performance and the results that Futuremark 3DMark gives. When the microarchitecture of Core i7 processors was transferred from Sandy Bridge to Ivy Bridge, 3DMark scores increased by 2 to 7 percent. The introduction of the Haswell design and the release of the Devil's Canyon processors added an additional 7-14 percent to the performance of the older Core i7. However, then the appearance of the Core i7-5775C, which has a relatively low clock speed, somewhat rolled back the performance. And the latest Core i7-6700K, in fact, had to take the rap for two generations of microarchitecture at once. The increase in the final 3DMark rating for the new Skylake family processor compared to the Core i7-4790K was up to 7 percent. And in fact, this is not so much: after all, Haswell processors have been able to bring the most noticeable performance improvement over the past five years. The latest generations of desktop processors are indeed somewhat disappointing.

Application Tests

In Autodesk 3ds max 2016 we are testing the final rendering speed. Measures the time it takes to render at 1920x1080 resolution using the mental ray renderer for one frame of a standard Hummer scene.

Another test of the final rendering is carried out by us using the popular free 3D graphics package Blender 2.75a. In it, we measure the duration of building the final model from Blender Cycles Benchmark rev4.

To measure the speed of photorealistic 3D rendering, we used the Cinebench R15 test. Maxon recently updated its benchmark, and now it again allows you to evaluate the speed of various platforms when rendering in the latest versions of the Cinema 4D animation package.

The performance of websites and online applications built using modern technologies is measured by us in the new Microsoft Edge browser 20.10240.16384.0. For this, a specialized WebXPRT 2015 test is used, which implements the algorithms actually used in Internet applications in HTML5 and JavaScript.

Graphics performance testing takes place in Adobe Photoshop CC 2015. The average execution time of the test script, which is a creatively reworked Retouch Artists Photoshop Speed ​​Test, which involves the typical processing of four 24-megapixel images taken by a digital camera, is measured.

Due to numerous requests from amateur photographers, we conducted a performance test in the graphics program Adobe Photoshop Lightroom 6.1. The test scenario includes post-processing and export to JPEG at 1920x1080 resolution and maximum quality of two hundred 12-megapixel RAW images taken with a Nikon D300 digital camera.

Adobe Premiere Pro CC 2015 tests non-linear video editing performance. Measures rendering time to H.264 Blu-ray for a project containing HDV 1080p25 footage with various effects applied.

To measure the speed of processors during information compression, we use the WinRAR 5.3 archiver, with the help of which we archive a folder with various files with a total volume of 1.7 GB with the maximum compression ratio.

The x264 FHD Benchmark 1.0.1 (64bit) test is used to estimate the speed of transcoding video to H.264 format, based on measuring the time it takes x264 encoder to encode source video to MPEG-4/AVC format with resolution [email protected] and default settings. It should be noted that the results of this benchmark are of great practical importance, since the x264 encoder is the basis of numerous popular transcoding utilities, such as HandBrake, MeGUI, VirtualDub, and so on. We periodically update the encoder used for performance measurements, and version r2538 took part in this testing, which supports all modern instruction sets, including AVX2.

In addition, we have added a new x265 encoder to the list of test applications, designed to transcode video into the promising H.265/HEVC format, which is a logical continuation of H.264 and is characterized by more efficient compression algorithms. To evaluate the performance, the original [email protected] Y4M video file that is transcoded to H.265 format with medium profile. The release of the encoder version 1.7 took part in this testing.

The advantage of the Core i7-6700K over its early predecessors in various applications is beyond doubt. However, two types of tasks have benefited most from the evolution that has taken place. Firstly, related to the processing of multimedia content, whether it be video or images. Secondly, final rendering in 3D modeling and design packages. In general, in such cases, the Core i7-6700K outperforms the Core i7-2700K by at least 40-50 percent. And sometimes you can see a much more impressive improvement in speed. So, when transcoding video with the x265 codec, the latest Core i7-6700K gives exactly twice as much performance as the old Core i7-2700K.

If we talk about the increase in the speed of performing resource-intensive tasks that the Core i7-6700K can provide compared to the Core i7-4790K, then there are no such impressive illustrations of the results of the work of Intel engineers. The maximum advantage of the novelty is observed in Lightroom, here Skylake turned out to be one and a half times better. But this is rather an exception to the rule. In most multimedia tasks, however, the Core i7-6700K offers only a 10 percent performance improvement over the Core i7-4790K. And with a load of a different nature, the difference in speed is even less or even absent.

Separately, a few words need to be said about the result shown by the Core i7-5775C. Due to the low clock speed, this processor is slower than the Core i7-4790K and Core i7-6700K. But do not forget that its key characteristic is efficiency. And it is quite capable of becoming one of the best options in terms of specific performance per watt of electricity consumed. We will easily verify this in the next section.

power usage

Skylake processors are manufactured on a modern 14nm process with second-generation 3D transistors, however, despite this, their TDP has increased to 91W. In other words, the new CPUs are not only “hotter” than 65-watt Broadwells, but also outperform Haswells in terms of calculated heat dissipation, produced using 22-nm technology and coexisting within the 88-watt thermal package. The reason, obviously, is that initially the Skylake architecture was optimized with an eye not to high frequencies, but to energy efficiency and the possibility of using it in mobile devices. Therefore, in order for the desktop Skylake to receive acceptable clock frequencies lying in the vicinity of the 4 GHz mark, the supply voltage had to be turned up, which inevitably affected power consumption and heat dissipation.

However, Broadwell processors did not differ in low operating voltages either, so there is a hope that the 91-watt Skylake thermal package was received due to some formal circumstances and, in fact, they will not be more voracious than their predecessors. Let's check!

The new Corsair RM850i ​​digital power supply used by us in the test system allows us to monitor the consumed and output electrical power, which we use for measurements. The following graph shows the total consumption of systems (without a monitor), measured "after" the power supply, which is the sum of the power consumption of all components involved in the system. The efficiency of the power supply itself in this case is not taken into account. To properly assess energy consumption, we have activated the turbo mode and all available energy-saving technologies.

In the idle state, a qualitative leap in the efficiency of desktop platforms occurred with the release of Broadwell. The Core i7-5775C and Core i7-6700K have noticeably lower idle consumption.

But under the load in the form of video transcoding, the most economical CPU options are Core i7-5775C and Core i7-3770K. The latest Core i7-6700K consumes more. His energy appetites are at the level of the older Sandy Bridge. True, the new product, unlike Sandy Bridge, has support for AVX2 instructions, which require quite serious energy costs.

The following diagram shows the maximum consumption under the load created by the 64-bit version of the LinX 0.6.5 utility with support for the AVX2 instruction set, which is based on the Linpack package, which has exorbitant energy appetites.

Once again, the Broadwell generation processor shows the wonders of energy efficiency. However, if you look at how much power the Core i7-6700K consumes, it becomes clear that progress in microarchitectures has bypassed the energy efficiency of desktop CPUs. Yes, in the mobile segment with the release of Skylake, new offerings have appeared with an extremely tempting performance-to-power ratio, but the latest desktop processors continue to consume about the same as their predecessors consumed five years before today.

conclusions

Having tested the latest Core i7-6700K and compared it with several generations of previous CPUs, we again come to the disappointing conclusion that Intel continues to follow its unspoken principles and is not too eager to increase the speed of desktop processors aimed at high-performance systems. And if, compared to the older Broadwell, the new product offers about a 15 percent improvement in performance due to significantly better clock frequencies, then compared to the older, but faster Haswell, it no longer seems to be as progressive. The difference in performance between the Core i7-6700K and Core i7-4790K, despite the fact that these processors are separated by two generations of microarchitecture, does not exceed 5-10 percent. And this is very little so that the older desktop Skylake could be unambiguously recommended for updating existing LGA 1150 systems.

However, it would take a long time to get used to such insignificant steps taken by Intel in increasing the speed of processors for desktop systems. The increase in the speed of new solutions, which lies approximately in such limits, is a long-established tradition. No revolutionary changes in the computing performance of Intel desktop-oriented CPUs have been happening for a very long time. And the reasons for this are quite understandable: the company's engineers are busy optimizing the developed microarchitectures for mobile applications and, first of all, think about energy efficiency. Intel's success in adapting its own architectures for use in thin and light devices is undeniable, but the adherents of classic desktops only have to be content with small increases in performance, which, fortunately, have not yet completely disappeared.

However, this does not mean at all that the Core i7-6700K can only be recommended for new systems. Owners of configurations based on the LGA 1155 platform with processors from the Sandy Bridge and Ivy Bridge generations may well think about upgrading their computers. Compared to the Core i7-2700K and Core i7-3770K, the new Core i7-6700K looks very good - its weighted average superiority over such predecessors is estimated at 30-40 percent. In addition, processors with Skylake microarchitecture boast support for the AVX2 instruction set, which has now found wide use in multimedia applications, and thanks to this, the Core i7-6700K is much faster in some cases. So, when transcoding video, we even saw cases when the Core i7-6700K outperformed the Core i7-2700K in speed by more than twice!

Skylake processors also have a number of other advantages associated with the introduction of the new LGA 1151 platform accompanying them. And the point is not so much in the support for DDR4 memory that has appeared in it, but in the fact that the new chipsets of the hundredth series have finally received really high-speed connection with the processor and support for a large number of PCI Express 3.0 lanes. As a result, advanced LGA 1151 systems boast numerous fast interfaces for connecting drives and external devices without any artificial bandwidth limitations.

Plus, when evaluating the prospects for the LGA 1151 platform and Skylake processors, one more thing needs to be borne in mind. Intel will not be in a rush to bring the next generation of processors known as Kaby Lake to market. According to the available information, representatives of this series of processors in versions for desktop computers will appear on the market only in 2017. So Skylake will be with us for a long time, and the system built on it will be able to remain relevant for a very long period of time.

Late 2000s Technique in a Modern Environment

Today we will continue testing "historical" platforms, which is interesting for the reasons already mentioned (and repeatedly): when their owners are no longer satisfied with the current level of performance, it is still interesting to compare it with the demonstrated new computers - at least in order to understand what is worth switching to (and whether it is worth it). Testing absolutely everything is unrealistic, but some "iconic" processor models are worth it if possible. Last time we dealt with AMD's first "integrated" platform - FM1, whose representatives also allow us to evaluate the level of speed and processors of the Athlon II line for AM3 with a fairly high accuracy. And with a little less - Intel processors for the LGA775 platform: somewhere from Pentium E5x00 to Core 2 Quad Q9500. Today we will clarify the "limits of the possible" for the latter a little, having studied the models of processors for the LGA1156 platform.

Why is it interesting in itself? If FM1 was AMD's first integrated offering, then Intel's earlier LGA1156 shaped the market altogether. In fact, it was the first two-chip solution (where only the southbridge remained from the chipset, and everything else moved under the processor cover) and the first platform with graphics integrated into processors. Back then, graphics were very weak (not far removed from Intel's "chipset" IGPs), were found only in some processors (only in dual-core models), and in today's systems it is not applicable: the last OS supported by Intel is Windows 7. But do not forget that it was not even 2011 (when AMD "rolled out" FM1, and Intel updated to LGA1155), but 2009-2010. The principles of building mass computer systems have not changed since then. Since then, Intel has retained not only the cooler mounting system (it's been identical for the entire 115x line for the eighth year now), but also the name of the processors. Core i7, however, were announced a year earlier (as part of LGA1366), but it was in 2009 that quad-core Core i5 first appeared on the market, and since 2010, dual-core Pentium, Core i3 and Core i5. And the basic principles by which processors fall into one of the listed families also do not change for many years. Hearing the name "Core 2 Quad", almost everyone understands that we are talking about something outdated ... But "Core i5" - what is it? Yes, the first generation Core processors are old, but they still work for some users. And technologically, from the point of view of the architecture of the processor cores, they, in general, differ little from Core 2. The extension of the instruction set, the ring bus, etc., etc. - all this debuted in Sandy Bridge. Accordingly, if initially the Core i5-750 roughly corresponded to the older Core 2 Quad models (being a little faster than the Q9650, but lagging behind the extreme Q9770), then no change in this ratio could occur during a software update. All in all, the Core i5s also show what you can expect from the LGA775. And Core i7 - what can be expected from quad-core processors for LGA1366, since the differences between the 800th and 900th line are even smaller. So it is all the more useful to test these processors, although they are of interest in and of themselves.

Test stand configuration

CPU	Intel Core i5-680	Intel Core i5-760	Intel Core i7-880
Kernel name	Clarkdale	Lynnfield	Lynnfield
Production technology	32/45 nm	45 nm	45 nm
Core frequency std/max, GHz	3,6/3,87	2,8/3,33	3,06/3,73
Number of cores/threads	2/4	4/4	4/8
L1 cache (total), I/D, KB	64/64	128/128	128/128
L2 cache, KB	2×256	4×256	4×256
L3 cache, MiB	4	8	8
RAM	2×DDR3-1333	2×DDR3-1333	2×DDR3-1333
TDP, W	73	95	95
Graphics	HDG	-	-
EU quantity	12	-	-
Frequency std/max, MHz	733	-	-

With the Core i5-760 and i7-880, everything is clear - these are the fastest processors on a 45-nanometer Lynnfield crystal, at one time one of the fastest on the market in their classes. With dual-core models for this platform, things are more complicated. Its start was not so easy, so we never saw the originally planned "monolithic" Havendale at 45 nm - with some delay (relative to older models), the "hybrid" 32/45-nm Clarkdale entered the market. The most popular on this chip were Core i3 - inexpensive models positioned to replace the Core 2 Duo and perfectly cope with this task. But buyers didn’t like the dual-core Core i5 at once - after the quad-core ones! And they differed from the significantly cheaper Core i3 only in clock speeds and Turbo Boost support. However, we have not been able to find the Core i3 at present, but the top-end (in its line) Core i5-680 has succeeded. Note that this particular model was generally very expensive - in fact, it even left the market at a recommended price of $305, i.e. higher than that of the younger Core i7 models! But if this family did not enjoy the love of end buyers who professed the DIY principle, then computer equipment manufacturers reacted to it very vividly. The reason for this was the presence of some kind of graphics core and a relatively low TDP level of 73 watts. Now you won’t surprise anyone with a “liter” computer with support for desktop socket processors - and in those years, even a Mini-ITX board with a socket, and not with a soldered surrogate solution, was a fresh and original product. However, we repeat, buyers looked more closely at the Core i3, but the older i5 will not be out of place for us - in order to evaluate the upper limit of Clarkdale performance.

CPU	AMD Athlon X4 880K	Intel Pentium G4400	Intel Core i3-6320
Kernel name	Godavari	skylake	skylake
Production technology	28 nm	14 nm	14 nm
Core frequency std/max, GHz	4,0/4,2	3,3	3,9
Number of cores/threads	2/4	2/2	2/4
L1 cache (total), I/D, KB	192/64	64/64	64/64
L2 cache, KB	2×2048	2×256	2×256
L3 cache, MiB	-	3	4
RAM	2×DDR3-2133	2×DDR3-1600 / 2×DDR4-2133	2×DDR3-1600 / 2×DDR4-2133
TDP, W	95	54	51
Graphics	-	HDG510	HDG530
EU quantity	-	12	24
Frequency std/max, MHz	-	350/1000	350/1150
Price	T-13582517	T-12874524	T-12874328

Since the study is more theoretical (if someone still uses processors of that time, then, in general, everything suits him - he will go to the store only when the computer “covers” physically), in choosing benchmarks for comparison, we a little more than completely free :) That's why we took this trio: the fastest Athlon X4 (ideologically similar just to the dual-core Core i5), the younger Pentium and the older (at the moment) Core i3 of the modern line, since all of them have already been tested before together with a discrete graphics card based on the Radeon R9 380 and 16 GB of RAM. Three of the subjects worked under similar conditions: it is no longer possible to fully use the integrated graphics of Clarkdale / Arrandale, and there was no such thing in Lynnfield yet.

Test Methodology

The technique is described in detail in a separate article. Here we briefly recall that it is based on the following four pillars:

• Methodology for measuring power consumption when testing processors
• Methodology for monitoring power, temperature and processor load during testing

And the detailed results of all tests are available in the form of a complete table with the results (in Microsoft Excel 97-2003 format). Directly in the articles, we use already processed data. In particular, this applies to application tests, where everything is normalized relative to the reference system (as last year, a laptop based on Core i5-3317U with 4 GB of memory and SSD, with a capacity of 128 GB) and grouped by the areas of application of the computer.

iXBT Application Benchmark 2016

Considering that the Pentium G4400 often outperforms the Sandy Bridge-based Core i3, we were not surprised by its superiority over the i5-680, and the fact that modern Athlon X4s are able to keep up with quad-core Core i5s under LGA1156 was also ready based on slightly earlier testing. However, they also led to the fact that the lag of one of the best once Core i7 (by the way, only six-core models have long been recommended prices) from the banal (albeit the best in the line) Core i3 did not shock us too much either :) But this a group of applications is the very case when the number of supported computation threads is comparable in importance to their quality - it will only get worse further.

For example, when processing photos, Photoshop already supports AVX, which affects not only modern Pentiums, but also processors for outdated platforms. And in general, there have been so many architectural improvements over such a period that even in the "greedy for cores" Lightroom i7-880 is at least a little, but it lags behind the i3-6320. But a little. But it's already behind. In general, over time, any carriage definitely becomes a pumpkin - if it doesn’t break earlier :)

The "quantity" here, as you know, is useless, while the "quality" of the cores of the first generation Core (almost identical, we recall, Core2) is such that it is already somewhere on the level of Athlon X4. It's not high by itself, but it can get worse.

The program is more or less capable of utilizing "additional" code streams, but it does not do it very actively - as a result, the i5-680 turned out to be (due to the clock frequency) from the "antiques" at least a little, but faster than the i5-760. And when comparing processors of different generations, this only allowed the Core i7-880 to overtake the youngest modern Pentium, which does not need comments at all.

However, on the old integer code, the “oldies” can still “turn on the heat”. Relative, of course - for parity with modern processors, they need to have about twice as many cores. This does not win in any way, but it tactfully hints that the older (ideologically, of course) the software used, the less incentive to replace the old computer with a new one. Or you won't be able to get by with "little blood" - even the best of modern Core i5 in this task are still at best only equal to such an ancient Core i7. What's faster? Modern Core i7 or so.

The above also applies to data packing, but this program (and many others similar) “expands” archives into one stream, which greatly affects the final result. But, in general, the then best Core i7 still managed to slightly overtake the best modern Core i3.

Recall that chipsets for LGA1156 support only SATA300, which naturally affects the speed of disk operations - especially when copying data. But we also recall that 100 points of the reference system were obtained just on the SATA600 and the corresponding SSD. But slower than we use in the main line of tests. And here, despite the limitations of the interface, it turned out faster. Output? Do not be afraid of the lack of support for new versions of SATA by the system - if you want to "spur" it on by installing a solid state drive, it's worth it. With hard drives, in any case, there is no comparison.

As we have already said, this program does not favor SMT technologies too much - the seemingly opposite effect is more related to the difference in frequency. Therefore, the battle of "clean" cores. And it is clearly seen that in high-quality optimized modern programs one 2015 core is fully compliant two 2009. If we also recall what was said above about the technological parity of the first generation Core with Core2, this also gives a good answer to the place, for example, of the Core 2 Quad in the modern world: about a Pentium of the same frequency. Alas, such is the fate of any high-tech products - they are guaranteed to lose the prefix “high” over time.

In general and average, the result is also the same - Core i5 (and Core 2 Quad) keep at the Pentium level, and Core i7 - the current Core i3. Six-core models for LGA1366 are at best like the older modern Core i5. But, of course, the alignment is different - for example, in old multi-threaded applications, the "oldies" look better than in new ones. When the load falls on one or two threads, everything is bad, regardless of the age of the program. But anyway - the newer, the worse :) Actually, at the same time, the answer to whether you need to suffer from "versomania": in order to use the capabilities of new platforms, you will have to. And for a person who continues to use, for example, Windows XP and software from the beginning of the decade, the new system will not give so much. It may even, on the contrary, cause problems when trying to “fasten” that same XP to yourself.

Note that these results were obtained in the normal mode of operation of all processors, while the ease of overclocking on outdated platforms by some users (a small but very vocal group of them) is often considered an advantage of the latter. With what in terms of pure performance it is difficult to argue - indeed: an increase in clock frequency increases the speed of work. True and power consumption too. And what about him at least in normal mode?

Energy consumption and energy efficiency

Clarkdale/Arrandale processors were dual-chip, and the actual "processor" chip in them was manufactured according to 32 nm standards - as a result, the Core i5-680 does not show us anything so terrible. In fact, power consumption from a system with a Core i3-2120 (with the same video card and memory) differs by only about 10 watts, and from modern dual-core Intel models by 20 watts. And this is better than AMD's achievements at the moment - if, of course, we evaluate only power consumption without reference to performance (more on that below). But the "old" quads, also made according to the standards of 45 nm, do not differ in some kind of economy - rather, the opposite. Although, again, the position is comparable with processors for FM2+, but this comparison is from the category of "both are worse" against the background of modern Intel platforms. And it is clear that overclocking can only exacerbate the matter. Although those who are not worried about this level of energy consumption, they are unlikely to be very upset by its increase.

Adjusted for performance, it all looks something like this. It is clearly seen that in terms of efficiency, even Clarkdale was already a step forward. Especially if you remember that in some cases these processors made it possible to do without discrete video. So, despite the initially not too amazing speed of work, there was a point in releasing such processors in 2010. Now they look almost as pale as the 45nm models. And it's not even that all the processors of that time work too slowly, or consume too much energy - that's half the battle. Worse than the other - in the aggregate, all this leads to the fact that they spend this energy extremely inefficiently. Which, of course, is not yet a reason to run to throw away the old computer for which the money was paid, but you should not ignore this state of affairs either.

iXBT Game Benchmark 2016

The almost complete coincidence of the results in both resolutions clearly shows that the performance "closely rests" on the processors - as expected. However, this does not cause any particular problems in practice: you can play comfortably.

In the case of "ships" everything is even more fun - the result is approximately equal to the maximum possible (recall that in this game the frame rate is limited from above).

The "bottleneck" is once again in the processors, but you can play. Which is not surprising - after all, the game was also originally almost from those times.

But even with a much more modern racing simulator, all test participants cope well - first of all, the requirements for the video system have increased (as is usually the case).

In FHD, in general, everything is determined by R9 380, in HD there is a difference between the participants, but from a practical point of view, it is insignificant.

Which applies to this game. However, as we have already noted, it is generally more demanding on the video card in any conditions.

It's funny that the old dual-core (albeit with NT) processor loses to the modern Pentium, although the game as a whole already has a "bad attitude" to the representatives of this family. But not unexpectedly: number flows must be assessed in conjunction with quality. Not apart.

In this case, all models for LGA1156 turned out to be worse than Pentium G4400, although they are still not inferior even to the newest Athlon X4. There are no special reasons for joy, but in practice this means that they are still suitable for a gaming computer (even if we are talking about the entry level).

And two, where the difference between the processors is already very noticeable. But this happens at such absolute values ​​that they can be ignored. Thus, from the point of view of game application, such systems (if they are already available) can still be considered acceptable. Of course, unlike the 2009-2010 Core i5-750/760, you can’t even call it a stretch best processors for gaming, however, with a good video card, the range of gaming applications available to the user will be very wide and representative. Tellingly, video cards of that era (even the best ones) can already be considered completely outdated, while processors can still work. Not only those, but also earlier ones - with the exception, perhaps, of the dual-core Core 2 Duo (especially the first series), which have no odds in the number of cores, and the quality of the latter is already too low. And the Core 2 Quad or the first representatives of the Core i5 and i7 families can at least do something.

Total

As we already noted in the last "historical" article, the use of computer systems released after 2006 does not present great difficulties today. We calmly equipped the boards with 16 GB of memory and a modern video card, installed Windows 10 and got access to any modern software. Yes, of course, the platform no longer has enough support for modern interfaces, and it's not always convenient to "add" something because chipsets support only PCIe 1.1. However, for a discrete USB 3.0 controller, for example, this is enough, and you can ignore the limited speed of disk operations - after all, computers are still sold (and even more so - used) only with mechanical drives, and this definition of a lower level of performance.

In short, if such a computer is already available, it is not surprising that it will be used until it breaks: after all, it has long been "paid" for it, and any replacement of equipment requires money. Against this background, it is not so critical that even the fact that each core of the sample of the end of the “zero” is only half of the modern one in terms of performance, and at the same time eats for three - quickly “beat off” a new purchase on electricity bills still will not work. Another question is if the level of performance provided is already low and / or the “large” computer is tired - there are, at first glance, a lot of replacement options. True, if you look closely, it turns out that, for example, even the best of today's mini-PCs are still somewhat slower - even if you do not consider gaming applications, where the use of only integrated graphics will still be very difficult to achieve gaming comfort. Top laptops are faster, but also quite expensive. Thus, the economic sense to continue operating the old PC (no matter how pale it may look against the background of modern ones) still remains - and will continue to be so long as the old computer continues to work. Of course, the owner's performance requirements may increase over time, but it seems to us that those who have them have already resolved the issue of modernization, and for a long time.