AMD FX processors. New generation amd processor line The best AMD processor based on Kaveri architecture

All six available models in one article

With the first implementation of the Bulldozer microarchitecture, everything has long been clear to everyone - including AMD :) Over the past year, we managed to improve the yield of suitable processors, which made it possible to completely cope with the shortage and even begin to carefully manipulate prices. But faster processors than the FX-8150 have not yet appeared. And it is obvious that it is impossible to achieve a significant increase in performance without architectural improvements. They are exactly planned for this year, but in stages. That is, the first processors with Piledriver cores were recently released, but they became Trinity laptop APUs, where at the moment progress is most important for AMD (especially taking into account the appearance of improved integrated graphics in Ivy Bridge by the competitor). Desktop Trinity will have to wait until the end of summer (in any case, such forecasts now seem most likely), but they will not bring anything to the high-performance segment - a configuration with a pair of modules, despite all the architectural improvements, is only a competitor to dual-core Ivy Bridge processors parts. However, in the same third quarter, Zambezi is expected to be replaced by Vishera, which promises increased performance at the same clock speed and the same four modules in the maximum configuration. The names of the first processors are already roughly known: FX-8350, FX-6300 and FX-4320 - respectively, four, three and two new “improved” modules. We'll evaluate how significant the improvements to Piledriver will be.

What to do now? Reducing prices is an obvious step: they looked somewhat overpriced even during Sandy Bridge, and the new technical process allowed Intel to “tighten up the performance a little more” for the same money. But we also need to somehow release new processor models to increase interest in our platform, which, as we wrote above, is very difficult without architectural improvements. More precisely, this is very difficult in the older segment, where the frequencies are already so high that there is no reserve for the thermal package. But two- and three-module models were rated at 95 W. On the one hand, there is also a lot. On the other hand, it’s still impossible to compete with Intel in this parameter, so why not take advantage of the 30 W reserve to improve performance? This is how the FX-4170 and FX-6200 were born, which are accelerated versions of the FX-4100 and FX-6100. But there is nothing new in them - in fact, they are nothing more than guaranteed factory overclocking. Moreover, for an additional payment of about 20 dollars compared to its predecessors, i.e. thrifty overclockers, in general, will not receive anything new - all multipliers are unlocked, so both core frequencies and UnCore can be tuned independently (and may very well be to a greater extent). degrees than with factory overclocking). On the other hand, these processors are quite worthy of attention. Even if only as express testing objects - so that you can get a more complete picture of the current AMD FX line.

We originally wanted to do a full review of the entire FX line, which now consists of seven models. Moreover, in fact, it was divided into two separate parts - processors with a thermal package of 95 W (4100, 6100 and 8100) and faster, but also more “gluttonous” models designed for 125 W (4170, 6200, 8120 and 8150). However... However, the stumbling block turned out to be the FX-8100 - currently the only “full” (i.e., four-module) processor with a TDP of 95 W. This unique combination of characteristics makes it highly desirable (because eight cores, eight gigs, gaming video card, and with a cooler for $10) from large computer manufacturers - to the detriment of retail chains. In general, we have not yet been able to “get” this processor. Even resorting to the help of the manufacturer, nothing came of it. Perhaps, over time, the shortage will “resolve” (especially after another improvement in production - it’s not for nothing that AMD’s plans even included reducing the thermal package of the FX-8120 to the same 95 W), but at the moment the idea of ​​​​an FX benefit had to be abandoned. However, it’s not really scary - if we couldn’t get the processor, it’s unlikely that buyers will have an easier time. Perhaps those who purchase, but among the regular readers of this line of articles, it seems to us, there are relatively few of them (and there are even fewer of those who are eager to purchase an “eight-core” “bulldozer”).

Test bench configuration

The most interesting model is the FX-4170: in fact, this is the first desktop processor in which the base frequency exceeds 4 GHz, i.e. AMD has kept its long-standing promise. Unfortunately, as has been known for a long time, the efficiency of the new architecture is not yet so high, which prevents us from enjoying high frequencies, but... In low-threaded software (of which there is still a huge amount), this processor should be no worse than the top-end FX-8150. Maybe even a little better. Yes, and the FX-6200 can turn out to be a reasonable compromise, being built exactly between the 6100 and 8120: in multi-threaded applications, the processor can compete with the second due to its higher clock frequency, and in low-threaded applications, due to its same speed, it can be the fastest. The situation, of course, is greatly spoiled by the relatively high TDP level - after all, 4100, 6100 and even 8100 “fit” into 95 W, however, in principle, it is not so terrible in reality. And it may even be that in practice it will turn out to be unattainable. Moreover, the buyer of small-module processors now also has a choice: faster and hotter or cooler and cheaper - the main thing is to use this opportunity correctly. That is why we decided not to complicate the picture, but to limit ourselves to just six processors: two main characters, two of their “energy-efficient” brothers and two top models. In the end, there have already been many comparisons of the FX family with other architectures (and with very different ones), but what the layout looks like inside the line is interesting.

Testing

Traditionally, we divide all tests into a number of groups, and show the average result for a group of tests/applications in diagrams (you can find out more about the testing methodology in a separate article). The results in the diagrams are given in points; the performance of the reference test system from the 2011 sample site is taken as 100 points. It is based on the AMD Athlon II X4 620 processor, but the amount of memory (8 GB) and video card () are standard for all tests of the “main line” and can only be changed within the framework of special studies. Those who are interested in more detailed information are again traditionally invited to download a table in Microsoft Excel format, in which all the results are presented both converted into points and in “natural” form.

Interactive work in 3D packages

As one would expect, in this low-threaded group all subjects were ranked according to clock frequency, and it was the highest for the FX-4170. The flagship of the line follows a little behind it, and then all the rest, and the best of the “rest” is the FX-6200. On the one hand, the victory of “turbocharged” models, on the other hand, is not so serious. That is, for such loads, the optimal choice is the slowest FX-4100: it is the cheapest and fits into 95 W, lagging behind the leader by less than 10%. Multi-module processors are obviously the worst suited for such tasks - even Turbo Core cannot completely save the situation.

Final rendering of 3D scenes

But in this group they are out of competition for obvious reasons, although frequency is also important when all actuators can be loaded with work. Taking this into account, even “almost” out of competition - the result of the FX-8120 is 144 points: only 1 (!) more than the FX-6200. But the latter is cheaper, and the heat package of these devices is the same. So it's obvious that the 6200 would easily beat the "economical" and hard-to-find 8100. All in all, not such a bad result. The FX-4170, of course, is incapable of such feats, but for dual-module processors, rendering is not exactly the target area of ​​application from the start.

Packing and Unpacking

But here the FX-4170 managed to amaze us, taking third place in the family - after the older FX-8150 and 8120 (and losing only 1 point to the latter). Which is not surprising - multi-core models here are only “stretched” by data packaging using 7-Zip, while the three remaining subtests are quite content with one or two computation threads.

Audio encoding

And even in this group, which traditionally gravitates towards multi-threaded processors, our today's heroes look very good. Yes, of course, they don’t set records, but they fully justify their positioning (primarily in terms of price).

Compilation

A similar picture, only a little more “torn”. However, we initially didn’t have any particular doubts about this state of affairs: these two groups are very sensitive to the number of integer computation threads, and this is the best thing we can come up with for Bulldozer. In turn, the dependence of performance on the clock frequency is an axiom regardless of the specific architecture (the main thing is that it be the same), which allows “accelerated” small-core systems to occupy an intermediate position between normal small- and multi-core systems. However, the dependence of the released thermal power on the clock frequency also does not disappear, so in a particular situation you need to choose what is more important: performance, price or heat dissipation. And here AMD is not original - the same trends are observed in the range of Intel processors.

Mathematical and engineering calculations

The two main characters turned out to be the same. And of all the FX, they only lagged behind the 8150, and the size of this lag can be a complaint against the flagship, and not vice versa.

Raster graphics

Another victory of frequency over quantity, which, in fact, there was no particular doubt about initially: there are multi-threaded tests in this group, but there are also single-threaded ones. Therefore, previously FX suffered precisely because of low single-threaded performance, and what could not be solved architecturally can be corrected using brute force.

Vector graphics

Moreover, here, where even two FX-41x0 modules are redundant. As a result, the FX-4170 won everyone, and the FX-6200 took an honorable third place.

Video encoding

Applications in this group traditionally gravitate toward multi-threading (not all to the same extent, of course, but this is a common phenomenon), but it is quite possible to compensate for the “shortage” of cores with an increased clock frequency. So the performance of the FX-6200 can be considered the best - it lags only behind the top-end FX-8150, overtaking two other four-module models (we did not test the FX-8100, but there is no doubt that it is slower than the 8120). And the FX-4170 was just a little short of being able to repeat this feat - it is literally one and a half percent slower than the FX-6100. But slower. But by one and a half percent :) But the TDP is higher. But cheaper. In general, each approach has its own advantages and disadvantages, so the main thing is to properly manage the freedom of choice that has emerged.

Office software

The only thing that somehow saves four-module processors is the multi-threaded FineReader. And even then, this is true only for the FX-8150, and the 8120 is already behind both the 6200 and, even more so, the 4170.

Java

The JVM simply loves multi-core processors, however, it is also ready to trade cores for frequency. In any case, the increase in clock frequencies allows us to make up at least half of the gap from “more nuclear” processors and make the “ladder” more even.

Games

If you compare only the 6000 and 8000 families, you get the feeling that games really need multi-threading. However, in reality this is not so - the failure of the FX-6100 and FX-6200 is due to only one reason: the F1 cannot stand any “six-threaded” processors - both Phenom II and FX (and the LGA2011 platform in this game seems to be saved only by Hyper -Threading, bringing the number of threads to 12). Moreover, this game is accelerated best of all at 8000, which allows these processors in general to slightly break away from 4000. If you close your eyes to the atypical “Formula”, you get an interesting result - the FX-6200 is approximately equal to the FX-8150, and the FX-4170 faster than both of them. Moreover, it is also the cheapest :) In general, a good gaming computer on AMD FX is possible - just don’t run after older models.

Multitasking environment

This experimental test has recently demonstrated good stability and predictability, so we once again decided to use it to look at the subjects from this point of view. Moreover, in their case, the question of “many slow” or “few fast” was very poorly understood until recently.

The picture turned out to be similar to the Java machine - with the same architecture, more cores are always better, but the performance of each one matters. In general, and from the point of view of such “refined multithreaded” applications, the release of the FX-4170 and FX-6200 is quite justified: they remarkably filled the gaps between the 4100, 6100 and 8120.

Total

For years now the following songs have been heard in various forums: A computer is not bought for one year, and over time the degree of utilization of multithreading by software will only increase, so you need to buy a multi-core processor. And for a year now we have been observing that the bulk of application software remains generally single-threaded. Yes, of course, over the past five years the number of programs that can use multithreading well has grown, but they are still far from 100% firstly, and “many” often means “two” or at best “four” in -second. And some applications that are interesting to the mass user, we repeat, generally remain single-threaded. Entirely or in large part. Well, since our testing methodology is based specifically on real applications, and not at all on synthetics, its overall score reflects the situation on average quite well. Due to resource-intensive applications, of course, multi-core processors are ahead, but the increase cannot always justify the difference in price. The cheaper FX-4170 finished neck and neck with the more expensive FX-6100, and the FX-6200 was only slightly behind the FX-8120 (and the latter does not have an advantage in terms of TDP).

And even this overall result is largely due to the presence of well-paralleled applications, which, as a rule, are not widely popular. What happens in “domestic” applications is clearly visible in some of the diagrams: there, multi-cores have no advantages at all. It’s clear why - if only a portion of the potential computation threads are involved, everything comes down to single-threaded performance. It can be increased either by architectural improvements that allow more work to be done per clock cycle, or... By the usual extensive method - increasing the clock frequency. And from this point of view, the release of the FX-4170 and FX-6200 is absolutely the right decision. Indeed, as has been said more than once, the Bulldozer architecture cannot yet boast of high efficiency, but it is designed for high clock frequencies of 4 GHz and above. And at such frequencies it unfolds in full force. Unfortunately, for multi-module processors, power consumption and heat dissipation increase too much, but younger models previously had a certain reserve. Which AMD managed in the right way.

Should you be afraid of high TDP levels? It seems to us that the most timid ones still look towards AM3+ and have not looked before :) Moreover, in a modern home there are many places where savings are more justified. A discussion of our recent testing of household light bulbs showed that even in 7520 from the creation of the world, many continue to use infrared incandescent heaters to illuminate their homes. And if a person is not interested in the opportunity to save several hundred watts on a banal chandelier, then what does an extra 30 watts from the processor matter to him? With heat release, the situation is a little more complicated, but far from hopeless - we are still not talking about something unprecedented. Only 125 W, which is quite capable of many coolers for AM3+ and previous AMD platforms (especially considering that some Phenom IIs were officially capable of 140 W). Moreover, in terms of heat removal, the huge (by today’s standards) Zambezi crystal is an advantage, not a disadvantage: “removing” at least 125, at least 150 W from 300 mm² is much easier than cooling an overclocked Ivy Bridge.

In short, the new processors are a step in the right direction. If you can’t yet increase the performance of older models, it makes sense to “pull up” the younger ones in order to strengthen your position in the $100-$150 segment. Moreover, Intel will not offer anything new in it until the end of summer. Of course, strictly speaking, there is nothing so “new” in these two new FX - all multipliers in the entire family are unlocked, so an economical overclocker can get an analogue of the FX-4170 by purchasing a cheaper FX-4100. And this could have happened before too. On the other hand, such “factory overclocking” is at least interesting as a guide. And for the vast majority of users who do not overclock anything on their own, an additional degree of freedom when choosing a processor will also not be superfluous.

Despite the imminent release of the first AMD Ryzen chips and the subsequent expansion of this family, Advanced Micro Devices Corporation is not going to say goodbye to AMD FX series processors anytime soon. In a message to partners, AMD announced that Ryzen and FX will exist in the company's product family, with the latter being offered for both entry-level systems and relatively mainstream PCs.

The FX suffix rose to prominence between 2003 and 2006 when AMD used it to mark its highest-end desktop processors. At that time, AMD Athlon FX were truly unsurpassed in terms of performance, leaving both the Pentium 4 and Pentium Extreme Edition behind in the vast majority of applications. The FX suffix was not used during the AMD Phenom era, but it returned as a separate brand in late 2011 when AMD released the first processors based on the Bulldozer microarchitecture. The latter was unable to live up to the expectations placed on it in the segment of high-performance CPUs for desktop PCs and servers, and therefore the development of the corresponding processors was curtailed in 2012. However, in October of the same year, the company released processors based on the Piledriver microarchitecture (improved by Bulldozer) and Vishera/Seoul crystals - AMD FX 4300/6300/8300 and Opteron 4300/6300.

Subsequently, AMD optimized the power consumption/frequency potential of the mentioned chips, and also selected chips with the highest frequency potential, but in fact, the evolution of Vishera/Seoul stopped in mid-2014. In order to maintain the relative competitiveness of the AMD FX family, Advanced Micro Devices has been reducing processor prices and also working with motherboard manufacturers to release new platforms with support for M.2 connectors, USB Type-C and other technologies. Apparently, AMD believes that FX processors are still attractive in certain market segments, despite the high heat dissipation and lack of support for a number of modern instructions. Thus, sales of chip-based processors introduced almost five years ago will continue this year.

It is logical to expect that older models of AMD Ryzen processors will be positioned by the manufacturer as solutions for the most powerful PCs, while FX will be aimed at other segments. For example, AMD FX 8350 and higher chips are proposed to be contrasted with Intel Core i5 6400 (Skylake) and more powerful processors (see table). Other FX will compete with the Core i3 and Pentium generations of Skylake and Haswell. It is noteworthy that older AMD FX processors are recommended to be equipped with AMD Radeon R9 Fury and AMD Radeon R9 390 graphics cards based on Fiji and Hawaii chips, which have not been produced since the second quarter of 2016.

For partners who are interested in the possible motives of buyers to choose AMD FX after the release of Ryzen, the developer has released a special brochure describing the advantages of FX. So, it says that AMD FX processors can operate at huge clock speeds thanks to a free multiplier (which means easy overclocking) and have a higher number of cores than similarly priced competitors. Of course, high clock speeds of multi-core processors mean increased power consumption and heat dissipation, which requires investment in an advanced motherboard and cooling system and may reduce demand for older AMD FX models among potential customers. AMD hardly understands that sales of powerful FX will decline significantly immediately after the release of Ryzen, and therefore does not make any promises regarding the lifespan of older FX models on the market. The only thing AMD is promising is that Ryzen and FX will coexist for some time.

Considering the fact that AMD Ryzen requires new motherboards, memory and, often, cooling systems, many owners of systems based on the lower FX may prefer to buy an older model instead of upgrading to a new platform. Therefore, it is in AMD's best interest to continue offering top-end FX for some time to come.

Taking into account that Vishera crystals have been produced for almost five years, their yield and frequency potential are at a very good level (due to the fact that the 32 nm SOI technological process has been brought to perfection in six years), which means the cost of such microcircuits is relatively low . Thus, the CPU developer should not have problems offering powerful AMD FX. AMD confirms this with its statements to partners and says that it will continue to supply existing processors for motherboards with AM3+ socket in stock.

GIGABYTE 970-Gaming: the latest generation AMD FX motherboard

Today it is difficult to say what the demand for AMD FX products will be as the Ryzen family expands. Apparently, AMD intends to first find out this, and only then curtail the production of FX, logic sets for them, and the production of motherboards. In the meantime, the demand situation remains uncertain, AMD FX will not disappear from store shelves.

Was an AMD chip FX-8350. Characteristicsthis solution, its actual level of performance, reviews about it and other important information will be further provided regarding such a CPU within the framework of this material. Comparative testing with competing semiconductor products will also be conducted.

Positioning of the processor device

The CPUs were divided as follows within the AM3+ computer platform:

The most budget PCs were based on chips from the Septron and Athlone series. They had very modest parameters, minimal operating frequencies and a reduced cache. But these solutions had a corresponding price tag.

Entry-level system units were based on solutions from the FX-43XX series. This series of processors necessarily included 4 computing modules, had 3-level “fast” memory and an unlocked multiplier. But the price in this case increased significantly.

Mid-level gaming systems, in turn, were based on FX-63XX chips. Their only difference from the previous series of CPUs is the presence of 6 modules for processing code instead of 4.

Premium PCs were based on microprocessors of the FX-9ХХХ and FX-83ХХ series. These devices already included 8 code and data processing clusters. It was to this line of semiconductor solutions that the one considered in this review belonged. AMD FX-8350. Characteristicshe had truly the best within this platform.

The AM3+ processor socket debuted in 2012 and is currently morally and physically outdated. It was replaced in 2017 by a more advanced socket coded AM4. It is this that provides the greatest performance among AMD solutions.

Types of configurations. What does it include?

The AMD FX-8350 processor can be found in two possible configurations. The characteristics of one of them indicate the absence of a cooling system. In this case, thermal paste and cooler are purchased separately. It is indicated in the Trail price lists. Otherwise, the manufacturer included the following in the supply list:

CPU.

Sticker - logo with the name of the processor model.

Brief instructions for use.

Warranty card.

BOX is the second possible type of configuration of this microprocessor device. The manufacturer included a standard cooling system made from a proprietary cooler and thermal paste, which made it possible to useAMD FX-8350 specifications. Model overviewindicates the presence of an unlocked multiplier. Therefore, for overclocking, it is recommended to buy the first delivery option in combination with an advanced and improved heat dissipation system. If the processor is planned to be used in nominal mode, then the BOX option in this case is beyond competition.

Chip architecture

The presence of 8 computing units was emphasized by the characteristics of the AMD FX-8350. The photo of the semiconductor crystal and its organization given indicate a not entirely standard layout of this solution. The number of integer processing blocks is actually 8 pieces. But there are only 4 processing modules on this silicon basis. As a result, we can note that this chip is equipped with only 4 paired cores, each of which has 2 integer devices and one floating point device. This organization significantly reduces the performance that the AMD FX-8350 processor can boast of. Its performance characteristics indicate that it is faster than the i3 series, but slower than the i5. Devices of the first series can process 8 instructions per clock (2 cores of 4 instructions), and the second - 16 (4 cores of 4 instructions). The hero of this article can also process 16 instructions (8 cores with 2 instructions each). But the program code is optimized specifically for Intel chips, and as a result, they show better results in tests. But all this is true for the nominal mode. If you overclock the FX-8350, it will be able to compete on equal terms with the 4th or even 5th generation i5. The code name of the microprocessor architecture of this solution is Vishera. This CPU was produced using 32nm technology.

Clock speeds

The frequency formula in the AMD FX-8350 is organized in a rather unusual way. Its technical characteristics indicate a base value of this parameter of 4.0 GHz. But this CPU also features TurboCore technology, which automatically increases this value to 4.4 GHz. In addition to this, the multiplier of this chip is unlocked. As a result, it can be overclocked with virtually no significant problems. But for this, as noted earlier, the cooling system must have an improved design. In this case, the CPU can be overclocked to 5.0-5.2 GHz. These values ​​are valid for air heat removal systems. If we use solutions based on liquid nitrogen for this, then such chips are quite capable of overcoming 5.5 GHz.

RAM memory

The outdated layout of the subsystem organization is indicated by the characteristics of the AMD FX-8350. Owner reviews indicate a decrease in computing system performance for this reason. Unlike modern PCs, the RAM controller is located outside the CPU chip and is included in the motherboard chipset. It can address 32 GB of memory (4 modules with a nominal size of 8 GB) and is capable of operating in 2-channel mode. Chip type - DDR3. Their clock speed can be 1066, 1333, 1600 or even 1866 MHz. You can also use faster solutions, but their frequency will not exceed the maximum permissible 1866 MHz. But this is true for the nominal mode. If you overclock the PC, the frequency of the RAM subsystem modules may exceed this limit.

Cash

Indicates a fairly good cache organization review AMD FX-8350. CharacteristicsThis microprocessor focuses on its three-level organization. The first of them is the most complexly organized. It consists of 4 64 kB clusters, which can only store processor instructions, and 8 16 kB blocks for data. In total, this allows you to store 192 kb. The second level includes 4 clusters of 2 MB each. At the same time, there is no strict specialization regarding the type of stored memory in this case. Its total size is 8 MB. The third level of very “fast” volatile memory is most simply organized in AMD FX-8350. Characteristicsit indicates the presence of 8 MB. Moreover, in this case it consists of only one segment. Again, there are no restrictions regarding the type of information stored in this case.

Energy consumption. Temperature

The review of the AMD FX-8350 indicates very high power consumption. The processor characteristics indicate a thermal package of 125 W. This is very important today. Its modern analogues have a thermal package of no more than 95 W. The overestimated power consumption of the hero of this review is due to the outdated technological process for producing a silicon crystal, which corresponds to 32 nm. Current chips are manufactured with tolerances of 14 nm. The temperature regime of this chip should not exceed 61 o C. In nominal mode, this value ranges from 50 to 57 o C. After overclocking with a standard cooling system, problems may arise with overheating of the microprocessor and the temperature may exceed the maximum permissible of 61 o C. Therefore In case of overclocking, the computer system must be equipped with an improved cooling system. If this condition is met, then you can count on overclocking such a CPU to a frequency of 5.0-5.2 GHz and the temperature will not exceed 57 o C.

Microprocessor overclocking

As has been noted many times before, it boasts a high level of overclocking potentialAMD FX-8350. Characteristics of WRAITH- the cooler, which comes with the BOX configuration, allows this chip to be overclocked to 4.5-4.6 GHz. Again, the case must provide improved air circulation and be equipped with several additional fans. If you purchase the Trail CPU option in combination with an improved fan and advanced thermal paste, you can easily count on reaching 5.2 GHz. It is most optimal to use nitrogen-based liquid cooling for overclocking. Such systems are much more expensive, but you can already count on the fact that the limit of 5.5-5.6 GHz will be reached. In this form, this microprocessor comes close to the results of the 4th generation Intel i7 series CPUs.

Performance

Now let's test the AMD FX TM - 8350. The characteristics of this chip indicate that it can compete on equal terms with the i3 and i5 series microprocessors. As a result, as an opponent from Intel, it is most optimal to choose the solution of the latest series with the index 3570K. For comparison, on the AMD side, it is most optimal to choose the FX-8150 and Phenom II with the 1100T index. These are the main competitors of the microprocessor discussed in this material. In the PCMark7 synthetic test, these chips received the following scores:

3570K - 6873.

FX-8350 - 6539.

FX-8150 - 6057.

1100T - 5402.

This test is optimized for single-threaded execution, and due to this feature, the Intel solution outperforms any AMD chip. In turn, AMD microprocessors are distributed quite standardly in order of improving the microarchitecture that underlies them. In the 3DMark11 test package, the balance of power in synthetic scores undergoes certain changes:

FX-8350 - 8502.

3570K - 8082.

FX-8150 - 7623.

1100T - 7082.

The AMD flagship outperforms the middling Intel in this test package, and there is a simple explanation for this - the software is optimized for multi-threaded execution and the improved architecture of the “blue” company in this case loses all its advantages. When processing multimedia information or archiving, AMD chips look better, but in games, Intel products look preferable. As an example, you can give the results in conditional points in the WinRAR archiver:

FX-8350 - 7112.

FX-8150 - 6268.

3570K - 5571.

1100T - 5470.

Multi-threaded optimization in this case makes itself felt and the Intel chip lags far behind the two AMD flagships. Now here are the tests in the game F1 2012 with a resolution of 1920X1080 and high image quality in FPS:

3570K - 97.

FX-8350 - 81.

FX-8150 - 67.

1100T - 63.

All CPUs provide an acceptable level of fps, but the more frames per minute, the more comfortable the gameplay. Therefore, the average product from Intel in this case is beyond competition. A similar arrangement of chips in the game R.U.S.E. The number of fps in a similar mode is obtained as follows:

3570K - 47.

FX-8350 - 39.

FX-8150 - 37.

1100T - 26.

The advantage of Intel's i5 already reaches 22 percent over the hero of this article. Again, most games are optimized for single-threaded execution and the solutions of the “blue” company in this case are simply better and preferable.

Price

The characteristics of the AMD FX-8350 are quite acceptable today. Reviews from owners only confirm this feature of this microchip. But at the same time, its cost is quite affordable. Today, the Trail delivery option can be purchased for 7,000 rubles. But Vox costs a little more - about 8,000 rubles. In the first case, it is necessary to add the cost of the cooling system to the cost of the chip. If you purchase an advanced air cooling system, you need to add 3,000 rubles. When using liquid cooling, this amount will have to be increased to 5,000 rubles. As a result, it should be noted that the BOX option is most optimally suited for operation in the nominal operating mode, and the second is better suited for overclocking the CPU.

As has been noted more than once, the AM3 platform is a noble “old man”. Some users managed to replace quad-core Phenom IIs with six-core models that came out later, and then replaced them with older versions of the FX series. The first generation of new AMD FX processors was met with mixed reactions, but that's understandable. Most “household” tasks require high processor speed in simple applications that are not always able to work with a large number of threads. The new architecture is initially aimed at multi-threaded programs, where it showed a good percentage of growth compared to previous Phenom II models.

We had to compete with older models for the Intel LGA 1155 platform in the productivity segment, which was extremely difficult to do before. It turned out well, although in a number of tests there was an obvious advantage of the Intel CPU, in other tasks the FX-8150 managed to outperform its competitors, while having a lower price - that is, it turned out to be a classic “good for the money” option. Still, any fan of AMD processors expected more.

Platform and architecture

Intel continues to release processors “tick-tock” from year to year, transferring the old architecture to new technological standards, and then releasing a new one using them. At AMD, the process of changing architectures until recently was more protracted. However, Bulldozer processors have only recently appeared, and we are already seeing Piledriver.