Processors amd phenom ii x6 x4. AMD Phenom II processor: characteristics, description, reviews. Description of test systems

AMD has removed the X2, X3, and X4 core count suffixes from the logo, instead changing the part number: the 9000 models have four cores, while the upcoming triple-core models will have the number 7000.

It's been a difficult year for AMD. Not only did the Phenom processor, which everyone had been waiting for so long, come out at significantly lower clock speeds (2.3 GHz instead of 3 GHz), but also an unpleasant error was discovered in the current Barcelona core stepping. You can bypass it, but only updated stepping will allow AMD to continue releasing quad-core processors for the server segment. And the fact that AMD's quad-core processor doesn't have enough performance to compete with Intel in the high-end segment doesn't help either. As a result of all these problems, AMD had to change its product promotion strategy and position the processor along with the new Spider platform on the mass market. However, despite all the problems, the Phenom is not as bad as many people think, as you will see from this comparison between the Phenom and the Athlon 64 X2.

In fact, AMD has quite a few significant advantages over Intel when it comes to upgrading current systems to a quad-core processor. While Intel very quickly releases new platforms for each new generation of processors due to changing requirements, AMD has not changed the Socket AM2 specifications at all. Therefore, it is technically possible to install a quad-core Phenom processor on a Socket AM2 motherboard, replacing an Athlon 64 or Athlon 64 X2, all you need is a BIOS update. However, this is not always true either - some motherboards cannot handle the Phenom's power consumption (95 or 125 W), but most enthusiast motherboards can be upgraded to a quad-core processor. At least in the future, because for now we have only been able to install Phenom on two "old" motherboards out of ten .

The upgrade situation does require some attention, as AMD and Intel are planning their next major technology update in about six months. AMD will introduce Socket AM3, which will support DDR3 memory, and Intel's next-generation processors, codenamed Nehalem, will finally bring the memory controller to the processor. Considering all this, even the upcoming Core 2 Duo E8000 or Core 2 Quad Q9000 lines can only be considered as intermediate products on the way to the next generation, even if they are about 10% faster than the existing Core 2 products.

November 17 AMD launched two Phenom models: Phenom 9500 and 9600, at 2.2 and 2.3 GHz, respectively. They both have a TDP of 95 W, which is close to the 105 W Intel claims for the Core 2 Quad Q6600 (2.4 GHz) and Q6700 (2.66 GHz). All faster models, which are planned to be released in the first quarter of 2008, will operate with a thermal package of 125 W. Towards the end of 2008, a Black Edition may appear, friendly to overclockers, but not higher than the top frequency of 2.3 GHz. But AMD has unlocked the multiplier to provide ideal conditions for overclocking, and this version should not be more expensive than the usual one.

You will be able to install the Phenom processor in almost any Socket AM2 motherboard on the market when all problems will be solved. Even cheap motherboards support a standard 95W TDP, but for 125W versions you need to use an enthusiast platform, which is true if you plan to significantly overclock the Phenom. The situation with BIOS updates is still far from ideal, so installing Phenom on existing Athlon boards is not as easy as AMD promised. Technically, this is the same socket with a 1000 MHz HyperTransport channel, but problems exist.

The Phenom microarchitecture was codenamed K10, but was later renamed Stars. The most significant difference, which mainly affects the number of transistors, is the L3 cache, which is an extension to the AMD64 two-level cache design. While each core has its own L1 cache for data and instructions (64 KB each), as well as 512 KB of L2 cache, L3 provides an additional 2 MB of fast storage for all Phenom cores.

This is not the first desktop processor to come with L3 cache: the 3.2-, 3.4-, and 3.46-GHz Intel Pentium 4 Extreme Edition models, all of which were built on the 130 nm Gallatin core, also included 2 MB L3 cache (along with 512 KB L2 cache). But, unlike the Pentium 4 EE's L3 cache, the Phenom's L3 cache works as a buffer for writing data to RAM.

AMD has also made some improvements to the branch prediction process, as the so-called sideband stack optimizer updates the ESP (enhanced stack pointer) without consuming CPU time. And the memory prefetcher is capable of loading data exclusively into the L1 cache, bypassing the L2 cache (that is, without unloading data from there). We also note the 128-bit width of SSE calculations, as well as the 32-byte instruction fetch unit. AMD has had virtualization technology for several months, and it is included in every Phenom processor.

Support for the 1.8 GHz HyperTransport 3.0 protocol is the latest performance enhancement feature that has been added to the Phenom. While HT 2.0 at 1.0 GHz supports speeds of 8.0 GB/s in both directions, HT 3.0 provides up to 20.8 GB/s. This will be especially important in the future when four or more cores need to provide access to other cores, for example to retrieve data from memory or to operate a PCI Express device such as a graphics card.

We were quite intrigued by AMD's claim that the Phenom is 25% faster per clock than current Athlon 64 X2 processors. Considering that there are no such architectural revolutions as the one that Intel accomplished by moving from NerBurst to Core, a 25% increase in performance per clock is very significant. It’s sometimes even hard to believe, which is why we were interested in taking a closer look at the new processor. We compared the Athlon 64 X2 and Phenom 9900 at a base clock of 2.6 GHz using only one core.

Phenom processors
Name	Clock frequency	L2 cache	L3 cache	TDP
AMD Phenom 9700	2.4 GHz	4x 512 kbytes	2 MB	125 W
AMD Phenom 9600	2.3 GHz	4x 512 kbytes	2 MB	95 W
AMD Phenom 9500	2.2 GHz	4x 512 kbytes	2 MB	95 W

All Phenoms look similar: this is our engineering sample with an unlocked multiplier.

Old discounted versus new cheap

We have already mentioned more than once the sale of processors of previous generations organized by AMD. So many times that there was reason to think: why don’t we have exact results for either of the two Phenom II X4s, which in the current conditions look like almost the best offers on the budget product market? Yes, of course, we have already tested the extreme ones in the 910 and 980 family, and it’s easy to estimate the performance of any intermediate model (including 955 or 965) using approximation, but many readers are simply too lazy to do it. And then: approximation by two points is an extremely unreliable thing. It would be advisable to add a third, which we recently did for a couple of Athlon II families, and now we’ll move on to the Phenom II.

But there will be no completely new AMD processors in testing. But from Intel we will take a couple of models that appeared not so long ago, which, however, are also part of long-studied families. In short, today we have the usual routine testing of five processors on our agenda. Not for the purpose of any scientific discoveries, but to clarify existing information.

Test bench configuration

CPU	Phenom II X4 955	Phenom II X4 960T	Phenom II X6 1075T
Kernel name	Deneb	Zosma	Thuban
Production technology	45 nm	45 nm	45 nm
Core frequency std/max, GHz	3,2	3,0/3,4	3,0/3,5
	4/4	4/4	6/6
L1 cache (total), I/D, KB	256/256	256/256	384/384
L2 cache, KB	4×512	4×512	6×512
L3 cache, MiB	6	6	6
UnCore frequency, GHz	2	2	2
RAM	2×DDR3-1333	2×DDR3-1333	2×DDR3-1333
Video core	-	-	-
Socket	AM3	AM3	AM3
TDP	125 W	95 W	125 W
Price	N/A(0)	N/A(0)	N/A(0)

So, three AMD Phenom II processors. Regarding the 955, everything has been said above - its wholesale price since the fall is only $81, so until old stocks are exhausted, this processor is very competitive. More precisely, other models in this price class are not very competitive, with the exception, perhaps, of the no less “sale” A6-3670K, where the weaker processor part is compensated by good graphics. But the buyer of a discrete video card is not interested in it, which makes the Phenom II X4 955 practically no alternative within the AMD range. Intel, for the money, only has dual-core Pentiums - older models, of course, but even an older Pentium is just a Pentium: two computation threads are no longer enough for many modern applications (including gaming ones). But more than four are not necessary.

We need one more processor, namely the Phenom II X6 1075T, primarily for the reason mentioned above (but there are others, which are discussed below) - this is the third point of approximation for the Phenom II X6. And the Phenom II X4 960T is interesting in its own right. The processor is based on, in fact, the same Thuban, but two cores in Zosma are initially locked. As a result, this OEM model was at one time extremely popular among those who liked to take risks: if successful, they got a cheaper Phenom II X6 than if they bought it initially. True, the probability of success was far from 100%; this processor was sold at retail in small quantities, and inexpensive six-core processors (such as 1035T/1055T) greatly undermined the idea of ​​saving money - why risk it for just 50 dollars? To be fair, our copy was unlocked without any problems - just changing one item in UEFI Setup was enough. But we still won’t say that there are no problems at all: the processor was not tested in this mode. Yes, this is not very interesting: unlocking a pair of cores turns the 960T into an almost complete analogue of the 1075T - only the frequency in turbo mode is 100 MHz lower. But its performance in normal mode is very interesting to us: a priori we can assume that when all four cores are loaded it should be slightly lower than that of the 955, and in low-threaded applications - at the level of 965. In any case, this is how the frequencies of these processors correlate. Let's see how much practice confirms the theory. And AMD’s six-core capability itself is now rarely of practical importance, be it innate or “unlocked”: Thuban processors have recently been present in AMD’s assortment only nominally, and it is extremely difficult to find them in retail. And the model range has not been updated for a long time, so having the results of three models (the previously tested 1035T and 1100T and the current 1075T), you can determine the performance of any other with fairly high accuracy using approximation by clock frequencies.

CPU	Pentium G2120	Core i3-3220	Core i5-3330
Kernel name	Ivy Bridge DC	Ivy Bridge DC	Ivy Bridge QC
Production technology	22 nm	22 nm	22 nm
Core frequency std/max, GHz	3,1	3,3	3,0/3,2
Number of cores/threads	2/2	2/4	4/4
L1 cache (total), I/D, KB	64/64	64/64	128/128
L2 cache, KB	2×256	2×256	4×256
L3 cache, MiB	3	3	6
UnCore frequency, GHz	3,1	3,3	3,0/3,2
RAM	2×DDR3-1600	2×DDR3-1600	2×DDR3-1600
Video core	HDG	HDG 2500	HDG 2500
Socket	LGA1155	LGA1155	LGA1155
TDP	55 W	55 W	77 W
Price	N/A()	$149()	$219()

Initially, we did not plan to include previously tested processors in the list of today's participants, but we decided to make an exception for the Pentium G2120. For two reasons. Firstly, the other two Intel processors in today's conditions are not direct competitors of the Phenom II X4 955 in price, but the Pentium somehow can. Secondly, at the moment this is the youngest Ivy Bridge “by far,” so it’s interesting to compare it with the younger Core i3 and the younger Core i5 on the same architecture. As for the i3-3220, there is nothing special about it - we have already tested its older brother (3240), and these processors differ only in clock frequency, and only by 100 MHz.

The release of the Core i5-3330 was somewhat unexpected. It would seem that the lower price level was clearly fixed at $184 wholesale back in the summer - when the Core i5-3470 replaced the older i5-3450. And then suddenly Intel releases three cheaper Core i5s! The 3350P model does not raise any special questions - as can be seen from the index, the video core is blocked here. Most likely, this is simply the disposal of a “total waste” in the video part. But only $177 dollars wholesale, both in OEM supplies and in retail packaging, plus a TDP of 69 W - an excellent offer for those who are going to use discrete graphics. That is, first of all, naturally, for small assemblers of ready-made systems, but also for individual buyers, $18 (the difference between the “boxed” versions of 3350P and 3470) will not be superfluous. Everything is clear with the 3330S - it is supplied only through OEM channels and costs $7 cheaper than the 3470S: just a little, but for a large batch of all-in-one PCs or compact desktops (where processors with a TDP of 65 W are used) the savings can be significant . But the Core i5-3330... It’s not clear - for whom? The “boxed” version costs only 8 dollars cheaper than the 3470, the OEM version is even 2 (two!) dollars cheaper. At the same time, the processors differ only in frequency, but the “floor” for the 3470 (3.2 GHz without turbo, which in practice will be a rare occurrence, since even with a load on all four cores the processor can overclock to 3.4 GHz) is the “ceiling” for 3330 (where this frequency is achieved only in turbo mode, and no more than at half load). And the maximum frequency of the video core has been reduced by 50 MHz - to the level of Core i3/Pentium.

In short, an incomprehensible processor. The only explanation is a retail (fortunately the “boxed” prices are the same) replacement of the Core i5-23xx line, which it was decided to “shoot” entirely. We wouldn’t buy it for ourselves :) But for testing, naturally, the processor is interesting. Firstly, because this is the youngest quad-core Ivy Bridge. Secondly, this is another processor with a nominal frequency of 3.0 GHz and turbo mode, i.e., formally the same as the Phenom II X4 960T and X6 1075T. Its maximum frequency, however, is the minimum (pardon the pun) in this trio, but the architecture is the most modern. With the Pentium G2120 and Core i3-3220, again, it’s interesting to compare.

As we have warned more than once, we have not yet used the ability of Ivy Bridge to work with DDR3-1600 in the main line of tests. However, increasing the memory frequency gives almost nothing to the top-end Core i7-3770K (when using a discrete video card, of course), so it would be difficult to expect record yields in relation to the Core i5, i3 or, especially, Pentium (we recently received for representatives this class of processors is only 2% on average of replacing DDR3-1066 with DDR3-1333, but a further transition to DDR3-1600 will not give that much). However, in testing according to the next version of the test methodology (the transition to which is not far off), we will stop “leveling” the environment for processors under LGA1155, but for now we will keep today’s practice unchanged (otherwise we would have to retest a considerable number of already studied processors of the Ivy Bridge family).

Testing

Traditionally, we divide all tests into a number of groups and show on diagrams the average result for a group of tests/applications (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 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 expected, the 960T was slightly faster than the 955, but slower than the 1075T - a low-threaded group of tests in which Turbo Core technology can be fully developed. However, this “power” itself, as we see, is not enough - Intel processors with the same or even slightly lower frequencies are much faster. And that they also stay in a tight group is understandable - as we have already installed, Hyper-Threading in this group only interferes, and additional “honest” cores are simply not needed.

Final rendering of 3D scenes

These subtests are already capable of loading any reasonable number of computation threads with work, so the Phenom II X6 1075T has almost caught up with the Core i5-3330. Achievement? Not very - average six-core processor almost caught up with the younger quad-core. Well, quad-core models with such initial data are naturally capable of competing on equal terms only against two cores with Hyper-Threading. And the only thing that saves the situation here is that the second one is more expensive. And for the same money, Intel offers only two regular cores, which are significantly slower.

From a less global perspective - as expected, with such a load the 955 is slightly faster than the 960T: Turbo Core does not work when the cores are fully loaded.

Packing and Unpacking

Multi-threading is supported in only one subtest out of four, so the 960T is slightly faster than the 955 and both are behind the Pentium G2120. But the 1075T is capable of competing with the Core i3-3220 - in general, it’s also a pretty funny comparison :)

Audio encoding

In terms of the type of load, this group of tests is similar to rendering, so the results are consistent. Not too happy for the Phenom II, the X4 is, of course, capable of overtaking conventional dual-core processors, but these are found only among budget products. But “two cores, four threads” at comparable clock frequencies is no worse in performance than four “real” old-style cores. Well, six of them, obviously, are hardly able to argue with four more modern ones. Yes, we remember that the 1075T is not the oldest Phenom II X6, but there were two models faster than it. And the Core i5-3330 is the slowest desktop quad-core Ivy Bridge.

Compilation

Compiler tests have always been Phenom's strong point, but at the moment their victory here is starting to turn into a purely nominal one: yes, somewhat faster, but whom faster? A couple of years ago, the same 1075T easily overtook the fastest Core i5, and the Phenom II X4 remained at a level comparable to the latter. So compare this with the current state of affairs.

Mathematical and engineering calculations

You can do without detailed comments - as we see, these types of loads have a bad effect on Intel processors (since Pentium, Core i3 and Core i5 “hang out” at the same level despite different prices), and for the Phenom II they are generally like death (since here and the comparison with Pentium would be politically incorrect).

Raster graphics

There is some multi-threaded optimization in some programs, but it only allows you to arrange Intel processors in the correct sequence and allows the Phenom II X6 to overtake the X4. That's all - two practically non-overlapping worlds.

Vector graphics

Two streams are enough, which leads to a certain chaos in the range of products for LGA1155, but Phenom helps little. The difference between the three models taken today is completely determined by the Turbo Core (or the lack of this technology in the 955) and does not allow any of them to fully compete with the older Pentiums. However, we note once again that the younger Core i5 also manages this with difficulty, which is why Intel has to artificially restrain the frequencies of dual-core budget models: there is a lot of software similar to these two programs on the market.

Video encoding

On the one hand, there is room for multi-core processors to develop; on the other hand, as we have said more than once (including quite recently), for video codecs the number of cores is important, but not the only parameter of processors. Accordingly, all that the Phenom II X4 955 and 960T managed to do was to overtake “simple” dual-core processors, and the Phenom II X6 1075T was enough to compete with also dual-core, but four-threaded ones. Again, let us remind you that a couple of years ago everything looked completely different: in video encoding, only Core i7 could handle X6, and X4 performed on an equal footing with older Core i5. Now everything is different. Because AMD has the same processors as they did then, while Intel has only the same family names that remain old :)

Office software

And again the same thing! Nothing unexpected, of course - most of the tests in this group are generally single-threaded. Just another illustration of the fact that you need to choose processors based on the number of cores very carefully - not all of them will necessarily be used by software. And selecting software “for multi-cores” is a simple task only for testers: there are a lot of “inconvenient” applications among the popular ones. As if not even the majority - if by “popular” we mean massively used.

Java

But in some specific niches, the oldies, of course, perform well. Relatively good - compared to other applications, and not at all in terms of absolute results. From their point of view, as we said above, the victories of an average six-core processor over a junior quad-core processor or a once good quad-core processor, at best, over a Core i3 do not cause much optimism.

Games

As we have said more than once, modern games require four computation threads in all cases where the video card is not the bottleneck. However, as we see, “in general” a fast dual-core processor (like a Pentium) is quite capable of keeping up with slow quad-core ones (like a Phenom II). If you look at the detailed results, it is noticeable that some applications still “like” the latter a little more. But there is no longer talk of any clear superiority. With the same architecture, we can say for sure that four cores are better than two in games (and any, even “flavored” with Hyper-Threading, not to mention “regular”), but with different ones, anything can happen.

Multitasking environment

As we have said more than once, there is no exclusivity in the results of the test with the simultaneous launch of several programs - we simply emulated another multi-threaded application. And the result is appropriate: the low-end quad-core Phenom II X4 is 25% faster than the dual-core Pentium, but approximately equal to the Core i3, and the average six-core Phenom II X6 1075T is just a little faster than the low-end third-generation Core i5. The resulting cores in the Ivy Bridge family are so efficient that they win not by numbers, but by skill.

Total

This, in fact, is the answer to the question why the Phenom II X4 955 is at the Pentium level. Yes, because its performance is on average at the same level! There are no miracles that many thrifty buyers hope for - the price of each item is determined by how much it can be sold for. And for processors, the latter depends on performance and power consumption. Could the 955 now cost over $100 like it did in the summer? Of course not - for that kind of money there are already more attractive offers. But for “about 100” it’s already a very good processor, capable (with a multi-threaded load) of competing with a Core i3. But, note, not with Core i5, where there are the same four cores - quantity does not always translate into quality. So it is precisely this (and not at all concern for the low-income segments of the population) that explains the price declines. And the disappearance of Thuban from retail chains while formally continuing supplies is also due to them: for market success, all six-core AMD models (including top ones) should cost no more than $150, and the company has neither the desire nor the ability to produce them with such initial data ( if you remember the crystal size of 346 mm² - more than two (!) times larger than that of quad-core Ivy Bridge). Of course, in some specific areas of application, multi-core Phenom IIs still look very good, but no less often (and just in widely demanded mass-market applications) they are outperformed by budget Intel processors. Developments on a new microarchitecture (both APUs and updated ones) are a much less sad sight, while the “classic” Athlon and Phenom have definitely reached a dead end.

Thus, for assembling a new system, Phenom II, despite the price reduction, is not of particular interest (except in the case of a “crazy programmer” who compiles something 24 hours a day, generating electricity using a personal windmill). However, there are users who can benefit from the ongoing “sale”: Phenom II X4 955 and 965 are perfect for upgrading a system on some Athlon II, not to mention older AMD processors (the latter, of course, only if technically possible) . The “hundred-buck upgrade” will be especially interesting to owners of large amounts of DDR2 memory: so what if the performance is far from the maximum on the market - but this is the only way not to change both the memory and the motherboard along with the processor. AMD is also aware of this. And I don’t mind (despite the established reputation of Robin Hood as a defender of the poor and oppressed) making extra money on it: only 955 and 965 have fallen in price, but for slightly faster models they are asking $140-160.

However, since all Phenom II X4 currently sold belong to the Black Edition family, ways to combat this injustice have long been known. Yes, yes: cobblestone dispersal is a weapon of the proletariat. In the same way, you can “defeat” AMD’s reluctance to reduce prices on the Phenom II X6: the Phenom II X4 960T can still be found on sale, and (if you have a suitable motherboard) you can also unlock a couple of cores for it. There is, of course, a risk that it won’t work out, but the end result, it seems to us, is worth the risk. Moreover, in case of failure, you will end up with a processor with performance approximately similar, as we see, to the Phenom II X4 955, which, taking into account the minimal difference in the price of these processors, is quite normal. But if everything goes well, you will get an almost complete analogue of the Phenom II X6 1075T. Not only much more expensive, but also in a different performance class.

And in any case, we should not forget that all the advantages of multi-core Phenom II can be experienced in practice only if there are a large number of programs optimized for multi-threaded processors among constantly used applications. If there is no confidence in this, then there is little point in having four to six cores either. One or two computation threads are the realm of the Pentium, in which these processors can easily compete on equal terms with the Core i3/i5, not to mention the Phenom II. And the video part in them is noticeably better than in the old (technologically; no matter what is still sold) integrated AMD chipsets, and the power consumption of such models is noticeably lower.

However, a sale is always a good thing because there are ways to take advantage of it. As well as the gradual transition of processors for LGA1155 to Ivy Bridge is also good: they are better than their predecessors, which, in general, will be noticeable to all their customers. Although this transition sometimes takes strange paths, sometimes giving rise to very strange models, such as the Core i5-3330. Until recently, the 2320 of the previous generation remained nominally the cheapest Core i5, but now Intel has apparently decided to make a replacement for it (and, by the way, a little faster than the i5-2400). But the practical implementation let us down: compared to the 3470, the processor was too slow, and the real retail prices of these models in Moscow often differ by only 100 rubles, or even less. 2320 or the older 2310 allow you (if you look hard enough) to save about 300 rubles, which is much more interesting when money comes first. In general, why he was born like this is absolutely unknown to us. On the other hand, its availability for sale, in general, does not bother anyone, and it may be useful for assemblers of ready-made systems. The main thing is not to buy it inadvertently. Why, in fact, we took the time to test it: forewarned is forearmed.

This time we will limit ourselves to the briefest possible theoretical introduction: we learned about the ideas contained in the AMD K10 core and Phenom processors long before the release of the processors themselves, several years ago. Many (note, by no means only fans who just want their favorite company to win), but also technically very knowledgeable specialists in matters of processor architecture, were waiting for these processors. It is justified (from a theoretical point of view) to expect, albeit not devastating for the competitor, but at least interesting results: somewhere there is a gain, at least thanks to the extended floating point calculation unit and the native quad-core design, somewhere equality, somewhere - then, of course, a lag, but overall a competitive result. After all, competitors have different architectural approaches, but they have their own trump cards.

After the release of Phenom, whose results were clearly lower than expected, at first many wondered: why exactly is this so? Then, as they say, everyone got used to it, moreover, now Phenom processors are quite well accepted by the market, are in demand, and many users are probably even happy that, due to price wars, these processors have received such affordable prices, which, at a minimum, are justified by their performance. In Phenom II, as we also learned long before the release of the processors themselves, changes that affect the performance seem to promise little: the third-level cache has been tripled and frequencies have been raised thanks to the transition to a 45 nm process technology. There is, however, a mention of architectural optimizations, although which ones are not specified. If such announcements were made regarding a long-polished processor core, from which all the juice had already been squeezed out during the release of numerous revisions, one would hardly expect anything interesting. But in this case, the thought quite naturally arises: what if these measures turn out to be enough to unlock potential that has not been adequately realized before? Let's see what actually happened.

We tested the older model with a frequency of 3.0 GHz and an unlocked multiplier, and at the same time a processor with index 920 with a frequency of 2.8 GHz was announced. The processors are installed in the Socket AM2+ socket, that is, they are completely oriented towards the platform formed for Phenom processors. The boards only require a BIOS update, and most manufacturers made the corresponding versions publicly available back in November, or even October, last year.

The recommended price of the Phenom II X4 940 is $275, so as a competitor for comparison in tests, it is tempting to take the results of the Core i7 920, whose recommended price is only $5 higher. Moreover, exactly in the configuration that was used in testing, with Turbo Boost and Hyper-Threading technologies enabled. Using the auto-overclocking function may not seem entirely fair, since Phenom also has overclocking potential and the ability to separately control processor multipliers for cores, but we will assume that this factor is balanced by installing 3 GB of memory, while other processors are tested at 4 GB. After all, our goal is to get as close as possible to the real conditions in which the processors themselves will operate, and it is unlikely that any of the Core i7 users will disable Turbo Boost in practice, while everyone will probably try to use a three-channel controller, but will immediately splurge on a kit in 6 GB will probably only be accepted by users of the extreme version, and not by the younger version in the line.

But, it should be borne in mind that even with this limitation, the platform as a whole, including the board and compatible memory, for the Core i7 is still very, very expensive, so in practice, most likely, most users will compare the Phenom II with the much more popular Core 2 Quad, so we chose a processor based on the Yorkfield core (Q9300) as our second competitor. From a research point of view, of course, it will be interesting to see how the new top model looks against the background of older representatives from the Phenom line (9850) and historical competitors on the Kentsfield core (Q6600). It should also be noted that in a number of tests dual-core processors still demonstrate very good performance, sometimes showing results at the level of more expensive quad-core processors. However, it is not entirely correct to compare these results directly; or rather, they are valid for synthetic (or rather sterile) conditions of test benches, when both dual-core cores can be guaranteed to be used to solve the test problem. In reality, background processes, if they do not take up significant resources, at least, to one or another, weakly predictable degree, “interfere” with their data in the cache memory. At the same time, both Phenom, Phenom II, and Core i7 (especially models with an unlocked multiplier) have excellent capabilities for selective overclocking of processor cores, so turning them into high-frequency dual- or triple-core processors, if a specific task requires it, is not a problem. difficulties.

Test bench configuration

CPU	Phenom X4 9850 Black Edition	Phenom II X4 940	Core 2 Quad Q6600	Core 2 Quad Q9300	Core i7 920
Kernel name	Agena	Deneb	Kentsfield	Yorkfield	Bloomfield
Production technology	65 nm	45 nm	65 nm	45 nm	45 nm
Core frequency, GHz	2,5	3,0	2,4	2,5	2,66 (***)
Number of cores	4	4	4	4	4
L1 cache, I/D, KB	64/64	64/64	32/32	32/32	32/32
L2 cache, KB	4 x 512	4 x 512	2 x 4096	2 x 3072	4 x 256
L3 cache, KB	2048	6144	-	-	8192
RAM (*)	DDR2-1066	DDR2-1066	-	-	DDR3-1066
Multiplication factor	12,5 (**)	15 (**)	9	7,5	20
Socket	AM2+	AM2+	LGA775	LGA775	LGA1366
TDP	125 W	125 W	95 W	95 W	130 W
Price	N/A(0)	N/A(0)	N/A(0)	N/A()	N/A()

(*) maximum frequency supported by the memory controller in the processor, it is permissible to set a lower frequency provided by this memory standard (for example, DDR2-667 and DDR2-800 for processors supporting DDR2-1066), for processors with LGA775 socket the frequency and type of memory are determined chipset used
(**) unlocked for user overclocking capability
(***) when the Turbo Boost “auto-overclocking” function is activated (which is implied by default), the actual frequency of individual cores increases relative to the nominal up to 2.8–2.93 GHz, depending on the load, so it is incorrect to directly compare this value with fixed frequencies of other processors

• memory capacity on stands: 4 GB (3 GB for Core i7 920);
• hard drive: Samsung HD401LJ (SATA-2);
• coolers: Thermaltake TMG i1, TMG a1;
• power supply: Cooler Master RS-A00-EMBA.

Testing

The performance testing methodology (list of software used and testing conditions) is described in detail in the article. For ease of perception, the results in the diagrams are presented as percentages (the result of the Intel Core 2 Quad Q6600 in each test is taken as 100%). Detailed results in absolute values ​​are available in the form of a table in Microsoft Excel format, which shows all previously tested processors.

3D modeling packages

Even looking at the first diagram, one can assume that Phenom II is quite serious about fighting for its place in the sun and, in any case, being a worthy competitor to the Core 2 Quad. If you look at the detailed results, you get the idea that the matter will not stop there. For example, in Lightwave, rendering takes less time than on the Core i7 920, and in terms of rendering speed in Maya, the Phenom II is faster than the Core 2 Extreme QX9770 (here, however, the Core i7 wins back). In a word, there is no longer talk of any “one-sided game”, and we will not be surprised if in some tests the Phenom II competes not only with competitors approximately equal in price, but also with more expensive ones.

CAD/CAM packages

A similar arrangement, with the only difference being that the “staircase” has become flatter. And if we take into account that this group of tests is quite conservative and weakly uses more than two cores, therefore the automatic overclocking (Turbo Boost) of the Core i7 gets the opportunity to shine. It is quite natural to assume that by similarly overclocking a couple of cores in the Phenom II, you can further reduce the existing difference. Fortunately, the ability to independently control multipliers of processor cores is available in Phenom initially, although auto-overclocking is not implemented at the hardware level, but thanks to proprietary utilities it is implemented very conveniently (including, the user can determine the desired level and method of overclocking, which will be automatically selected when starting it or other application). This requires a little more fiddling during the initial setup, but is quite a fun experience in itself, and the result can be more interesting from a performance point of view than any automatic method. We examined the topic of overclocking the Phenom II in detail in the corresponding article, but now let's continue testing at the standard frequency.

Compilation

Moreover, here awaits us the first convincing victory over both rivals, without any acceleration.

Professional work with photographs

However, it is still too early for AMD fans to get the champagne. Traditionally favoring Intel products, the graphics editor Adobe Photoshop is simply obliged to support at least the Core i7, which it successfully does. However, in a confrontation with the Q9300, the Phenom II continues to confidently control the situation.

Science and mathematics packages

In this subgroup, Phenom II takes first place among all previously tested processors in Maple, and in Mathematica it is at the same level as the leaders. But then we look at the MATLAB results, and they are what make the overall result not so impressive. We have already written in detail about the problems with this test. In this case, testing is carried out on the same version of the library for all processors (mkl.dll), since this is the solution used in the next version of this program (2008b), that is, recommended by the developers themselves, although it is obvious that this approach is far from optimal. At the same time, it cannot be said that the built-in benchmark in this test completely measures the weather on Mars, although the spread between the results taken from different series of measurements is somewhat large for a reliable comparison of processors with similar performance. Also, it has not yet been possible to establish to what extent it reflects at least typical tasks for MATLAB users themselves. But these are questions that obviously relate not to the topic of this testing, but to improving the methodology. From a practical point of view, you just need to keep in mind that in the other two tests, the results of the Phenom II X4 940 are close to the Core i7 920, and there is no talk of lagging behind the Q9300, even formally. So the potential of the Phenom II X4 940 as a mathematical “solver” is quite good.

Web server

In this category of tasks, AMD processors have performed successfully before; it is easy to see that the Phenom 9850 has the best result in this subtest among the results in all other categories. And Phenom II is actively developing this success. At the same time, it is in this test that the Q9300 is only formally superior to the Q6600, hence the maximum gap between the Phenom II X4 940 and the Q9300, also in comparison with the results in all other subgroups.

Overall "professional" score

To be precise, the result of the Phenom II X4 940 was lower than that of the Core i7 by 4.38%, but it managed to outperform the Q9300 more significantly - by 7.55%.

Archivers

If you look at the detailed results, we find that the Phenom II and Core i7 are equal in the 7-Zip and WinRAR archivers, and the advantage of the Intel processor recorded in the final rating is ensured by a slight difference (less than 10 seconds) in single-threaded Ultimate ZIP, where Turbo Boost manifests itself to the maximum. So from a practical point of view, these processors can be considered equivalent and have secured a significant lead over the rest of the “group of pursuers”.

Media encoding

We observe almost the same slender ladder as in the very first diagram. And what’s typical is that the detailed results again give reason to rejoice for those who are reviewing reviews of new processors, hoping to see that the new product will reach “new heights” in some tests, that is, demonstrate a result that is clearly superior to its competitors. In Canopus ProCoder, the honor of the Core i7 (and Intel processors in general) can now probably be defended only by the older models from this line. Of course, there are also tests where the Phenom II is not so strong (it even formally lags behind the Q9300 in one test (XviD). Well, on average, the result presented in the diagram is obtained.

The situation is very pleasant for a tester, because, strictly speaking, the point of reviewing processors appears precisely when there are, on average, equally strong competing models on the market. But they differ in architectural and technological parameters, and due to this very difference, they have their own characteristics, which allow us to say that this processor is especially good for those who spend all day and night in Photoshop, and the other is simply a “must have” for those who are partial to games . By the way, what do we get in games...

Games

But it turns out just great for the Phenom II! However, victory in gaming tests, unlike other successes, was perhaps the easiest to predict. After all, the Phenom 9850, upon closer examination, cannot at all be called some kind of unambiguous outsider, and many testers noted a funny phenomenon (pardon the pun) when even in cases where the Phenom lost at low resolutions, as the graphics settings and resolution increased There was not only a completely natural limitation of performance to the resources of the video card, but also a small but noticeable advantage of the AMD platform. And if you remember the comparisons between Phenom and Athlon, it was in games that the advantages of the new architecture were very clearly manifested. It is now obvious that the potential of the K10 architecture of the Phenom itself, for some reason, was far from fully revealed, and seeing how this potential begins to be demonstrated in the case of the Phenom II, it is quite logical to expect that tangible progress will be found on the gaming front. At the same time, for the Core i7, games turned out to be the weak point, where the new core shows minimal advantage over the previous one.

Perhaps AMD already has something to congratulate in the new year, which is even somehow symbolic, considering that last year the company’s products that stood out the most in tests were graphics processors from the Radeon HD4800 series. And for AMD’s gaming platform to take on an ideologically complete form, a processor was required that, like the HD4850/HD4870, would be positioned for a mid-range budget, but would provide gaming comfort at the level of more expensive competitors. Of course, by Phenom II in this case we mean the line as a whole, since there is reason to assume that younger quad-core, and possibly triple- and even dual-core models will be attractive for gaming computers (of course, in combination with video cards of different levels, since For a gaming computer, the issue of balancing the performance of these components is fundamental). As for the Phenom II X4 940 under consideration, it will be difficult for extreme versions of the Core i7 to significantly outperform this processor, so many who want to get maximum performance in games will also choose the Phenom II (probably not without thoughts of overclocking) , and the amount saved will significantly facilitate the purchase of components of some 3-Way SLI or Quad CrossFire.

Amateur work with photographs

Probably, Phenom II liked building dynamic images in games so much that when processing the same array of photos with five different photo editors, he got bored and lost! But seriously, it is quite expected to see less expressive behavior in this subtest, because even for Phenom from the “first edition” the result here is not impressive, and, apparently, no fundamental differences were introduced at the microarchitecture level in Phenom II. At the same time, the Core i7, which has already demonstrated its skills when working with photographs in Photoshop, gets the opportunity to demonstrate a similar advantage here. What can I say? To some extent, this subtest still imitates professional work (after all, it is not typical for amateur photo editing to process gigabytes of photos in batch mode). As for any simple operations on single photographs taken from any megapixel camera, this is all performed in these graphic editors in real time, that is, instantly, on any processors participating in testing, and even weaker ones. Of course, this in no way diminishes the merits of the Core i7 from a testing point of view, and, on the contrary, shows that in a number of tasks that are explicitly tied to integer calculations, it is objectively difficult to “butt heads” with this processor. Most likely, AMD will not try to compete in such applications head-on, increasing the frequency, cache, and even more so, urgently redesigning the core, which is very successful in other aspects, and will simply “go around”, as a plugin for Photoshop is already being prepared that will allow use video card resources. Which, of course, for AMD itself, as a manufacturer of graphics processors, is very tempting, and as a video card that will accelerate such calculations at an unprecedented speed, they promise that a very average one, that is, inexpensive, will do. Let's see.

Overall amateur score

But the “average temperature in the hospital” looks so calm and, in contrast to the results in individual subgroups, everyday, and since the author has run out of original comments, those who wish can offer their options on the forum, the most successful one will be added to the article :)

conclusions

First of all, testing showed that, very gratifying for users and testers, and for the IT industry in general, the fact that competition in the central processor market is again becoming very interesting. Phenom II was clearly a success, and in a number of problems we can talk about this without even linking the assessment to conversations about cost.

However, AMD is not going to be greedy, that is, the recommended cost of the older Phenom II model is set lower than that of the younger Core i7, although, as already noted, if we take into account the cost of the platform (motherboard and memory), comparison with processors from the line is much more appropriate Core 2 Quad, and here AMD’s advantage is obvious (and on average remains even if we use the more expensive Q9400/Q9450 as processors). And, for example, in games, only “extreme” models from Intel, the price of which is 4-5 times higher, can compete with the new AMD processors. Moreover, last year AMD worked very fruitfully to improve the functionality of chipsets under its brand (especially with integrated graphics), which we wrote about in detail. And with the expansion of the choice of processors with attractive characteristics, these developments will be able to be appreciated by a larger number of users. Of course, the high results of Phenom II will please those who have already purchased a computer on the Socket AM2+ platform (with an Athlon or Phenom processor) and will think about an upgrade in the future.

The processor is also interesting for overclocking enthusiasts (we will return to this issue in more detail); it should also be noted that the average heat dissipation has clearly decreased due to the transition to a 45 nm process technology. AMD claims that by 35-50% depending on the load (for processors with a declared TDP = 125 W from the previous and new line), for the sake of experiment we installed a box cooler from the Phenom 9550, designed for a 95 W thermal package, and were able to run a complete set of tests, and only in rare cases did the cooler reach maximum speed. Of course, this is a purely rough test, if only because the cooler control algorithm can be adjusted independently, but we must keep in mind that in general any power consumption tests based on the results of testing a single processor instance are for reference only. The main practical conclusion is that even for older Phenom II models, obviously, it will not be difficult to organize inexpensive, low-noise cooling, including during moderate overclocking (and extreme sports enthusiasts, as always, are free to find completely silent, some kind of liquid etc). Most users will be quite satisfied with the performance of the standard cooler (by the way, copper, with heat pipes) included in the package. And, apparently, AMD will not have any difficulties or delays in converting processors to a 95 W thermal package, which is planned simultaneously with the release of the Socket AM3 platform and the expansion of the model range.

Since we are talking about plans, it is logical to expect, simultaneously with the next wave of chipsets, the release of boards with the Socket AM3 connector and, accordingly, processors that differ from the current ones by supporting dual-channel DDR3-1333. Moreover, support for DDR2 will also remain, that is, these processors can be installed on boards with the Socket AM2+ connector, accordingly, it is not difficult to assume that the migration will be even smoother than the transition from Socket 939 to AM2. Most likely, the actual benefits of the new type of memory will appear only in certain applications. And it’s even more likely that the incentive to choose AM3 when choosing components for a new computer will be, for example, some functional advantages of new chipsets (associated, for example, with an integrated video core) and simply interesting new board models. At the same time, it is not at all surprising if owners of modern high-quality motherboards for AM2+ do not rush to upgrade the board and memory when purchasing an AM3 processor. By the way, the smooth migration described above seems self-evident, because most readers interested in the topic of processors and platforms have, of course, heard about it more than once, since it became known long before the release of the first version of Phenom. In fact, maintaining the electrical and logical compatibility of the connectors, and even more so the support of different types of memory by the processor controller, certainly implies many original technical solutions. And we, perhaps, will even be able to appreciate all this. Another reason to be pleased with the results of Phenom II, since all the advantages associated with the convenience of migration only make sense if the subject of the upgrade itself is interesting.

In this article we will talk about choosing the optimal video card for AMD AM3 and FM1 processors:

• Phenom X6 1035T, 1045T, 1055T, 1065T, 1075T, 1090T, 1100T
• Phenom X4 910, 920, 925, 940, 945, 955, 960T, 965, 970, 975, 980
• Athlon II X4 620, 630, 635, 640, 645, 655
• Athlon II X4 631, 641, 638, 651, 651K

Due to the unstable economic situation, many PC users do not want or do not have the opportunity to change the platform, “sitting” on the old one for as long as possible. Therefore, many people face the question of choosing the optimal combination of an old multi-core CPU and a more or less modern video card. We will try to select the most comparable solutions from those available on the market.

Video card for AMD Phenom X6 1035T, 1045T, 1055T, 1065T, 1075T, 1090T, 1100T and AMD Phenom X4 910, 920, 925, 940, 945, 955, 960T, 965, 970, 975, 980

These processors are close in performance to solutions from the AMD FX-4000 and FX-6000 lines. Consequently, older four- and six-core models, when overclocked, will be able to work in tandem with video cards of the levelAMD Radeon R7 370/RX 460 And NVIDIA GeForce GTX 750 Ti. We recommend using the younger ones together with solutions of the levelAMD Radeon R7 360 And NVIDIA GeForce GTX 750.

Video card for AMD Athlon II X4 620, 630, 635, 640, 645, 655, 631, 641, 638, 651, 651K

We recommend using the more productive solutions listed above in conjunction with video adapters of the level AMD Radeon R7 360 And NVIDIA GeForce GTX 750. As for models with low frequencies, somewhat outdated ones are best suited to them. AMD Radeon R7 250/R7 250X And NVIDIA GeForce GTX 650 / GT 740.

Test bench and software configuration

The Intel Core i5 750 and Phenom II X4 925 were chosen as opponents for the Phenom II X6 1055T in today's testing. The choice of the first is obvious, since the processor has a very close retail price and is one of the best (if not the best) options for building a home high-performance PC. The Intel Core i5-750 has excellent overclocking potential and often exceeds the 4000 MHz mark when using inexpensive air coolers. The Phenom II X4 925 is included in testing to determine performance scalability when increasing the number of processing cores from four to six, as well as to evaluate the gain from using the Turbo Core in applications that cannot boast of multi-thread optimization. It is worth noting that Intel Core i7 processors with Hyper-Treading support are significantly more expensive than the Phenom II X6 1055T, and therefore cannot be considered as direct competitors. The main characteristics of the test participants are shown in the table:

Name	AMD Phenom II X6	AMD Phenom II X4	Core i5
Model	1055T	925	750
Core	Thuban	Deneb	Lynnfield
Stepping	E0	C3	B1
Technical process, nm	45nm SOI	45nm SOI	45 high-k
Connector	AM3	AM3	LGA1156
Rated frequency, MHz	2800	2800	2666
Maximum frequency, MHz	3300*	2800	3200**
Factor	14-16,5*	14	20-24**
HyperTransport/QPI, GT/s	4000	4000	4800
L1 cache, KB	6x128	4x128	4x(32+32)
L2 cache, KB	6x512	4x512	4x256
L3 cache, KB	6144	6144	8192
Supply voltage, V	1,125-1,40	0,90-1,40	0,65-1,40
TDP. W	125	95	95
Limit temperature, °C	62	71	72,5
Set of instructions		ISC, IA32, x86-64, NXbit, MMX, 3DNow!, SSE, SSE2, SSE3, SSE4a	RISC, IA32, XD bit, MMX, EM64T, SSE, SSE2, SSE3, SSE4.2

* - with Turbo Core technology enabled
** - with Turbo Boost technology enabled

To test AMD processors, a test bench was assembled:

• processor: AMD Phenom II X4 925 (2800 MHz, 4 cores), AMD Phenom II X6 1055T (2800 MHz, 6 cores);
• motherboard: MSI 890FXA-GD70 (AMD890FX+SB850, BIOS 1.60 from 05/18/2010);
• video card: PowerColor Radeon HD5850 1GB (850/4500 MHz);
• sound: Creative Audigy 4;
• power supply: FSP600-80GLN;
• body: Cheiftec CH01-B-SL.

The Intel processor was tested as part of the configuration:

• processor: Intel Core i5-750 (2666 MHz, 4 cores);
• cooling system: Xigmatek-HDT1284S;
• motherboard Gigabyte GA-P55-UD3R (Intel P55, BIOS F4 from 11/20/2009)
• memory: Take-MS, 2x2GB PC-10660;
• video card: PowerColor Radeon HD5850 1Gb (850/4500 MHz);
• sound: Creative Audigy 4;
• drive: WD1001FALS (1000 GB, 7200 rpm);
• power supply: FSP600-80GLN;
• body: Cheiftec CH01-B-SL.

Both systems ran Microsoft Windows 7 Enterprise 64 bit (90-day trial) with the latest updates. AMD Catalyst 10.4 SB plus AHCI drivers for the AMD test bench and INF Update Utility 9.1.1.1025 for the Intel platform were installed. The video card was running the ATI Catalyst 10.4 driver.

AMD Phenom II X6 1055T and Intel Core i5-750 processors were tested in nominal operating mode and overclocked. During overclocking, Turbo Core and Turbo Boost technologies were disabled. Due to abnormally hot weather, the overclocking of the Intel processor had to be limited to 3800 MHz. AMD Phenom II X4 925 was tested only at the standard frequency. For ease of understanding, all the main system settings are summarized in the table:

CPU	Processor frequency, MHz	Memory frequency, MHz	Basic delays (CL-tRCD- tRP- tRAS-CR)	Uncore frequency for Intel, NB for AMD, MHz	QPI frequency for Intel, NT for AMD, MHz	Vcore, V
Phenom II X6 1055T	2800	1600	9-9-9-28-1T	2000	2000	1,425
	3710	1412	8-8-8-24-1T	2385	2385	1,46
Phenom II X4 925	2800	1333	8-8-8-24-1T	2000	2000	1,425
Intel Core i5-750	2666	1333	8-8-8-24-1T	2130	2400	1,125
	3800	1520	8-8-8-24-2T	3040	3040	1,325

Test results

Today's testing opens with the memory subsystem performance test, which is part of the Lavalys Everest 5.50 information and diagnostic utility. This application allows you to measure bandwidth with high accuracy, as well as determine the latency of access to RAM.

Alas, the miracle did not happen, and in terms of performance of the RAM subsystem, the AMD Phenom II still lags behind the Intel Core i5 750. Even the long-awaited support for DDR3-1600 does not save the AMD processor from defeat. But you should not be upset, since in real applications the balance of forces can be very different from synthetics.

In the Super Pi discipline, Intel processors traditionally lead, and this time the winner is the Core i5-750. It should be noted that Super Pi is a single-threaded application, and there is no benefit from using additional processing cores. This test is sensitive to clock frequency and the Phenom II X6 1055T is 15% ahead of the “equal-frequency” X4 925 precisely thanks to the Turbo Core.

But the Wprime application has native support for multi-core processors. In this test, the X6 1055T is significantly ahead of its predecessor X4 925 and easily crushes its competitor from Intel, and the latter is not saved by overclocking to 3800 MHz!

Testing in the Fritz Chess Benchmark app will be especially interesting for chess fans. Others can simply compare the relative performance of today's test participants in calculating chess combinations.

Chess calculations scale well with increasing number of computational threads. In nominal mode, the newcomer easily outperforms its competitors, but when overclocked, the results of the X6 1055T become completely unattainable. Complete victory for X6 1055T!

The PC Mark Vantage test package offers universal tools for assessing the performance of all major subsystems of a personal computer. In our today's review we will compare the results of the Memory, TV and movie, Music and Communication scenarios.

The memories scenario includes tests on simultaneously working with images and transcoding DV video into a format for portable devices. In this scenario, the X6 1055T and i5-750 at stock frequency demonstrate a similar level of performance, and the X4 925 loses to both of them. Overclocking an Intel processor makes it an absolute leader. The TV and Movie script emulates intensive work with video content, such as simultaneous transcoding and playback of high-definition video. At the nominal frequency, the six-core processor has a slight advantage. Intel is a little behind, and the X4 925 deservedly takes last place. But the performance of the X6 1055T does not scale very well with increasing frequency, but the i5-750 receives good dividends from overclocking and takes the lead. The Music script includes audio encoding tasks and emulates working in Windows Media Player. The X6 1055T processor famously outperforms the X4 925, which is quite natural. But the reason for such low results from Intel at the standard frequency remains a mystery to us. There is no error here, since the tests were repeated three times. Overclocking the Intel processor puts everything in its place and again gives the Core i5-750 an advantage. But the Communication test scenario, which emulates working with WEB applications, gives preference to the new product from AMD, and overclocking the 1055T only strengthens its position. Looking at the results, we can note a similar level of performance between the Core i5-750 and the Phenom II X6 1055T at the standard frequency, but the Phenom II X4 925 looks like an outsider.

From synthetic applications we move on to applied tasks and will start with one of the most common ones - data archiving. Today's test involves the WinRAR archiver, as one of the most common representatives of this class of software, and 7-Zip, a very powerful and completely free archiver. Measurements were carried out using built-in performance testing tools.

In nominal mode, the WinRAR archiver runs fastest on a Core i5-750. And, if the X4 925 cannot compete with the Intel processor, then two additional processing cores already allow the X6 1055T to fight the competitor “on an equal footing”. However, as the frequency increases, the performance of the i5-750 increases so much that it leaves no chance for its rivals from the AMD camp.

A slightly different picture is observed in 7-Zip. This archiver works great on multi-core processors and scales well in frequency. In nominal terms, the X6 1055T is significantly ahead of other participants, while the X4 925 and Core i5-750 processors demonstrate comparable results. In overclocking, the X6 1055T continues to hold the lead, ensuring the unconditional victory of the six-core AMD architecture!

Another typical task that users very often encounter is video encoding. We tested performance when processing HD MPEG-4 using the x264 HD Benchmark.

Very interesting results are obtained with two-pass compression of a video file using the H.264 codec. On the first pass of encoding, the Core i5-750 processor is faster, while both AMD processors are slightly behind. But when performing the second, final pass, the X6 1055T demonstrates all the advantages of six-core processors and confidently outperforms its rivals. And with the increase in frequency, the new Phenom became completely unattainable for its competitor.

The following test reflects the performance of processors when rendering images in 3D editors. It's no secret that home PCs are often used to complete freelance tasks, and for such users, time is money. To evaluate the speed of work in such tasks, the Cinebench 11.5R application was used.

Rendering 3D images is one of those tasks that scales well with an increase in the number of computational threads. In multi-threaded mode, the X6 1055T easily beats its rivals, and even overclocking the Core i5-750 only allows it to catch up with the younger six-core AMD processor. It is noteworthy that the single-threaded mode shows a noticeable increase from using Turbo Core. It is thanks to Turbo Core that the X6 1055T outperforms its younger brother, the X4 925, which is deprived of this useful function.

From synthetic applications and applied tasks, we smoothly move on to studying the performance of the Phenom II X6 1055T in games. But first, let me show you the results in 3DMark Vantage.

The overall winner was the Intel Core i5-750, but look how close the Phenom II X6 1055T comes to it. And in the CPU test, where physics and artificial intelligence are calculated, the new AMD processor leaves no chance for its opponent, both in overclocking and at standard frequencies. The Phenom II X4 925 has the hardest time, since its not the most advanced architecture and low clock speed do not allow it to demonstrate good results.

Our today's performance study ends with testing in modern games: FarCry 2, S.T.A.L.K.E.R. Call of Pripat, Tom Clancy`s HAWX and World in Conflict: Soviet assault. Testing was carried out in a resolution of 1680x1050 with high image quality settings. For S.T.A.L.K.E.R. CoP used the official benchmark; in all other cases, the game's built-in performance measurement tools were used.

Judging by the test results, the Intel Core i5-750 wins in this discipline with a minimal advantage. The Phenom II X4 925 shows the lowest result, and the X6 1055T occupies the second step of the pedestal. It was very difficult for the six-core processor to get second place, and for this we should thank the Turbo Core technology rather than the two additional cores. But this does not mean that the Phenom II X4 925 or Phenom II X6 1055T cannot provide a comfortable fps level in games. On the contrary, the performance of any of the considered processors is quite enough for a comfortable game, and with an increase in resolution and detail, the difference will disappear altogether. The fact is that modern games (with rare exceptions) cannot use more than two computing cores, so programmers have something to work on in terms of multi-threaded optimization...

conclusions

It's safe to say that with the release of the Phenom II X6 1055T, AMD has strengthened its position in the middle-end segment. The new processor offers excellent performance in applications optimized for multi-threaded execution. Thanks to the introduction of Turbo Core technology, the newcomer copes well with tasks that do not have multi-threaded optimization. Moreover, in most optimized programs, the increase from two additional processing cores was close to 50%. In most application tasks in general, the Phenom II X6 1055T outperforms the Core i5-750, but is slightly behind it in modern games. Therefore, if you often deal with 3D modeling, process large amounts of video content, or make extensive use of applications optimized for multi-threaded computing, then the Phenom II X6 1055T is your choice. It will also provide an acceptable level of performance in any task.

If your priority is performance in modern games, then the Intel Core i5-750 will provide better performance. As for the AMD Phenom II X4 925, this processor demonstrated the lowest level of performance. But don’t forget that the price of the X4 925 is about 25% lower than that of other test participants, and the overclocking potential allows you to boost frequencies to 3600-3800 MHz. Therefore, many will choose this option with a good price/performance ratio. In the meantime, we can confidently say that by releasing their six-core processors for the mass market, AMD is moving in the right direction.

The MSI 890FXA-GD70 motherboard was provided by the company for testing