Processors. Mobile processors AMD V, Athlon II, Turion II, Phenom II: background information Which is better: hairdryer or athlone
As the famous Russian proverb says, “Hand washes hand.” This proverb applies very well to various figures in the IT industry. And that's why. Manufacturers of software and hardware are actively interacting with each other so that the people, that is, you and me, change their hardware to something more recent as much as possible and as often as possible. And every time, as a result of such joint work, marketers from different companies declare the necessity and revolutionary nature of new technologies. A reasonable question arises: what should those who do not want or cannot pay exorbitant amounts of money for new top-end hardware do? Will I really have to sit on Celerones or Semprons all my life? No! You won't have to. AMD presented to the public a couple of its new developments - processors AMD Athlon II X4 630 and A MD Athlon II X4 620. The new items compare favorably with their “phenomenal” counterparts at an extremely attractive price. Using the Athlon II X4 as an example, I propose to figure out what we get as a result from AMD.
First of all, let's look not only at the processor itself, but also at its technical characteristics in comparison with other models of modern AMD processors:
| Characteristics | AMD Athlon II X4 630 | AMD Phenom II X3 720 Black Edition | AMD Phenom II X4 955 | |
| Volume L1 Cache (Instructions/data),KB | 4 x 64 | 4 x 64 | 3 x 64 | 4 x 64 |
| L2 Cache volume,KB | 4 x 512 | 4 x 512 | 3 x 512 | 4 x 512 |
| L3 Cache volume, MB | - No - | - No - | 6 | 6 |
| Factory multiplier | 13 | 14 | 14 | 16 |
| Number of active cores by default | 4 | 4 | 3 | 4 |
| Rated frequency, MHz | 2600 | 2800 | 2800 | 3200 |
| Supported connectors | AM2+|AM3 | AM2+|AM3 | AM2+|AM3 | AM2+|AM3 |
Technical process, nm | 45 | 45 | 45 |
So, having analyzed the presented table, let’s try to set the vector of this material. Let’s ask ourselves a number of questions, the answers to which will help you navigate when choosing a new processor:
- How well do the new processors overclock compared to older Phenoms? The question is perhaps the most pressing, since in our opinion, a processor with such an attractive price (the recommended retail price of the AMD Athlon II X4 620 is around the $120 mark) is aimed primarily at overclockers, and don’t feed them (that is, you) bread , just let me speed it up.
- What is better - 4 inexpensive cores without a third-level cache or 3 slightly more expensive cores with a 6MB third-level cache? Based on the characteristics of both CPUs, it is impossible to draw an unambiguous conclusion in favor of one or another solution. You need to look at the test results carefully.
First of all, let's finally get acquainted with the processor itself and the hardware that will act as a test bench:
The latest version of CPU-Z 1.55.2 at the time of testing recognized the processor and accurately showed its technical characteristics. The following components were used as a test platform:
- Processors: AMD Athlon II X4 620, AMD Phenom II X3 720
- CPU cooling system: ProModz Cooled Silence Extreme Packge.
- Motherboard: Gigabyte MA-790GP-DS4
- RAM: Kingston HyperX 2x2048MB DDR-2 1066 @ 5-5-5-15
- Video card: AMD Radeon HD 4870 512MB (PowerColor)
- Hard drive: Western Digital Raptor 36GB (SATA, 10,000 rpm)
- Power supply: IKONIK Vulcan 1000 Watt
- Operating system: Microsoft Windows Vista x64 SP1 Home Basic Edition
Overclocking
When you start overclocking a system, you must remember that overclocking one of the system components is not enough. The maximum performance increase in both 2D and 3D tests is possible only if all system components are overclocked (wisely :)). The AMD Athlon II X4 620, unlike its older brothers Phenom II X4 Black Edition, has a multiplier that is locked to increase, and this automatically introduces a number of additional factors that can prevent successful overclocking of the system. For those who are still learning to overclock computers, we list the factors on which overclocking depends:
- Operating temperature of the processor and other system components (directly depends on the cooling systems and TDP of the CPU);
- Operating voltage of CPU, RAM, NB/SB motherboard, etc. (determined by the capabilities of the BIOS, software and user hands)
- Sometimes the increase in the frequency of one of the system components depends on the overclocking ability of another component (for example, the absence of memory reduction dividers can limit CPU overclocking if the RAM cannot reach a higher frequency)
Let us draw your attention to the third item on our list. Successful overclocking and maximum results with the AMD Athlon II X4 620 processor can only be achieved if your motherboard overclocks well on the bus and the RAM scales well in frequency (in this regard, using DDR-3 significantly simplifies the task).
When starting practical overclocking tests of the new processor, we never expected that the test motherboard would begin to behave inappropriately out of nowhere. As we noted above, to successfully overclock the AMD Athlon II X4 620 it is necessary that " M o B o" allowed stable operation at high bus frequencies (preferably up to 300 MHz). While searching for the ceiling of our board from Gigabyte, after another reboot... The open test bench became a source of an unpleasant smell of "scorched iron" and the system did not start! The maximum frequency that managed to conquer before this incident, was 270 MHz. Thank God that we tested the new processor at nominal BEFORE overclocking the system. It’s unpleasant, but there is no other choice. Today we will NOT show the results of overclocking the new budget quad-core processor from AMD. However, we will try supplement this material with tests of this processor on another motherboard, which, in addition, will work with DDR-3 memory.It is on this that the CPU will be re-overclocked.
Since nothing good came out today with overclocking, let's see what happened at face value. Let's move on to evaluating the testing results of the AMD Athlon II X4 620.
EVEREST Ultimate Version 5.0
| CPU | Memory Read | Memory Write | Memory Copy | Memory Latency | CPU Queen | CPU PhotoVorxxx | CPU Zlib | FPU Julia | FPU Mandel | FPU SinJulia | |
| 8315 | 6875 | 9698 | 51.3 | 16181 | 19654 | 69005 | 7003 | 43987 | 2274 | ||
| AMD Phenom II X3 720 | 7958 | 6584 | 9610 | 53.2 | 13286 | 16805 | 56034 | 5657 | 3615 | 1829 |
Judging by the data obtained, the Phenom II X3 lost the battle in this test. The additional amount of third-level cache memory “spoiled” the performance results in the memory subsystem tests, but in the remaining tests the AMD Athlon II X4 620 took the lead thanks to the “extra” core. 1:0 in favor of the newcomer :)
Super Pi 1M
Since Super Pi is a single-threaded test, the presence of a fourth core inside the AMD Athlon II X4 62 does not affect the final score. But the Phenom II X3 720 makes full use of its trump card in the form of a 6MB third-level cache and a 200 MHz higher nominal frequency.
WinRAR 3.90 x64
WinRAR, unlike the synthetic Super Pi, is the most “real” application. The latest versions of this archiver implement not only multithreading support, but also support for the x86-64 instruction set. Judging by the results, the presence of an additional core did not allow the AMD Athlon II X4 620 to overtake the Phenom II X3 720, which took the lead in this test thanks to a higher core frequency and third-level cache.
Cinebench R10 x64
The rendering speed in multi-threaded mode is higher in the AMD Athlon II X4 620. The higher frequency and L3 cache do not give the Phenom II X3 720 the required boost when the test actively uses 4 threads. But in single-threaded mode, the new product is slightly inferior to the Phenom processor. In truth, the victory in this test should be given to the AMD Athlon II X4 620, since now it is unlikely that anyone will carry out final rendering on a single core.
Today there are many manufacturers of computer equipment, which primarily includes AMD. This company has been engaged in the production of central processors and video cards for many years.
Today, many buyers who want to buy a central processor are thinking about which model is best to choose. Today we will look at several processor options - amd athlon x2 and amd phenom. As a rule, in terms of their technical characteristics, both of these processors are very high quality and productive, but each of them has its own purpose.
The phenom processor is designed to handle a huge variety of tasks that are directly related to graphics, meaning it has a lot of power and multiple cores. Thanks to this, this processor can perform the most complex graphics tasks without any slowdowns. It belongs to the classification of professional processors. It is made of high-quality materials, due to which it practically does not heat up when performing various complex tasks, thus ensuring quiet and reliable operation of this or that computer. It also has extremely low power consumption, which prevents the power supply and motherboard from overheating. The downside of this product is that it is intended exclusively for professional use, which means you cannot play powerful games on such a computer.
Also on topic: Why is there no sound on my computer?
Another thing with the central processor amd athlon x2. This product is also designed on the basis of two cores, due to which the number of tasks performed can also be huge. This product is more relevant to the number of gaming processes, since it operates at a very powerful frequency and has great performance. That is why it is able to pull off the latest gaming innovations from well-known gaming companies. This processor also has a lot of power consumption and, as a result, it gets hot. As a result, this requires a powerful cooler and radiator, which are needed to cool it.
Each of the above processors has a huge number of advantages and some disadvantages. For example, if you buy a processor for games, then you should pay attention to amd athlon x2, and if the device is purchased for various 3D design and graphics display, then you need to buy amd phenom.
Closing the circle of “historical testing”, today we will deal with a platform that formally remains among the living and well, although ideologically even older than the previously reviewed AMD FM1 and Intel LGA1156. How does she do this? We have already dealt with this issue: Socket AM3+ 2011 is practically no different from “just” AM3 2009, which was obtained by switching from DDR2 to DDR3 from AM2/AM2+ from 2006, and these, in turn, are practically nothing more than Socket 939 from the summer of 2004, but with DDR2, and not with “simple” DDR. It would be more correct, however, to talk even about 2003, when Socket 940 appeared: Socket 939 is its simplification, without support for multiprocessor configurations. During this time, not only memory standards, of course, but also some other interfaces managed to change, but conceptually in the form of AM3+ we have a classic platform of the 2000s - three-chip and a relatively low degree of integration. It is also worth noting that the latest microarchitectural updates of processors produced for it date back to the end of 2012, i.e., from this point of view, even the latest modification of AM3+ is already history (to the same extent as LGA1155, for example). However, on other platforms, AMD ships no more than two-module processors (supporting, accordingly, only four computation threads) with a significant bias towards integrated graphics, so the most powerful AMD processors are still devices for AM3+. They have not been updated for a long time, but their final obsolescence is planned only for the second half of this year - in connection with the transition to a single (finally!) AM4 socket, for which both high-performance processors without integrated graphics and relatively budget ones with such will be produced. It is easy to notice that this is not yet an analogue of LGA1155 and subsequent Intel platforms - rather, a repetition of LGA1156, since when choosing a fast processor “for the load” you will have to use a discrete video card. But this is still much better than what has happened to the company’s product range over the past five years, when various FMx and the same long-outdated AM3+ were simply incompatible with each other.
How did the company manage to keep AM3+ afloat without updating processors? Yes, very simply: due to the price. In any case, we had to long ago forget about competition for lovers of high performance, but for about the same money a buyer can buy either an eight-thread FX-8350/8370 or a four-thread Core i5-6400. Yes, of course, price comparison in this case is not entirely correct, since it does not take into account other features of the platforms and, first of all, the opportunity to save on a video card in the case of the Intel platform. However, if you still need to purchase a video accelerator (for example, when you are interested in games - we were and continue to be of the opinion that a full-fledged gaming computer is still impossible without a discrete video card), this problem disappears. And at first glance, it doesn’t matter that the same FX-8350 appeared back in 2012: advertising in its case generally talks about eight cores (forgetting to clarify that these are slightly different cores from those in other processor architectures, even AMD itself), i.e. . e. creates the impression of a processor made by Intel Stubucks. This is the right approach, the wrong one - but it works. It’s useful to check how. In the end, as mentioned above, this year we will finally be able to meet new AMD processors - so in any case they will have to be compared with the old ones. Today we will create an “information reserve” on old and even very old processors, fortunately such an opportunity has presented itself.
Test bench configuration
| CPU | AMD Phenom II X6 1075T | AMD FX-8370 |
| Kernel name | Thuban | Vishera |
| Production technology | 45 nm | 32 nm |
| Core frequency std/max, GHz | 3,0/3,5 | 4,0/4,3 |
| Number of cores/threads | 6/6 | 4/8 |
| L1 cache (total), I/D, KB | 384/384 | 256/128 |
| L2 cache, KB | 6×512 | 4×2048 |
| L3 cache, MiB | 6 | 8 |
| RAM | 2×DDR3-1333 | 2×DDR3-1866 |
| TDP, W | 125 | 125 |
| Graphic arts | - | - |
| Qty EU | - | - |
| Frequency std/max, MHz | - | - |
| Price | - | T-11149970 |
There will be two main characters. The FX-8370 processor is relatively new - it appeared at the end of 2014, but it differs from the FX-8350 (the first-born of the Vishera family) only in the turbo mode clock frequency. Note that formally the top representatives of the family are the FX-9370 and FX-9590, but the latter exist only formally: a TDP of 220 W not only scares off many in itself, but also leads to compatibility problems with many motherboards, and also a thoughtful approach to choosing a cooling system. Well, if all this doesn’t scare you, then don’t forget that any processors of the FX family have fully unlocked multipliers, allowing arbitrarily fine tuning - including frequency. This, by the way, is another reason that the platform still has a certain popularity among those users who do not care about the result - the main thing is the process itself. Which in this case is also facilitated by the huge crystal of the processor produced using the 32 nm process technology - it is very easy to provide heat dissipation for it (sometimes disadvantages can become advantages). Moreover, equipping “boxed” processors with updated coolers allows one to count on good results even in this option, which may also turn out to be cheaper than the “traditional” approach with an OEM processor and some kind of “super cooler”. Overall, for those on a budget enthusiast The platform is interesting, despite its archaic nature.
But since testing this platform is still an excursion into history, we decided to use a new method (including studying energy consumption issues) to test an even older processor belonging to the Phenom II X6 family. Before the release of the first FX in 2011, it was the top product in the company’s range. Moreover, this is forever the best solution for old boards with “regular” AM3 and even AM2+. Moreover, as our tests showed, for processors of the Phenom II family the use of DDR3 is not so necessary, so we will not be surprised if such systems continue to be used somewhere (after all, even Pentium D owners regularly run through the Conference - until now :)). The top-end 1100T would have been the best choice for us, but we couldn’t find one, and the existing 1075T, alas, is not the Black Edition, so it doesn’t correctly turn into an older model. However, even with the possibility of overclocking with a multiplier, it is still unknown how correct this is from the point of view of measuring power consumption, and the line itself is so old (2010!) that, as it seems to us, there is no big difference anymore - testing 1100T or 1075T . Therefore, there will be a second one - since it already exists.
| CPU | AMD Athlon X4 880K | Intel Core i5-6400 | Intel Core i7-880 | Intel Core i7-3770 |
| Kernel name | Godavari | Skylake | Lynnfield | Ivy Bridge |
| Production technology | 28 nm | 14 nm | 45 nm | 22 nm |
| Core frequency std/max, GHz | 4,0/4,2 | 2,7/3,3 | 3,06/3,73 | 3,4/3,9 |
| Number of cores/threads | 2/4 | 4/4 | 4/8 | 4/8 |
| L1 cache (total), I/D, KB | 192/64 | 128/128 | 128/128 | 128/128 |
| L2 cache, KB | 2×2048 | 4×256 | 4×256 | 4×256 |
| L3 cache, MiB | - | 6 | 8 | 8 |
| RAM | 2×DDR3-2133 | 2×DDR3-1600 / 2×DDR4-2133 | 2×DDR3-1333 | 2×DDR3-1600 |
| TDP, W | 95 | 65 | 95 | 77 |
| Graphic arts | - | HDG 530 | - | HDG 4000 |
| Qty EU | - | 24 | - | 16 |
| Frequency std/max, MHz | - | 350/950 | - | 650/1150 |
| Price | T-13582517 | T-12873939 | - | T-7959318 |
Who will we compare with? It’s not for nothing that we mentioned the Core i5-6400 above - the junior quad-core processor of the modern Intel line directly competes in price with older AMD models (taking into account, of course, the remark about the video card). According to some readers, last time it was necessary to compare it with solutions for LGA1156, and not the dual-core Core i3-6320, which has a similar price and performance, but still. Therefore, today we will add to the list of test subjects the best processor for the mentioned platform, namely the Core i7-880, since the first FX were created, among other things, to compete with them. Unfortunately, however, they came out later than was necessary to ensure this - already at the time of processors for LGA1155. We have currently tested one of these models (albeit the third, and not the second generation Core) - we will add it to the list of test subjects to complete the picture. And, at the same time, the fastest Athlon X4 for FM2+ - for mass appeal. Moreover, for fans of AMD products, these are also, to some extent, direct competitors: the FX-8370 is certainly “cooler,” but it is also more expensive. Plus, it’s an archaic platform. And among those tested, let us remind you, there is also a Phenom II X6 1075T, so it will be interesting to see how six, but old, cores compare with modern, but two modules. It is clear that four is more interesting, but the transition from Phenom II (not necessarily six-core) will be simple and inexpensive only if you have a board with AM3+. If there is only AM2+, then change everything anyway. But if such a board, for example, has some kind of Athlon II installed, the performance of which is already not enough, the question of finding a Phenom II on the secondary market or changing the platform is not at all idle.
As for other testing conditions, all subjects worked on a system with a discrete graphics card based on Radeon R9 380 and 16 GB of RAM. The type and frequency of the latter were the maximum supported processors - for all, with the exception of the Phenom II X6 1075T, which we tested with DDR3-1600, which does not cause problems (however, it also has almost no effect on performance).
Testing methodology
The technique is described in detail in a separate article. Let us briefly recall here that it is based on the following four pillars:
- Methodology for measuring power consumption when testing processors
- Methodology for monitoring power, temperature and processor load during testing
And detailed results of all tests are available in the form of a complete table with results (in Microsoft Excel 97-2003 format). In our articles, we use already processed data. This especially applies to application tests, where everything is normalized relative to the reference system (like last year, a laptop based on a Core i5-3317U with 4 GB of memory and a 128 GB SSD) and grouped by areas of application of the computer.
iXBT Application Benchmark 2016
As we can see, if modular architecture had appeared in 2010, its “life” would have been significantly simplified: a couple of modules are no longer inferior to the Core i5 of that time, and four can convincingly surpass even the quad-core Core i7. But, unfortunately (or fortunately), in 2011, when developing processors for LGA1155, Intel managed to significantly improve all the characteristics of its products, and so dramatically that since then such “feats” have not been observed for five years. As a result, the older FX had to be positioned not in the segment between i5 and i7, but at the level of the former. So their price is quite consistent with their performance, but nothing more. Moreover, it is clearly noticeable that the company did not have any other options - the transfer of Phenom to a more sophisticated production process was unlikely to significantly “spur them on”: in order to bypass six old cores, two modules are often enough, not three or four .
Especially when the software cannot always fully utilize a large number of computation threads, but is demanding of their quality - including support for modern instruction sets, etc. As a result, even the older FX are now lagging behind the younger Core i5, but it could have been worse - as Phenom showed us. Actually, as has already been said more than once, intensive architectural improvements usually have their effect not at all in the generations of processors in which they are implemented. But the further you go, the more important they become.
But here, nothing matters: there would be one fast flow. In such conditions (which is no secret), AMD processors have a hard time, but it’s easy to see that they had a chance to be the fastest on the market in 2010.
But in this case, there was no hypothetical. However, judging by the small difference between FX and Phenom (and not even the older one), it is clear that no one worked on optimizing such work scenarios at all: the performance was still not bad for those times.
As we have written more than once, relatively old integer code is the best that AMD modular processors can encounter in life. And it is clearly noticeable that, in general, they were largely developed for such applications: after all, the six-core Phenom II in 2010 could no longer compete with the quad-core Core i7 in such tasks, but for the four-module FX this was a feasible task. Unfortunately, at the end of 2011 (when the first processors of this family finally appeared physically) it became significantly more complicated.
Actually, the aria is from the same opera - as we have already noted, data packaging is similar in logic to text recognition. And according to the results too.
The clear outsider here is the Core i7-880, but simply because LGA1156 only supported SATA300. As we have already noted, for the difference to become noticeable at all, you need to use a fast SSD, which was difficult in those years. Now it’s no longer there, so it’s not much, but it has an effect. But AMD endowed its chipsets with support for the new interface even then, so in this case there were no rough edges at all.
As we have already mentioned more than once, various SMT technologies are “alien” to the program, but the number of “hardware” cores and their quality are relevant, which, for example, results in the fact that the modern junior Core i5 is faster than the old Core i7. And not even that fundamentally old - not only the 880, but also the 3770 was left behind. The first one also lagged behind the FX-8370, which is a common thing. But the six very old architectural cores in the Phenom II... They can overtake two modules of modern AMD processors, but with great difficulty - they can’t cope with three.
What do we have in general? The FX-8370 is about one and a half times faster than the Athlon X4 880K - a normal increase due to doubling the cores and the addition of third-level cache. But, unfortunately, this is not enough to compete with modern Intel processors, which equal prices do not fully compensate for. If only because the buyer of the Core i5-6400 can do without a discrete video card, but those who choose FX cannot. But if he plans to buy it anyway, the result is something close to parity - until now. True, prices are not its cause, but rather a consequence - it’s not for nothing that they have been declining over the years.
Why the situation turned out to be exactly like this can, in principle, be guessed based on the results. We don’t know exactly what years the bulk of the development of modular architecture took place, but we can assume that it was earlier than 2011 - after all, it was then (and after several delays) that the first processors for AM3+ already began to be sold. If this had happened a year earlier, when quad-core processors such as the Core i7-870/880 cost around three to five hundred dollars, the effect would have been noticeable - comparable to the release of the first Athlon. At the same time, to replace the quad-core Phenom or Core 2 Quad, two-module processors (including models with an integrated GPU) would be suitable, and three-module processors would look normal against the background of the Phenom II X6 (or instead of them) and Core i5. But in the end, the processors had to compete not with models for LGA1366 or LGA1156, but with the brand new (at that time) LGA1155, which is still not bad compared to newer Intel platforms. Which, however, have become even better, while the old FX still live on the market without major changes since 2012. Which has to be compensated for by prices, which at first were between the Core i5 and i7, then at the level of the older i5s, then the middle ones, and now the younger ones. Since the consumer characteristics of processors approximately correspond to such prices. Only Core i5 are very cheap processors to produce, and FX are expensive. So it’s time to break this vicious circle - the further it goes, the more difficult it becomes. Let's hope everything works out this year.
Energy consumption and energy efficiency
However, as far as power consumption is concerned, not everything was smooth with it in those years, and from the point of view of modern times, 200 W is very frightening. It is clear that this includes what “passed” through the board to power the video card - but it is the same for everyone. But the “gluttony” of the three-chip platform is in its purest form its peculiarity and “hello from the 2000s”: modern ones are much more economical. However, if you pay attention to the actual needs of the processor, then there too it reached 140 W, i.e. for AMD exceeding the TDP level is just a common thing (although some are still trying to blame Intel for this in the old fashioned way). But the Phenom II X6 looks better at first glance. But don’t forget that this is not at all the top model in the line, firstly, and that power consumption only makes sense in conjunction with performance, secondly.
And from this point of view, modular architecture was a clear step forward. We also note that FX behave better than Athlon - if only because the shared third-level cache (which is not present in processors for FM2/FM2+) has a positive effect on performance, but is not too power-hungry. True, it takes up a lot of space, which is why its implementation in processors with integrated GPUs turned out to be impossible. But in general, it becomes clear why the company did not make an FX shrink to the 28 nm process technology: in the APU it made it possible to increase graphics power, but would not give anything or almost nothing to the processor cores. And the alarm bell sounded five years ago: it was possible to achieve the performance level of 45-nanometer Intel processors, but at the cost of excessive power consumption (who said "NetBurst"?). And then the situation only got worse.
iXBT Game Benchmark 2016
Can these processors work well in a gaming computer? Generally speaking, yes - after all, the main load falls on the video card. But how many of the latter’s capabilities will be “lost” due to the processor? This question, by the way, is especially important for users of boards with AM2+ or “regular” AM3, where the Phenom II X4/X6 is the best available without changing the platform, and the once popular Athlon II with the so-called. modern times no longer “pull” anything at all.
A case where “single-threaded performance” is critical, which puts all AMD processors in an awkward position. The performance of even the (already) inexpensive R9 380 is “restrained” by all test subjects. But you can play comfortably on everyone.
And here everyone copes close to the maximum possible. And, by the way, pay attention - the old Phenom IIs are noticeably better than the new Athlons.
It’s worse here, however, again, the Phenom II is no worse than any Core 2 Quad or Core i5/i7. And FX are already capable of “butting heads” with newer i5/i7.
But in the newer game in the series, Phenom II is on an equal footing (already on an equal footing) only with the Athlon. Which, however, is quite enough for practical use - but it could be better. At least at the FX level, which in FHD already allows the selected video card to “give it its all.”
But here everything is approximately the same - there are differences only in the mode with a reduced resolution. And what’s funny is that they are more in favor of AM3+ than vice versa.
When everything is determined by the video card, processors from five or six years ago are also good. The most powerful of them, of course. But a little later they started to cost very little.
The FX is performing well, but the Phenom II's time is unfortunately up. On the other hand, if such a processor already exists, then it is not at all necessary to change it in a gaming computer - there will be no noticeable effect. It would be better to install an even more powerful video card.
Thief clearly “votes” for powerful platforms - and considers only the modern Intel range as such. On the one hand. On the other hand, one cannot say that something does not work at all. There are about 40 frames - if you want to save on changing platforms, this can be considered sufficient.
In this pair, the frame rate already depends on processor performance. But, actually, so what? The absolute results of all subjects are more than sufficient for a comfortable game. So, in the end, we come to the conclusion that for an inexpensive gaming computer, “the old oak will still make noise.” Naturally, if it already exists (or can be purchased very cheaply). And, naturally, given the fact that even for budget modern video cards such a processor can turn out to be a “limiting factor.” Not in the sense that you won’t be able to play, but in the sense that the performance will still be lower than potentially possible. But this will still not always happen.
Total
In principle, we didn’t get anything unusual in the end - the platform is formally “live” and relevant, but in fact it has not been updated for a long time. Whether updates are needed or not is a debatable question. Some people, for example, don’t like the fact that Intel is constantly upgrading something without changing the performance of processors. On the other hand, for the same money, productivity is constantly (albeit slowly) growing, and the need to change platforms is primarily due to their functionality. As a result, some top-end motherboard from five years ago, for example, looks dull and pale compared to even the most budget modern offerings, priced five times lower. If you don’t touch anything, then the performance will not increase, and the rest of the computer’s characteristics will remain typical for five to seven years ago. Another question is that in many cases this is quite enough, and in the case of a reasonable pricing policy, “historical” platforms turn out to be quite suitable for practical use until they physically disappear from service, which will happen, obviously, even later than the end of sales. 
Introduction
I’ll start on a sad note, I’m a supporter of AMD processors, primarily for their reasonable prices, but unfortunately this time I felt disappointed and it’s really significant.
The “Red Army” from the AMD camp has been trying to catch up with Intel for a long time, managing to release its new products as opposed to Intel processors that have already become old.
This was almost always the case, with the exception of the successful Athlon 64 processor in various variants. One can give hundreds of examples of this multi-level confrontation. Take, for example, the example of the Core 2 Quad that AMD provided us in response - the Phenom 8xxx\9xxx series are not at all competitive. Further, Intel releases a new Nehalem architecture with the not-so-successful i7 920, AMD, in turn, upgrades Phenom to the second, thereby catching up with the Core 2 Quad and, in part, the new i5 line of the 6xx\7xx series, thereby confirming its lagging step.
With the release of a new architecture codenamed Bulldozer, as AMD engineers promised us, the situation should change for the better; perhaps a loophole will appear to get to the processor throne.
Buyers' interest was diluted by some rather interesting features of the new central processors.
To begin with, AMD is changing the names of processors, now the FX suffix appears again, immediately recalling the times of Athlon 64" and FX 64. Further, the line will contain 4, 6, 8 core processors, which is quite relevant today.
At the announcement date, the following models were presented:
FX 8150- the oldest processor in the line of 8-core AMD processors, which managed to “light up” several times with its excellent overclocking potential and became a Guinness World Record holder.
FX 6100- 6-core brother, somewhat reminiscent of the Phenom II X6;
FX 4100- 4-core competitor to the Intel mill;
I would like to dwell in detail on the FX4100, which came into my hands in OEM packaging.
There are several reasons, the relative availability of this processor in stores, the price is at the level of the Phenom II 965, while we must get a processor on a new technical process, with high potential overclocking and greater performance per watt, so let’s check that.
AMD FX 4100 The AMD FX-4100 quad-core processor belongs to the Zambezi family and is manufactured according to 32 nm process standards. The nominal frequency is 3.6 GHz; a feature of the new architecture is support for a wider set of instructions, as evidenced by the cpu-z utility. 
The new product also supports Turbo Core 2.0 technology (which has been available since the Phenom II X6 series), which is to some extent an analogue of Turbo Boost from Intel. At the same time, the user receives a small increase in performance by increasing the clock frequency of half of the CPU modules when solving “heavy” tasks.
The architecture of the Bulldozer family of processors is almost completely different from its predecessor. The AMD FX-4100 cache is distributed as follows. Level 1 cache: 16 KB per core is allocated for data with four associative channels, while there is 64 KB for instructions per module with 2 associative channels. Level 2 cache: 2 MB per processor module with 16 associative channels. The Level 3 cache is common to the entire processor and is 8 MB with 64 associative channels. 
To evaluate the increased performance, I would like to test the FX4100 with a previous generation processor, namely the Phenom II X4 955.
Test bench
AMD FX
*Motherboard: Asus M5A78L LE (Bios version 0503)
*CPU: AMD FX 4100 3.6GHz, 8MB
AMD Phenom II
*Motherboard: Gigabyte GA-MA790FXT-UD5P (Bios version F8n)
*CPU: AMD Phenom II X4 955 3.2GHz, 6MB
Other components:
*Video card: Asus GTX560 (336core\810\1620\4008)
*RAM: Samsung 2x4GB DDR3 1600MHz
*SSD: Corsair Force F60
*HDD: WD 320Gb
*Power unit: FSP Everest 600W 85Plus
Overclocking
FX4100- the multiplier was fixed at x18, which gives a nominal 3600 MHz.
Phenom II 955- I succumbed to slight overclocking; also, using an unlocked multiplier, in order not to disrupt testing, the mark was raised to x18, which amounted to an increase in processor frequency to the level of FX4100.
Test applications
FPS measurements were carried out by the Fraps program version 3.2.3. All tests were carried out at a resolution of 1280x1024.
The graphs show both average fps values and minimum/maximum values.
Alan Wake- Dx9; settings - High, AA8x;
Battlefield 3- Dx11; installations - Ultra, HBAO16x;
Crysis 2- Dx11; installations - Extreme;
Mafia 2- Dx9; settings - High, AA16x;
Metro 2033- Dx10; settings - Very High, AA4x;
Resident Evil 5- Dx10; settings - High, AA16x;
Test results, conclusions
conclusions
So what did AMD end up with? the same thing, but only under the FX brand + taken from the Turbo Core from Phenom II X6 and various multimedia instructions. Of course, overclocking, yes it will be more... yes it will be faster... but the output is a hot processor that eats a lot... obviously those who are on the Phenom II should not move to the "new" platform, those who will assemble a new computer on AMD - it’s still worth taking a closer look at the FX, as it’s new and cheaper than the same 955, but it’s not inferior to it.
Personally, I see only a failure, for the reason - a new architecture, at the level of the old one... but with the hope that AMD will improve the processors, as was the case with the Phenom 8xxx\9xxx series.
In conclusion, I would like to add about one nuance related to the launch of this processor; personally, on the Asus M5A78L LE motherboard it was not possible to do this the first time, the problem was in the old version of the BIOS, in order to bypass this “sin” you must have an AM3 socket processor at hand.
PS. I’m slowly getting used to the new style of personal pages, a test of the pen, so to speak, comments can be left here, starting from here.
How important is L3 cache for AMD processors?
Indeed, it makes sense to equip multi-core processors with dedicated memory that will be shared by all available cores. In this role, a fast third-level (L3) cache can significantly speed up access to data that is requested most often. Then the cores, if possible, will not have to access slow main memory (RAM).
At least in theory. Recently AMD announced the Athlon II X4 processor, which is a Phenom II X4 model without L3 cache, hinting that it is not that necessary. We decided to directly compare two processors (with and without L3 cache) to test how the cache affects performance.
How does the cache work?
Before we dive into the tests, it's important to understand some basics. The principle of how the cache works is quite simple. The cache buffers data as close to the processing cores of the processor as possible to reduce CPU requests to more distant and slow memory. On modern desktop platforms, the cache hierarchy includes as many as three levels that precede access to RAM. Moreover, caches of the second and, in particular, third levels serve not only to buffer data. Their purpose is to prevent the processor bus from becoming overloaded when the cores need to exchange information.
Hits and misses
The effectiveness of cache architectures is measured by hit rate. Data requests that can be satisfied by the cache are considered hits. If this cache does not contain the necessary data, then the request is passed further along the memory pipeline, and a miss is counted. Of course, misses lead to more time required to obtain information. As a result, “bubbles” (idles) and delays appear in the computing pipeline. Hits, on the contrary, allow you to maintain maximum performance.
Cache entry, exclusivity, coherence
Replacement policies dictate how space is freed up in the cache for new entries. Because data written to the cache must eventually appear in main memory, systems may do so at the same time as writing to the cache (write-through), or may mark the data areas as "dirty" (write-back) and write to memory. when it is evicted from the cache.
Data in several cache levels can be stored exclusively, that is, without redundancy. Then you won't find the same data lines in two different cache hierarchies. Or caches can work inclusively, that is, the lower cache levels are guaranteed to contain data present in the upper cache levels (closer to the processor core). AMD Phenom uses an exclusive L3 cache, while Intel follows an inclusive cache strategy. Coherency protocols ensure the integrity and freshness of data across different cores, cache levels, and even processors.
Cache size
A larger cache can hold more data, but tends to increase latency. In addition, a large cache consumes a considerable number of processor transistors, so it is important to find a balance between the transistor budget, die size, power consumption and performance/latency.
Associativity
Entries in RAM can be directly mapped to the cache, that is, there is only one cache position for a copy of data from RAM, or they can be n-way associative, that is, there are n possible locations in the cache where this data may be stored. Higher degrees of associativity (up to fully associative caches) provide greater caching flexibility because existing data in the cache does not need to be rewritten. In other words, a high n-degree of associativity guarantees a higher hit rate, but it also increases latency because it takes more time to check all those associations for a hit. Typically, the highest degree of association is reasonable for the last level of caching, since the maximum capacity is available there, and searching for data outside of this cache will result in the processor accessing slow RAM.
Here are some examples: Core i5 and i7 use 32 KB of L1 cache with 8-way associativity for data and 32 KB of L1 cache with 4-way associativity for instructions. It's understandable that Intel wants instructions to be available faster and the L1 data cache to have a maximum hit rate. The L2 cache on Intel processors has 8-way associativity, and the Intel L3 cache is even smarter, since it implements 16-way associativity to maximize hits.
However, AMD is following a different strategy with the Phenom II X4 processors, which uses a 2-way associative L1 cache to reduce latency. To compensate for possible misses, the cache capacity was doubled: 64 KB for data and 64 KB for instructions. The L2 cache has 8-way associativity, like the Intel design, but AMD's L3 cache operates with 48-way associativity. But the decision to choose one cache architecture over another cannot be assessed without considering the entire CPU architecture. It is quite natural that test results have practical significance, and our goal was precisely a practical test of this entire complex multi-level caching structure.
Every modern processor has a dedicated cache that stores processor instructions and data, ready for use almost instantly. This level is commonly referred to as Level 1 or L1 cache and was first introduced in the 486DX processors. Recently, AMD processors have become standard with 64 KB L1 cache per core (for data and instructions), and Intel processors use 32 KB L1 cache per core (also for data and instructions)
L1 cache first appeared on the 486DX processors, after which it became an integral feature of all modern CPUs.
Second-level cache (L2) appeared on all processors after the release of the Pentium III, although the first implementations of it on packaging were in the Pentium Pro processor (but not on-chip). Modern processors are equipped with up to 6 MB of on-chip L2 cache. As a rule, this volume is divided between two cores on an Intel Core 2 Duo processor, for example. Typical L2 configurations provide 512 KB or 1 MB of cache per core. Processors with a smaller L2 cache tend to be at the lower price level. Below is a diagram of early L2 cache implementations.
The Pentium Pro had the L2 cache in the processor packaging. In subsequent generations of Pentium III and Athlon, the L2 cache was implemented through separate SRAM chips, which was very common at that time (1998, 1999).
The subsequent announcement of a process technology up to 180 nm allowed manufacturers to finally integrate L2 cache on the processor die.
The first dual-core processors simply used existing designs that included two dies per package. AMD introduced a dual-core processor on a monolithic chip, added a memory controller and a switch, and Intel simply assembled two single-core chips in one package for its first dual-core processor.
For the first time, the L2 cache began to be shared between two computing cores on Core 2 Duo processors. AMD went further and created its first quad-core Phenom from scratch, and Intel again used a pair of dies, this time two dual-core Core 2 dies, for its first quad-core processor to reduce costs.
The third level cache has existed since the early days of the Alpha 21165 processor (96 KB, processors introduced in 1995) or IBM Power 4 (256 KB, 2001). However, in x86-based architectures, the L3 cache first appeared with the Intel Itanium 2, Pentium 4 Extreme (Gallatin, both processors in 2003) and Xeon MP (2006) models.
Early implementations simply provided another level in the cache hierarchy, although modern architectures use the L3 cache as a large, shared buffer for inter-core data transfer in multi-core processors. This is emphasized by the high n-degree of associativity. It is better to look for data a little longer in the cache than to end up with a situation where several cores are using very slow access to main RAM. AMD first introduced L3 cache on a desktop processor with the already mentioned Phenom line. The 65 nm Phenom X4 contained 2 MB of shared L3 cache, and the modern 45 nm Phenom II X4 already has 6 MB of shared L3 cache. Intel Core i7 and i5 processors use 8 MB of L3 cache.
Modern quad-core processors have dedicated L1 and L2 caches for each core, as well as a large L3 cache shared by all cores. The shared L3 cache also allows for the exchange of data that the cores can work on in parallel.
Our comparison involved two different AMD processors, which will help compare the benefits of the additional L3 cache in a quad-core processor.
Click on the picture to enlarge.
On one side we had the new AMD Athlon II X4 620, AMD's entry-level quad-core processor. By the way, the Athlon II X4 620 was the first quad-core processor available for $100 (unfortunately, not in Russia), so we get a new level of performance for that price. However, we must not forget that the impressive performance of the 620 only applies to serious multi-threaded applications, and even then not always, since the Athlon II X4 is completely devoid of L3 cache. For comparison, we took the Phenom II X4 965 processor.
Click on the picture to enlarge.
The positioning of the two products is completely different. The Phenom II is AMD's current leader in the top-of-the-line Black Edition line, while the "junior" Athlon II X4 is aimed at the entry-level market.
However, the processors are very similar in architecture. The Athlon II X4 cores, including their L1 and L2 cache, are identical to the Phenom cores. AMD didn't even change the cache associativity. The only real change is that AMD has disabled the Athlon II X4 cache on processors where the L3 cache had validation issues. (This is only true for the early Athlon II X4. In the future, more and more processors will be based on a completely different and more cost-effective silicon.)
We were able to make a 1:1 comparison by lowering the Phenom II X4's clock speed from 3.4 GHz to just 2.6 GHz - exactly the stock clock speed of the Athlon II X4 620.
Test configuration
| Performance Test Hardware | |
| Motherboard (Socket AM3) | Gigabyte MA790FXT-UD5P (Rev. 1.0), chipset: AMD 790GX, SB750, BIOS: 5c (04/01/2009) |
| DDR3 memory (two channels) | 2 x 2 GB DDR3-1600 (Corsair CM3X2G1600C9DHX) 2 x 1 GB DDR3-1600 (Crucial BL12864BA1608.8SFB) in DDR3-1066 mode |
| General hardware | |
| CPU AMD I | AMD Phenom II X4 965 (45 nm, 3.4 GHz, 4 x 512 KB L2 cache and 6 MB L3 cache, TDP 140 W, Rev. C2) |
| CPU AMD II | AMD Athlon II X4 620 (45 nm, 2.6 GHz, 4 x 512 KB L2 cache, TDP 95 W, Rev. C2) |
| Video card | Zotac GeForce GTX 260², GPU: GeForce GTX 260 (576 MHz), video memory: 896 MB DDR3 (1998 MHz), stream processors: 216, shader frequency: 1242 MHz |
| HDD | Western Digital VelociRaptor, 300 GB (WD3000HLFS), 10,000 rpm, SATA/300, 16 MB cache |
| Blu-ray drive | LG GGW-H20L, SATA/150 |
| power unit | PC Power & Cooling, Silencer 750EPS12V 750 W |
| System software and drivers | |
| operating system | Windows Vista Enterprise Version 6.0 x64, Service Pack 2 (Build 6000) |
| AMD Chipset Drivers | Catalyst Control Center 9.4 |
Tests and settings
| Far Cry 2 | Version: 1.0.1 Far Cry 2 Benchmark Tool Video Mode: 1280x800 Direct3D 9 Overall Quality: Medium Bloom activated HDR off Demo: Ranch Small |
| GTA IV | Version: 1.0.3 Video Mode: 1280x1024 - 1280x1024 - Aspect Ratio: Auto - All options: Medium - View Distance: 30 - Detail Distance: 100 - Vehicle Density: 100 - Shadow Density: 16 - Definition: On - Vsync: Off In-game Benchmark |
| Left 4 Dead | Version: 1.0.0.5 Video Mode: 1280x800 Game Settings - Anti Aliasing none - Filtering Trilinear - Wait for vertical sync disabled - Shader Detail Medium -Effect Detail Medium - Model/Texture Detail Medium Demo: THG Demo 1 |
| Audio and video encoding | |
| iTunes | Version: 8.1.0.52 Audio CD ("Terminator II" SE), 53 min. Convert to AAC audio format |
| Lame MP3 | Version 3.98 Audio CD "Terminator II SE", 53 min convert WAV to MP3 audio format Command: -b 160 --nores (160 Kbps) |
| TMPEG 4.6 | Version: 4.6.3.268 Video: Terminator 2 SE DVD (720x576, 16:9) 5 Minutes Audio: Dolby Digital, 48000 Hz, 6-Channel, English Advanced Acoustic Engine MP3 Encoder (160 Kbps, 44.1 kHz) |
| DivX 6.8.5 | Version: 6.8.5 == Main Menu == default == Codec Menu == Encoding mode: Insane Quality Enhanced multithreading Enabled using SSE4 Quarter-pixel search == Video Menu == Quantization: MPEG-2 |
| XviD 1.2.1 | Version: 1.2.1 Other Options/Encoder Menu - Display encoding status = off |
| Main concept Reference 1.6.1 | Version: 1.6.1 MPEG2 to MPEG2 (H.264) MainConcept H.264/AVC Codec 28 sec HDTV 1920x1080 (MPEG2) Audio: MPEG2 (44.1 kHz, 2 Channel, 16 Bit, 224 kbps) Codec: H.264 Mode: PAL (25 FPS) Profile: Settings for eight threads |
| Adobe Premiere pro CS4 | Version: 4.0 WMV 1920x1080 (39 sec) Export: Adobe Media Encoder == Video == H.264 Blu-ray 1440x1080i 25 High Quality Encoding Passes: one Bitrate Mode: VBR Frame: 1440x1080 Frame Rate: 25 == Audio == PCM Audio, 48 kHz, Stereo Encoding Passes: one |
| Grisoft AVG Anti Virus 8 | Version: 8.5.287 Virus base: 270.12.16/2094 Benchmark Scan: some compressed ZIP and RAR archives |
| Winrar 3.9 | Version 3.90 x64 BETA 1 Compression = Best Benchmark: THG-Workload |
| Winzip 12 | Version 12.0 (8252) WinZIP Commandline Version 3 Compression = Best Dictionary = 4096 KB Benchmark: THG-Workload |
| Autodesk 3D Studio Max 2009 | Version: 9 x64 Rendering Dragon Image Resolution: 1920 x 1280 (frame 1-5) |
| Adobe Photoshop CS4 (64-Bit) | Version: 11 Filtering a 16MB TIF (15000x7266) Filters: Radial Blur (Amount: 10; Method: zoom; Quality: good) Shape Blur (Radius: 46 px; custom shape: Trademark symbol) Median (Radius: 1px) Polar Coordinates (Rectangular to Polar) |
| Adobe Acrobat 9 professional | Version: 9.0.0 (Extended) == Printing Preferred Menu == Default Settings: Standard == Adobe PDF Security - Edit Menu == Encrypt all documents (128 bit RC4) Open Password: 123 Permissions Password: 321 |
| Microsoft PowerPoint 2007 | Version: 2007 SP2 PPT to PDF Powerpoint Document (115 Pages) Adobe PDF-Printer |
| Deep Fritz 11 | Version: 11 Fritz Chess Benchmark Version 4.2 |
| Synthetic tests | |
| Version: 1.02 Options: Performance Graphics Test 1 Graphics Test 2 CPU Test 1 CPU Test 2 |
|
| PCMark Vantage | Version: 1.00 PCMark Benchmark Memories Benchmark |
| SiSoftware Sandra 2009 | Version: 2009 SP3 Processor Arithmetic, Cryptography, Memory Bandwith Benchmark Results: Sandra 2009, PCMark Vantage |
Performance Notes
We typically measure power consumption at idle and under maximum load, and then evaluate system efficiency by tracking the energy required to run a specific load (typically a PCMark Vantage run). This allows us to calculate efficiency as performance per watt. However, in this case we had to take several steps that are not typical for real conditions. We lowered the Phenom's clock speed and had to disable Cool'n'Quiet to allow the Phenom II X4 965 to run at 2.6GHz instead of the stock 3.4GHz. Since the slowest Phenom II X4 starts at 3.0 GHz, it's unlikely that anyone will run the processor at low clock speeds. Additionally, we downclocked the Phenom II memory to DDR3-1066 to match AMD's specifications for the Athlon II X4.
We then got a noticeable power advantage for a processor without L3 cache. The cache itself takes up about a third of all processor transistors. This becomes obvious from the energy consumption data. AMD claims a TDP of 95 to 140 W for the Phenom II, while the Athlon II X4 operates at 95 W. Our test system with the 3.4 GHz Phenom II X4 965 processor reached a peak power consumption of 226 W, while the 2.6 GHz Athlon II X4 peaked at 170 W.
In idle mode, we see very similar results. We got 84 W for the Athlon II X4 620 and 85 W for the same system with the Phenom II X4 965 processor. In these cases, Cool'n'Quiet technology was active, so both processors reduced their frequency to 800 MHz and also lowered the voltage. Since most of the processor units are idle and turned off, the idle power consumption of our two processors is very similar.
Test results
We see a 5% advantage in the 3DMark Vantage CPU test, but in the overall score and GPU test we see no gain at all. Let's see what the gaming performance will be like.
Frame rates increased by 8% in Far Cry at medium detail settings when we swapped an entry-level quad-core Athlon II X4 processor for a similarly clocked Phenom II X4.
The advantage in GTA IV 5.7% is also not very much. The L3 cache has a fairly weak impact on performance.
In Left 4 Dead the results are completely different, a processor with 6 MB of L3 cache gives almost 20% higher frame rates.
Creating a PDF using Adobe Acrobat 9 from a Microsoft PowerPoint document does not benefit much from the L3 cache.
The WinRAR archiver is very sensitive to memory performance, so it takes 16% less time to complete its work.
But WinZip was not so critical of the lack of L3 cache. The L3 cache was 9.2% faster.
The performance of filters in Photoshop CS4 benefits little from the presence of the L3 cache in the Phenom II. The difference of three seconds is tiny.
iTunes needs a higher clock speed to improve audio transcoding performance. Therefore, the tiny difference between processors with and without L3 cache did not come as a surprise to us.
Here the results are generally the same, which is not surprising.
DivX transcodes a movie from MPEG-2 format to the Phenom II X4 just a little faster.
Xvid encoding benefits a little more, although this operation takes significantly longer than converting an MPEG-2 video to the DivX format.
MainConcept gets performance from the number of cores and their clock speed. We don't see any noticeable benefit from having an L3 cache.
We decided to create a performance index that would take into account the results of all tests. Since CPU-intensive applications require the most performance, we weighted them at 50%, games at 25%, and PCMark Vantage and 3DMark Vantage at 12.5% each. We ended up with a 5.8% performance advantage for the Phenom II X4 over the Athlon II X4, or a 5.5% performance hit if you use the Phenom II X4 as a base. Of course, you may have different PC usage priorities, so it's important to mention the minimum and maximum differences. In some tests we benefited from the presence of an L3 cache of 20%, and in some tests the processors gave absolutely identical performance despite the presence/absence of an L3 cache. In general, it seems to us that it is better to focus on the difference in performance from 5% to 6%, which we calculated based on the results of all tests.
Conclusion
A comparison of prices and performance clearly suggests that “budget” users should not look at the Phenom II X4 at all. The Phenom II X4 945 processor (3.0 GHz) starts at $170 (), and the new Athlon II X4 processor for $100 () gives very similar performance, all other things being equal. AM2+ models of Phenom II X4 processors may sell for less, but they do not offer DDR3 memory support.
Overall, the main performance difference between the Athlon II X4 and Phenom II X4 is related to the clock speed. Simply increasing the clock speed of the Athlon II X4 by 200 MHz will allow it to match the performance of the Phenom II X4, despite the latter having 6 MB of L3 cache. Knowing this, you will probably understand why there will be no Athlon II processors on the market that will equal (or even exceed) the Phenom II in frequency.
Of course, you need to take into account different market segments, which we have rather blurred in our article. The Phenom II is a processor for the upper end of the mass market, which sells for $150 to $250, while the Athlon II X4 is aimed at the "budget" audience, who are willing to shell out no more than $100 for a CPU. In any case, it is clear that the Athlon II X4 provides excellent performance/price ratio, especially for those users who plan to overclock the processor.
Finally, it should be noted that L3 cache is required to achieve high levels of performance. At 2.6GHz CPU speeds this may not be as obvious, but at 3GHz and above we see the Phenom II's performance scale much better than the Athlon II X4.