Iris Pro 6200 vs. Radeon R7 vs. HD Graphics vs. Discrete Radeon R7 250X

The publication of our first article on desktop processors of the Broadwell family, among other things, caused a couple of fair comments regarding testing the graphics core in gaming applications. Indeed: there are tests, but for comparison, only the GPU HD Graphics 4600 was taken, with which everything is clear anyway. But how the success of the new "graphics top" Intel look against the background of AMD processors or inexpensive discrete video cards - from a practical point of view, the question is more important. Moreover, the C-series processors are more expensive than similar Haswell dollars by $ 100, and this is quite enough to purchase a Radeon R7 250X or something close, that is, not a very slow solution.

Today we will remove all the questions.

Testbed configuration

CPU	Intel Core i5-4690K	Intel Core i5-5675C	Intel Core i7-4770K	Intel Core i7-5775C
Kernel name	Haswell	Broadwell	Haswell	Broadwell
Prospect technology	22 nm	14 nm	22 nm	14 nm
Core frequency, GHz	3,5/3,9	3,1/3,6	3,5/3,9	3,3/3,7
# Of cores / threads	4/4	4/4	4/8	4/8
L1 cache (sum), I / D, KB	128/128	128/128	128/128	128/128
L2 cache, KB	4 × 256	4 × 256	4 × 256	4 × 256
L3 (L4) cache, MiB	6	4 (128)	8	6 (128)
RAM	2 × DDR3-1600	2 × DDR3-1600	2 × DDR3-1600	2 × DDR3-1600
TDP, W	88	65	84	65
Graphics	HDG 4600	IPG 6200	HDG 4600	IPG 6200
Number of EU	20	48	20	48
Std / max frequency, MHz	350/1200	300/1100	350/1250	300/1150
Price	N / A (0) T-10887398	N / A (0) T-12645002	$412() T-10384297	N / A (0) T-12645073

There will be two pairs of Intel processors - to clearly understand where the Core i7 has preferences over the Core i5, and where is one vanity of vanities and vexation of spirit... Comparison will go in gaming applications, of course, and with a discrete video card. However, we have already investigated this question, but there the i5 and i7 were of different frequencies, and today we have equalized them in this parameter. In principle, it would be possible to take Broadwell of the same frequency, but it is only available in the form of Xeon, that is, not to say that it is a mass solution. So there will be no direct intersections - just both socket models for household use.

CPU	AMD A10-6800K	AMD A10-7850K
Kernel name	Richland	Kaveri
Prospect technology	32 nm	28 nm
Core frequency std / max, GHz	4,1/4,4	3,7/4,0
Number of cores (modules) / threads	2/4	2/4
L1 cache (sum), I / D, KB	128/64	192/64
L2 cache, KB	2 × 2048	2 × 2048
L3 cache, MiB	-	-
RAM	2 × DDR3-2133	2 × DDR3-2133
TDP, W	100	95
Graphics	Radeon HD 8670D	Radeon R7
Number of GPs	384	512
Std / max frequency, MHz	844	720
Price	$138() T-10387700	$162() T-10674781

We decided to take two AMD processors so that it wouldn't be boring. Besides, it is also interesting to evaluate the progress of graphics here, and do not forget that the A10-6800K also has a twin brother in the form of Athlon X4 760K. And which of the "Athlons" to choose when using a discrete video card (760K or 860K) is an interesting question from a practical point of view. Moreover, the 760K will work on a motherboard with a "regular" FM2. Maybe the user is no longer satisfied with some old A6-5400K, and he decided to change the processor and add a discrete video card? It may well. So let's see if it makes sense to change the motherboard in this situation.

As for the other test conditions, they were equal, but not the same: the operating memory frequency was the maximum supported by the specifications, but they differ slightly. But its volume (8 GB) and system storage (Toshiba THNSNH256GMCT with a capacity of 256 GB) were the same for all subjects. All tests were carried out using the integrated video core (which all six processors have), and in conjunction with the discrete Radeon R7 250X.

Testing technique

Since we have already established that a specific video card has very little effect on programs from the iXBT Application Benchmark 2015 set, we limited ourselves to the iXBT Game Benchmark 2015 gaming technique. All results were obtained in 1920 × 1080 (Full HD) resolution at minimum quality settings and in 1366 × 768 at maximum settings. Why such a choice? Maximum settings at FHD resolution are too tough not only for integrated video adapters, but also for many inexpensive discrete solutions. But many people want to improve the quality - even at the cost of lowering the resolution. Moreover, the decrease is not always so radical - users still have old monitors up to those supporting a maximum of 1280x1024 pixels. So why not check the "low" modes. In addition, with settings for maximum quality, the specific share of the load on the GPU increases, and today we are just interested in GPUs. And even if they do not cope with the work, it will turn out to be a stress test that well demonstrates the actual capabilities of graphics.

Minimum high definition quality

As you can see, HD Graphics in Haswell does not cope with this task, you can already play on both A10s, but on the verge, and Broadwell leaves no doubt with the Iris Pro. But if we talk about using a discrete video card, then all processors are equal. The price of Athlon X4 is several times lower than that of any Core i7. The same state of affairs will be in other games with low requirements for processor performance, but high - for graphics.

But WoT, however, is the exact opposite of the one formulated above - here the graphics are needed insofar as. If only it did not interfere. HD Graphics 4600 is obviously not enough. The rest - enough so that adding a discrete video card does not increase the performance, and may even decrease.

Another processor-dependent game, which is sufficient for the HDG 4600 for the selected mode. However, faster graphics, even with a weak processor part, allows you to achieve better results. And the discrete video adapter shows that the L4 cache in some cases actually makes Broadwell-C a much faster solution than Haswell. However, there is little practical benefit from this - 200 or 300 frames is no longer important. Here, obviously, the quality needs to be improved, which we will do a little later.

The game is difficult for all systems, but first of all - video cards. As you can see, only Broadwell integrated graphics, and in the older modification (GT3e), generally allow you to play in this mode: Haswell GT2 traditionally lags behind twice, and the best AMD IGPs - one and a half times. However, when using an inexpensive discrete video card, everything suddenly becomes equal: both cheap Athlons (and disabling the graphics part in the A10 will transform processors in this way), and expensive Core i7.

In the previous version of Metro, the situation is similar. True, here the A10 is already approaching the threshold of playability, but without exaggeration, only Broadwell-C and the like are suitable. Discrete (even such a relatively weak one, like 250X), on the other hand, already depends on the performance of processors. Another question is that there will still be enough "athletes", and ten frames per second can be neglected.

Once again, Hitman is similar to Metro 2033 with minor variations. For example, here two A10s of different generations behave very differently even when using discrete, i.e. optimization in Kaveri is not an empty phrase. However, no matter how you optimize, Core i5 is much faster. As for the integrated solutions, here again only Broadwell-C are suitable without any exaggeration - the rest will have to reduce the resolution.

A very difficult game that even Iris Pro cannot handle! However, as we can see, here the 250X is enough without much margin - paired with slow processors, it is on the verge of playability.

As we have said more than once, in minimal mode, Tomb Raider works great on everything (or almost everything). However, the new Broadwells still have something to praise for, since they are not so far behind the budget, but discrete video card :)

In this game, no discretion can be found anywhere. Interestingly, the Iris Pro 6200 is, as usual, twice as fast as the HDG 4600, but AMD's solutions are already slightly ahead. Apparently, the main load is on shader and other units, and they cannot be accelerated using eDRAM. Let's see how this manifests itself with an increase in quality.

There are more or less new A10s, Broadwell-C is enough without a stretch, there is nothing to catch Haswell (except for the R-series, also equipped with the GT3e video core). But ... but it will be cheaper to put a discrete graphics card.

So what do we have in the minimum quality mode? Broadwell-C cope with almost all games in our set, except one. The performance of the Broadwell GT3e is about twice as high as that of the Haswell GT2, and the integrated graphics from AMD are one and a half times bypassed. But it is better, of course, to use a discrete video card if possible - it may even come out cheaper. And always at least not slower.

Low resolution but high quality

A discrete graphics card allows you to play even with an inexpensive processor, the integrated graphics are still unusable. None.

With great difficulty and straining Core i5-5675C got out at 30 FPS. A cheaper combination of Athlon X4 760K or 860K and R7 250X gains almost 40 without straining. Comments are unnecessary.

This is where the Iris Pro 6200 looks very good. Let the discrete video card be a little faster, but not significantly. Worse, it is not always possible to use it, so the emergence of powerful integrated video is a great boon for those in such an environment.

Low-end discrete cards are not enough, which means that you can forget about integrated solutions in practice. From the point of view of theory, it is curious that here they are quite close to each other, which is not surprising: when the main load falls on the GPU itself, no tricks in terms of memory work anymore help.

Even more pronounced than in the previous case. The only curious thing is that the HDG 4600 is faster than the Radeon HD 8670D. However, this is not practically significant.

Again, even a discrete card fails, and its gap from integrated solutions increases up to three to five times. With the minimum quality, recall, there were sometimes less than two. Those. the higher the requirements for the GPU, the greater the difference between the integrated and discrete versions of the latter. Which is more than expected, but not taken into account by everyone.

If you have a discrete video card, you can play, but an integrated one is not enough at all, and any one. A similar picture was at the minimum FHD settings, only here it became even clearer. But nothing surprising - in general, for this game, cards with a minimum level of Radeon R7 265 and higher are desirable. And there are not so few such games.

If at minimum settings this game is very gentle on the video system, then an increase in quality can bring to its knees and much more powerful solutions than we are considering today. Those. The room for maneuver is huge, but only the owners of discrete video cards can use it successfully.

Sleeping Dogs behaves in a similar way, only the advantages of a discrete solution are even more visible. But the benefits from eDRAM are fading away even more noticeably, since the matter does not even reach the texturing speed: the GPUs themselves are still too weak. But they are weak in different ways, so that the integrated Radeon R7 can even overtake the Iris Pro. In practice, however, this does not matter, since both are still too slow.

And one more similar case confirms the above hypothesis :)

In general, as we can see, attempts to use modes with high picture quality (even with a lower resolution) only on integrated graphics are usually doomed to fiasco.

Total

So what do we see? Low quality modes lend themselves well to modern integrated graphics. At least the best representatives of the latter. The idea with eDRAM is correct and logical - it helps to alleviate the lack of memory bandwidth. Actually, thanks to this, the solutions of the Iris Pro line become the fastest in their class. Not necessarily Broadwell - Haswell is not much worse, but such modifications of the latter are not installed in the socket, which imposes its own specifics.

But can a gamer be satisfied with low-quality modes? Perhaps not. In any case, if he is interested in modern games at all, at the minimum settings, "modernity" easily disappears, often resembling a picture of ten years ago. Especially if you remember the high cost of Intel processors with GT3e - for this money you can buy something simpler, but with a good discrete graphics card. AMD solutions are much more affordable, and with an increase in the quality of the picture, they "sag" in performance weaker, since the graphics processors themselves are still more powerful (and eDRAM cannot fix this), but ... But this does not fundamentally change anything - all the same, the final performance too low, so gamers do not have to rely heavily on the graphics capabilities of the AMD APU.

What awaits us in the near future? According to forecasts, Skylake line processors will eventually acquire graphics cores such as GT4e, where there will be more executive devices than before (in fact, GT with the usual numbers will "grow up", but much less noticeably, but the appearance of a new modification directly hints at radical changes) , and eDRAM. Moreover, support for DDR4 will increase the memory bandwidth - albeit not immediately, maybe. However, it does not follow from this that even such processors will cope with high-quality game modes from our method even at low resolutions - for this, performance needs to be increased 3-5 times, which is unlikely to work. They will be able to overtake junior discrete video cards more often, but mostly only where there is either “enough” or “still fundamentally not enough”, so the fact of higher or lower performance is not very important in itself.

In general, progress in the field of integrated graphics is clearly visible. But so far, from the point of view of a gamer, it is still insufficient to fundamentally change the state of affairs. A full-fledged gaming computer, as before, must have a discrete video card, and more expensive than a processor. Which, by the way, makes Broadwell-C a bad gaming solution anyway (even with a discrete video card) - you can make sure that the advantages of L4 cache are not great enough to justify the higher prices. If instead of 250X we used 290X (for example), they would be more noticeable, but all the same, it's better to spend this money on a video card and spend it - the return will be much higher. In addition, the limited thermal package also interferes - the Core i5 is often slightly faster than the Core i7, which operates at a higher clock speed, which is not even close when comparing 4690K and 4770K. In general, Broadwell-C is initially a niche solution, perfectly suitable for compact computers, but in a "regular" modular desktop it has nothing special to do: there is no need to "squeeze" 65 W and you can use powerful video cards, or save a lot if high no video performance required.

Introduction In the development of all computer technology in recent years, the course towards integration and the accompanying miniaturization is well traced. And here we are talking not so much about the usual desktop personal computers, but about a huge park of "user-level" devices - smartphones, laptops, players, tablets, etc. - which are reborn in new form factors, absorbing more and more new functions. As for the desktops, it is this trend that affects them in the last turn. Of course, in recent years, the vector of user interest has slightly deviated towards small-sized computing devices, but it's hard to call this a global trend. The basic architecture of x86 systems, which assumes the presence of separate processor, memory, video card, motherboard and disk subsystem, remains unchanged, and this is what limits the possibilities for miniaturization. It is possible to reduce each of the listed components, but a qualitative change in the dimensions of the resulting system in total will not work.

However, in the course of the last year, it seems, there has been a certain turning point in the environment of personal computers. With the introduction of modern semiconductor technological processes with "finer" standards, developers of x86 processors are able to gradually transfer the functions of some devices that were previously separate components to the CPU. So, no one is surprised anymore that the memory controller and, in some cases, the PCI Express bus controller have long become a part of the central processor, and the motherboard chipset has degenerated into a single microcircuit - the south bridge. But in 2011, a much more significant event happened - a graphics controller began to be built into processors for productive desktops. And we are not talking about some kind of frail video cores that are only capable of ensuring the operation of the operating system interface, but about completely full-fledged solutions that, in terms of their performance, can be opposed to discrete entry-level graphics accelerators and probably surpass all those integrated video cores that were built into systems logic sets earlier.

The pioneer was Intel, which released Sandy Bridge processors with integrated Intel HD Graphics for desktop computers at the beginning of the year. True, she thought that good integrated graphics would be of interest primarily to users of mobile computers, and for desktop CPUs, only a stripped-down version of the video core was offered. The incorrectness of this approach was later demonstrated by AMD, which released Fusion processors with full-fledged graphics cores of the Radeon HD series on the market of desktop systems. Such proposals immediately gained popularity not only as solutions for the office, but also as the basis for inexpensive home computers, which forced Intel to reconsider its attitude towards the prospects for CPUs with integrated graphics. The company has updated its Sandy Bridge line of desktop processors by adding faster Intel HD Graphics to its desktop offerings. As a result, now users who want to build a compact integrated system are faced with the question: which manufacturer's platform is more rational to prefer? After conducting comprehensive testing, we will try to give recommendations on the choice of a particular processor with an integrated graphics accelerator.

Terminology question: CPU or APU?

If you are already familiar with the integrated graphics processors that AMD and Intel offer for desktop users, then you know that these manufacturers are trying to distance their products as much as possible from each other, trying to instill the idea that their direct comparison is incorrect. The main "confusion" is brought by AMD, which refers its solutions to a new class of APUs, and not to conventional CPUs. What's the difference?

APU stands for Accelerated Processing Unit. If we turn to detailed explanations, it turns out that from a hardware point of view, this is a hybrid device that combines traditional general-purpose computing cores with a graphics core on a single semiconductor chip. In other words, the same CPU with integrated graphics. However, there is still a difference, and it lies at the program level. The graphics core included in the APU must have a universal architecture in the form of an array of stream processors capable of working not only on the synthesis of three-dimensional images, but also on solving computational problems.

That is, the APU offers a more flexible design than simply combining graphics and computing resources within a single semiconductor chip. The idea is to create a symbiosis of these disparate parts, when some of the calculations can be performed by means of the graphics core. True, as always in such cases, software support is required to tap into this promising opportunity.

AMD Fusion processors with a video core, known under the codename Llano, fully meet this definition, they are precisely APUs. They integrate the graphics cores of the Radeon HD family, which, among other things, support the ATI Stream technology and the OpenCL 1.1 programming interface, through which calculations on the graphics core are really possible. In theory, a number of applications can get practical benefits from running on an array of Radeon HD stream processors, including cryptographic algorithms, rendering of 3D images, or post-processing tasks for photos, sound and video. In practice, however, everything is much more complicated. Implementation difficulties and dubious real performance gains have held back widespread support for the concept so far. Therefore, in most cases, an APU can be viewed as nothing more than a simple CPU with an integrated graphics core.

Intel, by contrast, has a more conservative terminology. It continues to refer to its Sandy Bridge processors, which contain the integrated HD Graphics, by the traditional term CPU. Which, however, has some ground, because the OpenCL 1.1 programming interface is not supported by Intel graphics (compatibility with it will be provided in the next generation Ivy Bridge products). So, Intel does not yet provide for any joint work of dissimilar parts of the processor on the same computational tasks.

With one important exception. The fact is that in the graphics cores of Intel processors there is a specialized Quick Sync unit, focused on hardware acceleration of the video stream encoding algorithms. Of course, as in the case of OpenCL, it requires special software support, but it is really capable of improving the performance when transcoding high-definition video by almost an order of magnitude. So, in the end, we can say that Sandy Bridge is to some extent also a hybrid processor.

Is it legal to compare AMD APUs and Intel CPUs? From a theoretical point of view, an equal sign cannot be put between an APU and a CPU with a built-in video accelerator, but in real life we ​​have two names for the same. AMD Llano processors can accelerate parallel computing, and Intel Sandy Bridge can only use graphics power when transcoding video, but in fact, both of these features are almost never used. So, from a practical point of view, any of the processors discussed in this article is a regular CPU and a video card, assembled inside a single microcircuit.

Processors - Test Participants

In fact, you shouldn't think of processors with integrated graphics as some kind of special offer aimed at a certain group of users with atypical requests. Universal integration is a global trend, and such processors have become the standard offer in the lower and middle price range. Both AMD Fusion and Intel Sandy Bridge have ousted CPUs without graphics from the current offerings, so even if you are not going to rely on an integrated video core, we can not offer anything other than focusing on the same processors with graphics. Fortunately, no one forces the built-in video core to be used, and it can be turned off.

Thus, starting to compare a CPU with an integrated GPU, we came to a more general task - comparative testing of modern processors with a cost of $ 60 to $ 140. Let's see what suitable options in this price range AMD and Intel can offer us, and what specific processor models we were able to involve in the tests.

AMD Fusion: A8, A6 and A4

To use desktop processors with an integrated graphics core, AMD offers a dedicated Socket FM1 platform that is compatible exclusively with the Llano family of processors - A8, A6 and A4. These processors have two, three or four general-purpose Husky cores with a microarchitecture similar to Athlon II, and a Sumo graphics core, inheriting the microarchitecture of the younger representatives of the five thousandth Radeon HD series.

The line of processors of the Llano family looks quite self-sufficient, it includes processors of different computing and graphics performance. However, there is one regularity in the model range - the computing performance is correlated with the graphics performance, that is, the processors with the largest number of cores and with the maximum clock frequency are always supplied with the fastest video cores.

Intel Core i3 and Pentium

Intel can oppose the AMD Fusion processors with its dual-core Core i3 and Pentium, which do not have their own collective name, but are also equipped with graphics cores and have a comparable cost. Of course, there are graphics cores in more expensive quad-core processors, but they play a clearly secondary role there, so the Core i5 and Core i7 were not included in the actual testing.

Intel did not create its own infrastructure for low-cost integrated platforms, so Core i3 and Pentium processors can be used in the same LGA1155 motherboards as other Sandy Bridges. To use the integrated video core, you will need motherboards based on special H67, H61 or Z68 logic sets.

All Intel processors that can be considered competitors for Llano are based on a dual-core design. At the same time, Intel does not place much emphasis on graphics performance - most CPUs have a weak version of HD Graphics 2000 graphics with six executive devices. An exception was made only for the Core i3-2125 - this processor is equipped with the most powerful graphics core in the company's arsenal, HD Graphics 3000 with twelve executive devices.

How we tested

After we got acquainted with the set of processors presented in this testing, it's time to pay attention to the test platforms. Below is a list of components from which the composition of the test systems was formed.

Processors:

AMD A8-3850 (Llano, 4 cores, 2.9 GHz, 4 MB L2, Radeon HD 6550D);
AMD A8-3800 (Llano, 4 cores, 2.4 / 2.7 GHz, 4 MB L2, Radeon HD 6550D);
AMD A6-3650 (Llano, 4 cores, 2.6 GHz, 4 MB L2, Radeon HD 6530D);
AMD A6-3500 (Llano, 3 cores, 2.1 / 2.4 GHz, 3 MB L2, Radeon HD 6530D);
AMD A4-3400 (Llano, 2 cores, 2.7 GHz, 1 MB L2, Radeon HD 6410D);
AMD A4-3300 (Llano, 2 cores, 2.5 GHz, 1 MB L2, Radeon HD 6410D);
Intel Core i3-2130 (Sandy Bridge, 2 cores + HT, 3.4 GHz, 3 MB L3, HD Graphics 2000);
Intel Core i3-2125 (Sandy Bridge, 2 cores + HT, 3.3 GHz, 3 MB L3, HD Graphics 3000);
Intel Core i3-2120 (Sandy Bridge, 2 cores + HT, 3.3 GHz, 3 MB L3, HD Graphics 2000);
Intel Pentium G860 (Sandy Bridge, 2 cores, 3.0 GHz, 3 MB L3, HD Graphics);
Intel Pentium G840 (Sandy Bridge, 2 cores, 2.8 GHz, 3 MB L3, HD Graphics);
Intel Pentium G620 (Sandy Bridge, 2 cores, 2.6 GHz, 3 MB L3, HD Graphics).

Motherboards:

ASUS P8Z68-V Pro (LGA1155, Intel Z68 Express);
Gigabyte GA-A75-UD4H (Socket FM1, AMD A75).

Memory - 2 x 2 GB DDR3-1600 SDRAM 9-9-9-27-1T (Kingston KHX1600C8D3K2 / 4GX).
Hard disk: Kingston SNVP325-S2 / 128GB.
Power supply: Tagan TG880-U33II (880 W).
Operating system: Microsoft Windows 7 SP1 Ultimate x64.
Drivers:

AMD Catalyst Display Driver 11.9;
AMD Chipset Driver 8.863;
Intel Chipset Driver 9.2.0.1030;
Intel Graphics Media Accelerator Driver 15.22.50.64.2509;
Intel Management Engine Driver 7.1.10.1065;
Intel Rapid Storage Technology 10.5.0.1027.

Since the main goal of this test was to study the capabilities of processors with integrated graphics, all tests were carried out without using an external graphics card. The built-in video cores were responsible for displaying the image on the screen, 3D functions and accelerating HD video playback.

It should be noted that, due to the lack of DirectX 11 support in Intel graphics cores, testing in all graphics applications was carried out in DirectX 9 / DirectX 10 modes.

Performance in common tasks

Overall performance

To assess the performance of processors in common tasks, we traditionally use the Bapco SYSmark 2012 test, which simulates the user's work in common modern office programs and applications for creating and processing digital content. The idea of ​​the test is very simple: it produces a single metric that characterizes the weighted average speed of a computer.

As you can see, AMD Fusion series processors look just shameful in traditional applications. AMD's fastest quad-core Socket FM1 processor, the A8-3850, barely outperforms the dual-core Pentium G620 at half the price. All the other representatives of the AMD A8, A6 and A4 series are hopelessly behind Intel competitors. In general, this is a quite natural result of using the old microarchitecture, which migrated there from the Phenom II and Athlon II, in the basis of the Llano processors. Until AMD implements processor cores with a higher specific performance, even a quad-core APU of this company will find it very difficult to fight against current and regularly updated Intel solutions.

A deeper understanding of the SYSmark 2012 results can provide insight into the performance scores obtained in various system use cases. The Office Productivity script simulates typical office work: preparing word, processing spreadsheets, working with e-mail, and surfing the Internet. The script uses the following set of applications: ABBYY FineReader Pro 10.0, Adobe Acrobat Pro 9, Adobe Flash Player 10.1, Microsoft Excel 2010, Microsoft Internet Explorer 9, Microsoft Outlook 2010, Microsoft PowerPoint 2010, Microsoft Word 2010, and WinZip Pro 14.5.

The Media Creation scenario simulates the creation of a commercial using pre-shot digital images and video. For this purpose, popular packages from Adobe are used: Photoshop CS5 Extended, Premiere Pro CS5 and After Effects CS5.

Web Development is a scenario within which the creation of a website is modeled. Applications used: Adobe Photoshop CS5 Extended, Adobe Premiere Pro CS5, Adobe Dreamweaver CS5, Mozilla Firefox 3.6.8 and Microsoft Internet Explorer 9.

Data / Financial Analysis Scenario is dedicated to statistical analysis and forecasting of market trends that are performed in Microsoft Excel 2010.

3D Modeling Script is all about creating 3D objects and rendering static and dynamic scenes using Adobe Photoshop CS5 Extended, Autodesk 3ds Max 2011, Autodesk AutoCAD 2011 and Google SketchUp Pro 8.

The last scenario, System Management, is used to create backups and install software and updates. Several different versions of Mozilla Firefox Installer and WinZip Pro 14.5 are involved here.

The only type of application that AMD Fusion processors can achieve with acceptable performance are 3D modeling and rendering. In such tasks, the number of cores is a weighty argument, and the quad-core A8 and A6 can provide higher performance than, for example, Intel Pentium. But up to the level set by Core i3 processors in which support for Hyper-Threading technology is implemented, AMD's offerings fall short even in the most favorable case.

Application performance

To measure the speed of processors when compressing information, we use the WinRAR archiver, with the help of which we archive a folder with various files with a total size of 1.4 GB with the maximum compression ratio.

We measure performance in Adobe Photoshop using our own benchmark, which is a creatively reworked Retouch Artists Photoshop Speed ​​Test including typical processing of four 10-megapixel images taken with a digital camera.

When testing the audio transcoding speed, the Apple iTunes utility is used, with the help of which the contents of a CD-disc are converted to AAC format. Note that a characteristic feature of this program is the ability to use only a couple of processor cores.

To measure the speed of video transcoding into H.264 format, the x264 HD test is used, which is based on measuring the processing time of the original MPEG-2 video recorded in 720p resolution with a 4 Mbps stream. It should be noted that the results of this test are of great practical importance, since the x264 codec used in it underlies numerous popular transcoding utilities, for example, HandBrake, MeGUI, VirtualDub, etc.

Testing the final rendering speed in Maxon Cinema 4D is performed using the specialized Cinebench benchmark.

We also used the Fritz Chess Benchmark, which evaluates the speed of the popular chess algorithm used in the programs of the Deep Fritz family.

Looking at the diagrams, you can once again repeat everything that has already been said in relation to the SYSmark 2011 results. AMD processors, which the company offers for use in integrated systems, can boast of any acceptable performance only in those computing tasks where the load is good. is parallelized. For example, in 3D rendering, video transcoding, or when iterating over and evaluating chess positions. And then, the competitive level of performance in this case is observed only in the senior quad-core AMD A8-3850 with a clock frequency that is increased to the detriment of power consumption and heat dissipation. Still, AMD processors with a 65-watt thermal capacity fall behind any of the Core i3s, even in the most favorable case for them. Accordingly, against the background of Fusion, representatives of the Intel Pentium family look quite decent: these dual-core processors perform about the same as the three-core A6-3500 with a well-parallelized load, and outperform the older A8 in programs like WinRAR, iTunes or Photoshop.

In addition to the conducted tests, to check how the power of the graphics cores can be used to solve everyday computing tasks, we conducted a study of the video transcoding speed in Cyberlink MediaEspresso 6.5. This utility has support for computing on graphics cores - it supports both Intel Quick Sync and ATI Stream. Our test consisted of measuring the time it took to transcode a 1.5GB 1080p video to H.264 (which was a 20-minute episode of the hit TV series) at downscaling for viewing on an iPhone 4.

The results are divided into two groups. The first includes Intel Core i3 processors, which have support for Quick Sync technology. Numbers speak better than words: Quick Sync transcodes HD video content several times faster than any other toolkit. The second large group unites all other processors, among which CPUs with a large number of cores are in the first place. The Stream technology promoted by AMD, as we can see, does not manifest itself in any way, and the Fusion series APUs with two cores show no better result than Pentium processors, which transcode video exclusively by the computational cores.

Graphics core performance

The group of 3D gaming tests opens with the results of the 3DMark Vantage benchmark, which was used with the Performance profile.

A change in the nature of the load immediately leads to a change in leaders. The graphics core of any AMD Fusion processor is in practice superior to any Intel HD Graphics option. Even the Core i3-2125, equipped with the HD Graphics 3000 video core with twelve execution units, is able to reach only the performance level demonstrated by the AMD A4-3300 with the weakest integrated graphics accelerator Radeon HD 6410D among all presented in the Fusion test. All the rest of Intel's processors are two to four times worse than AMD's in terms of 3D performance.

Some compensation for the drop in graphics performance can be the results of the CPU test, but it should be understood that the speed of the CPU and GPU are not interchangeable parameters. We should strive to balance these characteristics, and as is the case with the compared processors, we will see further, analyzing their gaming performance, which depends on the power of both the GPU and the computing component of hybrid processors.

To study the speed of work in real games, we selected Far Cry 2, Dirt 3, Crysis 2, the beta version of World of Planes and Civilization V. Testing was carried out at a resolution of 1280x800, and the quality level was set to Medium.

In gaming tests, a very positive picture for AMD's proposals is developing. Despite the fact that they have rather mediocre computational performance, powerful graphics allow them to show good (for integrated solutions) results. Almost always, representatives of the Fusion series allow you to get a higher number of frames per second than Intel platform with processors of the Core i3 and Pentium families gives.

Even the fact that Intel began to build in a productive version of the HD Graphics 3000 graphics core did not save the situation of the Core i3 processors. The Core i3-2125 equipped with it turned out to be faster than its counterpart Core i3-2120 with HD Graphics 2000 by about 50%, but the graphics embedded in Llano, even faster. As a result, even the Core i3-2125 can only compete with the cheap A4-3300, while the rest of the Sandy Bridge microarchitecture carriers look even worse. And if we add to the results shown in the diagrams the lack of support for DirectX 11 in the video cores of Intel processors, then the situation for the current solutions of this manufacturer seems even more hopeless. Only the next generation of the Ivy Bridge microarchitecture can fix it, where the graphics core will receive both much higher performance and modern functionality.

Even if we disregard specific numbers and look at the situation qualitatively, AMD's offerings look like a much more attractive option for an entry-level gaming system. The older Fusion A8 series processors, with certain compromises in terms of screen resolution and image quality settings, allow you to play almost any modern games without resorting to the services of an external video card. We cannot recommend any Intel processors for cheap gaming systems - various HD Graphics options have not yet matured for use in this environment.

Energy consumption

Systems based on processors with integrated graphics cores are gaining more and more popularity not only due to the opening possibilities for miniaturizing systems. In many cases, consumers opt for them, guided by the opening opportunities to reduce the cost of computers. Such processors allow not only to save on a video card, they also allow you to assemble a system that is more economical to use, since its total power consumption will obviously be lower than the consumption of a platform with discrete graphics. A concomitant bonus is quieter modes of operation, since a decrease in consumption translates into a decrease in heat generation and the possibility of using simpler cooling systems.

That is why developers of processors with integrated graphics cores try to minimize the power consumption of their products. Most of the CPUs and APUs reviewed in this article have an estimated typical heat dissipation, which lies in the 65W range - and this is an unspoken standard. However, as we know, AMD and Intel approach the TDP parameter somewhat differently, and therefore it will be interesting to assess the practical consumption of systems with different processors.

The graphs below show two energy consumption values. The first is the total system consumption (without a monitor), which is the sum of the energy consumption of all components involved in the system. The second is the consumption of only one processor through a dedicated 12-volt power line. In both cases, the efficiency of the power supply is not taken into account, since our measuring equipment is installed after the power supply and records the voltages and currents entering the system via 12-, 5- and 3.3-volt lines. During the measurements, the load on the processors was created by the 64-bit version of the LinX 0.6.4 utility. The FurMark 1.9.1 utility was used to load the graphics cores. In addition, to correctly estimate idle power consumption, we have activated all available energy-saving technologies, as well as Turbo Core technology (where supported).

At rest, all systems showed the total energy consumption, which is approximately at the same level. At the same time, as we can see, Intel processors practically do not load the processor power line in idle mode, while competing AMD solutions, on the contrary, consume up to 8 watts per 12-volt dedicated line on the CPU. But this does not mean that the representatives of the Fusion family do not know how to fall into deep energy-saving states. The differences are caused by the different implementation of the power scheme: in Socket FM1 systems, both the computational and graphic cores of the processor and the north bridge built into the processor are powered from the processor line, while in Intel systems the north bridge of the processor takes power from the motherboard.

Maximum compute load reveals that the power efficiency issues inherent in the Phenom II and Athlon II are not gone with AMD's 32nm process. Llano uses the same microarchitecture and loses to Sandy Bridge in the same way in terms of the ratio of performance per watt of electricity consumed. Older Socket FM1 systems consume about twice as much as systems with LGA1155 Core i3 processors, despite the fact that the computing performance of the latter is clearly higher. The gap in power consumption between Pentium and the younger A4 and A6 is not that huge, but nevertheless, the situation does not change qualitatively.

Under the graphics load, the picture is almost the same - Intel processors are significantly more economical. But in this case, a good excuse for AMD Fusion can be their significantly higher 3D performance. Note that in gaming tests, the Core i3-2125 and A4-3300 "squeezed out" the same number of frames per second, and in terms of consumption under the load on the graphics core, they also went very close to each other.

The simultaneous load on all blocks of hybrid processors allows you to get a result that can be figuratively represented as the sum of the two previous graphs. The A8-3850 and A6-3650 processors, which have a 100-watt thermal package, seriously break away from the rest of the 65-watt offerings from AMD and Intel. However, even without them, Fusion processors are less economical than Intel solutions in the same price range.

When using processors as the basis of a media center, busy with playing high-definition video, an atypical situation arises. Computing cores are mostly idle here, and the decoding of the video stream is assigned to specialized blocks built into the graphics cores. Therefore, platforms based on AMD processors manage to achieve good energy efficiency; in general, their consumption does not greatly exceed the consumption of systems with Pentium or Core i3 processors. Moreover, the lowest-frequency AMD Fusion, the A6-3500 offers the best economy in this use case.

conclusions

At first glance, summing up the test results is easy. AMD and Intel processors with integrated graphics have shown completely dissimilar advantages, which allows us to recommend either one or the other depending on the planned use of the computer.

So, the strong point of the AMD Fusion family of processors is the integrated graphics core with relatively high performance and compatibility with DirectX 11 and Open CL 1.1 software interfaces. Thus, these processors can be recommended for those systems where the quality and speed of 3D graphics is not the least important. At the same time, the processors included in the Fusion series use general-purpose cores based on the old and slow K10 microarchitecture, which translates into their low performance in computational tasks. Therefore, if you are interested in options that provide the best performance in common non-gaming applications, you should look towards Intel's Core i3 and Pentium, even though such CPUs are equipped with fewer processing cores than competing offerings from AMD.

Of course, in general, AMD's approach to the design of processors with an integrated video accelerator seems to be more rational. The APU models offered by the company are well balanced in the sense that the speed of the computing part is quite adequate to the speed of the graphics and vice versa. As a result, the older A8 series processors can be considered as a possible basis for entry-level gaming systems. Even in modern games, such processors and the Radeon HD 6550D video accelerators integrated into them can provide acceptable playability. With the younger A6 and A4 series with weaker versions of the graphics core, the situation is more complicated. For universal gaming systems of the lower level, their performance is no longer enough, therefore, it is possible to rely on such solutions only in those cases when it comes to creating multimedia computers, which will run extremely graphically simple casual games or network role-playing games of previous generations.

However, whatever is said about balance, the A4 and A6 series are poorly suited for demanding computing applications. Within the same budget, Intel Pentium line-ups can offer significantly faster computing performance. To tell the truth, against the background of Sandy Bridge, only the A8-3850 can be considered a processor with an acceptable speed in common programs. And even then, its good results are far from being manifested everywhere and, moreover, are provided with increased heat dissipation, which will not please every computer owner without a discrete video card.

In other words, it's a shame that Intel still can't offer a graphics core worthy of performance. Even the Core i3-2125, equipped with the fastest Intel HD Graphics 3000 graphics in the company's arsenal, works at the level of AMD A4-3300 in games, since the speed in this case is limited by the performance of the built-in video accelerator. All the other Intel processors are equipped with a one and a half times slower video core, and in 3D games they appear very faded, often showing a completely unacceptable number of frames per second. Therefore, we would not recommend at all to think of Intel processors as a possible basis for a system capable of working with 3D graphics. The Core i3 and Pentium video core does an excellent job of displaying the operating system interface and playing high-definition video, but it is not capable of more. So the most suitable application for Core i3 and Pentium processors is seen in systems where the computing power of general-purpose cores is important with good energy efficiency - in these parameters, no AMD offers with Sandy Bridge can compete.

Well, in conclusion, it should be reminded that Intel's LGA1155 platform is much more promising than AMD Socket FM1. When purchasing an AMD Fusion series processor, you should be mentally prepared for the fact that it will be possible to improve a computer based on it within very limited limits. AMD plans to release only a few more Socket FM1 models from the A8 and A6 series with a slightly increased clock frequency, and their successors coming out next year, known under the codename Trinitу, will not be compatible with this platform. Intel's LGA1155 platform is much more promising. Not only can the much more computationally productive Core i5 and Core i7 be installed in it today, but the Ivy Bridge processors planned for next year in motherboards purchased today should work.

19.04.2014 0 25889

There have been times when a PC couldn't run any decent game if it didn't have discrete graphics card... Today, most off-the-shelf computers and almost all laptops rely on graphics solutions integrated into central processing units... Yet the discrete graphics market continues to thrive. If you don't play heavy AAA games, is the graphics card a worthwhile upgrade? To find out the answer, let's compare the performance of integrated and discrete GPUs.

AMD and Intel significantly improved the quality integrated graphics... AMD's Kaveri APUs use the same powerful GCN graphics core found in their top-of-the-range Radeon series discrete graphics cards.

Intel also updated the features and capabilities of its HD-series graphics, which are built into the 4th generation Core processors (codenamed Haswell). They currently provide broader support for Microsoft DirectX 11.1, can support multiple displays (including 4K resolution), and are compatible with most games.

To determine the benefits of a discrete graphics card, two computers were assembled. One runs on a Kaveri A8-7800 with an integrated Radeon R7 series GPU and the other on an Intel Core i7- 4670 Haswell processor with an integrated Intel HD 4600. Tests were then run with and without a discrete graphics card on board each system.

The argument for discrete graphics

Per discrete graphics says its performance. All but entry-class graphics cards have a much more powerful GPU than those integrated into processors. Moreover, a separate graphics card will provide the GPU dedicated pool of high-speed memory... An integrated GPU must be content with sharing system memory and data bus. Usually, with a discrete card, you can set the graphics settings in games higher than with integrated solutions.

There are other benefits to using discrete graphics cards as well. On current generation Nvidia graphics cards, users can use proprietary technologies Shadowplay and PhysX... ShadowPlay optimizes the use of the video encoding engines built into NVIDIA GPUs to record and stream games in real time, with little impact on frame rates. This is a key feature of the Nvidia Shield portable gaming device.

PhysX is a proprietary physics simulation technology that makes objects in games behave more closely to reality. PhysX is not supported by all games, but it can have a huge visual impact on supported ones.

Games aren't the only applications that benefit from the performance of a discrete GPU. GPUs from AMD and Nvidia are made up of thousands of processors that can perform multiple operations at the same time. Any application can benefit from parallel processing, be it image editing programs like Photoshop, data encryption, or distributed computing projects like Folding @ Home or SETI @ Home.

Discrete video cards can accelerate the mining of cryptographic currencies Bitcoins, Litecoins and others. The miners bought the latest graphics cards from AMD, as the Radeon architecture here turned out to be more efficient than Intel processors and Nvidia graphics cards. Where the Intel Haswell Core i7-4770K processor is capable of processing about 93K hashes per second, the AMD Radeon R9 290X makes about 880K hashes per second.

The argument against discrete graphics

Discrete video cards also have drawbacks, and the main one is the price. Buying a video card will cost from a couple of thousand rubles to 30 thousand or more. AMD recently announced the most powerful graphics card ever. The Radeon R9 295X2 has two Tahiti XT GPUs on one card and costs $ 1,500.

AMD and Intel have almost completely ditched processors without integrated graphics (only AMD's FX series and Intel's Ivy Bridge-E chips don't), and motherboards that support these processors have integrated video output.

The discrete graphics card also adds complexity to the system. The motherboard must have a free PCIe x16 slot for installing a video card. Usually it is available in the system unit, although some ready-made small computers may not have it, or the card may not fit inside the case. Or the power supply will not be able to support the requirements of the card. This is because the PC makers did not anticipate, or simply did not care, that the end user could upgrade.

Installing a discrete graphics card with Intel processors, technologies such as the Quick Sync video encoding engine can be difficult to use. Quick Sync is associated with the Intel Integrated Graphics and installing a discrete card may disable it. Fortunately, it can be activated again.

But you have to pay for everything. An external video card will increase the level of power consumption, generate heat, which requires a fan to remove it (some cards even have three fans), and this will increase the noise level of the system as a whole. There are passive cooling systems, but they are only suitable for entry-level cards and are more expensive.

Moving on to numbers

Two computers were assembled: an AMD A8-7600 APU with a Radeon R7 iGPU on an Asus A88X-Pro motherboard, and an Intel Core i5-4670 with an Intel HD 4600 on a Gigabyte Z87X-UD5 TH board. Both systems were equipped with 16GB of memory, a Samsung 840 Pro SSD and a 1000W Silverstone power supply, and the operating system was Windows 8.1 Pro 64-bit.

A series of tests were run, including games and content creation applications, using only integrated GPUs. After that, a Radeon R9 280X video card from XFX was installed in the systems and the tests were repeated.

As you can see from the graphs, the presence of a discrete graphics card improves performance in almost all directions, and not only in games. PCMark 8, for example, launched the Home and Work versions with OpenGL support. This interface uses all available computing resources of the computer, both the central processor and the graphics. The addition of a discrete graphics card increased system performance in this benchmark by 3-19% (Figure 1).

In the multi-threaded Cinebench test, the video card had little effect, but with OpenGL in a system with an Intel processor, the video card gave a performance increase by 79%, in an AMD system - 42% (Figure 2).

Many people think that people who play simple games like Farmville, Angry Birds, etc. - will not get any benefit from discrete graphics. But the addition of a graphics card gave a significant performance boost in the HTML5-oriented Fishbowl benchmark. This test is limited to 60 frames per second (the refresh rate of most monitors), and this value was achieved in three of the four tests with a discrete card (Figure 3). "Casual" games are becoming more and more complex, and accordingly their requirements for video cards are growing.

Speaking of complex games, the graphics cards showed a noticeable boost in BioShock Infinite at 1920 x 1080 pixels (Figure 4) and the synthetic gaming benchmark 3DMark Fire Strike.

There is an area where the addition of a discrete video adapter did not have a significant impact: video playback. There was very little CPU impact when running both YouTube videos (HTML5) and H.264 files in an MKV container.

Takeaway: Almost every desktop user can benefit from a graphics card. They are useful not only for gamers, although, of course, they get the main benefit.

P.S. If you have any problems with the equipment, please contact our computer service, or order a visit

An integrated GPU plays an important role for both gamers and undemanding users.

The quality of games, movies, watching videos on the Internet and images depends on it.

Principle of operation

The graphics processor is integrated into the motherboard of the computer - this is how the integrated graphics looks like.

As a rule, they use it to remove the need to install a graphics adapter -.

This technology helps to reduce the cost of the finished product. In addition, due to their compactness and low power consumption, such processors are often installed in laptops and low-power desktop computers.

Thus, integrated GPUs have filled this niche so much that 90% of laptops on US store shelves have just such a processor.

Instead of a conventional video card, the computer's RAM itself is often an auxiliary tool in integrated graphics.

True, this solution somewhat limits the performance of the device. Yet the computer itself and the GPU use the same bus for memory.

So this "neighborhood" affects the performance of tasks, especially when working with complex graphics and during gameplay.

Kinds

Built-in graphics have three groups:

1. Shared memory graphics are a device based on shared memory management with the main processor. This significantly reduces the cost, improves the energy saving system, but degrades the performance. Accordingly, for those working with complex programs, this kind of integrated GPU is more likely to be unsuitable.
2. Discrete graphics - a video chip and one or two video memory modules are soldered on the motherboard. Thanks to this technology, the image quality is significantly improved and it also becomes possible to work with 3D graphics with the best results. True, you will have to pay a lot for this, and if you are looking for a high-power processor in all respects, then the cost can be incredibly high. In addition, the electricity bill will rise slightly - the power consumption of discrete GPUs is higher than usual.
3. Hybrid discrete graphics - a combination of the two previous types, which ensured the creation of the PCI Express bus. Thus, access to the memory is carried out both through the unsoldered video memory, and through the operational one. With this solution, manufacturers wanted to create a compromise solution, but it still does not eliminate the disadvantages.

Manufacturers

As a rule, large companies are engaged in the manufacture and development of integrated graphics processors, and, but many small enterprises are also involved in this area.

This is not difficult to do. Find Primary Display or Init Display First. If you don't see something like that, look for Onboard, PCI, AGP or PCI-E (it all depends on the buses installed on the motherboard).

Choosing PCI-E, for example, you enable the PCI-Express video card, and disable the built-in integrated one.

Thus, to enable the integrated video card, you need to find the appropriate parameters in the BIOS. The start-up process is often automatic.

Problems registering on the site? CLICK HERE ! Do not pass by a very interesting section of our site - the projects of visitors. There you will always find the latest news, anecdotes, weather forecast (in the ADSL newspaper), a TV program of terrestrial and ADSL-TV channels, the freshest and most interesting news from the world of high technologies, the most original and amazing pictures from the Internet, a large archive of magazines in recent years , delicious recipes in pictures, informative. The section is updated daily. Always the latest versions of the best free programs for everyday use in the Necessary programs section. There is almost everything you need for your daily work. Start gradually abandoning pirated versions in favor of more convenient and functional free counterparts. If you still do not use our chat, we highly recommend you to get acquainted with it. You will find many new friends there. It is also the fastest and most efficient way to contact project administrators. The Antivirus Updates section continues to work - always up-to-date free updates for Dr Web and NOD. Didn't have time to read something? The full content of the creeping line can be found at this link.

AMD Trinity for desktop. Part 1. Graphics core

Announcement: High hopes were pinned on APU Llano, but they were only partially justified - in the laptop segment. AMD has not lost hope: the new Trinity APUs are already out for mobile platforms, and now they are also available for desktops. We have tested the second generation of desktop APUs and in this review we will talk about both their architecture and graphics performance.

There is no doubt that AMD is not the supplier of the fastest processors for personal computers today. And this situation did not develop yesterday. Since Intel moved from the Pentuim 4 to a variety of Core processors, AMD's offerings have slipped into second place. In fact, all of today's processor products of this company are either budgetary or some highly specialized solutions that are of little interest to a wide range of users, who put high performance at the forefront. However, the low performance indicators of the manufactured processors, as well as the decline in market share, are not at all a reason to put an end to the results of the work of AMD's processor division.

The engineers of this company are famous for being able from time to time to come up with some original ideas that allow AMD not only to maintain its market position, but also to have a significant impact on the entire industry as a whole. You don't have to go far for examples of such ideas: 64-bit extensions of the x86 microarchitecture, multi-core design of the CPU, integration of the memory controller and the north bridge of the chipset into the processor - all these solutions were first developed and implemented by AMD, and not the current leader in processor construction.

That is why we continue to closely monitor what innovations are ripening in the depths of AMD. And it seems that by now the company has once again groped for a fruitful vector of development, which is capable of giving a positive impetus not only to itself, but to the entire processor market as a whole. This vector - APU (Accelerated Processing Unit, "accelerated processing unit") - an ideology that provides for the combination of traditional computational cores with a powerful graphics core on one semiconductor chip. Moreover, not a simple neighborhood, but a symbiosis - the possibility of combining their resources to solve common problems.

The APU class includes several different AMD offerings released back in 2011. Of greatest interest among them are the A-series hybrid processors, codenamed Llano, which serve as the basis for the Lynx and Sabine platforms and are aimed at a wide range of desktop and mobile systems. Despite the fact that these processors and platforms serve only as a "trial balloon", since they are only used to test the principles of APUs, they were quite warmly received by the market. Llano was especially in demand in the mobile segment, which was immediately reflected in the increase in the presence of AMD products in modern laptops. And this is really visible to the naked eye. If a couple of years ago, AMD mobile platforms were found in very few offers, today it is not difficult to buy a laptop based on a processor from this manufacturer. In any computer store, you can easily find a huge number of offers equipped with AMD APUs.

However, the increased interest in AMD processors seen in the mobile computer market did not arise at all because of their hybridity. Rather, it should be understood as a side effect. In reality, the fact is that a sufficiently powerful graphics core, combined with computational cores that are acceptable by the standards of mobile solutions, is exactly what Intel's product range is lacking. And if we take into account the very affordable prices set by AMD for its APUs, then it is absolutely not surprising that they fit perfectly into inexpensive laptops, thereby giving them the opportunity to assemble modern computers without installing discrete video accelerators in them and the additional costs associated with this. ...

As a result, the concept of APU itself was popular with the people. Its preachers from AMD, communicating with software developers, were able to rely on relevance and prevalence, and in the end, real programs appeared at the disposal of users, designed to fully utilize the resources of hybrid processors. The May update of the A-series of AMD mobile processors with the Trinity design, which increased the performance of both the computing and graphics parts of the APU, became an additional argument in favor of the viability and attractiveness of the concept. So in the future, the share of laptops with the AMD Vision logo will only increase.

A completely different story happened with AMD desktop APUs. The demands of desktop users are significantly different from those of laptop owners, and they were not particularly interested in the APU topic from the very beginning. The driving force behind the penetration of the first generations of hybrid processors into laptops was quite powerful graphics, but when using it in desktop computers, this epithet must be abandoned. The fact is that desktops are characterized by much higher screen resolutions, in which AMD A-series processors do not develop an acceptable level of 3D performance. In other words, from the point of view of desktop users, the graphics core of Llano processors is qualitatively little different from the integrated graphics of Intel's offerings: both options are almost equally bad for an entry-level gaming system. The processing power of the AMD APUs is significantly lower than that of Intel processors, and this closes Llano's path to a whole range of home or office systems. Even as the heart of the media center, AMD's APUs have little chance of competing offerings. In this case, too high heat dissipation and the lack of means to accelerate the encoding of high-resolution video content let them down.

However, the most ambitious obstacle on the way of Llano to desktop computers was the Socket FM1 platform, specially designed for them, with completely unclear prospects. It is impossible to install any other processors, except for Llano, and this makes it a “thing in itself”, on the one hand, not prone to a subsequent upgrade, and on the other, with a very limited lifespan. It is quite natural that it is almost impossible to interest a solution with such a combination of characteristics of desktop users, because the market is flooded with competing LGA1155 offers for every taste and wallet with a much longer life cycle.

But giving the market of integrated desktop processors to the power of a competitor, which, seeing the perspective of the APU concept, hastily increases the capacity of its own graphics cores, is clearly not included in AMD's plans. Therefore, about a year after the appearance of Llano, the company is ready to offer the second generation of desktop A-series processors, revised and revised. The design of the new desktop APUs is not specialized or utilitarian. This is Trinity, and it has already been tested on mobile systems, where it has been successfully used since the beginning of the summer. However, for desktop systems, the frequencies of the computing and graphics components have been seriously increased, which allows the manufacturer to assure the public that the fresh APUs, unlike their predecessors, should appeal to many desktop users, including enthusiasts.

In general, we are almost ready to believe in AMD's words: at least in terms of design, Trinity is definitely better than Llano. As we have already seen with the example of mobile APUs, the computational cores used in Trinity, which are based on the Piledriver microarchitecture, work faster than the Husky cores from Llano, whose microarchitecture roots go back to the distant past. The performance of the graphics core has also significantly improved, the structure of which has been radically redesigned. And most importantly, a new Socket FM2 platform is now offered for Trinity desktop processors, which should be devoid of all the old flaws. AMD is ready to guarantee its stability for the next several years, and the model range of processors in a compatible performance will include a wide range of offerings of different levels.

In other words, if we compare Trinity and Llano, then the new processors are obviously better. However, are they good enough to effectively push the APU concept to desktop systems, whose users are still very skeptical about such solutions? In our material, we will try to partially answer this question, for which we will thoroughly test the graphics component of the new generation desktop hybrid processors and try to understand whether its power is enough for use in entry-level gaming systems.

Unfortunately, we have to postpone a detailed consideration of the second part of Trinity - computing cores - for some time. However, this is not our fault. The fact is that AMD has not yet officially announced its new A-series processors for desktop systems. Therefore, our hands are partially bound by non-disclosure obligations, so this article will be followed by a second one, which includes tests of a different kind. However, no one forbids us to operate with the available information about the Trinity microarchitecture, so first, let's analyze what work AMD engineers have done to make the new APUs a reality.

Trinity design

In accordance with the original concept, any APU consists of three main parts. In this regard, Trinity does not bring any changes: the new generation of APUs include processor cores, an integrated graphics accelerator and a small but very important component - the unified northbridge. It is he who turns the sum of dissimilar cores into a single system and, including the DDR3 SDRAM controller, is responsible for the interaction of the computing and graphics cores with each other and with the system memory, ensuring the possibility of their joint work with the same data.

In general, the general structure of Trinity has remained exactly the same as that of Llano, but at a lower level, all components have been reworked. At the same time, all the changes were made in such a way as not to inflate the semiconductor crystal: AMD's production technology has not been updated, the company continues to use the 32nm Globalfoundries process with SOI, and the manufacturer is not going to raise the cost of APUs positioned as fairly affordable proposals. As a result, the area of ​​the Trinity crystal compared to Llano has increased by only 8 percent - up to 246 mm 2. The number of transistors also changed very slightly and reached 1.303 billion pieces (it was - 1.178 billion). Moreover, even the division of the transistor budget between computing and graphics resources has not changed much: they occupy approximately the same area on the chip in both cases.

Nevertheless, talk about the similarity of Llano and Trinity can end there. Computing cores, for example, have changed dramatically with the release of a new generation of APUs. Now, hybrid processors are based on (and will be used in the future) the Bulldozer microarchitecture, and more specifically, its second generation - Piledriver. Trinity dual-core and quad-core processors include one or two conditionally called dual-core modules, which, recall, contain two sets of executive devices and can process two threads simultaneously, but at the same time have a common cache memory module, an instruction fetch unit, and their decoder and a floating point block. At the same time, in Trinity, in comparison with FX class processors based on the Bulldozer microarchitecture without integrated graphics, not only the number of cores is reduced, but also there is no third-level cache.

But the second generation Bulldozer microarchitecture used in the new APUs, which has not yet been represented in any other processor family, offers a number of small improvements aimed at increasing performance, reducing leakage currents and ensuring stability at high clock speeds. The front end of the pipeline has received a more accurate branch predictor, as well as a larger instruction window. The executive devices gained an improved scheduler, and they themselves learned to execute individual instructions a little faster, for example, integer and real-number division. In addition, the developers talk about increasing the L1 TLB capacity and improving the arbitration and prefetching algorithms for L2 cache data. All this is estimated at approximately 25 percent superiority (according to AMD calculations) of Trinity processors over Llano in computing performance.

The unified north bridge has also been drastically changed. First of all, engineers revised the system of priorities for access to shared memory, giving priority to computational cores, which, as practice shows, generate a relatively small part of requests. In addition, AMD has taken care of support for new types of memory, including DDR3-1866 in standard mode or DDR3-2400 when overclocked. The internal data buses were expanded, in particular, the graphics core got the ability to work with the memory controller via the 256-bit dedicated Radeon Memory Bus, and outside the chip, all connections now use the PCI Express protocol, which replaced Hyper-Transport.

However, the most interesting are the changes that have occurred with the graphics core. The fact is that without a significant increase in the transistor budget and without a radical revision of the architecture, AMD managed to significantly increase its performance, that is, in fact, to increase the density of useful blocks in the GPU by eliminating some surplus. This find, perhaps, deserves a separate discussion, especially since it is the graphics integrated into Trinity that are in the focus of our attention today.

Graphic core Devastator

The most intriguing fact about the design of the Devastator - which is the codename for the GPU built into the Trinity processors - is that it is based on the VLIW4 architecture. Considering that the Llano graphics core was based on the VLIW5 architecture, such a move by AMD seems somewhat strange, and we would rather expect to see the CGN architecture in Trinity, which is typical for the latest versions of discrete accelerators. However, in fact, it is VLIW4 that allows you to increase the specific efficiency of the graphics core, which is artificially limited by the number of transistors. AMD has already done this trick with its own Radeon HD 6900 series graphics cards, and then its results were more than satisfactory.

The bottom line is that the VLIW5 grouping of ALUs of five per streaming VLIW processor turns out to be not very effective, and one of the ALUs is simply idle in a large number of cases. Therefore, the VLIW4-layout of the Devastator, which assumes the presence of four ALUs in the VLIW stream processor, entails a more rational use of available resources. Of course, the downside is a decrease in the total number of executive devices and a decrease in the theoretical peak performance of the core, but the practical specific performance per square millimeter is growing. And for a die of a hybrid processor, on which, in addition to the graphics core, there are computational cores, this is the most correct way of optimization.

In total, the Trinity graphics core provides six SIMD engines, each of which consists of four texture units and sixteen VLIW stream processors. In total, this gives the presence of 384 ALUs in the core, and this is 16 pieces less than the Sumo graphics core of the Llano processors had at the disposal. However, simple arithmetic is not entirely appropriate here, the Devastator ALUs are usually more loaded with work than their predecessors, and, in addition, the relative simplicity of VLIW stream processors allows the graphics core to be set at higher clock speeds. For example, while in the older version of Llano the graphics worked at 600 MHz, in Trinity the video core speed can reach 800 MHz.

Considering that Devastator has 24 texture units (4 TMUs for each SIMD engine) and 8 raster operations (ROPs) at its disposal, we can conclude that this graphics core actually represents about one quarter of the GPU of the Radeon HD 6970 class. That, even taking into account the correction for a slightly lower operating frequency and the absence of a dedicated high-bandwidth memory bus, is very good. In other words, AMD is not disingenuous when it says that Trinity processors are equipped with integrated graphics of the "discrete" class. Indeed, you can expect very good 3D performance from the new generation APUs.

It is hardly surprising that the Trinity graphics core is compatible with the modern DirectX 11, OpenCL 11 and DirectCompute 11. These features were available in the Radeon HD 6900 video cards based on the same architecture, and in the predecessors of Trinity, the Llano processors. But at the same time, the new embedded graphics inherited some features of very modern solutions, in which the CGN architecture found its place. In particular, Devastator has an improved tessellation unit, as well as support for all current types of full-screen anti-aliasing: SSAA, EQAA and MLAA.

Special attention in the Trinity graphics is paid to media capabilities relevant to hybrid processors. The graphics core has a specialized unit AMD HD Media Accelerator borrowed from the latest versions of the GPU, which includes engines for hardware video decoding (UVD3) and hardware encoding of video content in H.264 format (VCE). The latter feature is very important for Trinity's successful competition with Intel's APUs, which have long received Quick Sync technology for high-speed high-definition video transcoding. Now there is something similar in AMD processors, but at the moment we have not been able to make sure that the VCE engine is working due to problems with its support in drivers and in existing software.

Introducing its new hybrid processor to the desktop market, AMD also thought about so that its users do not feel deprived compared to the owners of discrete graphics cards in terms of monitor connectivity. This is expressed in the fact that up to four independent displays can be connected to an integrated system with a Trinity processor simultaneously, while all types of connections are supported: analog - VGA - and digital - DVI, HDMI and Display Port 1.2, as well as four independent audio streams. True, the number of physical outputs is limited to three, and to connect four displays, you will need to connect a pair of monitors with a "chain" via Display Port.

Even more impressive, Trinity graphics also support Eyefinity technology. Of course, in order to find some game that can work with an acceptable FPS level on three or four monitors connected to the Devastator, you have to work hard, but the very existence of such a possibility speaks of the attention with which AMD developers approached equipping the second generation APU before its introduction to the mass market.

Trinity lineup

Speaking about the graphics core of Trinity desktop processors, it is necessary to touch upon the composition of their model range. The fact is that different representatives of the A-series with the Trinity design can be equipped with different variants of the Devastator core. Their differences are formed in a standard way: trying to introduce segmentation of their products into different price categories, the manufacturer in the lower modifications disables one or more SIMD engines. As a result, only the older APU modifications have the set of resources described in detail in the previous section, which includes 384 actuators.

The Trinity model nomenclature is as follows. The fastest models with a full-fledged Devastator core, which has the marketing name Radeon HD 7660D, belong exclusively to the new flagship A10 series. All other modifications with graphics cores with a reduced number of stream processors and lower frequencies belong to the "simpler" A8, A6 and A4 series, replacing processors with the old Llano design.

The complete lineup of APUs based on the Trinity design is shown in the table:

Trinity APU Specifications
Model number	A10-5800K	A10-5700	A8-5600K	A8-5500	A6-5400K	A4-5300
Integrated graphics	HD 7660D	HD 7660D	HD 7560D	HD 7560D	HD 7540D	HD 7480D
TDP, W	100	65	100	65	65	65
Number of unified shader processors	384	384	256	256	192	128
GPU frequency, MHz	800	800	760	760	760	723
Number of Cores	4	4	4	4	2	2
CPU frequency, GHz (base / turbo)	3,8 / 4,2	3,4 / 4,0	3,6 / 3,9	3,2 / 3,7	3,6 / 3,8	3,4 / 3,6
L2 cache, MB	4	4	4	4	1	1
Maximum memory frequency	DDR3-1866	DDR3-1866	DDR3-1866	DDR3-1866	DDR3-1866	DDR3-1600

Even the version of the graphics core installed in the A8-class processors is theoretically more than 35 percent slower than the full-fledged Devastator. What can we say about the even slower A6 and A4. This means that the A10-5800K and A10-5700 processors are primarily of interest for use as a gaming solution. It is them that you can try to imagine in entry-level gaming systems, devoid of a discrete video card. Processors of the lower series, perhaps, are very poorly suited for universal gaming computers, so they are recommended to be used in multimedia centers or in home entertainment systems that are not aimed at running resource-intensive 3D gaming applications.

That is why in this material we focused on testing the oldest hybrid processor - A10-5800K, with a built-in Radeon HD 7660D graphics core. This processor has at its disposal two Piledriver modules, due to which it is recognized by diagnostic utilities and the operating system as a quad-core. However, we also note the existence of an alternative opinion, according to which this processor is dual-core, but with the ability to execute four threads. Actually, this opinion, although it contradicts the statements of AMD itself, just more accurately reflects the positioning of the A10-5800K. At its cost, this APU falls into the same price category as Intel's Core i3, which, as you know, are dual-core, but with support for Hyper-Threading technology.

The operating frequency of the processor in question, taking into account its support for Turbo Core 3.0 technology, should range from 3.8 to 4.2 GHz. In practice, however, we have seen that under load, the A10-5800K spends most of its time in an intermediate state - at a frequency of 4.0 GHz.

The Radeon HD 7660D graphics core built into the A10-5800K operates at 800 MHz, and when there is no 3D load, it drops to 300 MHz. Despite the fact that AMD promised the operation of the turbo mode for the graphics core, in reality, its frequency does not rise above the 800 MHz specified in the specifications.

How we tested

As part of this material, we set ourselves the goal of examining the graphics core performance of new AMD APUs and, based on the results obtained, answer the question: can the most modern processors with integrated graphics be used as part of entry-level gaming systems without adding discrete graphics cards.

In testing, the AMD A10-5800K processor with Radeon HD 7660D graphics core was contrasted with other integrated chips on the market that have 3D graphics with an acceptable level of performance. Firstly, these are AMD Llano, although they are outdated with the advent of Trinity, but still relevant, presented in our tests by the older processor of this family, AMD A8-3870K with the Radeon HD 6550D video core. Secondly - representatives of the Intel Ivy Bridge family, the maximum version of the graphics core of which, HD Graphics 4000, has promising (according to its developers) 3D performance. Intel graphics defended the dual-core processor Core i3-3225. We chose it, and not a quad-core of the Core i5 family, since AMD's APUs are positioned as an alternative to Intel's dual-core processors by the manufacturer itself. In particular, according to preliminary data, the cost of AMD A10-5800K will be about the same as that of the younger members of the Core i3 family.

In addition, we should not forget about the findings of our past research, showing a higher specific efficiency of the cores of Intel processors. Quad-core processors with Sandy Bridge microarchitecture quite successfully withstood the eight-core Bulldozer processors, and it is unlikely that with the release of new generations of Ivy Bridge and Piledriver microarchitectures, this situation has somehow changed. This can be confirmed by the relative results of the SYSmark 2012 test, which show the generally used performance of processors.

Although the AMD A10-5800K received a noticeably higher performance than the AMD A8-3870K, it lags behind the Core i3-3225 and Core i3-2125 processors, not to mention its significant loss in computing performance to the four-core Core i5-3330. So the opposition of AMD's quad-core APUs in graphics tests to a dual-core Core i3 is quite justified. In addition, the differences in the graphics performance of the most powerful Intel Core i7 and the Core i3 we chose boil down to a 100-MHz difference in the frequency of the integrated video core: 1.05 GHz for our test subject versus 1.15 GHz for the flagship processor for Socket LGA1155. So, no other Intel processor will be able to show a fundamentally better result than the Core i3-3225 in graphics tests.

In order for us to be able to judge the performance level of the integrated graphics cores of modern processors relative to discrete graphics cards, a variant equipped with external graphics was also added to the number of tested configurations. The reference point was the video card Radeon HD 6570, the cost of which today in the version we use with GDDR5 memory is about $ 70. It was tested on a system with an A10-5800 processor.

As a result, the following hardware and software components were used in the tests:

• Processors:
  • AMD A10-5800K (Trinity, 4 cores, 3.8-4.2 GHz, 4 MB L2, Radeon HD 7660D);
  • AMD A8-3870K (Llano, 4 cores, 3.0 GHz, 4 MB L2, Radeon HD 6550D);
  • Intel Core i3-3225 (Ivy Bridge, 2 cores + HT, 3.3 GHz, 3 MB L3, HD Graphics 4000).
• Motherboards:
  • ASUS P8Z77-V Deluxe (LGA1155, Intel Z77 Express);
  • ASUS F2A85-V Pro (Socket FM2, AMD A85);
  • Gigabyte GA-A75-UD4H (Socket FM1, AMD A75).
• Video card: AMD Radeon HD 6570 1 GB GDDR5 128-bit.
• Memory: 2 x 4 GB, DDR3-1866 SDRAM, 9-11-9-27 (Kingston KHX1866C9D3K2 / 8GX).
• Disk subsystem: Crucial m4 256 GB (CT256M4SSD2).
• PSU: Corsair AX1200i (80 Plus Platinum, 1200W).
• Operating system: Microsoft Windows 7 SP1 Ultimate x64.
• Drivers:
  • AMD Catalyst 12.8 Driver;
  • AMD Chipset Driver 12.8;
  • Intel Chipset Driver 9.3.0.1019;
  • Intel Graphics Media Accelerator Driver 15.26.12.2761;
  • Intel Management Engine Driver 8.1.0.1248;
  • Intel Rapid Storage Technology 11.2.0.1006.

When testing the platform based on the AMD A10-5800K processor, the operating system patches KB2645594 and KB2646060 were installed to adapt the scheduler behavior to the Bulldozer microarchitecture.

The main emphasis in this testing was quite naturally placed on gaming applications of the integrated processor graphics. Therefore, the bulk of the benchmarks we used are games or specialized gaming tests. Moreover, if we take into account the goals set, we were primarily interested in the performance of various graphic solutions in the de facto standard for desktop systems Full HD-resolution 1980x1080. Therefore, most of the tests were carried out in it with a low or medium level of image quality.

3D performance

The 3DMark test results are very popular for assessing the weighted average gaming performance of video cards. Therefore, we turned to 3DMark in the first place. First, let's take a look at the performance in the Vantage version, which uses DirectX version 10.

The significant progress that has occurred with AMD's APUs in the transition from the Sumo graphics core to the new Devastator design is immediately evident. The advantage of the Trinity processor over the flagship of the Llano family is about 40 percent. As a result, the system based on the A10-5800K approaches the graphics performance of the platform with a discrete AMD Radeon HD 6570 graphics card.

A more recent version of 3DMark is focused on measuring DirectX 11 performance. Previously, Intel processors could not take part in such tests, leaving AMD's APUs alone, but the Intel HD Graphics 4000 graphics core implemented in Ivy Bridge finally received support for all modern software interfaces, so the Core i3-3225 processor is also present on this diagram.

3DMark 11 came up with an extremely interesting result. According to this benchmark, the graphics core built into the A10-5800K was able to bypass the discrete Radeon HD 6570. This is an excellent illustration of the high efficiency of the VLIW4 architecture used in Devastator. Recall that the video card Radeon HD 6570 is based on an 800 MHz Turks graphics processor with VLIW5 architecture and at the same time has 480 stream processors against 384 in the Devastator. However, a larger number of executive devices, as we see, does not always translate into best practical indicators, from which we can conclude that the choice for Trinity VLIW4 design is a very correct decision.

Aliens vs. Predator (2010)

Despite the fact that in the synthetic benchmark 3DMark 11, the graphics core of the A10-5800K processor was able to overtake the discrete video card Radeon HD 6570, in a real gaming application - Aliens vs. Predator - the situation is completely different. Here the discrete video accelerator is seriously ahead of any version of integrated graphics, including the Radeon HD 7660D. It is obvious that the weak point of any processor video accelerators is the memory bus, which clearly has insufficient bandwidth. It should be noted, however, that we are comparing the Radeon HD 7660D here with the Radeon HD 6570 equipped with high bandwidth GDDR5 memory. But if a "simple" discrete video card with DDR3 SDRAM was used in the tests, then it would probably have been defeated by the Devastator core.

Batman: arkham city

The difference in performance between the old and new graphics cores used in AMD's APUs in Batman: Arkham City is about 30 percent. So from the point of view of graphics performance, the transition from Llano design to Trinity design is a completely justified decision, bringing tangible dividends. At the same time, such a step was made not at all due to heightened competition with Intel: even the newest and fastest GPU of the microprocessor giant looks very dull against the background of AMD's proposals. Obviously, AMD is aiming at signing a death sentence for budget graphics cards with DDR3 memory, such as the Radeon HD 6570 or GeForce GT 630.

Battlefield 3

Of course, the Radeon HD 7660D is not at all the same as a high-end or mid-range discrete graphics card. The performance of this solution is significantly lower. However, as we can see, the new integrated graphics core from AMD allows you to play quite decently the most modern games in Full HD-resolution, including Battlefield 3. Sometimes this requires setting low quality settings, but the average number of frames per second is at an acceptable level. Doesn't demonstrate Radeon HD 7660D and obvious drawdowns. For example, when tested in Battlefield 3, the minimum instant performance with low quality settings was quite decent, albeit not quite playable, 18 frames per second.

Borderlands 2

Even the newest first-person shooter Borderlands 2 runs without any problems on the A10-5800K. Of course, you will have to forget about the "niceties", but the fresh APU from AMD, unlike Intel processors with integrated graphics, makes it possible to play Borderlands 2 in 1920x1080 resolution without installing a discrete video accelerator.

Car racing games are usually not very graphics intensive. The behavior of F1 2012 is typical in this regard - this game runs on integrated systems with good performance even when choosing Full HD-resolution and high image quality. At the same time, although the advantage of the Radeon HD 7660D over the graphics from the Llano processor is close to 35 percent, the discrete Radeon HD 6570 still shows a slightly higher result. However, compared to the graphics core of competing processors, Intel HD Graphics 4000, any of AMD's integrated offerings look great. In F1 2012, the A10-5800K processor outperforms the Core i3-3225 by about 60 percent.

Far cry 2

We are deliberately not leaving Far Cry 2 out of the test suite. The presence of this four-year-old shooter allows you to see firsthand that in the games of the previous generation, the modern APU of the Trinity class works with simply outstanding performance. For example, in Far Cry 2, for example, we were able to set the resolution to 1920x1080 with the maximum available image quality and at the same time got an average of more than 30 frames per second. At the same time, the minimum FPS recorded in testing was quite acceptable 23 frames per second.

Sleeping dogs

Unfortunately, in the most modern of our selected games, the graphics core of the A10-5800K processor again demonstrates its inability to withstand the full-fledged Radeon HD 6570 video card, lagging behind it by about 10-15 percent. The source of the APU problem is clear - it would benefit from higher bandwidth memory. That is why the proliferation of solutions like Trinity can greatly revive the DDR3 SDRAM market. In common applications, the speed of operation depends on the memory frequency quite subtly, but for systems with integrated graphics, a fast memory subsystem can be of fundamental importance. However, we will pay more detailed attention to this issue.

Sniper elite v2

The Devastator core in the Radeon HD 7660D version is the fastest available on the market for embedded GPUs. The results obtained in the Sniper Elite V2 benchmark confirm this once again. The new version of the integrated graphics core, developed by AMD, outperforms the previous Sumo by 26 and 43 percent, depending on the image quality settings. As a result, the superiority of the Radeon HD 7660D over Intel HD Graphics 4000 reaches double the value. In other words, in terms of GPUs built into the processor, AMD continues to significantly outperform its competitor. Moreover, to the progress made by Intel with the release of the Ivy Bridge microarchitecture, AMD found an equally impressive answer - Trinity. So that the current APUs of both companies again fall into completely different weight categories.

Cinebench R11.5

All the games we tested on are DirectX applications. However, we also wanted to see how accelerators cope with work in OpenGL. Therefore, to purely gaming tests, we have added a small study of performance when working in the professional graphics package Cinema 4D.

The balance of power is quite typical. Trinity's performance level in an OpenGL application does not qualitatively differ from its speed in gaming DirectX tasks. The Radeon HD 7660D accelerator integrated into the AMD A10-5800K processor is ahead of its predecessor and Intel's competitor, but lags behind the discrete Radeon HD 6570 graphics card. so absurd. Moreover, in the assortment of AMD there are even corresponding offers - "professional" Trinity processors, sold under the FirePro trademark.

GPGPU performance

AMD is relentlessly emphasizing that its Llano and now Trinity processors are APUs. This means that their architecture is optimized for solving problems of various classes by using not only traditional x86 cores, but also graphics core stream processors - they must work together. For the successful functioning of such a community of fundamentally different computing resources, of course, specialized software is required. And if a year ago it sounded like a verdict on the APU concept, now the situation has begun to actively change. Developers of a number of popular software products began to make concrete attempts to take advantage of hybrid solutions. Today, there is information that the computing capabilities of the graphics core may involve current or future versions of programs such as Adobe Flash 11.2, Adobe Photoshop CS6, GIMP, ArcSoft MediaConverter 7.5, CyberLink MediaEspresso 6.5, Handbrake and WinZip 16.5.

As part of this material, we do not yet have the right to resort to testing the Trinity processor in such software, nevertheless, we can evaluate the practical performance of the Devastator graphics core on a GPGPU load created through the OpenCL and Microsoft DirectCompute APIs. For this we used the SiSoftware Sandra 2012.10.18.74 test suite.

The computational performance of the Devastator graphics core looks very good. The use of the VLIW4 architecture in its basis allows achieving high efficiency of general-purpose computing, as a result of which the Radeon HD 7660D noticeably outperforms not only the previous version of the graphics accelerator from Llano and Intel's graphics core Intel HD Graphics 4000, but also the discrete graphics card Radeon HD 6570. in applications that support OpenCL, you can expect high levels of performance from Trinity.

The situation is similar in the cryptographic test. In other words, by placing high-performance graphics with VLIW4 architecture in new hybrid processors, AMD was striving to solve a very specific problem - to demonstrate the usefulness and prospects of combining general-purpose x86 cores and streaming graphics cores. Considering that software makers are starting to try hybrid processors in business, this is a very timely move. At this stage, AMD should not only demonstrate the potential of new approaches, but also prove their advantage in practice.

conclusions

The days when it was necessary to approach integrated graphics from the position of “just to work” are long gone. Since the introduction of graphics cores into central processors, AMD and Intel have been actively expanding their power, displacing budget graphics cards from the market and giving their processors new use models. In this race of integrated GPUs, AMD is leading the way, with the fastest GPUs from the Ivy Bridge processors so far not beating even Llano graphics, let alone the new Trinity. However, this state of affairs was not an excuse for AMD to slow down the pace of innovation. This company is fighting not with a specific competitor's product, but for changing the attitude towards hybrid processors in principle. This requires not a simple superiority over alternative products in benchmarks, but its other quality.

It is very likely that the new Trinity desktop processors that we met today are the very necessary qualitative leap. The AMD A10-5800K is not just a hybrid processor with the fastest graphics core to date. The important thing is that the speed of this core is already sufficient to provide acceptable performance in almost any modern 3D games in Full HD resolution. Of course, in this case, you have to set not the maximum quality settings, but the fact remains: Trinity looks quite decently on a par with discrete 3D-accelerators of the lower level, costing about $ 60-70, which a new hybrid processor can easily replace. In fact, today we can say that accelerators such as the Radeon HD 6570 and GeForce GT 630 with the Trinity distribution can be sent to a landfill, at least this applies to their DDR3 modifications.

Today we got acquainted only with the graphics component of the new promising AMD project. And this component is his strong point. In terms of general performance, Trinity probably won't be as impressive as it gets. Even the 25% speed increase promised by AMD itself is clearly not enough for the A10-5800K, like other products in the family, to be able to perform on an equal footing with Intel's Ivy Bridge generation processors. Of course, we can count on the fact that AMD will be able to push the APU concept, and the hybrid offerings of this manufacturer will receive a noticeable increase in performance due to the computing resources of the graphics core. However, if this does happen, it will obviously not happen very soon. Therefore, for now, you have to keep in mind that Trinity also has a weak side.

What is the bottom line? Think about it, most Intel desktop processor buyers, by and large, don't give a damn about their graphics performance. They are ready to put up with any level of it, since they are attracted by the high speed of x86 cores. Trinity, on the other hand, may well win the favor of consumers by going from the other side. If this APU offers an alluring level of 3D performance, is it worth worrying so much about the slower speed of the x86 cores than the competitor? The answer to this question, judging by the available data, may well be negative: for most typical tasks, the available Trinity performance is probably quite enough.

However, let's not rush to final conclusions and wait until the embargo on the publication of the full test results is lifted. While you are reading these lines, work on the continuation of the material is already underway.