Comparing AMD FM1 and FM2 sockets. AMD processors for the FM1 platform Are am3 and fm2 sockets compatible?

#Socket_FM2_plus #Socket_FM2

For entry-level gaming computers, AMD has developed processors called APUs. They combine medium-performance processor cores with very fast graphics cards by the standards of integrated video. Modern APUs use FM2 and FM2+ sockets.

Processors with the FM2 socket first appeared in 2012; they were completely incompatible with the first generation APUs on the FM1 socket. have Trinity and Richland cores.

FM2 was replaced in 2014 by . The new socket was also compatible with FM2, so when upgrading the system, you could first purchase a new one for use with the old processor, and then a new processor.

Socket FM2 plus	Socket FM2

If you plan to use integrated graphics, then you should be careful when choosing a processor. For FM2 and FM2+ processor sockets, there are models without a built-in video core, which usually belong to the Athlon series. If you do not pay attention to this, then, despite the presence of video outputs on , the system will not start without installing a discrete card.

Let's turn to the characteristics of the processors. For comparison, let’s choose the Richland (FM2) and Kaveri (FM2+) cores:

Property	Socket FM2	Socket FM2 plus
Core	Richland	Kaveri
Technological process, microns	0.032	0.028
Clock frequencies, MHz	3400-4400	3100-4000
System bus frequencies, MHz	5000	5000
Processor-chipset bus bandwidth		4 GB/s (2 GB/s one way)
Heat dissipation, W	25-100	45-95
L1 cache size, KB	96 x2	128 x2
Internal L2 cache size, KB	2048 x2	2048 x2
Number of conveyor stages	18~22	18~22
Maximum number of instructions per clock cycle	4 x2	4 x4
Memory Types Supported	DDR3, LV-DDR3, 2 channels	DDR3, LV-DDR3, 2 channels
Supported memory bus frequencies		800, 1066, 1333, 1667, 1600, 1866, 2133 MHz
Maximum supported memory size	64 GB	64 GB
Embedded video (title)	Radeon HD 8670D, Radeon HD 8570D, Radeon HD 8470D	Radeon R7 (GCN 1.1)

Unlike most other processors with different sockets, AMD solutions usually work with the same chipsets, and for this reason it makes no sense to compare them. Here are the characteristics of the latest generation chipsets.

Property	AMD A88X	AMD A78
Components of m/c	FCH A88X	FCH A78
Maximum power consumption, W	7.8	7.8
System bus	UMI x4 Gen2 5 GT/s	UMI x4 Gen2 5 GT/s
PCI support (version)	2.3	2.3
Maximum number of PCI slots	3	3
PCI Express support (version)	2	2
Maximum number of PCI Express slots		4 slots using up to 4 lines
SLI/Crossfire support	Crossfire	Crossfire
Number of USB ports	12	12
USB2.0 support		Yes (only for 10 of 12 ports)
USB 3.0 support	4 ports	4 ports
SerialATA support	8 SATA 6Gb/s channels with support for splitters	6 SATA 6Gb/s channels with support for splitters
RAID support	0, 1, 10, 5 JBOD from SATA devices	0, 1, 10, JBOD from SATA devices
AC support	Intel High Definition Audio	Intel High Definition Audio
Other Features		Built-in SD-Card controller

As you can see, the chipsets differ only in the number of SATA ports and supported RAID levels.

Let's look at the performance of older APUs for sockets FM2 (A10-6800K) and FM2+ (A10-7870K)

As you can see in the table, the performance of the processors has remained virtually unchanged. The main difference between the platforms is the built-in video core. The results of their performance measurements in the 3DMark test are in the table:

Test	A10-6800K, 4.1 GHz, 4 cores	A10-7870K, 3.9 GHz, 4 cores
FutureMark 3DMark 2013 Ice Storm 1920*1080 Preset Extreme	39956	27198
FutureMark 3DMark 2013 Cloud Gate 1920*1080 Preset Extreme	2209	3439
FutureMark 3DMark 2013 Fire Strike 1920*1080 Preset Extreme	770	1174

Based on the performance measurements, it can be seen that the new ones cope much better with the complex DX10 and DX11 graphics used in Cloud Gate and Fire Strike, showing a serious increase in performance, but in the DX9 (Ice Storm) test the result is better.

When choosing a processor, you should pay attention, in addition to performance, to its power consumption. The higher the power consumption, the higher the heating of the chip and, accordingly, the higher the requirements for the cooling system. We measured power consumption in two states: idle and full load.

The new ones have better energy efficiency, that is, with similar performance parameters they consume noticeably less energy. In addition, there is now support for the proprietary Mantle API, which allows games that support this API to get a performance boost.

Hi all. In this article you can fully familiarize yourself with the main characteristics of FM2 and FM2+ processors. The list includes all processors: From A4 series to A10 series. In the list of all FM2/FM2+ socket processors, the processors are arranged by decreasing performance (at least I tried to arrange them that way): from the most powerful to the weakest. In the processor frequency column, you can see the value in brackets - this is the processor turbo frequency (or frequency in Boost mode). Prices for processors were taken from the cheapest online stores and are constantly updated. Therefore, you don’t have to worry about the relevance of prices for FM2+/FM2 processors.

FM2+ processors								Processor name	Price	Cores	Frequency	Video card	Cache memory	Power	Technical process	A10-7890K	RUR 4,274	4	4.1(4.3) GHz	R7(866 MHz)	4 MB	95 W	28 nm	Athlon x4 880K	RUB 5,845	4	4(4.2) GHz	No	4 MB	95 W	28 nm	A10-7870K	RUB 3,785	4	3.9(4.1) GHz	R7(1100 MHz)	4 MB	95 W	28 nm	Athlon x4 870K	RUB 3,156	4	3.9(4.1) GHz	No	4 MB	95 W	28 nm	A10-7850K	RUR 3,486	4	3.7(4) GHz	R7(757 MHz)	4 MB	95 W	28 nm	Athlon x4 860K	RUB 1,894	4	3.7(4) GHz	No	4 MB	95 W	28 nm	A10-8750/A10 PRO-8750B	RUB 2,815	4	3.6(4) GHz	R7(757 MHz)	4 MB	65 W	28 nm	A10-7860K/A10 PRO-7850B	RUB 3,550	4	3.6(4) GHz	R7(757 MHz)	4 MB	65 W	28 nm	FX-770K	?	4	3.5(3.9) GHz	No	4 MB	65 W	28 nm	A10-7800/A10 PRO-7800B	RUB 2,761	4	3.5(3.9) GHz	R7(720 MHz)	4 MB	65 W	28 nm	A8-7680	?	4	3.5(3.8) GHz	R7(1029 MHz)	2 MB	45 W	28 nm	Athlon x4 845	RUB 2,827	4	3.5(3.8) GHz	No	4 MB	65 W	28 nm	A10-7700K	RUR 3,017	4	3.4(3.8) GHz	R7(720 MHz)	4 MB	95 W	28 nm	A8-8650/A8 PRO-8650B	RUB 1,975	4	3.2(3.8) GHz	R7(757 MHz)	4 MB	65 W	28 nm	Athlon x4 840	RUB 1,577	4	3.1(3.8) GHz	No	4 MB	65 W	28 nm	A8-7690K	?	4	3.7 GHz	R7(757 MHz)	4 MB	95 W	28 nm	A8-7670K	RUB 4,491	4	3.6(3.9) GHz	R7(758 MHz)	4 MB	95 W	28 nm	A8-7650K	RUB 3,525	4	3.3(3.8) GHz	R7(757 MHz)	4 MB	95 W	28 nm	Athlon x4 850	RUB 1,603	4	3.2 GHz	No	4 MB	65 W	28 nm	A8-7600/A8 PRO-7600B	RUR 3,086	4	3.1(3.8) GHz	R7(757 MHz)	4 MB	65 W	28 nm	Athlon x4 835	RUB 2,521	4	3.1 GHz	No	4 MB	65 W	28 nm	Athlon X4 830	?	4	3(3.4) GHz	No	4 MB	65 W	28 nm	A8-7500	?	4	3 GHz	R7	4 MB	65 W	28 nm	A6-7470K	?	2	3.7(4) GHz	R5(800 MHz)	1 MB	65 W	28 nm	A6-8550/A6 PRO-8550B	?	2	3.7(4) GHz	R5(800 MHz)	1 MB	65 W	28 nm	A6-7400K/A6 PRO-7400B	RUB 2,170	2	3.5(3.9) GHz	R5(800 MHz)	1 MB	65 W	28 nm	A4-8350	RUB 1,314	2	3.5(3.9) GHz	R5(757 MHz)	1 MB	65 W	28 nm	Athlon x2 450	RUB 1,775	2	3.5(3.9) GHz	No	1 MB	65 W	28 nm	A6-7480	?	2	3.5(3.8) GHz	R5(900 MHz)	1 MB	65 W	28 nm	A4 PRO-7350B	?	2	3.4(3.8) GHz	R5(515 MHz)	1 MB	65 W	28 nm	FM2 processors								Processor name	Price	Cores	Frequency	Video card	Cache memory	Power	Technical process	A10-6800K	RUB 2,885	4	4.1(4.4) GHz	8670D(866 MHz)	4 MB	100 W	32 nm	A10-6790K	RUB 3,551	4	4(4.3) GHz	8670D(866 MHz)	4 MB	100 W	32 nm	A8-6600K	RUB 2,170	4	3.9(4.2) GHz	8570D(844 MHz)	4 MB	100 W	32 nm	A10-5800K	RUB 2,578	4	3.8(4.2) GHz	7660D(800 MHz)	4 MB	100 W	32 nm	AMD FirePro A320	?	4	3.8(4.2) GHz	8570D(800 MHz)	4 MB	100 W	32 nm	Athlon X4 760K	RUB 1,030	4	3.8(4.1) GHz	No	4 MB	100 W	32 nm	FX-670K	?	4	3.7(4.3) GHz	No	4 MB	65 W	32 nm	A10-6700	RUB 2,419	4	3.7(4.3) GHz	8670D(866 MHz)	4 MB	65 W	32 nm	A8-5600K	RUB 2,367	4	3.6(3.9) GHz	7660D(760 MHz)	4 MB	100 W	32 nm	A8-6500	RUB 2,412	4	3.5(4.1) GHz	8570D(800 MHz)	4 MB	65 W	32 nm	Athlon x4 750K	953 RUR	4	3.4(4) GHz	No	4 MB	100 W	32 nm	Athlon x4 750	?	4	3.4(4) GHz	No	4 MB	100 W	32 nm	FirePro A300	?	4	3.4(4) GHz	7660D(760 MHz)	4 MB	65 W	32 nm	A10-5700	RUB 2,156	4	3.4(4) GHz	7660D(760 MHz)	4 MB	65 W	32 nm	A8-5500	RUB 2,269	4	3.2(3.7) GHz	7560D(760 MHz)	4 MB	65 W	32 nm	Athlon x4 740	887 RUR	4	3.2(3.7) GHz	No	4 MB	65 W	32 nm	Athlon x4 730	722 RUR	4	2.8 GHz	No	4 MB	65 W	32 nm	A8-6700T	?	4	2.5(3.5) GHz	8670D(758 MHz)	4 MB	45 W	32 nm	A8-6500T

A socket, as you know, is a connector on the motherboard for installing a central processor. Sockets differ in form factor, number of contacts and type of fastening. The use of sockets, in principle, is intended to facilitate system upgrades by simply changing the processor. However, the problem is that the release of almost every new processor from AMD or Intel is associated with a transition to a new platform, that is, with the appearance of a new socket.

This is clearly seen, in particular, in the example of the latest FM1 and FM2 sockets, designed to connect powerful AMD hybrid processors. The FM1 platform was developed for Llano processors, which were released not so long ago - in mid-2011. However, when developing the new Komodo and Trinity family of processors, AMD decided to abandon the use of the FM1 socket in favor of the new FM2 platform. In this short article we will try to figure out what this threatens users with and whether there are significant design differences between the FM1 and FM2 sockets.

Platforms FM1 and FM2

Socket FM1 is a processor socket with 905 pins. It was developed specifically for hybrid APU processors from AMD based on the Fusion architecture. We are talking, first of all, about Liano hybrid processors, which, due to the presence of an integrated graphics core, required not only a new design. Liano processors from AMD were presented in dual or quad-core versions with support for Direct X 11 GPU and DDR3 1600 RAM. All motherboards released with Socket FM1 for installing Liano processors adopted the UEFI system instead of the traditional BIOS. In the desktop segment, Liano processors and, accordingly, the FM1 platform debuted on June 30, 2011.

It seemed that the next generation of AMD hybrid processors would also be Socket FM1. However, the appearance of AMD Llano processors on the market was assessed ambiguously by computer enthusiasts and overclocking enthusiasts, for whom the new product was actually designed. Although the powerful integrated graphics core provided a good level of performance comparable to the performance of junior discrete video cards, the Liano processors did not bring the expected increase in frequency potential. And while AMD Llano solutions were quite competitive in the mobile segment, their popularity in desktop systems turned out to be low.

AMD has decided to rely on a new generation of Trinity hybrid processors with more powerful graphics and computing cores. Creating a more powerful processor for desktop systems required abandoning the existing FM1 platform. This is how socket FM2 appeared, which is structurally different from FM1 in a slightly different arrangement of contacts.

AMD's new Trinity processors are based on an improved Piledriver architecture and feature powerful integrated graphics. They have a dual-channel DDR3 memory controller that supports operation in modes up to DDR3 1866. One of the main differences between Trinity chips and their predecessor Liano processors is their higher clock speeds. If Liano processors managed to get close to the 3 GHz mark, then older Trinity models can already be overclocked to 3.8 GHz - 4.2 GHz.

Despite the fact that older Trinity models have slightly fewer shader units than Llano, this is more than compensated for by the use of VLIW4 multiprocessor units, acceleration of the tessellation processing unit and a higher clock frequency. The integrated Trinity graphics core has full support for DirectX 11 with ShaderModel 5.0, OpenCL 1.1 and DirectCompute 11. Solutions on the FM1 socket, by the way, did not provide the ability to use two graphics adapters in the system at once. The new FM2 platform with Trinity processors is aimed at a wide range of users who are interested in building fairly powerful multimedia desktop PCs.

Differences and compatibility of sockets FM1 and FM2

In general, the FM2 socket is a logical continuation of the FM1 platform, so the differences between the two connectors were not too significant. Upon careful examination, one can be convinced that even appearance The FM2 socket has not undergone radical changes compared to the previous platform. However, these changes still exist. Although the pin layout of both sockets looks similar, the FM2 is missing one of the pins in the central part. Thus, if the FM1 processor socket had 905 pins, then new platform– only 904.

In addition, the so-called “keys”, that is, areas without contacts, are located in different places on the substrate for Llano and Trinity processors. Unfortunately, a different location of the “keys” will not even allow installing the AMDTrinity processor into the old FM1 socket. Some other subtle changes to the FM2 socket are related to power delivery.

AMD representatives have for a long time given rather evasive answers to the question of whether the FM1 and FM2 platforms will ultimately be compatible. This was probably done so as not to indirectly reduce the demand for processors with socket FM1. But today it is already known that the new AMD hybrid processors have neither direct nor backward compatibility with the FM1 platform.

This means that desktop users with AMD Liano processors will need to purchase motherboards that support socket FM2 to upgrade to the latest Trinity processors. This incompatibility is understandable, because the new AMD processors are based on a completely different architecture, which required a transition to different power subsystems. This circumstance forced AMD to switch to the new Socket FM2 platform. However, owners of desktop PCs with the FM1 platform were unlikely to be satisfied with this decision.

Prospects for sockets FM1 and FM2

AMD has earned user recognition not only for its powerful and cost-effective solutions, but also for the fact that it has always strived to maintain the same design for several generations of its processors. This provided users with the ability to easily and quickly upgrade their PC by purchasing and installing a new processor. So the policy of frequently changing sockets has never been distinctive feature AMD. That is why the rejection of the FM1 platform actually gave rise to a lot of discontent among a serious part of AMD product supporters.

With the emergence of the new FM2 platform, the company's management de facto recognized Llano hybrid processors and accompanying motherboards with the FM1 socket as a “dead-end” solution. It is clear that the previous generation platform with the lack of upgrade options is unlikely to have any success among users. It can be assumed that the FM1 socket, which was released seemingly not so long ago, will have a short life on the market.

With the FM2 platform, as AMD assures us, everything will be different. This processor socket will not become “single serial”, as happened with FM1, but will be aimed at supporting several future generations of AMD processors. However, given the not very pleasant history with the release of first-generation hybrid processors, potential consumers may have concerns and questions for AMD about whether the FM2 platform is really serious and long-lasting. Perhaps in the near future, in connection with the development of new, more productive solutions, the company will again have to switch to a completely different processor socket.

Be that as it may, at present a number of manufacturers have already announced the release of motherboards with the FM2 socket for new AMD processors. This is, for example, flagship model GA-F2A85X-UP4 from Gigabyte and Hi-Fi A85W board from Biostar. Everything suggests that the choice of motherboards with FM2 connector will become quite wide in the near future.

Company AMD introduced the second generation of hybrid processors for desktop systems. Chips Trinity are based on the improved Piledriver architecture and also have a powerful integrated video core. Mobile versions New generation processors from AMD have been offered as part of laptops for almost six months now. An attractive combination of consumer parameters allowed the company to increase its share in this segment. Let's see if Trinity desktop versions designed for the new platform will be so successful Socket FM2.

What are the new hybrid processors with the code name? Trinity? In the maximum configuration, these chips include a quad-core x86 computing unit with the most advanced AMD architecture to date - Piledriver. This is a further development of the Bulldozer architecture, which is used for the fastest AMD FX series chips. In addition, the chip houses a graphics core, which the manufacturer classifies as Radeon series HD 7000.

Trinity, although they are successors to Llano processors, has practically nothing in common between them. Both the computing part and the graphic part in this case are not just improved, they are fundamentally different. Perhaps the only thing that connects both generations of APUs is the 32-nanometer process technology, which is also used for Trinity. Of course, a more advanced technological process would be preferable here, but GlobalFoundries production facilities are not yet ready for mass production of chips using technology thinner than 32 nm.

The Trinity die area is 246 mm² and contains 1.3 billion transistors, while the silicon wafer of the Llano chip occupies 228 mm² and carries 1.18 billion transistors (after a recent clarification of this figure by the manufacturer). The packaging density remained approximately the same, the area increased by about 8%, while the number of semiconductors increased by 10%. Taking into account the timing of the development of the 32-nanometer technical process, we assume that the cost of producing crystals has increased, if at all, but only slightly.

What's new in Trinity? The dual-channel DDR3 memory controller officially supports operation in modes up to DDR3-1866, and it is also possible to use modules with a reduced supply voltage (1.25 V). As you can see, almost half of the crystal is occupied by the graphic part. The built-in GPU has the architecture inherent in chips for discrete adapters of the family Northern Islands. An important innovation is the AMD HD Media Accelerator video encoding/decoding unit. The northbridge functions of the chipset are, of course, now integrated into the processor. As for computing power, Trinity has a pair of dual-core x86 modules. Within each of them, the cores are partially dependent, since they use some common resources, in particular, instruction prefetch and real number processing (FP) units. Each module has a dedicated 2 MB L2 cache segment. There is no third-level cache memory here - this is the prerogative of the AMD FX series CPU. For communication with external devices, the processor has at its disposal 24 PCI Express lanes. Let's celebrate the support HDMI interfaces, DisplayPort 1.2 and DVI.

Trinity processors initially operate at fairly high clock speeds. While Llano chips have just reached the 3 GHz bar, the older model from the new APU family operates at 3.8 GHz as normal, with the ability to accelerate to 4.2 GHz. Trinity received the latest modification of the dynamic acceleration mechanism AMD Turbo Core 3.0, which, depending on the nature of the load, can automatically increase the CPU frequency. Each processor model has its own range: from 200 to 600 MHz.

Integrated Graphics

Introducing a term APU(Accelerated Processing Unit), the company initially wanted to emphasize the importance of the built-in graphics unit. The integrated Trinity graphics core, called Devastator, uses architecture VLIW4, which was used for the Radeon HD 6900 Northern Islands family. Obviously, the developers have not yet managed to optimize the new GCN (Graphics Core Next) architecture for the needs of the APU, which is used in GPUs for discrete video cards of the Radeon HD 7000 series.

Let us remember that the graphics part of Llano chips has VLIW5 architecture. The compute units it includes can theoretically perform more operations in parallel than those with VLIW4. However, in real problems the latter turn out to be more effective. In addition, VLIW4 stream processors, all other things being equal, can operate at a higher clock frequency. It is quite difficult to draw parallels here, but some quantitative indicators are interesting. IN full version The Llano graphics core contains 400 computing units, while the Trinity GPU has 384, but in the latter case the standard operating frequency of the graphics unit is 800 MHz, while its predecessor has 600 MHz.

The Devastator core includes 24 texture units and 8 rasterization units. AMD emphasizes that in this case the tessellation processing unit is noticeably accelerated. A dedicated hardware unit is dedicated to working with video data. AMD HD Media Accelerator, which includes the most advanced video decoding module UVD3, inherited from the processor from Radeon HD 6000/7000. In addition, the processor contains a video transcoding unit AMD Accelerated Video Converter. Functionally, it is similar to Quick Sync, which Intel uses in its processors.

Overall, the Trinity graphics core has excellent functionality. It boasts full support for DirectX 11 with Shader Model 5.0, OpenCL 1.1 and DirectCompute 11. Moreover, the new APUs allow you to connect up to four independent display devices; moreover, support for Eyefinity technology is also announced. It is also worth noting the support AMD Steady Video 2.0, which allows you to improve video quality by helping to get rid of the effect of image shake obtained during handheld shooting.

Like their predecessors, Trinity processors have the ability to operate in Dual Graphics, combining the efforts of an integrated GPU with a discrete graphics card. However, in this case we are still talking about entry-level devices from the Radeon HD 6500/6600 lines.

To support the A10 chips, the manufacturer recommends using the Radeon HD 6670; for the A8 and A6, the Radeon HD 6570 is offered, while for the A4 – the HD 6450. In fact, it is possible to use the Dual Graphics mode, but in the current conditions such combinations are interesting in cases where The potential owner of a Socket FM2 system already has a video card that can be used as an additional accelerator. The deliberate purchase of an adapter of the required class for use in Dual Graphics mode, although it has the right to exist as a deferred upgrade option, however, in general it loses to the idea of ​​​​purchasing a faster graphics adapter, which will cost a little more, but in games will be noticeably more productive than the proposed combination.

Piledriver architecture

The Piledriver architecture is an upgraded version of the Bulldozer that is used for Zambezi (AM3+) chips.

The branch prediction and data prefetch units have been improved, the efficiency of working with the second level cache has been increased, the L1 TLB volume has been increased, and the work of the INT and FP module load scheduler has also been improved. Additionally, the new F16C instruction sets are now supported, as well as FMA3, which Intel plans to add to its Haswell chips. AVX sets are now available for new APUs, which were not supported by Llano chips. In general, Piledriver is not fundamentally different from the Bulldozer architecture; it is a modified version with a number of improvements and cosmetic optimizations.

APU Trinity lineup

At the time of the launch of the new platform, the line of chips Trinity includes six models. Two quad-core processors A10 and A8, as well as one each A6 and A4. As you can see, the name of the APU series does not reflect the number of x86 blocks in any way. At the same time, there is a dependence on whether the chip belongs to a particular line, which is determined by the number of computing cores of the integrated graphics: A10 – 384, A8 – 256, A6 – 192, A4 – 128. This is another clear example of how the manufacturer wants emphasize the importance of the graphic component.

The flagship of the line – A10–5800K– operates at 3.8/4.2 GHz, its built-in GPU contains 384 computers and operates at 800 MHz. The L2 cache capacity is 4 MB, and the declared power consumption level is 100 W. The second “ten” has the same characteristics, with the exception of the frequency formula. For A10-5700 The base clock is 3.4 GHz, and the dynamic auto-overclocking limit is 4 GHz. This was enough to reduce the TDP to 65 W. For A8 models, in addition to the reduced number of video core computing units from 384 to 256, its operating frequency is also reduced to 760 MHz. Formulas for x86 blocks: A8-5600K– 3.6/3.9 GHz, A8-5500– 3.6/3.8 GHz. Single-module A6 and A4 chips, in addition to losing two x86 blocks, have a common L2 cache of only 1 MB. The number of GPUs is reduced to 196 in the case of A6-5400K, and up to 128 – y A4-5300.

As for the cost of new APUs, Trinity chips play in virtually the same price segment, as its predecessors – $50–130. At the same time, the pricing system is interesting. Both A10s are priced at $122. Both the model with an unlocked multiplier and the chip with a lower clock frequency and a locked GPU, which nevertheless has a TDP of 65 W, instead of the flagship’s 100 W, have the same recommended price. The situation is exactly the same with the A8 line of APUs - both models are offered at the same price of $101. For some, higher performance is valuable, while for others, more economical options are preferable. For both, suitable processors will cost the same price.

As is the case with Llano processors, as well as devices from competitors, models with the “K” index have an unlocked multiplier. It’s interesting that now the most affordable model with this feature costs only $67, while the price for the previous generation APU with a free multiplier started at $80. However, the A6-3670K is a quad-core model, while the A6-5400K is equipped with only one module with a pair of dependent modules.

Processors with a disabled graphics core will also be available for Socket FM2, which will complement the Athlon chip line. Considering the general concept of the APU, it is obvious that separate crystals will not be produced for such models (although, given the area occupied by the GPU, this would make sense); for such processors, chips will primarily be used, with certain problems in the graphics part, and if there are fewer of them than the market demands, then full-fledged crystals with a deactivated GPU will be used.

Compatibility of Socket FM1 and Socket FM2

Unfortunately for owners of systems with first-wave hybrid chips, the new APUs have neither direct nor backward compatibility with the Socket FM1 platform. The processor socket, and accordingly the legs on the chip, visually have minimal differences (905 vs. 904), however, the different arrangement of the “keys” does not even allow Trinity to be installed in the old socket.

(left – APU Trinity, right – APU Llano)

For quite a long time, AMD gave evasive answers to questions about the compatibility of FM2 and FM1 sockets, so as not to indirectly reduce the demand for processors for the latter. Now there is no need for this. Considering that the new APUs are fundamentally different from their predecessors at the architectural level, it is not surprising that they have their own power subsystem features that were not taken into account within Socket FM1. It was this fact that forced AMD to change the platform.

Chipsets

Despite the fact that Socket FM1 and Socket FM2 are incompatible with each other, the chipsets used on previous generation platforms are quite suitable for the new one. Chips AMD A55, and AMD A75 we will see it included in motherboards for Socket FM2. In general, there is nothing surprising here. Given the fact that key functions chipsets are taken over by central processors, their role in modern platforms is largely reduced to maintenance peripheral devices. But here innovations don’t happen so often. If there are already certain complaints about the functionality of the AMD A55 (lack of SATA 6 Gb/s), then the AMD A75 cannot be called outdated. The latter became the first chipset in the industry with an integrated native USB 3.0 controller. And the rest of the body kit is quite up to par.

To make the Socket FM2 announcement even more exciting, AMD also introduced a new chipset that will be used for this platform - AMD A85X. One of its key differences from the A75 is its separation ability PCI-E bus x16 for two devices (x8+x8), and, as a result, the ability to create CrossFire configurations with a pair of discrete video cards. In addition, the A85X now supports 8, rather than 6, SATA 6 Gb/s ports and allows you to create RAID 5 disk arrays. It also provides FIS-Based Switching channel separation capabilities. There are no changes regarding USB bus support and configuration: 4 USB port 3.0, up to 10 USB 2.0 ports and up to two USB 1.1.

The Socket FM1 platform did not provide the ability to use two graphics adapters in the system. Such configurations are the lot of fairly enthusiastic gamers or experienced crunchers. It is obvious that in the case of Socket FM2, AMD wants to make the most universal platform that could interest users with different needs in terms of performance and functionality.

Upgrade prospects

Considering the experience with the release of the platform for the first generation APU, AMD hastened to assure potential buyers of new solutions that Socket FM2– this is serious and for a long time. At least one more generation of hybrid chips will use this connector, and accordingly, they will be able to be installed on motherboards that are now on sale.

The lack of upgradeability and the very short lifespan of Socket FM1 are important reasons for the generally muted enthusiasm for the previous generation platform. Yes, we can agree that this is not a segment in which the issue of modernization is paramount. However, for users who pay money for a new solution, the prospect of an upgrade is often important even if in reality the need for it does not arise until it becomes completely obsolete. With Socket FM2 everything should be fine in this regard. It will remain relevant for at least 2–3 years.

All motherboard manufacturers have already presented their solutions with Socket FM2 connectors. It is curious that vendors have focused on models with different chipsets. Some presented a whole set of devices based on the most affordable AMD A55 and several boards based on the top-end AMD A85X, without attracting the A75 at all, while others, on the contrary, relied on the latest chipset, maximally diversifying their offerings based on it. All this suggests that the range of devices for Socket FM2 will be very wide, so it will be easier for users to choose a device that suits their requirements. As for prices, in our opinion, the range here will be only slightly wider than in the case of boards for Socket FM1 – $50–120.

AMD A10-5800K processor

We received the top model of the new Trinity APU line for testing - AMD A10-5800K.

Gigabyte GA-F2A85X-UP4 motherboard

For research Socket platforms FM2 we used the older model in the current line of boards from Gigabyte - GA-F2A85X-UP4, based on the new AMD A85X chipset.

The board complies with the latest specification Ultra Durable 5, which involves the use of high-quality energy-efficient components. Eight-phase power stabilizer (6+2). IN power circuit powerful IR3550 assemblies are used, as well as chokes with ferrite cores. A digital controller is used to control VRM parameters.

The layout of slots for expansion cards is optimal. Three PCI-E x16, the same number of PCI-E x1 and one PCI. The latter does not require an additional controller, since support for this bus is still implemented in AMD chipsets. Considering the number of PCI Express lanes, nuances in the use of slots cannot be avoided. The first slot operates in full speed mode by default. When using two video cards, the first and second slots are switched to x8+x8 mode. The third full-length PCI-E x16 has x4 bandwidth, and if the nearest PCI-E x1 is used, the lower PCI-E x16 will also provide data transfer rates at the x1 level. Gigabyte GA-F2A85X-UP4 allows you to fully realize the advantages of the A85X chipset - the model allows you to create a configuration with two video cards on AMD chips that will operate in CrossFireX mode.

On board Gigabyte GA-F2A85X-UP4 There is a gentleman's set of overclockers - Power, Reset, Clear CMOS buttons, as well as an LED status indicator. The board is expectedly equipped with two BIOS chips, and the UEFI shell uses a graphical version of the 3D BIOS, which is already conceptually familiar to us from the manufacturer’s previous boards.

Among the interesting features of the model, we note the technology Dual Clock Gen. The board has a chip with an additional clock generator (the main one is in the chipset). According to the manufacturer, it allows for stable operation at higher bus clock frequencies (~135–150 MHz), which may be of interest to owners of APUs with locked multipliers who want to boost their processor. Although, of course, given AMD’s pricing policy for Trinity chips, it is better for enthusiasts to initially look towards models with the “K” index.

The board has a full set of video outputs: DVI, HDMI, DisplayPort and D-Sub. In this case, you can simultaneously connect up to three display devices with any combination of interfaces. Note that the DVI port operates in Dual-Link mode, allowing the use of monitors with resolutions up to 2560×1600.

The disk subsystem will allow you to connect 8 drives via SATA 6 Gb/s: seven internal and one via eSATA. As for peripherals, the user has six USB 3.0 ports at his disposal. Four of them are implemented using the chipset, another two use an additional Etron EJ168 controller.

Overall, the board leaves a pretty good impression. A decent set of functions for an older solution, nothing superfluous and at the same time a good foundation for the future.

Performance

To evaluate the possibilities AMD A10-5800K, we have selected worthy opponents for him. First of all, it's the processor. AMD A8-3850. This chip differs from the older model of the previous generation APU line (A8-3870K) only in a 100 MHz lower clock frequency and a locked processor multiplier, while the integrated graphics part uses the most powerful Radeon HD 6550D. A model of the same price category is presented from the main competitor - a dual-core processor Intel Core i3-3220 from the new line of 22nm Ivy Bridge chips. First of all, let's check how the CPU unit works.

Trinity's compute performance is on average slightly better than Llano's (+5-10%), although given the notable architectural differences, the difference may vary depending on the applications used. In some cases, first-generation APUs with four full-fledged cores can be even faster than a pair of dual-core modules running at significantly more high frequency. In application tasks, Trinity is not lost against the background of dual-core Intel Core i3, offering quite decent performance for its price. In single-threaded tasks, the Intel processor will definitely have an advantage; the phenomenal efficiency of the Intel Core architecture makes itself felt. But in multi-threaded tasks, the number of computational units matters a lot, and here quad-core CPUs from AMD have an advantage. Of course, Intel processors with the same number of cores are even more powerful, but they are significantly more expensive.

During the test of the new APU, we also decided to evaluate the effectiveness of the combination CPU+GPU in applied tasks, using for these purposes graphics editor Musemage, which uses graphics core resources to perform various operations. The list of stages also included the SVPMark benchmark, which can also connect graphics for video processing.

The range of programs interspersed with heterogeneous computing is gradually expanding. Moreover, this is not only synthetic software for tests, but also applied applications. The pace, of course, leaves much to be desired, but there is hope that such initiatives from developers will be encouraged in every possible way by hardware manufacturers. This is a rare case when the interests of both competitors coincide. Intel also places greater emphasis on the performance and capabilities of its integrated video with each subsequent architectural iteration. Ivy Bridge chips have significantly improved here compared to their predecessors, and in the expected Haswell, the graphics core should receive an even more significant increase in performance. In the meantime, AMD has a noticeably stronger position here.

In 3D synthetics, Trinity has a very solid performance increase of 40–45%. Of course, the overall ranking also takes into account the increased performance of the x86 unit, but this is not bad. 6000 points in 3DMark Vantage is almost the level of the Radeon HD 6570, that is, a discrete video card that is now offered for $50–60. The performance of the Intel HD Graphics 2500 looks noticeably more modest compared to the “built-in” ones from AMD.

Intel offers separate modifications of processors equipped with Intel HD Graphics 4000. In the case of dual-core models of the Ivy Bridge line, this is Core i3-3225. It also has an operating clock frequency, like the Core i3-3220 - 3.3 GHz, but is equipped with a full-fledged graphics module with 16 computing units (the HD Graphics 2500 has only six), although it costs $20–25 more. At the time of writing, we did not have such a model, however, in order to include in the review not only the results of the Intel HD Graphics 2500, but also the performance of Intel's most powerful integrated graphics solution at the moment, we used the Core i7-3770K. It only appears in gaming tests with embedded video. This will allow a more balanced assessment of the current position and potential capabilities of the integrated GPUs of both companies.

In real-world games, the A10-5800K again outperforms the A8-3850 very comfortably. The advantage is not as great as in the case of tests from Futuremark, but an increase of 25–35% can also be considered an excellent result. In addition, an average of 30 fps in a resolution of 1920x1080 already allows you to not only view pictures in not the most simple games.

Intel's decisions are expectedly less hasty, especially in the case of lightweight GPU options. It would seem that Intel HD Graphics 4000 had just managed to get even remotely close to Llano's performance when Trinity chips again made this mission impossible. We hope that with the release of Haswell there will be some intrigue here again.

The capabilities of integrated video seriously depend on the performance of the memory subsystem. Let's see how it is with the A10-5800K RAM bandwidth affects gaming performance.

If we compare AMD processors under such conditions, then, as we see, in most cases, the A10-5800K has a slight advantage (2–5%). Mafia II, in which the system with the new APU received a 10% increase, can rather be considered an exception. In addition, the opposite situations are also possible, as evidenced by the results in Lost Planet 2, where the A8-3850 outperformed the newcomer by almost 5%. However, in any case, the competition here is only between AMD chips. The results demonstrated by a PC with a dual-core Core i3-3220 processor are beyond their reach. The gap from the pursuers is 7–18%. Even despite the smaller number of computational units, the dual-core Ivy Bridge chip turns out to be extremely efficient in games, and here AMD processors cannot be helped even by twice the number of computational units. On the other hand, the difference does not look depressing and the discrete video card does most of the work here.

In general, the increase in Trinity computing performance is relatively small and averages 5–15%. Despite the fact that full-fledged Llano computing cores in some cases are still preferable to dual modules, due to internal improvements in the architecture, as well as higher frequencies, chips based on Piledriver manage to outperform their predecessors. The capabilities of integrated graphics were more pleasing. A 30% advantage over its predecessor, which before the advent of Trinity was a kind of benchmark in terms of the capabilities of the built-in GPU, inspires optimism.

Energy consumption

Having gotten a general idea of ​​the performance of the Trinity APU, we were also interested in assessing the level of power consumption of the new AMD processors. The declared TDP parameter for the A10-5800K is 100 W, let's look at the real performance in typical tasks.

When the computing units are under load (rendering in Cinebench), the consumption of Llano and Trinity is approximately the same level. But the increase in graphics core power did not go unnoticed. In games where the GPU is heavily loaded, the power consumption of the A10-5800K is 18 W higher than its predecessor. The manufacturing process remains the same, but higher clock frequencies make themselves felt. At the same time, it is worth noting that in rest mode, in which the processor often remains most of the time, the energy efficiency of new APUs is higher. However, here it is worth making allowance for the fact that both processors use different motherboards, which can affect the absolute figures.

Dual-core Intel Core i3s generally demonstrate exemplary efficiency. The CPU spends minimal energy on computational tasks, but when assessing performance in games, it is worth taking into account the significant difference in the performance of solutions.

Results

Platform Socket FM2 and processors Trinity are quite an interesting option for assembling fairly powerful multimedia PCs. Compared to its predecessors, the performance of computing units with Piledriver architecture has not increased so significantly, while the capabilities of integrated graphics have been improved by a third, reaching the performance of entry-level discrete video cards. At the moment, this is a serious advantage of AMD solutions. At the same time, the range of Trinity chips is exactly the same as that of Llano. Considering the balanced price, they will look very organic as part of inexpensive universal solutions “for everything”. And although recently for such tasks they are increasingly purchasing mobile systems, new desktop APUs will also find their buyers.

Every time we buy an AMD-based computer, we ask ourselves which processor and socket to choose? Especially now that AMD changes them almost every year. Will there be a prospect of replacing the processor in the future and what is the old processor good for? It is also important to know when there is a bunch of old hardware with different performance. And from all this you need to assemble a computer with tolerable performance. This table shows that there is a decent range for creativity. Especially for overclockers and gamers, a large amount of stale hardware accumulates. And it makes sense to rummage through the mezzanines and assemble, for example, a computer for the dacha, or for a younger brother/sister.

CPU	motherboards
		AM2	AM2+	AM3	AM3+	FM1	FM2	+ – Compatible; – Theoretically compatible, but compatibility in each specific case must be clarified on the motherboard manufacturer’s website; - - Absolutely incompatible.
	AM2	+	+	—	—	—	—
	AM2+		+	—	—	—	—
	AM3			+	+	—	—
	AM3+	—	—	—	+	—	—
	FM1	—	—	—	—	+	—
	FM2	—	—	—	—	—	+

It is clear from the table that, unfortunately, contrary to popular belief, sockets FM1 and FM2 are absolutely incompatible. Here you need to choose, go for the more expensive one motherboard And budget processor, or build a powerful PC, but on the previous socket. In my opinion, the solutions are equivalent. For example you purchased powerful computer on the outgoing socket, no problem, you will use it for several years. Although if you build a PC on a new socket, there is a prospect of installing a more powerful and more economical CPU in a year.