Presentation of Intel Sandy Bridge processors: model range and architectural features. From Sandy Bridge to Skylake: Comparison of specific performance

The image of the 28-nuclear crystal of the SkyLake-Sp processor. Then we noticed that the location of the kernels and interfaces has undergone significant changes. Yesterday, Intel was explained at one of the home events, with which these changes are connected in design. As it turned out, in the future Intel refuses (and already refused for sKYLAKE processors In Xeon versions and desktop solutions of higher performance) from the intra-processor ring tire.

Ring Bus (Ring Bus) was presented in 2008 along with the NEHALEM architecture and Westmere-EX processors. It was necessary due to the increase in the number of nuclei on the crystal. Intel's developers used three design options for processors (depending on the maximum number of nuclei on the crystal) with three variants of the ring bus. In the most difficult case, the processor inside was divided into two clusters, each of which was served by two annular tires. Between the tires were connected to bidirectional buffering switches (on the diagram above marked with gray).

As the number of cores grows, the annular tire has become an obstacle to increasing bandwidth and reduce delays. More precisely, it has become too much to consume that it can be scaled towards an increase in the data exchange rate. Therefore, in the SkyLake-SP processors, the Intel developers decided to apply a different structure to communicate with each other - well-tested in the architecture of Intel Xeon Phi (Knights Landing) with a cellular network.

Each kernel in the new architecture has its own buffer switch and is associated with any other core in the processor only through two nodes - the outgoing and incoming. This allows the cellular bus to work on relatively small frequencies and significantly reduce the overall intake of the interface without deterioration in the bandwidth and increase delays. In addition, such a structure of communications is very well scaled, allowing Intel in the future to increase the number of nuclei on the crystal without a noticeable increase in the energy costs for internal data transportation.

Explaining the essence of the new inner tire, as well as the appearance of an image of a 18-nuclear processor with a new design, also allows you to make sure that new processors really carry an integrated 6-channel memory controller, which is now spaced along the edge on the side of the crystal just above the middle.

The capabilities of the Sandy Bridge graphics processor are generally comparable to those of the previous generation of such intel solutions, Unless, in addition to the features of DirectX 10 added support for DirectX 10.1, instead of expected support for DirectX 11. Accordingly, not many OpenGL support applications are limited to hardware compatibility only with the 3rd version of the specification of this free API.

Nevertheless, there are many innovations in the Sandy Bridge chart, and they are focused mainly to increase productivity when working with 3D graphics.

The main emphasis when developing a new graphics core, according to Intel representatives, was made on the maximum use of hardware capabilities for 3D functions, and the same - for processing media data. This approach is radically different from a fully programmable hardware model adopted, for example, in NVIDIA, or in the Intel itself to develop Larrabee (with the exception of textural blocks).

However, in the implementation of Sandy Bridge, the departure from programmable flexibility has its undeniable advantages, due to it is achieved more important for integrated graphics benefits in the form of less latency in the execution of operations, better performance on the background of saving energy consumption, simplified programming model for drivers, and what is important, with savings of the physical size of the graphic module.

For programmable executive shader modules Sandy Bridge graphics, according to the traditions called in Intel "Executive Blocks" (EU, Execution Units), the increased size of the register file is characterized, which makes it possible to achieve the effective execution of complex shaders. Also in new executive blocks used branching to achieve better parallelization of executable commands.

In general, according to representatives of Intel, new executive blocks have a double-capacity double-standing compared to the previous generation, and the performance of calculations with transcendental numbers (trigonometry, natural logarithms, and so on) due to the accent on the use of hardware computing capabilities of the model will grow in 4 -20 times.

The internal set of commands, reinforced in the Sandy Bridge near new, allows you to distribute most of the DIRECTX 10 API instructions in the one to one mode, as in the case of the CISC architecture, which results in a significantly higher performance with the same clock frequency.

Quick access via a quick ring bus to a distributed L3 cache with dynamically configurable segmentation allows you to reduce latency, lift performance and at the same time reduce the frequency of graphical processor appeals to random access memory.

Ring tire

The whole history of the upgrade of the processor microarchitectures Intel in recent years is inextricably linked with consistent integration into a single crystal more and more modules and functions, previously located outside the processor: in the chipset, on the motherboard, etc. Accordingly, as the productivity of the processor and the degree of integration of the chip increases, the requirements for the capacity of internal intercomponent tires grew by a leading pace. For the time being, even after the introduction of a graphic chip in the Arrandale / Clarkdale chip architecture, it was possible to manage intercomponent tires with a familiar cross-axis topology - this was enough.

However, the effectiveness of such topology is high only with a small number of components participating in the exchange of data. In the Sandy Bridge microarchitecture to increase the overall performance of the system, the developers decided to refer to the ring topology of a 256-bit intercomponent tire (Fig. 6.1), made on the basis of new version QPI technologies (QuickPath Interconnect), advanced, improved and first implemented in the architecture of the NEHALEM-EX server chip (Xeon 7500), as well as planned to be applied together with the architecture of Larrabee chips.

Ring Interconnect in the Sandy Bridge architecture version for desktop and mobile systems It serves to exchange data between six key components of the chip: four CPU cores x86, a graphical core, L3 cache, now it is called the LLC (Last Level Cache) and the system agent. The bus consists of four 32-byte rings: data bus (Data Ring), query bus (Request Ring), status monitoring bus (SNOOP RING) and confirmation bus (ACKNOWEDGE RING), in practice it actually allows you to share access to the 64-byte interface Kesha last level into two different packages. Bus management is carried out using the communication protocol of distributed arbitration, while the conveyor processing of requests occurs on the clock frequency of the processor cores, which gives architecture additional flexibility during acceleration. The productivity of the ring bus is estimated at the level of 96 GB per second to the connection at a clock frequency of 3 GHz, which is actually four times higher than the Intel processor indicators of the previous generation.

Fig.6.1. Ring InterConnect

The annular topology and tire organization provides minimal latency when processing queries, maximum performance and excellent scalability of technology for versions of chips with different number of cores and other components. According to representatives of the company, in perspective to the ring bus can be "connected" to 20 processor cores on the crystal, and such a redesign, as you understand, can be made very quickly, in the form of a flexible and operational reaction to the current market needs. In addition, the physically ring bus is located directly above the L3 cache memory blocks in the upper metallization level, which simplifies the design layout and allows you to make a chip more compact.

Failure tires

Ring tire.It consists of spotted rings covering teeth from the vestibular side in the form of a strip, and is located in the occlusal part of the crown closer to the cutting edge (Fig. 26). From the paternal side, the ring expands and overlaps the dental tubercle. Rings, as a rule, are prepared from stamped crowns, but a solid design can be an option for such a tire. In the preparation of the teeth, interdental contact points on the thickness of the stamped crown to the lower edge of the ring are polished. For this, it is pre-on the diagnostic model with a chemical pencil, the boundaries of the rings, which subsequently serve as a reference with the preparation of teeth. Separation of contact surfaces facing each other is carried out on the thickness of two rings. The cutting edge is left open and this circumstance requires special care in determining the indications for the use of this tire. The pronounced vertical mobility of teeth, not closed from the cutting edge, may be the cause of cement resorption and tire fixation disorders. In addition, with a sharply expressed anatomical form of the lower front incisors, it takes a polishing of a rather significant layer of solid tissues from the contact surfaces to the lower edge of the ring, which makes it difficult to restore the contact surfaces on the stamped rings and reduces the accuracy of the rings to the surface of the tooth. It may also cause the resorption of the fixing cement and the development of caries.

Tire technology is as follows. In the first visit after a thorough examination and compilation of the dispute plan, it is necessary to remove the impression of alginate mass for the manufacture of diagnostic gypsum models. The parallelometer is determined by the topography of the center line, the models are fixed in the articulator and apply a pattern of the ring bus. On the same model there is a phantom preparation of tight teeth. The next visit to anesthesia displays teeth, strictly observing the boundaries of the phantom preparation. For the manufacture of the rings again remove the impression with the help of alginate mass. Removing the doublettage in patients with periodontal diseases may be difficult due to the mobility of individual teeth and the dangers of their removal. On the resulting prints, gypsum working models are cast, which are used to make stamped blanks of future rings. The resulting stamped crowns are used for the manufacture of rings that are checked in the oral cavity of the patient and if they meet the requirements with them remove the impression to translate the rings on the gypsum model. Before removing the impression, the contact surfaces of the rings facing each other are cleaned from the scale for the subsequent spike of the rings on the gypsum model without the preliminary removal, which ensures the accuracy of their mutual location in the manufacture of the tire. The finished tire after the spike rings is bleaching, polished and fixed in the oral cavity with special cement.

The disadvantages of the ring tire include: 1 - disruption of aesthetics of natural teeth, part of which is closed with a metal ring; 2 - the presence of solder often leads to its oxidation and a change in color in the form of darkening, especially often this is observed in patients with increased acidity of gastric juice; 3 - the absence of a pinning effect during vertical load; 4 - Tire requires the use of cements, very resistant to the effects of the oral fluid (if this condition is not respected, the danger of dental damage to the caries and tire fixation) occurs.

Fig. 26. Ring tire: A - view from the lip side; b - view from the paternal side; in - general view of the ring; G - Tooth Preparation Scheme: The dotted line indicates the edge of the rings; The left is shown to excessive removal of solid tissues from the contact surface; On the right - the correct preparation, when the rings protruding over the lower boundary of the rings removed exactly to the designated dotted line; D, E - Crawling Borders (front view and top)

Seawing tire.Constructively tire is built on the same principle as the ring. However, in order to increase the aesthetic properties of the tire, the middle part of the rings from the lip side is removed and, thus, the vestibular surface of the tooth in its middle part is released from the metal (Fig. 27). Thus, there remain short shoulders in the form of ribbon climers, covering their teeth completely from the paternal side and partly with the vestibule. The best busting effect is achieved when the complete reference crowns covering the extreme teeth - fangs are turned on into the tire. From a technological point of view, the tire is most practical in the manufacture of a solid structure, since the stamped semirings do not have the hardness needed to be reduced. In addition, currently there is an opportunity to cover cast semiring with decorative material - ceramics, which makes the tire very beneficial in aesthetic

Fig. 27. Semicolves Schip: A - view from the vestibular side; b - view from the paternal side

Cap tire.The system of paved caps covering the cutting edge, the contact surfaces of the tooth, and on the gear surface of the dental tuberca, is denoted as a cap tire (Fig. 28). The preparation edge and contact surfaces on the cap thickness are subjected. From the lip side, the edge of the cap can be located on top of the solid tooth tissues or end on a specially formed ledge. The second option is preferable, since the edge of the cap turns out to be lying on the same level with solid tissues adjacent to it, that is, flush. In the first version, the edge of the cap is often felt by patients, can injure the surrounding rolling mucous membrane of the oral cavity and requires the creation of a fold when moving the edge of the cap into solid tooth tissues. Caps can be manufactured in two ways: 1) from stamped crowns, 2) solid. The second option is considered more perfect because the accuracy of the entire pinning structure increases, which means that its shining effect increases, and, in addition, it becomes possible to favay the litter design with ceramics. For better stability, the bus is combined with complete crowns (metallometallic or metal-meal), covering the extreme most stable teeth - fangs or premolars. The sequence of manufacture is the same as when making a ring bus.

Fig. 28. Cap tire: A - view from the lip side; b - view from the paternal side; in - layer of removed solid tissues under the cap; g - stamped cap; d - preparation under the cast cap; C - Cast Cap Construction Cutting Cutting Edge

Tires used on vital teeth have one major advantage - the vitality of the pulp remains, which means there are no conditions for changing reactivity in periodontal tissues. However, it is often due to the proximity of the pulp, especially when erasing part of the cutting and chewing surfaces of the teeth, the use of a complex tire design that requires the formation of deep cavities, requires preliminary deputation of teeth. Of course, in the presence of depositated teeth, the manufacture of tires is much easier. Below we will look at such structures that are used on dsvitalizable teeth.

When using fixed tire designs, it is necessary to strictly follow the rules for placing the edge of the tires located near the desiye edge. The latter should not be injured with a tire. To this end, the edge of the crown should be minimally immersed in the gum-groove groove, and to prevent the possible pressure on the gum to apply the preparation method of teeth with a ledge almost at the level with it. Racking attitude towards patient periodontal when applying non-removable tires Beneficially affects the course of periodontal disease and is not an obstacle to conservative and surgical therapy. In addition, it is important in terms of preventing the injury of the desiye region is the method of obtaining implications. We consider the most optimal in this case to remove the prints for the manufacture of pin-structures with the most elastic alginate materials, allowing movable teeth to avoid random dental removal along with prints. The recommendations encountered in the special literature to shoot two-layer prints using silicone print materials, even with preliminary concrete, as shown observations are not acceptable, since the removal of two-layer impulsions may cause the removal of movable teeth.

Removable tires

In the question of the ways of sowing teeth there are different points of view. Some authors consider justified the preferential use of non-removable tires, while others, on the contrary, give the previously removable tires and the covered designs of removable prostheses. Moreover, the shining with removable structures can be used both with intact dental rows and in partial loss of teeth.

If necessary, replacing remote teeth with artificial restoration of the removable tire can be carried out without replacing the entire design.

Removable tires provide reliable stabilization primarily in the vestibulo-oral and mesio-distal direction. At the same time, the need for radical disubstation disintegration is eliminated, good conditions for hygienic care and medical surgical treatment are created both in the preparatory period and in the process of using the removable pinning structure.

With orthopedic treatment of periodontal diseases using removable tires, it is advisable to highlight two groups of patients:

with intact dental rows; Partial loss of teeth.

Removable Elbrecht bus.The tire is used for the preserved dental rows and is constructed by type of multi-sized clamps that ensure the immobilization of teeth in the horizontal plane, leaving them unprotected from the action of the vertical load, developing during chewing. Elements of cake clammers, occlusal linings and vestibular clawing processes allow you to achieve a good pinning effect.

Fig. 44. Removable tire Elbrecht: A - Elbrecht bus (explanation in the text); B - varieties of multi-sized (continuous) clammer: 1 - high position of the clammer (in the upper part of the gear surface) of the drop-shaped form; 2 - the location of the clammer in the middle part of the gear surface; 3 is the low position of the clammer (in the enhancing half of the germination surface); 4 - CLAMMER in the form of a wide strip

Removable tire with denot-alveolar clames in V.N. Copekin.The removable tire of Elbrecht was modified by V.N. Kopeykype, which suggested to enhance the retention properties and achieve a better aesthetic effect to use T-shaped clummers of rowcha. Multiple clammers in this design are lowered below the gauge edge and in the form of an arc there are a palate of alveolar processes of the front departments of jaws with vestibular and paternal sides. T-shaped clammers are deployed from them to each front teuba, whose shoulders are located in the zones of submution. The tire can be recommended for stable or movable 0-1 degrees of front teeth, when the pin-shaped properties of the holding of T-shaped clammers will not have a harmful effect on the patient periodontal (Fig. 45). To do this, it is necessary to place the shoulders of T-shaped clammers in such a way that they are outside the zone of undercutting. The fixing properties of the tire are provided by introducing into the zone of submuting those cast shoulders of the clammers, which are located on stable teeth with the least affected periodontail. This sewn as well as all other solid structures should be molded using refractory models. The removable tire of Elbrecht can be enhanced with arcs located on the paternal surface of the alveolar leaf of the lower jaw or the top of the sky (Fig. A, b). If such a pinning design is applied only to shining side teeth, a napseagittal stabilization is achieved (Fig. B, d).

Fig. Removable tires, reinforced arcs for the bottom (A) and the upper jaw (b). Tire design to create a para-sagittal stabilization: in - on the model; G - General Tire View

Fig. M.Removable spike with cast kappa for the front teeth: a - on the plaster model; b - frame of removable tire

Fig. 48. Removable shids for the front teeth; a - removable circular tire; B - removable spikes in the form of a continuous clamber with cogtevidymn process

In general, in the absence of several teeth and severe periodontal pathology, preference is given to removable prosthesis. The design of the prosthesis is selected strictly individually and requires several visits to the doctor.

Removable design requires careful planning and a certain sequence of actions:

Diagnosis and surveys of periodontal.

Preparation of the surface of the teeth and getting the casts for the future model

Studying Model and Tire Design Planning

Tire wax reproduction modeling

Obtaining a foundry and checking the accuracy of the frame on the plaster model

Checking tires (protein tires) in the oral cavity

Final finish (polishing) tires

Not all working steps are listed here, but even this list indicates the complexity of the procedure for manufacturing a removable tire (protein tires). The complexity of the manufacture explains the need for several sessions of the patient and the duration of time from the first to the last visit to the doctor. But the result of all efforts is always alone - the restoration of anatomy and physiology, leading to the restoration of health and social rehabilitation.

These days, Intel represents the world long-awaited processors Sandy Bridge., the architecture of which was painted in advance as a revolutionary one. But not only processors have become new in these days, and all the associated components of new desktop and mobile platforms.

So, this week is announced at 29 new processors, 10 chipsets and 4 wireless adapters For laptops and desktop workers and game computers.

Mobile novelties include:

intel Core i7-2920xm processors, Core i7-2820qm, Core i7-2720qm, Core i7-2630qm, Core i7-2620m, Core i7-2649m, Core i7-2629m, Core i7-2657m, Core i7-2617m, Core i5- 2540m, Core i5-2520m, Core i5-2410m, Core i5-2537M, Core i3-2310m;

intel QS67, QM67, HM67, HM65, UM67, HM65, UM67, HM65, UM67, EXPRESS;

wireless network controllers Intel Centrino. Advanced-n + WiMAX 6150, Centrino Advanced-N 6230, Centrino Advanced-N 6205, Centrino Wireless-N 1030.

The desktop segment will appear:

intel Core I7-2600K processors, Core i7-2600s, Core i7-2600, Core i5-2500k, Core i5-2500s, Core i5-2500T, Core i5-2500, Core i5-2400, Core i5-2400s, Core i5- 2390t, Core i5-2300;

intel P67, H67, Q67, Q65, B65 Express chipsets.

But it is worth noting that the announcement of the new platform is not one-piece for all models of processors and chipsets - from the beginning of January, only solutions of the Mainstream class are available, and most of the more massive and not such expensive will appear a little later. Together with the release of desktop processors, Sandy Bridge has a new processor connector for them LGA 1155.. Thus, the novelties do not complement the lineup Intel Core i3 / i5 / i7, and are a replacement for processors under LGA 1156, most of which are now becoming a completely non-prospective acquisition, because in the near future their release should stop at all. And only for enthusiasts by the end of the year Intel promises to continue the release of senior quad-core models on the Lynnfield core.

However, judging by the "Roadmap" platform, the Socket T (LGA 775) long-liver (LGA 775) will still remain relevant to at least the middle of the year, being the basis for the entry-level systems. For the most productive gaming systems and real enthusiasts, processors on the Bloomfield kernel on the LGA 1366 connector will be relevant to the end of the year. As you can see, the life cycle of dual-core processors with "integrated" graphic adapter At the Clarkdale kernel, it was very short, just one year, but they were "protusive" the track for the "today" of Sandy Bridge, after escaping the consumer to the thought that in the processor not only the memory controller, and the video card can be integrated into the processor. Now it's time not to simply release faster versions of such processors, seriously update the architecture to ensure a noticeable increase in their effectiveness.

The key features of the Sandy Bridge architecture processors are:

release in compliance with 32 nm technical process;

noticeably increased energy efficiency;

optimized Intel TURBO BOOST technology and Intel Hyper-Threading support;

a significant increase in the performance of the built-in graphics core;

implementing the new set of Intel Advanced Vector Extension (AVX) instructions to accelerate the processing of real numbers.

But all the above innovations would not provide opportunities to talk about a really new architecture, if all this were not implemented now within one nucleus (crystal), unlike processors on the Clarkdale kernel.

Naturally, all the processor nodes are agreed to be coordinated, it was necessary to organize a quick exchange of information between them - an annual architectural innovation was the ring bus Ring InterConnect.

Combines the same Ring InterConnect through L3 cache, now called LLC (Last Level Cache), processor kernels, graphics core and system agent (System Agent), which includes a memory controller, PCI Express bus controller, DMI controller, power management module and Other controllers and modules previously having the United Name "Uncore".

The Ring Interconnect ring bus is the next QPI tire development step (QuickPath InterConnect), which, after "running" in server processors with updated 8-nuclear architectures NEHALEM-EX, moved to the kernel of processors for desktop and mobile systems. Through Ring InterConnect, four 32-bit rings for data tires (Data Ring), query (Request Ring), Snoop Ring and Confirmation (ACKNOWLEDGE RING) are formed. The ring bus is functioning at the core frequency, so it bandwidthDelays and power consumption are completely dependent on the frequency of the processor computing blocks.

The third-level cache memory (LLC - Last Level Cache) is common to all computing cores, graphics core, system agent and other blocks. Wherein graphic driver Determines which data streams to place in the cache, but any other block can access all data in LLC. A special mechanism controls the distribution of cache memory so that the collisions do not occur. In order to accelerate the work for each of the processor nuclei, the "own" segment of the cache memory, to which it has direct access. Each such segment includes an independent access controller for the Ring InterConnect bus, but at the same time continuously interact with the system agent, which manufactures the overall management of cache memory.

The system agent (System Agent) is essentially built into the "Northern Bridge" processor and combines PCI Express, DMI tire controllers, RAM, video processing unit (media processor and interface management), power dispatcher and other auxiliary blocks. With the rest of the processor nodes, the system agent interacts through the annular tire. In addition to streamlining data streams, the system agent monitors the temperature and loading of various blocks, and through Power Control Unit provides the supply voltage control and frequencies in order to provide the best energy efficiency at high performance. Here it can be noted that for the power of new processors you need a three-component power stabilizer (or two, if the built-in video language remains inactive) - separately for computing kernels, system agent and an integrated video card.

The PCI Express bus embedded in the processor corresponds to Specifications 2.0 and has 16 lines to possibly increase the power of the graphic subsystem using a powerful external 3D accelerator. In case of using older sets system logic And the harmonization of licensing issues, these 16 lines can be divided into 2 or three slots in 8x + 8x or 8x + 4x + 4x modes, respectively, for NVIDIA SLI and / or AMD CrossFirex.

To exchange data with the system (drives, I / O ports, the periphery, the controllers of which are in the chipset) Using DMI 2.0 bus, allowing to pump up to 2 GB / s useful information In both directions.

An important part of the system agent is a two-channel DDR3 memory controller, nominally supporting modules at a frequency of 1066-1333 MHz, but when used in motherboardsah on the Intel P67 EXPRESS chipset without problems ensuring the operation of modules at a frequency up to 1600 and even 2133 MHz. Placing a memory controller in one crystal with processor cores (Clarkdale core consisted of two crystals) should reduce memory latency and, accordingly, increase system performance.

In part, thanks to the advanced monitoring of the parameters of all computing keys, cache and auxiliary blocks, which is implemented in Power Control Unit, improved Intel Turbo Boost 2.0 processors have appeared in the Sandy Bridge processors. Now, depending on the load and the tasks performed, the core of the processor at high need can be accelerated even with the excess of the thermal packet, as with the usual manual acceleration. But the system agent will follow the temperature of the processor and its components, and when the "overheating" of the nodes will be fixed to decrease gradually. However, in desktop processors, the operating time in supernatural mode is limited, because It is much easier to organize more efficient cooling at times than the "Boxing" cooler. Such a "overbust" will allow to obtain an increase in performance to the system critical for system, which should create the user's impression with a more powerful system, as well as reduce the time of the system's response. Also Intel Turbo Boost 2.0 ensures that desk computers Built-in video card has dynamic performance.

The architecture of Sandy Bridge processors implies not only changes in the structure of intercomponent interaction and the improvement of the possibilities and energy efficiency of these components, but also internal changes in each computing kernel. If you drop "cosmetic" improvements, then the most important will be the following:

return to the selection of cache for approximately 1.5 thousand decoded micro performations L0 (used in Pentium 4), which is a separate part of L1, which makes it possible to simultaneously provide more uniform loading of conveyors and reduce energy consumption due to increased pauses in the work of sufficiently complex operations decoders;

improving the efficiency of the branch prediction unit due to an increase in the capacity of the buffers of the results of branching results, team history, branch history, which increased the efficiency of conveyors;

increase the capacity of the reordrore command buffer (Rob - Reorder Buffer) and improving the efficiency of this part of the processor by implementing the physical register file (PRF - Physical Register File, too characteristic features Pentium 4) for data storage, as well as the expansion of other buffers;

doubling the capacitance of registers to work with streaming real data, which in some cases can provide a twice-greater speed of operations that use them;

an increase in the efficiency of enforcement of encryption instructions for AES, RSA and SHA algorithms;

introduction of new vector instructions Advanced Vector Extension (AVX);

• optimization of the operation of the Cash memory of the first L1 and the second L2 levels.

An important feature of the Sandy Bridge processor graphics core is that it is now in one crystal with other blocks, and its characteristics and condition tracking performs the system agent at the hardware level. At the same time, the media processing unit and signal generation for video outputs is submitted to this most system agent. Such integration provides closer interaction, smaller delays, greater efficiency, etc.

However, the architecture of the graphics core is not so many changes as I would like. Instead of expected support, DirectX 11 was simply added support for DirectX 10.1. Accordingly, not many OpenGL support applications are limited to hardware compatibility only with the 3rd version of the specification of this free API. At the same time, although it is stated on the improvement of computing blocks, but they remained as much - 12, and only for senior processors. However, an increase clock frequency Up to 1350 MHz promises a noticeable performance increase in any case.

On the other hand, create a built-in video language with really high performance and functionality for modern games at low power consumption is very hard. Therefore, the lack of support for the new API will affect only compatibility with new games, and performance with a really big desire to play comfortably to play with a discrete 3D accelerator. But the expansion of functionality when working with multimedia data, first of all when encoding and decoding video in the framework of Intel Clear Video Technology HD, can be found to the advantages of Intel HD Graphics II (Intel HD Graphics 2000/3000).

The updated media processor allows you to unload the processor kernel when encoding video in MPEG2 and H.264 formats, and also expands a set of post-processing functions with hardware implementation of algorithms for automatic adjustment adjustment (ACE - Adaptive Contrast Enhancement), color adjustments (TCC - Total Color Control) And improve skin display (SKIN TONE ENHANCEMENT). Enhances the prompt's use of the built-in video card implemented by the HDMI version 1.4 interface support, compatible with Blu-Ray 3D (Intel Intru 3D).

All the above, the listed architectural features provide a new generation of processors a noticeable superiority to speed over the models of the previous generation, both in computing tasks and when working with video.

As a result, the Intel LGA 1155 platform becomes more productive and functional, coming to shift LGA 1156.

If you summarize, then the processors of the Sandy Bridge family are designed to solve a very wide range of tasks with high energy efficiency, which should be done really mass in new productive systems, especially when more affordable models will appear in a wide range.

In the near future, 8 processors for desktop systems of different levels will be available to buyers: Intel Core i7-2600K, Intel Core i7-2600, Intel Core i5-2500K, Intel Core i5-2500, Intel Core i5-2400, Intel Core i5-2300 , Intel Core i3-2120 and Intel Core i3-2100. Models with index k are distinguished by a free factor and a faster integrated video adapter Intel HD Graphics 3000.

Also, energy-efficient (index S) and high-energy efficient (index T) of the model are also released for systems that are critical to energy consumption of systems.

To support new processors, parents are available today intel chipsets P67 Express and Intel H67 Express, and in the near future is expected on Intel Q67 Express and Intel B65 Express, oriented on corporate users and small business. All these chipsets finally began maintaining drives with sATA interface 3.0, although not all ports. But support seemingly even more sought-after USB 3.0 tires in them. Interesting features New chipsets for ordinary motherboards became the fact that they refused to support the PCI tire. In addition, now the clock generator is built into the chipset and control its characteristics without consequences for the stability of the system, only in a very small range, if only ± 10 MHz is lucky, and in practice and less.

It should also be noted that different chipsets are optimized for use with different processors in systems intended for various purposes. That is, Intel P67 Express from Intel H67 Express differs not only to the lack of support for working with embedded video, but also advanced features for "overclocking" and tuning performance. In turn, Intel H67 Express does not notice the free multiplier in models with the index of K.

But due to architectural features, the acceleration of the Sandy Bridge processors is still possible only by means of a multiplier if this is the K-series model. Although all models are inclined to some optimization and "overbust".

Thus, temporarily to create an illusion of work on a very powerful processor Even the models with a blocked multiplier are capable of noticeable acceleration. The time of such an acceleration for desktop systems, as mentioned above, is limited hardware, and not only the temperature as in the mobile PC.

After the presentation of all architectural features and innovations, as well as updated branded technologies, it remains only to summon once again than Sandy Bridge such innovative and remind of positioning.

For high-performance and mass productive systems, in the near future, the Intel Core i7 and Intel Core i5 series can be used, which are different with the support of Intel Hyper-Threading technology (for quad-core Intel Core i5 models, it is disabled) and the third-level cache. For more economical buyers are new intel models Core i3, which have 2 times less computing cores, albeit Intel Hyper-Threading, only 3 MB LLC cache, do not support Intel Turbo Boost 2.0 and all are equipped with Intel HD Graphics 2000.

In the middle of the year, Intel Pentium processors will be presented for mass systems (from this brand it is very hard to refuse, although a year ago it was propheted) based on a very simplified architecture of Sandy Bridge. In fact, these processors for "work horses" will be reminded by the possibilities of yesterday the current Core i3-3xx on the Clarkdale kernel, because Almost all functions inherent in senior models for LGA 1155, they will lose.

It remains to note that the release of Sandy Bridge processors and a whole desktop platform LGA 1155 became another "so" in the framework of the concept of Intel "Tik-so", i.e. A serious update of architecture for the release of the already debugged 32 nm technical process. About a year later, we will be waiting for the IVY Bridge processors with an optimized architecture and made by 22 nm in the process, which, for sure, will again have "revolutionary energy efficiency", but hopefully the processor connector LGA 1155 will not be abolished. Well, wait - see. In the meantime, we have a minimum year to study the architecture of Sandy Bridge and its comprehensive testing What is going to start in the coming days.

An article read 14627 times (a)

Subscribe to our channels

Term topology network means a way to connect computers to the network. You can also hear other names - network Structure or network configuration (This is the same). In addition, the concept of topology includes many rules that determine the placement of computers, the cable laying methods, methods for placing the binder equipment and much more. To date, several main topologies have been formed and established. From them can be noted " tire”, “ring"And" star.”.

Topology "Tire"

Topology tire (or, how else is often called total tire or highway ) It assumes the use of one cable to which all workstations are connected. The shared cable is used by all stations in turn. All messages sent by individual workstations are accepted and listened to all other computers connected to the network. From this stream, each workstation selects only messages addressed to it.

Advantages of the topology "tire":

• simplicity settings;
• relative ease of installation and low cost, if all workstations are located nearby;
• the failure of one or several workstations does not affect the operation of the entire network.

Disadvantages of the Top Topology:

• tire problems anywhere (cable breaking, the failure of the network connector) leads to non-working network;
• the complexity of troubleshooting;
• low performance - at each time only one computer can transmit data to the network, with an increase in the number of workstations network performance drops;
• bad scalability - To add new workstations, it is necessary to replace the sections of the existing bus.

It is on the topology "tire" built local networks on coaxial cable. In this case, the segments of the coaxial cable were protruded as a tire, connected by T-connectors. The tire was laid through all the rooms and approached each computer. The side output of the T confitor was inserted into the connector on the network card. That's what it looked: Now such networks are hopelessly outdated and everywhere replaced by the "star" on the twisted pair, but the equipment under coaxial cable You can still see in some enterprises.

Topology "Ring"

Ring - This is the topology of the local network, in which the workstations are connected in succession to each other, forming a closed ring. Data is transmitted from one workstation to another in one direction (in a circle). Each PC works as a repeater, relaying messages to the next PC, i.e. The data is transmitted from one computer to another as if on the relay. If the computer receives data intended for another computer - it transmits them further along the ring, otherwise they are not transmitted further.

The advantages of the ring topology:

• easy installation;
• almost complete absence of additional equipment;
• the possibility of sustainable work without a significant drop in the data transfer rate with intensive network load.

However, the "ring" has significant disadvantages:

• each workstation should actively participate in sending information; In case of failure, at least one of them or a cable break - the work of the entire network stops;
• connecting a new workstation requires a short-term network shutdown, because during the installation of the new PC, the ring must be open;
• the complexity of configuration and configuration;
• the difficulty of finding faults.

Ring network topology is quite rare. The main use she found in fiber optic networks Standard Token Ring.

Topology "Star"

Star - This is a local network topology, where each workstation is attached to the central device (switch or router). The central device controls the movement of packets on the network. Each computer through network card Connects to the switch with a separate cable. If necessary, you can combine several networks with the "Star" topology - as a result you will receive a network configuration with tree topology. The tree topology is distributed in large companies. We will not consider it in detail in this article.

The topology "Star" today has become the primary when building local networks. This happened due to its numerous advantages:

• the failure of one workstation or damage to its cable is not reflected in the work of the entire network as a whole;
• excellent scalability: To connect a new workstation, a separate cable from the switcher is sufficient;
• easy troubleshooting and cliffs in the network;
• high performance;
• easy setup and administration;
• additional equipment is easily embedded in the network.

However, like any topology, the "star" is not devastable:

• the failure of the central switch will result in the inoperability of the entire network;
• additional network equipment costs - a device to which all network computers will be connected (switch);
• the number of workstations is limited by the number of ports in the central switcher.

Star - the most common topology for wired and wireless networks. An example of star-shaped topology is a network with type cable twisted para, and the switch as a central device. Such such networks are found in most organizations.