Energy supply and cooling

These video cards can only play games that are old and have limited resources.

NVIDIA NVS 4200M
The business graphics card, which is based on the GeForce GT 520M, has special BIOS drivers that will benefit business add-ons.
Core – 810 MHz, shader – 48, DirectX 11. Memory – 800 MHz, 64-bit.

AMD Radeon HD 8350G
The AMD Richland (A4) processor has graphics without visible video memory.
Core – 514-720 MHz, shader – 128, DirectX 11.

AMD Radeon HD 8330
The graphics are based on GCN with 128 stream processors, but without video memory. Call me and go in pairs AMD processors Richland A4-5000 "Kabini".
Core – 500 MHz, shader – 128, DirectX 11.1.

AMD Radeon R3 (Mullins/Beema)
Graphics have been created that are based on the GCN architecture.
Core – 350 – 600 MHz, shader – 128, DirectX 11.2. Memory – 64-bit.

AMD Radeon HD 6510G2
Two graphics cards connected via asymmetric CrossFire - discrete Radeon HD 6430M/6450M/6470M and embedded in A-series 6480G processors.
Shadery – 400, DirectX 11.

AMD Radeon HD 7450M

Core – 700 MHz, shader – 160, DirectX 11. Memory – 1800 MHz, 64-bit.

NVIDIA GeForce 610M
Graphics cob vine, which is based on the older GeForce GT 520M or GeForce GT 520MX.
Core – 672-900 MHz, shader – 48, DirectX 11. Memory – 1800 MHz, 64-bit.

NVIDIA GeForce 705M
The graphics are of the same level, which is based on the GF119 chip, like its predecessors – GeForce GT 520M, 520MX and 610M.
Core – 775 MHz, shader – 48, DirectX 11. Memory – 1800 MHz, 64-bit.

AMD Radeon HD 6470M
Ear level graphics are based on the Seymore XT core and include a UVD3 video processor.
Core – 700/750 MHz, shader – 160, DirectX 11. Memory – 800 MHz, 64-bit.

AMD FirePro M3900
Top-level graphics for mobile workstations based on the AMD Radeon 6470M.
Core – 700-750 MHz, shader – 160, DirectX 11. Memory – 900 MHz, 64-bit.

NVIDIA GeForce GT 520M
Primary level graphics, which are based on the GF119 chip with a 64-bit memory bus or on the GF108 with a 128-bit memory bus, and with lower clock frequencies.
Core – 740/600 MHz, shader – 160, DirectX 11. Memory – 800/900 MHz, 64/128-bit.

AMD Radeon HD 7420G
The processor has built-in graphics, as can be found in processors of the Trinity A4 series (for example, A4-4300M). Based on the VLIW4 architecture of the Radeon HD 6900 desktop series.
Core – 480-655 MHz, shader – 128, DirectX 11.

Intel HD Graphics (Haswell)
Graphics are derived from Haswell Celeron and Pentium processors.
Core – 200-1000 MHz, shader – 10, DirectX 11.1.

AMD Radeon HD 6520G
Graphics are built into A6 series processors from the Llano line.
Core – 400 MHz, shader – 320, DirectX 11.

AMD Radeon HD 8310G
Graphics are built into AMD Richland ULV A4 series processors and do not contain any external video memory.
Core – 424-554 MHz, shader – 128, DirectX 11.

AMD Radeon HD 7400G
Graphics are built into Trinity A4 series processors (for example, A4-4355M). Based on the VLIW4 architecture of the Radeon HD 6900 series.
Core – 327-423 MHz, shader – 192, DirectX 11.

AMD Radeon HD 6480G
The graphics are built into the A4-Series Llano processor and do not contain any waste video memory.
Core – 444 MHz, shader – 240, DirectX 11.

NVIDIA GeForce GT 415M
The most advanced of the GT 400M series.
Core – 500 MHz, shader – 48, DirectX 11. Memory – 800 MHz, 128-bit.

NVIDIA GeForce 410M
Core graphics, which are based on the GF119 chip and can match the performance of the 520M, run at lower clock speeds.
Core – 575 MHz, shader – 48, DirectX 11. Memory – 800 MHz, 64-bit.

AMD Radeon HD 7370M
Renamed HD 6370M/HD 547.
Core – 750 MHz, shader – 80, DirectX 11. Memory – 1600 MHz, 64-bit.

AMD Radeon HD 6370M
HD 5470 renaming.
Core – 750 MHz, shader – 80, DirectX 11. Memory – 800 MHz, 64-bit.

AMD Radeon HD 8280
The graphics are built, which are based on the GCN architecture and do not use waste video memory. Consider pairing with AMD E2-3000 Kabini processors.
Core – 450 MHz, shader – 128, DirectX 11.1.

ATI Mobility Radeon HD 5470
Top-level graphics supported by GDDR5 memory, with more than 80 processor cores. Supports Eyefinity (up to 4 monitors) and eight-channel HD audio via HDMI port. Productivity can be equaled with the older GeForce 8600M GT.
Core – 750 MHz, shader – 80, DirectX 11. Memory – 1800 MHz, 64-bit.

AMD Radeon HD 6450M
Ear level graphics are based on the Seymore-PRO chip and support Eyefinity+.
Core – 600 MHz, shader – 160, DirectX 11. Memory – 800 MHz, 64-bit.

AMD Radeon HD 7430M
Renamed Radeon HD 6450M.
Core – 600 MHz, shader – 160, DirectX 11. Memory – 1800 MHz, 64-bit.

AMD Radeon R6 (Mullins)
The graphics are built into AMD Mullins processors and are based on the GCN architecture.
Core – 500 MHz, shader – 128, DirectX 11.2. Memory – 64-bit.

AMD Radeon HD 8240
The graphics are built, which are based on the GCN architecture and do not use waste video memory. Consider pairing with AMD E1-2500 Kabini processors.
Core – 400 MHz, shader – 128, DirectX 11.1.

AMD Radeon HD 8250
Graphics have been built into the AMD A6-1450 "Temash" processor. Based on GCN architecture.
Core – 300-400 MHz, shader – 128, DirectX 11.1.

ATI Mobility Radeon HD 5450
Graphics of the cob level with the same frequency and productivity as the HD 4570, but with lower energy consumption.
Core – 675 MHz, shader – 80, DirectX 11. Memory – 800 MHz, 64-bit.

AMD Radeon R2 (Mullins/Beema)
Graphics are based on AMD Beema and Mullins processors. Based on GCN architecture.
Core – 300-500 MHz, shader – 128, DirectX 11.2. Memory – 64-bit.

Intel HD Graphics 3000
Graphics, embedded in Intel processors Sandy Bridge (Core ix-2xxx).
Core – 350-1350 MHz, shader – 12, DirectX 10.1.

NVIDIA GeForce 405M
Renamed GeForce 310M / 315M, which is still based on the G2xx GeForce G210M architecture.
Core – 606 MHz, shader – 16, DirectX 10.1. Memory – 1600 MHz, 64-bit.

AMD Radeon HD 6430M
The most advanced graphics based on the Seymour chip is supported by the UVD3 and Eyefinity+ video processor.
Core – 480 MHz, shader – 160, DirectX 11. Memory – 800 MHz, 64-bit.

AMD Radeon HD 6380G
The graphics built into the E2 Llano series processors do not contain waste video memory.
Core – 400 MHz, shader – 160, DirectX 11.

ATI Mobility Radeon HD 5430
Based on chips with Park LP codenames, the most advanced of the HD 5400 series.
Core – 550 MHz, shader – 80, DirectX 11. Memory – 800 MHz, 64-bit.

AMD Radeon HD 8210
GCN-based graphics have been introduced to pair with AMD A4-1250 “Temash” and E1-2100 “Kabini” processors.
Core – 300 MHz, shader – 128, DirectX 11.1.

Intel HD Graphics 2500
Graphics are built into Ivy Bridge processors (Core ix-3xxx).
Core – 650-1150 MHz, shader – 6, DirectX 11.

Intel HD Graphics (Ivy Bridge)
Graphics are derived from Ivy Bridge Celeron and Pentium processors.
Core – 350-1100 MHz, shader – 6, DirectX 11.

The AMD company produces not only processors, but also video Radeon cards, characteristics that are sufficient to launch the most productive games.

This technology, especially released over the last two years, allows you to work with resource-intensive add-ons (3D graphics).

In order to select the required model and determine whether you have enough capabilities to finish your tasks, take a look at the parameters of the graphics processors in the report.

To get to know the installation characteristics more simply, you can put together a table outlining the basic values ​​that affect the productivity and functionality of the video card.

These include the bus parameters (frequency and capacity), the type of memory that is used during the production of the graphics processor, the processing speed, the speed of the data and the memory used.

You can also rely on the generated electrical energy, which can put pressure on the computer's power supply unit and the method of cooling the device.

Memory frequency and bus capacity

The memory frequency of the video card directly affects the speed of its work. The average value of this indicator is 1000 MHz for HBM memory and 6000-8000 for GDDR5.

The level of productivity of the card in relation to its frequency is always directly proportional, as compared to other indicators, to the throughput of the device and the bus capacity.

Based on the bus parameters, first of all, set the video card memory bandwidth.

The greater the width, the faster the data are collected graphics processor(GPU).

So, pay for 64 bit and it’s already practical to avoid vikorystvuyutsya in today's computers, if you want to sell it in online stores.

Most current models of video cards have a capacity of 128 and 256 bits, top versions - 512 bits and higher.

The top ten AMD models today are about to reach their lowest capacity:

  • RX 470, 480 and 380 series – 256-bit;
  • 390 series R9 - 512 bits;

  • The remaining models, R9 Fury and Nano, are equipped with a new type of memory – 4096 bits;
  • one of the quiet ones that are let out new technologies With the technological process of 18 nm models, RX, the bit capacity is only 128 bits, which results in low data transfer speed, although it costs very little, representing a budget option for gamers.

High capacity remaining video cards AMD, which eliminates the high cost of high-ball memory modules, allows a lower frequency, resulting in greater power.

In this case, the power consumption (1 Watt per 1 GB/s data transmission speed) is lower – R9 models with HBM memory can cope less electricians against other cards.

The main feature of the Radeon Fury and Nano is the ability to launch more powerful graphics add-ons and resource-intensive games. tall display FPS (frame rate).

Memory type

GDDR5 memory, which was recently introduced the shortest option The graphics card is starting to age.

Moreover, the producers say that they are able to reach their limits and are starting to find new solutions. One of them is HBM technology, which includes:

  • increased productivity;
  • less energy consumption;
  • the specific organization of the memory subsystem.

For these reasons, the current high-speed video cards R9 Fury, Fury X and Nano, running at a low frequency of 1000 MHz, perform 33% more than the flagship of the previous generation R9 390X – 512 GB/s instead of 384.

It’s so new, too budget model RX 460 at the highest frequency of 1212 MHz has 5 times less speed than the same pushing the model The device contains GDDR5 memory, and its capacity is 128 bits.

The amount of memory for current Radeon graphics devices is around 4096-8192 MB.

In this case, games require at least 4 GB of memory to run with normal settings.

Although this indicator is not so important for HBM memory, it is important to add the throughput capacity that it has, lower than that of GDDR.

Technical process

The main design elements of a processor, including a graphics processor, are transistors that pass or block electric strum to the singer directly.

These depend largely on the productivity of the video card, and this indicator, in turn, depends on the size of the transistors and the technology that goes into their production.

Most video card vendors, including AMD, use a 28 nm transistor technology.

This value applies to all current models, except the RX 400 series.

New generation graphics processors are built using 14 nm technology. And in the upcoming Radeon card, it will be launched using the 7 nm technology process.

It is expected that 14-nanometer technology will ensure graphics core Increases productivity by 2–3 times and supports up to 3 independently working monitors.

Throughput capacity

The transmission speed of additional video cards must first be determined by obtaining an effective frequency for their memory capacity.

The higher the value, the faster the information is transmitted, and therefore the games run faster.

For what it's worth new memory HBM discharge capacity is 8 times higher, and therefore the frequency may be lower.

For example, for the R9 Fury X model, the throughput is the same (4096/8) bytes * 1 GHz = 512 GB/s. This value is sufficient to launch any game at maximum settings.

An RX 460 video card with a capacity of 128 bits can transfer more than 112 GB/s of information (=7000*128/8).

Energy supply and cooling

The energy efficiency of various video cards depends on various factors:

  • technologies that are being developed during the development of the processor;
  • type of memory;
  • the tightness of the graphics board.

When looking at one series of cards, you can find models with high and low energy consumption.

So, for example, the R9 390 and 390X models deliver up to 275 W of power and deliver at least 500 W to the power supply.

The same performance is found in the more productive cards R9 Fury and Fury X. Similarly, the R9 Nano consumes only 175 W, although it does not compromise on productivity and instead surpasses them.

And the inexpensive RX 460 model delivers only 75 W, providing optimal power balance to energy efficiency.

Power up to 75 W is provided by one PCI Express slot.

The excess of this value is compensated by additional 8-pin sockets, which can supply up to 150 W.

This means that in order to ensure the energy of daily life AMD cards One PCI slot is not enough and the additional power required.

The design of the cooling system also depends on the energy consumption of the graphics processor:

  • Less productive models are cooled by a secondary fan system;

  • processors that run daily games, run and suffer serious cooling – of course. For example, the R9 Nano's ventilation system includes not only a cooler, but also an evaporation chamber with heat pipes. And the R9 Fury has a metal plate installed under the radiator.

Visnovki

AMD, as Nvidia's main competitor, continues to increase the characteristics of its video cards.

The 1st Fury series surpasses the previous generation in terms of performance (in addition to energy efficiency).

Although there is no need for older versions, budget RX graphics cards, created on the basis of the new 14 nm process technology, are comparable to older flagships and can be compared with inexpensive models of the previous generation.

When choosing a card for your PC, you will still have to pay attention to the financial side of the equation - it costs a lot to get the best characteristics.


Solid state storage devices with a volume of 1 TB on the market may even cost a lot due to their versatility - they are too expensive. That's why we rarely test solutions to this class. However, we are still testing - selecting the most available models at the time of writing. And we are trying to work hard so that we can spit on the hardware platforms ourselves, and not on the models themselves.

advertising

Yak painting on Narazi? The cheapest SSD models from Yandex.Market:
  • SmartBuy Ignition Plus – for 16.5 thousand rubles – is a representative of a rare hardware platform that supports Phison S11 and MLC 3D V-NAND Micron. Although the family of these hoarders has been on sale for a long time, the modification itself to 960 GB showed up literally quickly;
  • Micron MTFDDAV1T0TBN – worth 16.8 thousand rubles – under this toothy name there is a controller Marvell 88SS1074 and TLC 3D 32L V-NAND Micron, the essence is Micron M1100 (and also Crucial MX300). We tested this hardware configuration in Sichna;
  • WD Blue PC SSD – 17.1 thousand rubles – Marvell 88SS1074 and 15 nm planar TLC NAND SanDisk. This configuration was made a little more than fate. The term is inaccurate, but no fundamental changes have been made during this hour, and it can be taken into account that there is no fault to blame for these particular nutritional verses;
  • Samsung 850 Evo - 17,200 rubles. The model has been beaten up and down and is being tested by everyone who is not lazy. Sensu especially does not bother;
  • Crucial MX300 – 17.5 thousand rubles – already mentioned in a couple of paragraphs;
  • Intel SSDSCKKW010X6X1, or Intel 540s priced at 17.9 rubles - we also looked at that;
  • WD Blue 3D NAND SATA SSD - at a price of 17.9 thousand rubles - a new product that was only found on sale, since then we have been protesting the modification to 500 GB and have rejected statements about the potential of this modification ї;
  • AMD Radeon R3 – 18 thousand rubles – this has not yet been accumulated;
  • Kingston SSDNow UV400 - 18.5 thousand rubles - similar;
  • Toshiba HDTS896EZSTA - worth 19 thousand rubles - upgrade of Phison S10 and 15 nm planar TLC NAND Toshiba, known from the Toshiba TR150 test of the same date, in stock, they are already selling surplus, soon the hoarder is to blame sale, fragments of knowledge from production;
  • Corsair Force LE - priced at 19.5 thousand rubles - essentially the same as Phison S10 and 15 nm TLC NAND Toshiba.

Unknown to us in this category and with price tags that can still be called “budget” (up to 20 thousand rubles), there are only three accumulators: SmartBuy Ignition Plus, AMD Radeon R3 and Kingston SSDNow UV400. With the first - unfortunately, food is still in the suspended state (administrative nature), another number of meetings for this material, and with the third, we will try to improve nutrition.

Let me be clear: I have not seen any improvements in the M.2 form factor. On the right, such an accumulator can easily be “converted” to 2.5" - similar shaft adapter housings on AliExpress/eBay for a price of 200-250 rubles with delivery.

And the difference between M.2 and “output” 2.5" from the hardware point of view is extremely small. And now there is a larger layout with a risk of overheating, which seems to be true, with great advances this is no longer true e. For 2.5" cases, the multi-size arms of the boards have changed to The obvious view is that the manufacturers have recently switched to shortened payments in order to reduce production costs.

Drivers:

ASUS 8200 Ti200 - Detonator 11/23
ATI RADEON 8500 OEM – 4.13.9009.

At the time the testing was completed, these drivers themselves were the recommended versions (releases) for both cards for Windows 98/ME.

The system, of course, by any standards is clearly not “Top”, it is more than average or a little more than that, but on the other hand, video cards are also not the most expensive and cheap, so a certain balance is present here.

Tests were carried out only in 32-bit color (games in 16-bit color were not visible), all tests had VSync enabled. Other settings are subject to change, then. After installing the driver, there were no changes (except for enabling VSync). The sound was distorted in all tests.

All tests show resolutions of 1024x768, 1280x1024, and in these resolutions this video card will ensure a comfortable level of performance in all current games. In addition, in our opinion, most consumers in our minds are using 15"-17" monitors, for which only 1024x768 and 1280x1024 are provided with a completely “humane” refresh rate of 85-100Hz. And for those who are constantly gaining more and more popularity of LCD monitors, which is also the main thing.

All tests were carried out both in the normal mode and with anisotropic filtration. Why, please, with anisotropic filtration? On the right is that in modern games in the basic mode (with trilinear) these allow extortionate fps often give weaker and, apparently, less expensive cards, such as GeForce2 Titanium, ATI Radeon 7500, GeForce4 MX 440. Apparently, many active users of video cards of the GF3 and Radeon 8500 class are interested in playing on such powerful cards (even on their younger modifications) without additional consideration, as it has anisotropic filtration, all the same, you need to go bakery, and as you know, “our people don’t take taxis to the bakery” :).

Nutrition depends only on the type of desire you choose. We obviously turned off other types of vantagement, because (again, in our opinion), it’s far from in our minds Today's games You can play simply with antialiasing in the 4x mode (not that this mode is compromised by anisotropic filtering). The choice is actually between AA 2x and AF, from which we have deprived us of the remaining mode (which means, subjectively, that the 3D image is more acceptable in most modes). It’s understandable, it’s not possible to see what’s going on better than a solution, but for the minds of the time spent on testing, it is necessary to reinforce the main one in front of the other.

For tests, the 8500 LE is set to the mode with the maximum level of anisotropic filtration, for the Ti200 Anisotropic filtration level = 4, as a compromise between capacity and productivity.

To evaluate the effect of overclocking video cards, we tested the Radeon 8500 LE both in normal mode (250/250) and in overclocked mode (286/266). The Ti200 card was also tested in a standard (175/200) and a customized (200/230) mill. Zrozumil, Rosgin - Tse Roulette, Kapos Videokard Razznit, I will stupidly pratsuvati (and not tilki passes the peldow) on the Belsh of Vosic frequencies, especially the yakshcho vicoristovuvati, the abdominal abdominal, the volunteer of the resistor. We were confused by these numbers, so we thought that such an increase in guarantees for these video cards without additional inputs.

Due to the number of hours and the great number of tests, it was decided to carry out tests (at this stage) only under operating system Win98SE. In the near future we plan to prepare the 2nd part of this article with tests in Windows XP.

10 Ber. 2016

On this page below there is a message to download the rest of the free downloads Drivers for AMD video cards Radeon HD 8200 / R3, which is part of the Radeon HD 8000 series. Installation files are taken from the official website and are suitable for: Windows 7, 10, 8, 8.1, XP, Vista 32/64-bit (x86/x64).

To make it easier to select the required files, the version of your Windows and its bit level are indicated below.

Your computer works on:

  1. entice (153.5 MB / version 16.8.2 (Crimson Edition 16.8.2 Hotfix) / release date 08/12/2016)

    For Windows 7 32-bit

  2. entice (239.8 MB / version 16.8.2 (Crimson Edition 16.8.2 Hotfix) / release date 08/12/2016)

    For Windows 7 64-bit

  3. entice (134.8 MB / version 16.8.2 (Crimson Edition 16.8.2 Hotfix) / release date 08/12/2016)

    For Windows 10 32-bit

  4. entice (208.24 MB / version 16.8.2 (Crimson Edition 16.8.2 Hotfix) / release date 08/12/2016)

    For Windows 10 64-bit

  5. entice (205 MB / version 14.4 (Catalyst Software Suite) / release date 04/25/2014)

    For Windows 8 32-bit

  6. entice (260 MB / version 14.4 (Catalyst Software Suite) / release date 04/25/2014)

    For Windows 8 64-bit

  7. entice (154.21 MB / version 16.8.2 (Crimson Edition 16.8.2 Hotfix) / release date 08/12/2016)

    For Windows 8.1 32-bit

  8. entice (239.88 MB / version 16.8.2 (Crimson Edition 16.8.2 Hotfix) / release date 08/12/2016)

    For Windows 8.1 64-bit

  9. entice (179 MB / version 14.4 (Catalyst Software Suite) / release date 04/25/2014)

    For Windows XP 32 and 64-bit

  10. entice (151 MB / version 13.12 (Catalyst Software Suite) / release date 12/18/2013)

    For Windows Vista 32-bit

  11. entice (209 MB / version 13.12 (Catalyst Software Suite) / release date 12/18/2013)

    For Windows Vista 64-bit

Backup option - remove the drivers and get help from AMD Driver Autodetect

The Danish option is manual because the program AMD Driver Autodetect will select and capture the remaining working drivers which is suitable for your AMD video card and yours Windows version. The program does not require installation, it was created by AMD and the download of files is downloaded from their official servers.

Instructions:

  1. Run AMD Driver Autodetect and it will automatically select required files to install drivers.
  2. To download the files, click the “Download Now” button.
  3. Make sure you have downloaded your files and begin installation.