Gtx 650 green edition: Galaxy GTX 650 Green Edition Specs

Inno3D GeForce GTX 650 Green Rev. 2 review

39points

Inno3D GeForce GTX 650 Green Rev. 2

Inno3D GeForce GTX 650 Green Rev. 2

Why is Inno3D GeForce GTX 650 Green Rev. 2 better than the average?

• Thermal Design Power (TDP)?
  65Wvs199.56W
• Load GPU temperature?
  51°Cvs73.04°C
• Idle GPU temperature?
  27°Cvs31.4°C
• Power consumption while under peak load?
  231Wvs323.6W
• Load noise level?
  43.5dBvs47.58dB
• DVI outputs?
  2vs0.73
• Width?
  160mmvs260.4mm
• Mini DisplayPort outputs?
  1vs0.01

Which are the most popular comparisons?

Nvidia GeForce GTX 1650

vs

Nvidia GeForce RTX 3050 Laptop

Nvidia GeForce RTX 3050 Ti Laptop

vs

Nvidia GeForce RTX 3060 Laptop

Nvidia GeForce RTX 3050 Laptop

vs

Nvidia GeForce RTX 3050 Ti Laptop

AMD Radeon RX 5500M

vs

Nvidia GeForce GTX 1650

Nvidia GeForce GTX 1650

vs

Nvidia GeForce RTX 3050 Ti Laptop

Nvidia GeForce RTX 2060

vs

Nvidia GeForce RTX 3050

AMD Radeon RX 6400

vs

Nvidia GeForce GTX 1650

AMD Radeon RX 6650 XT

vs

Nvidia GeForce RTX 3060

AMD Radeon RX 6500 XT

vs

Nvidia GeForce GTX 1650

AMD Radeon RX 6750 XT

vs

Nvidia GeForce RTX 3060 Ti

Price comparison

User reviews

Performance

GPU clock speed

1058MHz

The graphics processing unit (GPU) has a higher clock speed.

GPU turbo

Unknown. Help us by suggesting a value.

When the GPU is running below its limitations, it can boost to a higher clock speed in order to give increased performance.

pixel rate

8.46 GPixel/s

The number of pixels that can be rendered to the screen every second.

floating-point performance

0.81 TFLOPS

Floating-point performance is a measurement of the raw processing power of the GPU.

texture rate

33.9 GTexels/s

The number of textured pixels that can be rendered to the screen every second.

GPU memory speed

1250MHz

The memory clock speed is one aspect that determines the memory bandwidth.

shading units

Shading units (or stream processors) are small processors within the graphics card that are responsible for processing different aspects of the image.

texture mapping units (TMUs)

TMUs take textures and map them to the geometry of a 3D scene. More TMUs will typically mean that texture information is processed faster.

render output units (ROPs)

The ROPs are responsible for some of the final steps of the rendering process, writing the final pixel data to memory and carrying out other tasks such as anti-aliasing to improve the look of graphics.

Memory

effective memory speed

5000MHz

The effective memory clock speed is calculated from the size and data rate of the memory. Higher clock speeds can give increased performance in games and other apps.

maximum memory bandwidth

80GB/s

This is the maximum rate that data can be read from or stored into memory.

VRAM (video RAM) is the dedicated memory of a graphics card. More VRAM generally allows you to run games at higher settings, especially for things like texture resolution.

memory bus width

128bit

A wider bus width means that it can carry more data per cycle. It is an important factor of memory performance, and therefore the general performance of the graphics card.

version of GDDR memory

Newer versions of GDDR memory offer improvements such as higher transfer rates that give increased performance.

Supports ECC memory

✖Inno3D GeForce GTX 650 Green Rev. 2

Error-correcting code memory can detect and correct data corruption. It is used when is it essential to avoid corruption, such as scientific computing or when running a server.

Features

DirectX version

DirectX is used in games, with newer versions supporting better graphics.

OpenGL version

OpenGL is used in games, with newer versions supporting better graphics.

OpenCL version

Some apps use OpenCL to apply the power of the graphics processing unit (GPU) for non-graphical computing. Newer versions introduce more functionality and better performance.

Supports multi-display technology

✔Inno3D GeForce GTX 650 Green Rev. 2

The graphics card supports multi-display technology. This allows you to configure multiple monitors in order to create a more immersive gaming experience, such as having a wider field of view.

load GPU temperature

A lower load temperature means that the card produces less heat and its cooling system performs better.

supports ray tracing

✖Inno3D GeForce GTX 650 Green Rev. 2

Ray tracing is an advanced light rendering technique that provides more realistic lighting, shadows, and reflections in games.

Supports 3D

✔Inno3D GeForce GTX 650 Green Rev. 2

Allows you to view in 3D (if you have a 3D display and glasses).

supports DLSS

✖Inno3D GeForce GTX 650 Green Rev. 2

DLSS (Deep Learning Super Sampling) is an upscaling technology powered by AI. It allows the graphics card to render games at a lower resolution and upscale them to a higher resolution with near-native visual quality and increased performance. DLSS is only available on select games.

PassMark (G3D) result

This benchmark measures the graphics performance of a video card. Source: PassMark.

Ports

has an HDMI output

✔Inno3D GeForce GTX 650 Green Rev. 2

Devices with a HDMI or mini HDMI port can transfer high definition video and audio to a display.

HDMI ports

Unknown. Help us by suggesting a value.

More HDMI ports mean that you can simultaneously connect numerous devices, such as video game consoles and set-top boxes.

HDMI version

Unknown. Help us by suggesting a value.

Newer versions of HDMI support higher bandwidth, which allows for higher resolutions and frame rates.

DisplayPort outputs

Allows you to connect to a display using DisplayPort.

DVI outputs

Allows you to connect to a display using DVI.

mini DisplayPort outputs

Allows you to connect to a display using mini-DisplayPort.

Price comparison

Which are the best graphics cards?

NVIDIA GeForce GTX 650 Variants Compared — Find The Best GTX 650 Card in Our Benchmarks

NVIDIA GeForce GTX 650 Variants Compared — Find The Best GTX 650 Card in Our Benchmarks — GPUCheck United States / USA

Performance Comparison of ASUS GTX 650 DirectCU TOP vs EVGA GTX 650 Superclocked vs EVGA GTX 650 Superclocked 2 GB vs ZOTAC GTX 650 AMP! Edition vs ASUS GTX 650 DirectCU OC vs SPARKLE GTX 650 OC Dragon Cyclone and 56 more variants of the NVIDIA GeForce GTX 650 Graphics Card.

Graphics Card	GPU Clock	Boost Clock	Memory Clock	Average 1080p FPS	Average 1440p FPS	Average 4K FPS
ASUS GTX 650 DirectCU TOP	1215 MHz	1215 MHz	1275 MHz
EVGA GTX 650 Superclocked	1202 MHz	1202 MHz	1250 MHz
EVGA GTX 650 Superclocked 2 GB	1202 MHz	1202 MHz	1250 MHz
ZOTAC GTX 650 AMP! Edition	1189 MHz	1189 MHz	1400 MHz
ASUS GTX 650 DirectCU OC	1137 MHz	1137 MHz	1250 MHz
SPARKLE GTX 650 OC Dragon Cyclone	1124 MHz	1124 MHz	1400 MHz
MSI GTX 650 Power Edition OC	1124 MHz	1124 MHz	1250 MHz
ZOTAC GTX 650 Destroyer TSI	1110 MHz	1110 MHz	1350 MHz
ZOTAC GTX 650 Destroyer TSI 2 GB	1110 MHz	1110 MHz	1350 MHz
Galaxy GTX 650 GC	1110 MHz	1110 MHz	1250 MHz
Galaxy GTX 650 GC 2 GB	1110 MHz	1110 MHz	1250 MHz
GIGABYTE GTX 650 OC 1 GB	1110 MHz	1110 MHz	1250 MHz
GIGABYTE GTX 650 OC 2 GB	1110 MHz	1110 MHz	1250 MHz
GIGABYTE GTX 650 WindForce 2X	1110 MHz	1110 MHz	1250 MHz
GIGABYTE GTX 650 WindForce 2X 2 GB	1110 MHz	1110 MHz	1250 MHz
GIGABYTE GTX 650 WindForce 2X Rev. 3.0	1110 MHz	1110 MHz	1250 MHz
KFA² GTX 650 EX OC	1110 MHz	1110 MHz	1250 MHz
Leadtek WinFast GTX 650 OC Rev. A	1100 MHz	1100 MHz	1250 MHz
Leadtek WinFast GTX 650 OC Rev. B	1100 MHz	1100 MHz	1250 MHz
Club 3D GTX 650	1098 MHz	1098 MHz	1250 MHz
SPARKLE GTX 650 OC	1098 MHz	1098 MHz	1250 MHz
MSI GTX 650 OC	1084 MHz	1084 MHz	1250 MHz
Gainward GTX 650 GS	1072 MHz	1072 MHz	1300 MHz
Palit GTX 650 OC	1072 MHz	1072 MHz	1300 MHz
ASUS GTX 650	1072 MHz	1072 MHz	1250 MHz
ASUS GTX 650 2 GB	1072 MHz	1072 MHz	1250 MHz
ASUS GTX 650 Formula II	1072 MHz	1072 MHz	1250 MHz
ASUS GTX 650 Formula II 2 GB	1072 MHz	1072 MHz	1250 MHz
MSI GTX 650 OC 2 GB	1072 MHz	1072 MHz	1250 MHz
MSI GTX 650 OC V1	1072 MHz	1072 MHz	1250 MHz
ZOTAC GTX 650 LP	1072 MHz	1072 MHz	1250 MHz
ZOTAC GTX 650 Thunder Edition	1072 MHz	1072 MHz	1250 MHz
ZOTAC GTX 650 Thunder Edition 2 GB	1072 MHz	1072 MHz	1250 MHz
ASUS GTX 650 DirectCU	1058 MHz	1058 MHz	1250 MHz
Colorful iGame GTX 650	1058 MHz	1058 MHz	1250 MHz
Colorful iGame GTX 650 ARES X	1058 MHz	1058 MHz	1250 MHz
Colorful iGame GTX 650 Flame Wars 2 GB	1058 MHz	1058 MHz	1250 MHz
ELSA GTX 650 S. A.C	1058 MHz	1058 MHz	1250 MHz
ELSA GTX 650 S.A.C V2	1058 MHz	1058 MHz	1250 MHz
EVGA GTX 650	1058 MHz	1058 MHz	1250 MHz
EVGA GTX 650 2 GB	1058 MHz	1058 MHz	1250 MHz
Gainward GTX 650	1058 MHz	1058 MHz	1250 MHz
Gainward GTX 650 2 GB	1058 MHz	1058 MHz	1250 MHz
Galaxy GTX 650 Green Edition	1058 MHz	1058 MHz	1250 MHz
GIGABYTE GTX 650 1 GB	1058 MHz	1058 MHz	1250 MHz
GIGABYTE GTX 650 2 GB	1058 MHz	1058 MHz	1250 MHz
Inno3D GTX 650 2 GB	1058 MHz	1058 MHz	1250 MHz
Inno3D GTX 650 Green	1058 MHz	1058 MHz	1250 MHz
Inno3D GTX 650 Green Rev. 2	1058 MHz	1058 MHz	1250 MHz
Inno3D GTX 650 HerculeZ 1000	1058 MHz	1058 MHz	1250 MHz
Leadtek WinFast GTX 650 Hurricane II	1058 MHz	1058 MHz	1250 MHz
MAXSUN GTX 650 Terminator	1058 MHz	1058 MHz	1250 MHz
MSI GTX 650 Power Edition	1058 MHz	1058 MHz	1250 MHz
Palit GTX 650	1058 MHz	1058 MHz	1250 MHz
Palit GTX 650 2 GB	1058 MHz	1058 MHz	1250 MHz
PNY XLR8 GTX 650 Performance Edition	1058 MHz	1058 MHz	1250 MHz
PNY XLR8 GTX 650 Performance Edition 2 GB	1058 MHz	1058 MHz	1250 MHz
ZOTAC GTX 650 Synergy Edition	1058 MHz	1058 MHz	1250 MHz
ZOTAC GTX 650 Synergy Edition	1058 MHz	1058 MHz	1250 MHz
ZOTAC GTX 650 Synergy Edition 2 GB	1058 MHz	1058 MHz	1250 MHz
NVIDIA GeForce GTX 650	1058 MHz	MHz	1250 MHz

comparison of video cards Nvidia GTX 650 and 1650S / Video cards and monitors / iXBT Live

The development of technology greatly simplifies our lives, and in order to make all the wishes come true, modern and productive parts are needed. After all, if the hardware performance did not change, then big-budget films were shot much longer, while spending huge sums, and the graphics component in games would be stuck in one time. That is why computer hardware manufacturers are constantly reducing technical processes, improving microarchitectures, introducing software optimizations, and much more.

But the approach to improvement can be different:

• In 2012, Intel released extremely successful Sandy Bridge processors and further engaged in their running, thinning, improving the microarchitecture of the cores, while not changing the design of the processors. And now, 9 years later, Rocket Lake processors came to desktops, which are essentially mega-improved Sandics. In new generations, Intel added several percent to the performance, but a total of 5% per generation resulted in a fairly large gap between the Intel Core i-2xxx and i-11xxx processors.
• AMD assembled its Ryzen processors in parts and therefore the differences between Zen, Zen2 and Zen3 are visible to the naked eye. And after the release of the processors, the red company looks where their weak points are and fixes them. An example of such a fix can be considered a cache change in R-5xxx.
• Nvidia is constantly adding performance, mastering new technologies (Ray-Tracing in RTX 2xxx) or changing the structure of the core (in RTX 3xxx there was a doubling of the number of FP32 blocks, which Nvidia designated as an increase in CUDA cores by 2 times).

Due to the constant increase in productivity and new technologies, the green company has firmly established itself in the market and does not seek to lose ground. This can be seen on the example of Steam’s monthly statistics, where the first video card from AMD is found in 10th place, in the top twenty there are only 3 solutions from the red company. And one of them is Vega8:

Contents

• Characteristics
• Tests
  • Synthetics and work tasks
  • Games
• Output

Features

Nvidia has three brands for graphics cards: Tesla, Quadro, and GeForce. They are aimed at completely different groups of people, and we are most interested in GeForce graphics cards aimed at gamers. And in order to avoid confusion within the GeForce brand, video cards are divided into lines: 90 cards are top, fat and overpriced, because you have to pay for the best; 80 cards are the best solution for games, because with slightly lower performance they are much cheaper than 90 ruler; 60 and 70 fall into the middle segment, and depending on the specific model, the card may be closer to the top or basic solution; 50 line is intended for budget gamers; well, 10-40 cards are plugs for office computers.

And it was budget video cards that I wanted to compare: the already old GTX 650 and the newest video card 50 of the 1650S line at the moment, because the 50 line based on the Amper architecture is not yet represented. And I would like to say right away that this will not be a fair comparison, because each of the video cards has three modifications: the GTX650 has the GTX650, GTX650Ti and GTX650Ti Boost, and the GTX1650 has the GTX 1650, GTX 1650 D6 and GTX 1650Super. It turns out that I will compare the weakest card among 650 and the most powerful of 1650. 9 169 150 150 160

The development of video cards was not free and the regular 1650 became 37.5% more expensive than the gtx 650, although Boosti1650 was cheaper than 1650 But after the release of the R7-250X, the recommended price for the GTX 650 TI Boost dropped to $149.

Modification of the GTX 650 that fell into my hands from ASUS. The card is equipped with one fan and occupies one expansion slot. The radiator consists of aluminum, there are no heat pipes in it, as well as a copper penny. Video outputs are HDMI, VGA and DVI-D. A graphics chip and 4 memory chips are hidden under the heatsink.

Benchmarks

Even without architecture improvements, the new card should be faster due to the increased number of CUDA cores and a 4x increase in the video buffer. But between these cards, Nvidia released 3 generations of graphics cards and 2. 5 architectures (GTX 700 is based on the updated Kepler architecture). And I propose to check how much the performance has grown in several games and a couple of tests.

Initially, CS:GO should have been included in the tests, for which this map would be perfect. But unfortunately, the game has been having constant server problems lately, and even on its own servers, monitoring can signal problems. And since both cards showed abnormally low results, it was decided to exclude the game from the test set.

The test system is a computer with the following characteristics:

• R3-3100 4 cores +SMT; 3.6-3.9GHz
• Gigabyte B450V S2H BIOS F60e
• 2x8GB RAM 3200 MHz (OS)

The computer has undergone memory overclocking and some software changes which can be found in this article. Nvidia GameReady Driver Pack 465.89.

Synthetics and workloads

AIDA GPGPU

For the first test, I decided to use the built-in test for video cards in AIDA64. And here you don’t even need to look closely to find the difference in performance:

3dMark ( Steam, overlay included )

In 3dMark, the most standard TimeSpy test was run, but with a reduced resolution to FHD. You can see the test settings in the screenshot below:

And here you can see a warning about the lack of video memory, because the minimum amount of 1.7 GB is specified in the test requirements. The test itself ran with an average fps value of about 3 and a final result of 619 parrots, with the same settings the 1650S gave out 7488 points.

And for fun, I decided to get a result similar to the GTX650 and set the resolution to 8k. At this resolution, the 1650S delivered a score of 381 parrots.

DaVinci Resolve 17.1.1

In order to dilute the synthetics, I decided to render in DaVinci. And the first problem with the GTX 650 came out right after opening the project: the project has GPU-accelerated elements and the preview of such elements was very slow. The second problem got out after the start of the render: the timer that shows approximately the time until the end of the render estimated its duration at 1.5 hours. The render was canceled, the program was restarted, but the situation repeated itself. Drawing parallels with the OpenCL processing mode on my video card, the GPU processing mode slider was checked and the OpenCL option was selected there. Rendering on CUDA for the GTX 650 turned out to be unavailable. After that, the winner in the comparison became obvious, but to understand the difference, the render was launched, although it ended unexpectedly. Processing was interrupted by a message about the impossibility to process the clip. 1650S on the current version of the program took 13:50 to complete the render.

Games

WD2 (UbisoftConnect)

And I want to start right away with something hard: WATCH_DOGS 2 in FHD, though at minimum settings. And on the GTX 650, the game immediately greets us with a notification of non-compliance with system requirements:

And with such settings and lack of memory, a motorcycle race was made along the main street. With this race, fps constantly kept below the 20 fps mark, giving out an average of 19fps with drawdowns of up to 15 fps. At the same settings, the 1650S averages 70 fps, sometimes hovering between 60 and 80 fps.

9003

FPS	650	1650	Growth in %
minimum	15.9	56.8
9,000 9,000 9,19,19,19,19,19,19,19,19,ETARIS	256
maximum	22.7	91.6	303.5

Tombraider 2013 (STEAM)

Well, let’s go more simply, for example, for example, the overhead of Lara Croft 2013. The game came out half a year after Nvidia introduced the Kepler architecture, so the results should be better. And indeed, even at high settings, the minimum fps indicator did not fall below 30 frames. And if you set the minimum graphics quality, then the average fps will overcome the mark of 60 frames.

278

348.6

9000 384

461.5

Minimum

4000.4000 9000 9000 98

58

210

249.5

2820002 average

maximum

32

39,000 39,4,0003

48

150

186.6

214

9000

373.6 9000 9000

282

9028OUS Unfortunately, they gave me the card only for a day, and then for maintenance, because in the current realities, no one wants to be left without a video card, even without such a plug. And the people who gave me this card thought that their video card would be used in mining if left for a longer time, but even from what I managed to test, general conclusions can be drawn. Over the seven years that these video cards separate, the company has not stagnated and developed its solutions, and even without optimizations for old games, new cards are taken by brute force and greater computing power.

Well, what can we say about the gtx 650 in the current realities? In multimedia tasks, this card can still prove itself, the ideal application for it is displaying an image, surfing the Internet. In the games of its time, the card performs quite well, but in new projects, the gtx 650 rests on 1GB of the video buffer and the old chip. The older models, 650 Ti and 650 TiBoost, are able to offer something else in games due to more memory and a more powerful chip.

The destiny of this card in the current realities is multimedia or e-sports games that are created in order to be played by a large number of people. Here I will recall the test results in CS:GO, which were not included in the selection. Even taking into account the problems with the game, the average frame rate at the minimum did not fall below 60 frames, but in smoke and around it, the FPS dropped to an unplayable 15-20.

Review and testing of ZOTAC GeForce GTX 650 Ti AMP video card! Edition GECID.com.

Page 1

::>Video cards
>2012
> ZOTAC ZT-61103-10M

09-10-2012

Page 1
Page 2
One page

On September 13, NVIDIA, having officially announced the NVIDIA GeForce GTX 660 and GeForce GTX 650 graphics processors, competed in the market with very successful products based on the AMD Radeon HD 7850 and Radeon HD 7870, but until recently a worthy competitor for the Radeon HD 7770 in «green» camp was not observed.

Today, October 9, NVIDIA officially unveiled the new NVIDIA GeForce GTX 650 Ti graphics processor, which will occupy its niche in the hierarchy of the 600 GTX models, between the GeForce GTX 660 and GeForce GTX 650, thereby not only

In the official presentation, a significant part of the material is devoted to comparing the new NVIDIA GeForce GTX 650 Ti graphics accelerator with exactly the solutions that have been on sale for a long time based on Based on the Radeon HD 7770. Taking as a basis for testing 25 popular games, some of which are shown in the screenshot above, NVIDIA specialists, after conducting comparative tests of their solution and that of their opponent, found out that the GeForce GTX 650 Ti outperforms the Radeon HD 7770 by an average of 15%. Whether this statement is true or not, we will find out in the relevant part of the review.

Also, NVIDIA specialists provided comparative graphs of the power of the modern GeForce GTX 650 Ti chip in comparison with the previous generation (GeForce GTX 550 Ti), as well as with the obsolete solution based on the GeForce 9600 GT. It is worth noting, just in time, just such a comparison, because the new generation is based on a 28-nm process technology, which can significantly reduce power consumption and heat dissipation, not to mention the benefits of the NVIDIA Kepler architecture. As for obsolete solutions based on GeForce 9600 GT is that the solutions of those years today are no longer able to satisfy the desires of users when watching high-definition video and the ability to run modern games in popular resolutions of 1920×1080 and 1900×1200 pixels with the activation of anti-aliasing and anisotropic filtering algorithms, not to mention the lack of DirectX API support 11, which did not allow the use of modern game effects. Since economical users, as a rule, upgrade their video subsystem at intervals of 3-4 years, replacing the GeForce 9600 GT on the GeForce GTX 650 Ti, according to NVIDIA, will be timely.

The above die diagram clearly shows that the NVIDIA GeForce GTX 650 Ti is based on the NVIDIA GK106 (GeForce GTX 660) graphics chip, but with one locked GPC (Graphics Processing Cluster) block, which includes two blocks SMX has 192 cores each and with one locked memory controller, resulting in 768 CUDA cores (in NVDIA terminology) and a 128 bit memory bus. It should also be noted that there is no support for GPU Boost technology, but all other innovative innovations first implemented by NVIDIA engineers in the flagship GeForce GTX 680 solution based on the GK104 chip, as well as later on the «full-fledged» GK106 chip (GeForce GTX 660), are fully supported. graphics adapter on NVIDIA GeForce GTX 660 Ti. First of all, we are talking about such options as Adaptive V-Sync, PhysX, 3D Vision Surround, support for FXAA and TXAA anti-aliasing algorithms, ultra-high resolution displays and PCI-E 3. 0 interface.

According to NVIDIA’s stated specification, the NVIDIA GeForce GTX 650 Ti reference card has a core frequency of 925 MHz, and an effective memory frequency of 5400 MHz with a peak power consumption of only 110 W, additional power is provided through a single 6-pin connector. The recommended cost of new items is $149.

Once again, we were pleased by the partners of the Californian giant, providing our test lab with their samples of new graphics accelerators based on the GeForce GTX 650 Ti chip in the shortest possible time.

ZOTAC was the first to please us by providing data on its model range, which includes:

• ZOTAC GeForce GTX 650 Ti 1GB — having 1 GB GDDR5 video memory with an effective frequency of 5400 MHz (like the «reference»), two DVI plus mini-HDMI connectors like the reference card and a slightly overclocked chip with a frequency of 941 MHz;

• ZOTAC GeForce GTX 650 Ti 2GB — having twice the amount of video memory with an effective frequency of 5400 MHz (as in the «reference»), two DVI and already two full-size HDMI connectors and a slightly overclocked chip with a frequency of 941 MHz;

• ZOTAC GeForce GTX 650 Ti AMP! Edition — having both doubled and overclocked video memory to an effective frequency of 6200 MHz, and a significantly overclocked chip to a frequency of 1033 MHz and two DVI plus two full-size HDMI connectors.

In our today’s review, we will focus on the ZOTAC GeForce GTX 650 Ti AMP! Edition as the strongest representative of its lineup.

Specification:

Model	ZOTAC GeForce GTX 650 Ti AMP! Edition (ZT-61103-10M)
Graphic processor	NVIDIA GeForce GTX 650 Ti (GK106)
Supported APIs	Microsoft DirectX 11, OpenGL 4.2
Proprietary technology support	Adaptive V-Sync, PhysX, SLI, NVIDIA Surround, 3D Vision, CUDA
Number of conveyors	768 unified
Core frequency, MHz	1033
Memory frequency (effective), MHz	1600 (6400)
Memory size, MB	2048
Memory bus bit	128
Memory type	GDDR5
Tire type	PCI-E 3. 0 x16
Maximum resolution	Up to 2560×1600 Dual-link DVI Up to 2048×1536 VGA (via adapter) Up to 2560×1600 DisplayPort Up to 2560×1600 HDMI
Outputs	2x DVI-D Dual-Link DVI 1x DisplayPort 1.2 1x HDMI
Support for HDCP and HD video decoding	Yes MPEG-2, MPEG-4, DivX, WMV9, VC-1 and H.264/AVC decoding. Optional NVENC hardware decoding unit.
Dimensions, mm	154 x 112
Recommended PSU W	400
Drivers	Fresh drivers can be downloaded: — GPU manufacturer website .
Manufacturer website	http://www. palit.biz/

From the specifications table, we can conclude that we have a significantly revised version of the graphics accelerator based on the NVIDIA GeForce GTX 650 Ti. The design of the printed circuit board, as well as the appearance of the cooling system, have been seriously changed, but there is no doubt that the “reference” sample was taken as the basis. As we already noted, the base frequency of the graphics core instead of the reference 925 MHz is 1033 MHz (an increase of 12%), and the effective frequency of the GDDR5 video memory standard was increased in the factory from 5400 to 6200 MHz (an increase of 15%), which can be called a rather serious result. like a factory overclock. We will consider the real efficiency of such overclocking in comparison with the “reference” a little later.

Now directly to the packaging.

ZOTAC GeForce GTX 650 Ti AMP! Edition is made in a laconic corporate style with a predominance of black and orange in the design.

The back of the box contains information about the advantages of factory overclocking, as well as the impact of some technologies on the gameplay. It also lists the requirements for the system in which this video accelerator is planned to be installed. Note that the power supply in such a system must have a power of at least 400 W and be capable of delivering 20 A along the + 12V line. Unfortunately, there is no description in Russian, apparently the reason for this is an early sample of the package supplied.

In addition to the ZOTAC GeForce GTX 650 Ti AMP! Edition, in the box we found:

• CD with drivers and utilities;

• Instructions for installing and operating a video card;

• More complete instructions for installing and operating a video card;

• Branded body sticker;

• HDMI-to-DVI adapter;

• Adapter from two 4-pin peripheral connectors to a 6-pin video card power connector.

Such a set of accessories cannot be called rich, however, in general, for most scenarios of using the ZOTAC GeForce GTX 650 Ti AMP! Edition will be enough.

ZOTAC GeForce GTX 650 Ti AMP! Edition is no different from what we saw on the «reference» design graphics card. The interface panel has the following connectors:

• 1x DVI-D Dual-Link DVI;

• 1x DVI-I Dual-Link DVI;

• 2x HDMI 1.4a.

For most users, this set of interfaces will be enough for everyday use. It’s nice to note that the HDMI connectors are full-sized, which eliminates the need for additional adapters even when connecting multiple displays. Also for ZOTAC GeForce GTX 650 Ti AMP! Edition, you can connect three displays simultaneously with support for NVIDIA 3D Vision technology (3D Vision Surround mode), however, you should remember that in this scenario, the load on the video card will increase many times, and the resulting effect may not meet expectations, due to not as much power as the flagship GeForce GTX 680 solutions.

As we mentioned above, the PCB design of the ZOTAC GeForce GTX 650 Ti AMP! Edition differs from the one proposed by NVIDIA engineers. The main changes relate to the location of memory chips, the use of a higher quality element base and a slightly redesigned power converter.

Elements of the 2+1-phase GPU power stabilization module and memory chips are soldered closer to the interface panel. The uPI Semiconductor uP1626P chip is used as a PWM controller.

In addition to the PCI-Express 3.0 connector, additional power for the ZOTAC GeForce GTX 650 Ti AMP! Edition is provided through a single 6-pin power connector mounted on the top corner of the graphics card.

The reverse side of the board is almost devoid of elements. Here we found only cooling system mounts, several components of the VRM module, four memory chips and one tantalum capacitor soldered close to the graphics core, the presence of which can positively affect the overclocking potential and stability of the ZOTAC GeForce GTX 650 Ti AMP graphics accelerator! Edition.

Unfortunately, the GPU-Z utility was unable to fully identify this graphics chip, but we know for certain that it was manufactured in compliance with the 28 nm process technology, includes 768 universal shader processors (or CUDA cores in NVIDIA terminology) and 16 units rasterization, and data exchange with video memory is carried out via a 128-bit bus. As we have already noted, the frequencies of the graphics core and memory in the case of the ZOTAC GeForce GTX 650 Ti AMP! Edition have been upgraded from the factory and are 1033 MHz and 6200 MHz respectively.

GDDR5 video memory with a total capacity of 2 GB is assembled using eight 256 MB chips manufactured by Hynix. The chips are marked H5GQ2h34FR ROC and, according to the documentation, can operate at an effective frequency of up to 6.0 GHz. On the ZOTAC GeForce GTX 650 Ti AMP! Edition video memory operates at a frequency of 6200 MHz, taking into account factory overclocking, so there is no hope for significant manual overclocking.

The cooling system consists of a small all-aluminum heatsink with no heatpipes that has direct contact with the graphics core and an 85cm fan in a nice translucent orange color. The air blown by this fan passes through the aluminum fins of the radiator and cools the memory chips and power system elements at the exit. However, unfortunately, only a small part of the air is blown out through the grille of the second slot of the system, and most of it remains inside the case. In addition, half of the memory chips are located on the back side of the textolite and are not ventilated at all.

To evaluate the efficiency of the applied cooling system, we performed our standard set of tests using the MSI Kombustor utility. The tests were carried out in two cooler operating modes: automatic control and maximum fan speed.

With automatic fan speed control in the maximum load mode, the graphics core heated up to 66 degrees, and the cooler, judging by the monitoring readings, worked at 25% of its power, while creating a fairly low noise level, which can be characterized as “quiet”, however, the operation of the chokes under load stands out noticeably against the background of all the other noises of the test system unit, and even at a distance of 3-4 meters from it, a “squeak” was noticeable.