In this article, we will look at the principles of operation of optical mouse sensors, shed light on the history of their technological development, and also debunk some myths associated with optical "rodents".
Who made you...
Optical mice familiar to us today trace their pedigree since 1999, when the first copies of such manipulators from Microsoft appeared on mass sale, and after a while from other manufacturers. Before the advent of these mice, and for a long time after that, most mass-produced computer "rodents" were optomechanical (the movements of the manipulator were tracked by an optical system associated with the mechanical part - two rollers responsible for tracking the movement of the mouse along the x and y axes; these rollers, in in turn, rotated from the ball rolling when the user moved the mouse). Although there were also purely optical models of mice that required a special rug for their work. However, such devices were not often encountered, and the very idea of development of such manipulators gradually came to naught.
The “view” of mass optical mice familiar to us today, based on the general principles of operation, was “bred” in the research laboratories of the world famous corporation Hewlett-Packard. More precisely, in its division of Agilent Technologies, which only relatively recently completely separated into a separate company in the structure of HP Corporation. To date, Agilent Technologies, Inc. - a monopolist in the market of optical sensors for mice, no other companies develop such sensors, no matter what anyone tells you about the exclusive IntelliEye or MX Optical Engine technologies. However, the enterprising Chinese have already learned how to “clone” Agilent Technologies sensors, so when buying an inexpensive optical mouse, you may well become the owner of a “left” sensor.
Where the visible differences in the operation of manipulators come from, we will find out a little later, but for now, let's start considering the basic principles of the operation of optical mice, or rather their movement tracking systems.
How do computer mice "see"
In this section, we will study the basic principles of operation of optical motion tracking systems, which are used in modern mouse-type manipulators.
So, "vision" is optical computer mouse obtains through the following process. With the help of an LED, and a system of lenses focusing its light, a surface area is highlighted under the mouse. The light reflected from this surface, in turn, is collected by another lens and enters the receiving sensor of the microcircuit - the image processor. This chip, in turn, takes pictures of the surface under the mouse at a high frequency (kHz). Moreover, the microcircuit (let's call it an optical sensor) not only takes pictures, but also processes them itself, since it contains two key parts: the Image Acquisition System (IAS) image acquisition system and an integrated DSP image processing processor.
Based on the analysis of a series of successive images (which are a square matrix of pixels of different brightness), the integrated DSP processor calculates the resulting indicators indicating the direction of mouse movement along the x and y axes, and transmits the results of its work outside via the serial port.
If we look at the block diagram of one of the optical sensors, we will see that the microcircuit consists of several blocks, namely:
- the main block is, of course, ImageProcessor- image processor (DSP) with built-in light signal receiver (IAS);
- Voltage Regulator And Power Control- a voltage adjustment and power consumption control unit (power is supplied to this unit and an additional external voltage filter is connected to it);
- Oscillator- an external signal is supplied to this block of the chip from the master crystal oscillator, the frequency of the incoming signal is about a couple of tens of MHz;
- Led Control- this is a LED control unit, with which the surface under the mouse is highlighted;
- Serial port- a block that transmits data about the direction of mouse movement outside the chip.
We will consider some details of the operation of the optical sensor chip a little later, when we get to the most advanced of modern sensors, but for now let's return to the basic principles of operation of optical systems for tracking the movement of manipulators.
It should be clarified that the optical sensor chip does not transmit information about mouse movement directly to the computer via the Serial Port. Data is sent to another controller chip installed in the mouse. This second "master" chip in the device is responsible for responding to mouse clicks, scroll wheel rotation, etc. This chip, among other things, already directly transmits information about the direction of mouse movement to the PC, converting the data coming from the optical sensor into signals transmitted via PS / 2 or USB interfaces. And already the computer, using the mouse driver, based on the information received via these interfaces, moves the cursor-pointer across the monitor screen.
It is precisely because of the presence of this “second” controller chip, or rather due to different types such microcircuits, already the first models of optical mice differed quite noticeably among themselves. If I can’t speak too badly about expensive devices from Microsoft and Logitech (although they were not at all “sinless”), then the mass of inexpensive manipulators that appeared after them did not behave quite adequately. When moving these mice on ordinary rugs, the cursors on the screen made strange somersaults, jumped almost to the floor of the desktop, and sometimes ... sometimes they even went on an independent journey across the screen when the user did not touch the mouse at all. It even got to the point that the mouse could easily bring the computer out of standby mode, erroneously registering the movement, when no one actually touched the manipulator.
By the way, if you are still struggling with a similar problem, then it is solved in one fell swoop like this: select My Computer\u003e Properties\u003e Hardware\u003e Device Manager\u003e select the installed mouse\u003e go to its "Properties"\u003e in the window that appears, go to the tab "Management power supply" and uncheck the box "Allow the device to bring the computer out of standby mode" (Fig. 4). After that, the mouse will no longer be able to wake the computer from standby under any pretext, even if you kick it with your feet :)
So, the reason for such a striking difference in the behavior of optical mice was not at all in the "bad" or "good" installed sensors, as many still think. Believe me, this is nothing more than a myth. Or fantasy, if you prefer :) Mice behaving in completely different ways often had exactly the same optical sensor chips installed (fortunately, there were not so many models of these chips, as we will see below). However, thanks to the imperfect controller chips installed in optical mice, we had the opportunity to strongly scold the first generations of optical rodents.
However, we have digressed somewhat from the topic. We return. In general, the mouse optical tracking system, in addition to the sensor chip, includes several more basic elements. The design includes a holder (Clip) in which an LED is installed and the sensor chip itself (Sensor). This system of elements is mounted on a printed circuit board (PCB), between which and the bottom surface of the mouse (Base Plate) a plastic element (Lens) is fixed, containing two lenses (the purpose of which was described above).
When assembled, the optical tracking element looks like shown above. The scheme of operation of the optics of this system is presented below.
The optimal distance from the Lens element to the reflective surface under the mouse should be between 2.3 and 2.5 mm. These are the sensor manufacturer's recommendations. Here is the first reason why optical mice feel bad “crawling” on plexiglass on the table, all kinds of “translucent” rugs, etc. And you should not glue “thick” legs on optical mice when the old ones fall off or are erased. The mouse, due to excessive “elevation” above the surface, can fall into a state of stupor, when it becomes quite problematic to “stir up” the cursor after the mouse is at rest. These are not theoretical fabrications, this is personal experience :)
By the way, about the problem of durability of optical mice. I remember that some of their manufacturers claimed that, they say, "they will last forever." Yes, the reliability of the optical tracking system is high; it cannot be compared with the optomechanical one. At the same time, there are many purely mechanical elements in optical mice that are subject to wear in the same way as under the dominance of the good old “optomechanics”. For example, the legs of my old optical mouse were worn out and fell off, the scroll wheel broke (twice, the last time irrevocably :(), the wire in the connecting cable frayed, the cover of the case peeled off the manipulator ... but the optical sensor works normally, as if nothing Based on this, we can safely state that rumors about the allegedly impressive durability of optical mice have not been confirmed in practice. And why, pray tell, optical mice "live" for too long? After all, new, more "Perfect models created on a new element base. They are obviously more perfect and more convenient to use. Progress, you know, is a continuous thing. Let's see what it was like in the field of evolution of optical sensors that interest us. Let's see now."
From the history of mouse vision
Development engineers at Agilent Technologies, Inc. they do not eat their bread in vain. Over the past five years, the company's optical sensors have undergone significant technological improvements and their latest models have very impressive characteristics.
But let's talk about everything in order. Chips were the first mass-produced optical sensors. HDNS-2000(Fig. 8). These sensors had a resolution of 400 cpi (counts per inch), that is, dots (pixels) per inch, and were designed for a maximum mouse movement speed of 12 inches / s (about 30 cm / s) with an optical sensor frame rate of 1500 frames in a second. Permissible (with preservation stable operation sensor) acceleration when moving the mouse "in a jerk" for the HDNS-2000 chip - no more than 0.15 g (about 1.5 m/s 2). 
Then optical sensor chips appeared on the market. ADNS-2610 and ADNS-2620. The optical sensor ADNS-2620 already supported a programmable frequency of "shooting" the surface under the mouse, with a frequency of 1500 or 2300 shots / s. Each picture was taken with a resolution of 18x18 pixels. For the sensor, the maximum operating speed of movement was still limited to 12 inches per second, but the limit on permissible acceleration increased to 0.25 g, with a surface “photographing” frequency of 1500 frames / s. This chip (ADNS-2620) also had only 8 legs, which made it possible to significantly reduce its size compared to the ADNS-2610 chip (16 pins), which looks similar to the HDNS-2000. At Agilent Technologies, Inc. set out to "minimize" their chips, wanting to make the latter more compact, more economical in power consumption, and therefore more convenient for installation in "mobile" and wireless manipulators.
The ADNS-2610 chip, although it was a “large” analogue of the 2620, was deprived of support for the “advanced” mode of 2300 shots / s. In addition, this option required 5V power, while the ADNS-2620 chip cost only 3.3V.
Chip coming soon ADNS-2051 was a much more powerful solution than the HDNS-2000 or ADNS-2610 chips, although outwardly (packaging) it was also similar to them. This sensor already made it possible to programmatically control the "resolution" of the optical sensor, changing it from 400 to 800 cpi. The variant of the microcircuit also allowed adjusting the frequency of surface shots, and it allowed changing it in a very wide range: 500, 1000,1500, 2000 or 2300 shots/s. But the size of these very pictures was only 16x16 pixels. At 1500 shots/s, the maximum allowable acceleration of the mouse during the "jerk" was still 0.15 g, the maximum possible movement speed was 14 inches/s (i.e., 35.5 cm/s). This chip was designed for a supply voltage of 5 V. 
Sensor ADNS-2030 designed for wireless devices, and therefore had low power consumption, requiring only 3.3 V power. The chip also supported energy-saving functions, such as the function to reduce power consumption when the mouse is at rest (power conservation mode during times of no movement), switch to sleep mode, including when the mouse is connected via USB interface, etc. The mouse, however, could also work in a non-power-saving mode: the value of "1" in the Sleep bit of one of the chip registers made the sensor "always awake", and the default value of "0" corresponded to the operating mode of the microcircuit, when after one second, if the mouse did not move (more precisely, after receiving 1500 completely identical surface shots), the sensor, along with the mouse, went into power saving mode. As for the other key characteristics of the sensor, they did not differ from those of the ADNS-2051: the same 16-pin package, movement speed up to 14 inches / s with a maximum acceleration of 0.15 g, programmable resolution of 400 and 800 cpi, respectively, snapshot rates could be exactly the same as the above-considered version of the microcircuit.
These were the first optical sensors. Unfortunately, they were characterized by shortcomings. A big problem that occurred when moving an optical mouse over surfaces, especially with a repeating small pattern, was that the image processor sometimes confused separate similar areas of a monochrome image received by the sensor and incorrectly determined the direction of mouse movement.
As a result, the cursor on the screen did not move as required. The pointer on the screen even became capable of impromptu:) - of unpredictable movements in an arbitrary direction. In addition, it is easy to guess that if the mouse was moved too fast, the sensor could generally lose any “link” between several subsequent surface shots. Which gave rise to another problem: the cursor, when moving the mouse too sharply, either twitched in one place, or “supernatural” phenomena occurred in general :) phenomena, for example, with the rapid rotation of the world around in toys. It was quite clear that for the human hand, the limitations of 12-14 inches / s in terms of the maximum speed of moving the mouse are clearly not enough. There was also no doubt that 0.24 s (almost a quarter of a second), allotted for accelerating the mouse from 0 to 35.5 cm / s (14 inches / s - the maximum speed) is a very long period of time, a person is able to move the brush much faster. And therefore, with sharp mouse movements in dynamic gaming applications with an optical manipulator, it can be hard ...
Agilent Technologies understood this too. The developers realized that the characteristics of the sensors needed to be radically improved. In their research, they adhered to a simple but correct axiom: the more pictures per second the sensor takes, the less likely it is that it will lose the “trace” of mouse movement when the computer user makes sudden movements :)
Although, as we can see from the above, optical sensors have evolved, new solutions are constantly being released, however, development in this area can be safely called “very gradual”. By and large, there were no cardinal changes in the properties of the sensors. But technological progress in any field is sometimes characterized by sharp jumps. There was such a “breakthrough” in the field of creating optical sensors for mice. The advent of the ADNS-3060 optical sensor can be considered truly revolutionary!
Best of
Optical sensor ADNS-3060, compared to its "ancestors", has a truly impressive set of characteristics. The use of this chip, packaged in a 20-pin package, provides optical mice with never-before-seen possibilities. Permissible maximum speed movement of the manipulator increased to 40 inches / s (that is, almost 3 times!), i.e. reached the "sign" speed of 1 m/s. This is already very good - it is unlikely that at least one user moves the mouse at a speed exceeding this limit so often that he constantly feels discomfort from using the optical manipulator, including gaming applications. The allowable acceleration, however, has increased, frighteningly, by a hundred times (!), and has reached a value of 15 g (almost 150 m/s 2). Now, the user is given 7 hundredths of a second to accelerate the mouse from 0 to the maximum 1 m / s - I think that now very few will be able to exceed this limitation, and even then, probably in dreams :) The programmable speed of taking surface images with an optical sensor in the new chip model exceeds 6400 fps, i.e. "beats" the previous "record" almost three times. Moreover, the ADNS-3060 chip can itself adjust the image repetition rate to achieve the most optimal operating parameters, depending on the surface over which the mouse moves. The "resolution" of the optical sensor can still be 400 or 800 cpi. Let's look at the example of the ADNS-3060 chip general principles operation of optical sensor chips.
The general scheme for analyzing mouse movements has not changed compared to more early models- microimages of the surface under the mouse received by the IAS sensor unit are then processed by the DSP (processor) integrated in the same chip, which determines the direction and distance of the manipulator movement. The DSP calculates the relative x and y offset values relative to the mouse's home position. Then the external mouse controller chip (what we need it for, we said earlier) reads information about the movement of the manipulator from the serial port of the optical sensor chip. Then this external controller translates the received data about the direction and speed of mouse movement into signals transmitted via standard PS / 2 or USB interfaces, which are already coming from it to the computer.
But let's delve a little deeper into the features of the sensor. The block diagram of the ADNS-3060 chip is presented above. As you can see, its structure has not fundamentally changed compared to its distant "ancestors". 3.3 Power is supplied to the sensor through the Voltage Regulator And Power Control block, the same block is assigned the voltage filtering function, for which connection to an external capacitor is used. The signal coming from the external quartz resonator to the Oscillator unit (the nominal frequency of which is 24 MHz, lower-frequency master oscillators were used for previous microcircuit models) serves to synchronize all computational processes occurring inside the optical sensor microcircuit. For example, the frequency of snapshots of an optical sensor is tied to the frequency of this external generator (by the way, the latter is not subject to very strict restrictions on permissible deviations from the nominal frequency - up to +/- 1 MHz). Depending on the value entered at a certain address (register) of the chip memory, the following operating frequencies for taking pictures by the ADNS-3060 sensor are possible.
| Register value, hexadecimal | Decimal value | Sensor snapshot rate, fps |
| OE7E | 3710 | 6469 |
| 12C0 | 4800 | 5000 |
| 1F40 | 8000 | 3000 |
| 2EE0 | 12000 | 2000 |
| 3E80 | 16000 | 1500 |
| BB80 | 48000 | 500 |
As you might guess, based on the data in the table, the sensor snapshot frequency is determined by a simple formula: Frame rate = (Generator master frequency (24 MHz) / Frame rate register value).
The surface images (frames) taken by the ADNS-3060 sensor have a resolution of 30x30 and represent the same matrix of pixels, the color of each of which is encoded in 8 bits, i.e. one byte (corresponding to 256 shades of gray for each pixel). Thus, each frame (frame) entering the DSP processor is a sequence of 900 bytes of data. But the "cunning" processor does not process these 900 bytes of a frame immediately upon arrival, it waits until 1536 bytes of pixel information are accumulated in the corresponding buffer (memory) (that is, information about another 2/3 of the next frame is added). And only after that, the chip starts to analyze information about the movement of the manipulator by comparing changes in successive surface images.
With a resolution of 400 or 800 pixels per inch, they are indicated in the RES bit of the microcontroller memory registers. Zero value this bit corresponds to 400 cpi, and a logical one in RES puts the sensor in 800 cpi mode.
After the integrated DSP processor processes the image data, it calculates the relative offset values of the manipulator along the X and Y axes, entering specific data about this into the memory of the ADNS-3060 chip. In turn, the microcircuit of the external controller (mouse) through the Serial Port can "scoop" this information from the memory of the optical sensor with a frequency of about once per millisecond. Note that only an external microcontroller can initiate the transfer of such data, the optical sensor itself will never initiate such a transfer. Therefore, the question of the efficiency (frequency) of tracking the movement of the mouse largely lies on the "shoulders" of the external controller chip. Data from the optical sensor is transmitted in 56-bit packets.
Well, the Led Control block, which the sensor is equipped with, is responsible for controlling the backlight diode - by changing the value of bit 6 (LED_MODE) at address 0x0a, the optosensor microprocessor can switch the LED to two operating modes: logical "0" corresponds to the state "diode is always on", logical "1" puts the diode in "on only when needed" mode. This is important, say, when using wireless mice, as it allows you to save their charge. autonomous sources nutrition. In addition, the diode itself can have several brightness modes.
On this, in fact, everything with basic principles operation of the optical sensor. What else can be added? The recommended operating temperature of the ADNS-3060 chip, as well as all other chips of this kind, is from 0 0С to +40 0С. Although Agilent Technologies guarantees the preservation of the working properties of its chips in the temperature range from -40 to +85 °С.
Laser future?
Recently, the web was filled with laudatory articles about the Logitech MX1000 Laser Cordless Mouse, which used an infrared laser to illuminate the surface under the mouse. It promised almost a revolution in the field of optical mice. Alas, having personally used this mouse, I was convinced that the revolution did not happen. But it's not about that.
I didn't understand mouse Logitech MX1000 (did not have the opportunity), but I am sure that our old friend, the ADNS-3060 sensor, is behind the "new revolutionary laser technology". For, according to the information I have, the characteristics of the sensor of this mouse are no different from those of, say, the Logitech MX510 model. All the "hype" arose around the statement on the Logitech website that with the help of a laser optical tracking system, twenty times (!) More details are revealed than with the help of LED technology. On this basis, even some respected sites have published photographs of certain surfaces, they say, as they see their ordinary LED and laser mice :) 
Of course, these photos (and thank you for that) were not the multi-colored bright flowers with which they tried to convince us on the Logitech website of the superiority of the laser illumination of the optical tracking system. No, of course, optical mice did not begin to “see” anything similar to the given color photographs with varying degrees of detail - the sensors still “photograph” no more than a square matrix of gray pixels that differ only in different brightness (processing information about the extended color the palette of pixels would be an exorbitant burden on the DSP).
Let's estimate, to get a 20 times more detailed picture, you need, sorry for the tautology, twenty times more details, which can only be conveyed by additional image pixels, and nothing else. It is known that the Logitech MX 1000 Laser Cordless Mouse takes pictures of 30x30 pixels and has a maximum resolution of 800 cpi. Consequently, there can be no question of any twenty-fold increase in the detailing of images. Where did the dog fumble :), and are such statements generally unfounded? Let's try to figure out what caused the appearance of this kind of information.
As you know, a laser emits a narrowly directed (with a small divergence) beam of light. Therefore, the illumination of the surface under the mouse with a laser is much better than with an LED. The laser operating in the infrared range was chosen, probably in order not to dazzle the eyes by the possible reflection of light from under the mouse in the visible spectrum. The fact that the optical sensor works normally in the infrared range should not be surprising - from the red range of the spectrum, in which most LED optical mice work, to the infrared - "within a stone's throw", and it is unlikely that the transition to a new optical range was difficult for the sensor. For example, the Logitech MediaPlay manipulator uses an LED, but also provides infrared illumination. Current sensors work without problems even with blue light (there are manipulators with such illumination), so the spectrum of the illumination area is not a problem for sensors. So, due to the stronger illumination of the surface under the mouse, we can assume that the difference between places that absorb radiation (dark) and reflect rays (light) will be more significant than when using a conventional LED - i.e. the image will be more contrast.
Indeed, if we look at real images of the surface taken by a conventional LED optical system and a system using a laser, we will see that the "laser" version is much more contrast - the differences between the dark and bright areas of the image are more significant. Of course, this can significantly facilitate the work of the optical sensor and, perhaps, the future belongs to mice with a laser illumination system. But it is hardly possible to call such "laser" images twenty times more detailed. So this is another "newborn" myth.
What will be the optical sensors of the near future? It's hard to say. They will probably switch to laser illumination, and there are already rumors on the Web about a sensor being developed with a “resolution” of 1600 cpi. We can only wait.
It is not the most important component of the entire computer as a whole, but no, without it, working on a PC turns into a very difficult, not enjoyable activity. World brands A4Tech, Logitech, Defender are constantly fighting with each other to create the most in the world. That's why for now different kinds computer mice are constantly undergoing changes for the better. If you constantly follow all the new products on the computer mouse market, and at the same time buy at least one of latest models, you can simply be left without money.
Surely many of you remember the first mice that recognized movements and coordinates thanks to the rubber ball inside. Everything old is always replaced by a new one, which is why today mechanical manipulators are remembered less and less. came to replace the mechanical one, absorbing all of its best qualities. However, in less than a few years, the laser mouse, the latest development of various input device companies, was already knocking on the door.
Top Choice: Laser Mouse or Optical Mouse?
While the guys from A4Tech have not yet come up with a new best principle for recognizing coordinates with a mouse, every user of a computer, laptop or netbook has a choice: a laser mouse or an optical one. That is why it is necessary to understand the advantages and disadvantages of a laser and optical mouse, in order not to experience any difficulties in the future when using one of the options presented.
Undoubtedly, a computer mouse, in addition to moving the cursor around the screen, has two important features - accuracy and speed. These words will be confirmed by any professional gamer. In the race for precision, a mechanical manipulator has no chance against new input devices. Therefore, whether it be optical or optical, they have gone far in the race for accuracy from a mechanical mouse.
By itself, the principle of operation of both types of mice is the same: the sensor takes a photo of the surface, and the chip inside the mouse analyzes this photo and determines the coordinates. When operating an optical mouse, as well as a laser mouse, the surface from below the manipulator is highlighted. This is done for a better and more accurate picture, which will be taken by a special reading element, only LEDs work in an optical mouse, while a laser directly works in a laser mouse. By the way, the laser illuminates the readable surface better, as a result of which the image quality of the laser image is much clearer than that of the LED. It turns out that the laser mouse is more accurate than the optical one, because the laser is several times more accurate than the LED and does not distort the image being read. This is the so-called slight difference between a laser mouse and an optical one.
However, in addition to accuracy in a good manipulator, resolution and speed are very important. Resolution is measured in units called dpi (dots per inch in Russian). Again, a laser mouse has a resolution of up to two thousand, while an optical mouse can boast only one thousand two hundred dots per inch. In truth, eight hundred dpi is considered the most appropriate and convenient extension for a pleasant mouse experience, but computer manipulator companies simply use these indicators as a small marketing ploy. If desired, the resolution of the mouse can be adjusted in the control panel, and then you will personally feel all the pros and cons high resolution manipulator.
Optical mice are available in two PS / 2 interfaces, while laser mice only with USB interface. USB technology is more narrow-profile, and may be smaller than PS/2. Therefore, the cursor will move around the screen not so smoothly.
Now that you've taken a closer look at input devices, try to decide if a laser or optical mouse is best for you, and be sure to try both options when buying.
The sensor is the main component of the mouse. It is the sensor that determines to the greatest extent how well the mouse will transmit the movements of your hand. This material is short review of the most popular optical sensors, In the first part we will look at the weakest of the gaming sensors on the market.
It should be remembered that manufacturers gaming devices do not produce sensors themselves, but use products from other companies. Therefore, as a rule, mice built on the same sensors, but from different manufacturers, will have similar characteristics, such as tracking quality, resolution (dpi), maximum speed, etc. However, some companies (especially large ones) may make significant modifications to the optical system of their device, may use specific firmware, microcontroller, etc. This must be borne in mind.
Note.
After Pixart bought Avago's (Agilent's) "touch" division, there is virtually one manufacturer in the gaming device market - Pixart.
WEAK SENSORS
Avago A5050
In fact, the A5050 is not a gaming sensor (it is a simplified version of the truly gaming A3050). However, the market is now great amount cheap mice with this sensor from different manufacturers who claim that their mouse is really gaming. THIS IS NOT TRUE. Behind the beautiful wrapping of a $10 mouse from aliexpress (or from a hypermarket shelf) is an ordinary office sensor, the main disadvantages of which are a very low speed limit (less than one meter per second: no more than 8g acceleration) and a high angular error. Of course, this is quite enough for fans of tanks and DotA, but the A5050 is definitely not suitable for a serious game. It is interesting that the official specification says that the maximum dpi for the A5050 is 1375. But in fact, manufacturers write what they want. You can find mice and 1600 dpi and 2400 dpi and 5500 dpi.
Conclusions: A5050 is one of the worst sensors on the gaming market
 |
| "Game" mouse Aula killing the soul. 2000 dpi. Sensor A5050. |
Pixart 3305
The history of this sensor is closely related to the SteelSeries Kana mouse. In the face of a new series of mice from SteelSeries (Kana and Kinzu v2), the entire gaming world was waiting for the killers of the reference Microsoft 1.1A at that time. As a result, we did not see anything unusual. Kana and Kinzu v2 received an extremely mediocre sensor - Pixart PAW3305. With a rather large 32x32 pixel matrix, the photographing speed in Pixart 3305, for some reason, turned out to be only 3600 frames / s, which ultimately made this sensor technically worse than the previous generation (A3060/A3080/S3888)
The standard PAW3305DK has a very low top speed of 1.5-2 m/s. This is not enough for professional players - the sensor "breaks" with sudden movements. As a solution to this problem, a variant of the PAW3305DK-H sensor with a weaker (0.5x) lens was proposed. This helped increase the top speed to an acceptable 3+ m/s. At the same time, dpi, of course, decreased by half and 3200 dpi on mice with the "H" modification is a myth.
The second issue is angular binding. It's not very big, but it's there.
Also in PAW3305 there is a slight positive acceleration and an incomprehensible, but small sensor lag on some models.
As a result, mouse movements with the PAW3305 sensor are perceived as slightly irregular and unnatural.
Interestingly, the former player of the NaVi Starix team played SteelSeries Kana (PAW3305-H) for some time, after which he soon left the team and took up the position of a coach.
Now Pixart 3305 is installed in A4tech V-series mice, as well as in almost any mouse that says 3200 dpi, of which there are thousands on the market now.
Conclusions: Pixart PMW3305 is quite suitable for novice players. But definitely not for real gamers.
 |
| A4tech Bloody V3. PAW3305. Convenient and inaccurate. |
Avago A3050
Now Pixart A3050. The cheapest and easiest gaming sensor from Pixart. Matrix 19x19 pixels and 6666 fps. Maximum 4000 dpi (2000 is often indicated on the box, sometimes 3500). But 4000 dpi is better not to use - absolutely not playable. The best dpi values for this sensor are 500 or 1000 (slightly worse). Often, mice on the A3050 have a high breakaway distance. Available adjustable (but not completely) angular binding. And in general, there are problems with calculating angles (which is natural for such a small matrix). In addition, the A3050 has quite strong acceleration. Of the pluses - a quick response of the sensor to the movements of the user, as well as a maximum speed - about 3 m / s, depending on the rug.
The A3050 sensor is very popular thanks to A4tech's A-series mice. It is also actively used by many other manufacturers. The most famous models on A3050 are SteelSeries Kinzu v3 (quickly withdrawn from the market in favor of Rival 100), CM Storm Xornet.
Conclusions: The Pixart A3050 is an unremarkable mid-range sensor. It is better to add a little and take something more worthy.
Conclusion
Unfortunately, we had to talk about sensors, the purchase of which does not promise anything good. Currently, in the same price range, there are many strong mid-range models, such as A3090, PMW3320, AM010, without any significant drawbacks, which have much best ratio price quality. As a result, the material from the series "how not to do it" turned out. However, the information provided will be extremely useful to those who are going to buy an inexpensive device.
The computer mouse is perhaps the most widespread and widespread computer device. Since its invention in 1963, the design of the manipulator has undergone fundamental technological changes. Forgotten are mice with direct drive from two perpendicular metal wheels. Nowadays, optical and laser devices are relevant. Which computer mouse is better - laser or optical? Let's try to understand the differences between these two types of mice.
Design
A modern mouse manipulator has a built-in video camera that takes pictures of the surface at an incredible speed (more than a thousand times per second) and transmits information to its processor, which, by comparing the pictures, determines the coordinates and displacement of the manipulator. To make the pictures better, the surface should be highlighted. Various technologies are used for this purpose:
Optical mouse
It uses an LED, the operation of which allows the sensor to receive better, and the processor to read information faster and, accordingly, determine the position of the device.
laser mouse
For contrast illumination of the surface, not an LED is used, but a semiconductor laser, while the sensor is tuned to capture the corresponding wavelength of this glow.
Photo: compress.ru
Resolution
The abbreviation dpi, which we often see on price tags in stores where mice are sold, means the number of dots per inch, i.e. resolving power. The higher it is, the better the sensitivity of the device. For normal work on a computer, 800 dpi is enough - an optical mouse is also suitable, but for amateurs virtual games and professional artists-designers need a higher resolution of the manipulator - so they are better off buying a laser computer mouse.
Optical mouse
For most of them, this figure is 800 dpi, while the maximum is 1200 dpi.
laser mouse
They have an average resolution of 2000 dpi, while the maximum exceeds 4000 dpi, and not so long ago, laser mice with a resolution of 5700 dpi appeared on the market, which also allow you to control the value of this indicator to save energy.
Price
Optical mouse
Cheaper - cost from 200 rubles.
laser mouse
Quite expensive: from 600 to 5000 rubles and more (top gaming models)
Speed and accuracy
A semiconductor laser that emits light invisible to the eye in the infrared range is more accurate, the reading of information is better, and therefore the positioning of the mouse is more accurate. Criteria such as speed and accuracy are improving. This is especially true for gamers, as well as for graphic designers - they are better off choosing a laser mouse.
Photo: www.modlabs.net
Power consumption
A laser mouse, compared to an optical LED mouse, consumes much less power. This is especially important when using a wireless mouse, where the issue of saving battery or battery power is vital. For wired manipulators, this factor is insignificant.
Working surface
Even the simplest member of the LED mouse class doesn't need a mouse pad as it works on almost all surfaces. The exceptions are transparent glass, glossy and mirror. Here, the LED mouse will operate with such malfunctions that you just have to put a mat under it. But laser illumination is practically indifferent to the material of the mouse movement plane, such devices can easily cope with any surfaces, including mirror ones. But, there is one nuance. For a laser mouse, close contact with the working reflection plane is very critical. The appearance of a gap of even 1 mm significantly complicates the operation of such a device, and the LED can even work on the knee.
Photo: www.engineersgarage.com
Backlight
Another drawback of the LED mouse, which is noted by many users, is the glow (more often red, less often blue or green) even when the computer is turned off, which is not always convenient and pleasing to the eye - for example, at night when you are trying to fall asleep, but with computer desk a fairly bright light shines. In lasers, there is no glow, because, as mentioned above, it emits infrared light invisible to our eyes.
Photo: topcomputer.ru
Such characteristics of a mouse manipulator as ergonomics, beauty, color, manufacturing material, tactile sensations, the number of additional buttons are purely personal and depend on human preferences.
Summing up: advantages and disadvantages
Optical LED mouse
Advantages:
- low price;
- the gap between the mouse and the work surface is not critical.
Flaws:
- does not work on mirror, glass and glossy surfaces;
- low accuracy and cursor speed;
- low sensitivity;
- distracting lighting;
- high power consumption in wireless version.
Optical laser mouse
Advantages:
- work on any working surfaces;
- high accuracy and cursor speed;
- high sensitivity and ability to control resolution;
- no visible glow;
- low power consumption in wireless design;
- the ability to use many additional function buttons.
Flaws:
- high price;
- criticality to the gap between the mouse and the work surface.
Which mouse is better to buy - laser or optical?
Based solely on specifications, then laser mice are better than optical LED devices in almost all respects. But does this mean that we must definitely get rid of the optical mouse? So far, she's done an excellent job.
The choice is always yours. For a laser mouse, you will have to pay a fairly large amount. Well, if you are a gamer or a designer, then the investment will quickly pay off (either materially or morally). If you ordinary user office programs and the Internet, then you most likely will not even notice any qualitative leap in the level of accuracy of the response of the manipulator. Another thing, if required wireless mouse- then it is better to buy a laser mouse instead of an optical one. By purchasing a laser, you will save a lot on batteries - it holds a charge several times longer than an optical one.
There are many types and designs of mouse pads. They can have a working surface made of fabric, soft or hard plastic, metal. The first option is the cheapest and, oddly enough, one of the best. For soft plastic, and for hard plastic too, there are options for optical and laser mice.For tests, we used regular mats Nova MicrOptic+ and Defender Ergo opti-laser. Appearance they are about the same:
According to both manufacturers, these mousepads are optimized to work with laser mice. Let's check.
First, zoomed surface shots:
There are some differences, but not particularly noticeable. Nova's grains are smaller and less pronounced. Does that mean he's worse?
Now let's look at the mats through the eyes of an optical sensor:
Agree that there is a difference and very cardinal. The high-contrast structure is clearly visible on the Nova mat, but the Defender gave some kind of "soap". Most likely, this is due to the size of the "granules". In laser sensors, unlike optical sensors, the visible window size is reduced. It seems that on the Defender mat the size of the granules is larger than the window and the sensor captures only part of them, constantly switching between monotonously light and dark areas. For comparison, I will give photos of the surface of the plastic.
The right figure is obtained from the left by increasing the contrast. The mouse sees this surface like this:
On such a surface, "office" optical mice do not work at all, but laser ones somehow manage to work very successfully.
Tear-off height
What do you do when the mouse reaches the edge of the pad? You pick up the mouse and move it to a new place, to the center of the mat. The optical sensor is highly sensitive and tries to maintain normal functioning constantly adjusting the parameters of the equipment. As a result, when the mouse is raised above the surface, the speed decreases. More precisely, the speed does not decrease, but rather the quality and reliability of motion detection drops quite sharply. Theoretically, when the surface quality drops below reasonable, the optical sensor should stop producing movement. That is, with some lifting of the mouse, he should not notice that the mouse has been raised, and if it is even slightly raised, then simply stop transmitting the movement. This is ideal, but in real mice, when the surface deteriorates, the quality of the movement transmitted by the mouse degrades. Moreover, this harmful effect depends on the speed of movement, which is why it is more difficult to get used to such a mouse.
The break-off height of LED mice is 1.5-2 mm, for laser versions the figure is larger and is already 2.5-4 mm. These are all numbers, but in reality such a mouse is inconvenient to use even for office applications, you have to lift it very high above the rug. According to my personal impressions, a stall height of 1.5-2 mm is quite comfortable. But what to do with laser mice and their 4 mm stall height?
Let's take one by the tail and look at the insides. Mice on the Avago sensor are now common (link to http://www.avagotech.com) ADNS-6010
In order not to be too smart, I took a picture from the documentation.
Explanations:
This figure shows the figure 2.4 mm - this is the optimal distance from the bottom of the optical system to the surface. One point - the bottom of the mouse has some thickness, so the distance from the surface to the bottom of the mouse will be less by the thickness of this bottom.
And what does the separation height depend on and why is this height smaller on optical mice? Let's see another picture:
I allowed myself to show amateur activity to paint some important elements of the design.
The lenses of the optical system are highlighted in yellow, the light flux of the laser is highlighted in gray. Green - visibility zone of the optical sensor. The zone of "visibility" of the sensor is determined only by its focus and the ability to work with a defocused image. The higher the speed of moving the picture, the worse the stability for unfocused objects should be. If you look at the test data, then it turns out. The stall height of 4 mm is not functional, I tried to reduce this value by slightly changing the principle of operation - the loss of the image by the sensor can be obtained not due to deterioration in focus, but due to the departure of the light spot from the sensor's visibility zone. This is how LED mice work. To do this, I increased the angle of the backlight beam from 21 degrees to about 50 degrees from vertical.
When the mouse is lifted, the backlight spot (gray beam) goes out of the visible sensor window (green zone).
The refinement technique is not particularly difficult - you need to cut the optical block along vertical line and not touch the lenses. In extreme cases, you can slightly damage the backlight lens, it is not so important. You can fasten the two components with hot melt adhesive, marked in brown in the figure.
It has sufficient rigidity and strength of the connection, while allowing multiple correction of the position of the glued parts of the optics. When the backlight is tilted, part of its design will go beyond the dimensions of the optics unit and it will have to be filed a little, marked in blue in the figure.
Unfortunately, the backlight unit must not only be tilted, but also moved down, which will cause the backlight lens to be below the level of the optics. This is bad, in the bottom of the mouse you will have to melt a small dent under the ledge. However, this is not difficult and does not interfere, because the lens goes beyond the dimensions quite a bit. The laser module was attached to the optics using a VCSEL Clip. Now it will have to be removed and secured with a drop of glue or sealant. Although, he's holding up pretty well. Such a construction has one feature - the illumination beam falls on the surface with a different angle than the angle of view of the sensor. As a result, an angle of about 15 degrees is formed between the surface plane and the reflection plane.
Black - a beam on an unmodified optical system, green - after completion. The surface for the modified case has been conditionally raised so that it does not merge with the normal mode. The sensor looks, as it were, from the side at the surface and sees all the irregularities on it more clearly. The additional slope of the backlight gives additional brightness modulation when passing through volumetric areas under the lens. Whether this is good or bad depends on the rug, the texture of its surface. By the way, if you take pictures of the surface of the Nova pad on this modified mouse, then the photo will not have such clear edges. And, most likely, it's not about focusing. The angle of view has simply changed and the clear structures of the rug have disappeared. Nova and Defender look almost the same on this mouse. However, the mouse walks well on both surfaces. Alas, there is also a clear drawback - due to the fact that the reflection surface is tilted relative to the surface of the rug, the overall level of illumination decreases and it becomes necessary to increase the backlight laser current. Usually it is a figure in the region of eight milliamps. After refinement, I had to increase the current to 12 milliamps. It's already too much, but within the reach.
If you are finalizing a regular, serial mouse, then it would be nice to help the circuit a little. automatic control laser current. The documentation for the ADNS-6010 sensor mentions the Rbin resistor from pin 13 of the microcircuit. Usually, its face value is 12.7 com. In order to correct the current, it is necessary to reduce its value. For my case, it would be good to increase the current by 1.5 times, which means soldering another resistor in parallel with this resistor with a rating of 2 times more, i.e. 24-27-30KOm. And a couple of surfaces - fabric and aluminum sheet. Quite often you hear recommendations to use these surfaces, they give very good results.
First, on a mouse with unmodified optics (W-Mouse 730). Textile:
Aluminum sheet:
And a mouse after modification of the optical block (W-Mouse 750).
Aluminum sheet:
On a surface with a three-dimensional relief, the modification of the optics leads to a greater visibility of this relief. But the picture from the aluminum sheet looks rather worse, but not so significantly. Nothing happens for free. They touched the optics - they got problems with focusing.
Recommendation - when repeating such a refinement, do not get carried away! It is hardly worth increasing the angle of the backlight so much, because the stall height turns out to be too small and unpleasant problems arise with pushing it into the case and increasing the laser current.
There is an easier way to reduce the height of the fall - put the button on the bottom of the mouse and, when it is raised, turn off, block the sensor. There are many means of influence, at first I tried to turn off the laser, but the controller in A4 is smart and if you just open the laser current, the controller notices this very quickly and turns off the mouse. Alas, it turns off completely, you have to poke the USB connector, you will have to act not so straightforward. There is a proposal to connect a pair of silicon diodes instead when turning off the laser, but this will require installation additional components. I acted differently - I acted on the resistor Rbin (see the documentation for the ADNS-6010 sensor), with an increase in its value, the auto-regulation system tries to set such a current. If Rbin is disconnected or made very large, then the laser will actually turn off, but this will not cause any problems within the regulation system.
I took the "button" itself from a 3.5" floppy disk drive. The force is small, but I had to loosen it a little. The idea worked well, you can choose the height you like, but the plastic pin of the button quickly wears off.