- Specifications, packaging
- Menu, controls, additional functions and software
- LCD testing
– Micrographs of the matrix
– Color rendering quality assessment
– Measurement of uniformity of black and white fields, brightness and power consumption
– Determining response time and output latency
– Measurement of viewing angles
PRODUCT NUMBER CMI-GM34-CW
SCREEN SIZE 34″
PANEL VA w/ Quantum Dot
RESOLUTION UWQHD 3440×1440
FRAME RATE 144Hz
CONTRAST RATIO 3000:1
BRIGHTNESS 400 nits
COLOR GAMUT 95% DCI-P3 (125% sRGB) (±3)
I/O – DP 2
I/O – HDMI 2
I/O – EARPHONE OUT 1
I/O – DP 1.4 CABLE 1
I/O – SPEAKER 2 (4Ω/3W)
DCR (DYNAMIC CONTRAST RATIO) Yes
PIP/PBP (SCREEN SPLIT) Yes
DISPLAY HDR400 COMPATIBLE Yes
HDR10 SUPPORT (W/ FREESYNC) Yes
FREESYNC FreeSync 2 Compatible, G-Sync compatible
RESPONSE TIME 1ms (MPRT)
EYE PROTECTION Low Blue Light
MODE SWITCH Standard, Graphic, Movie, Game
LIGHTING Purple (Steady & Blinking modes)
HEIGHT ADJUSTMENT 0~100mm
SWIVEL ADJUSTMENT -45°~+45°
TILT ADJUSTMENT -5°~+15°
NET WEIGHT 7.02kg
GROSS WEIGHT 8.96kg
WARRANTY 2 years
The Cooler Master GM34-CW design is strict, the game orientation in an explicit form almost does not give out anything, and only the backlight on the back panel suggests that this is not a simple office monitor. The panels of the screen unit are made of black plastic with a matte surface. The outer surface of the matrix is black, semi-matt (the specularity is well expressed), the outer layer of the matrix is hard.
The plane of the screen is curved along a cylinder so that the right and left edges are slightly pushed forward. The declared bending radius is 1.5 m. The uniformity of the screen curvature of this model is good, although in a vertical strip about 10-15 cm wide at the right and left edges the screen is almost flat, but this is typical for all curved LCD screens. From the front, the screen looks like a monolithic surface, bounded from below by a narrow bar, and along the perimeter by a narrow bezel. By displaying the image on the screen, you can see that in fact there are not wide margins between the outer borders of the screen and the actual display area.
On the bottom right, on the front plate, the button labels are applied in the corporate color. The buttons themselves are mechanical and are located under the icons at the bottom of the screen block. A status light shines down through a tiny hole to the right of the power button. Several bumps help to distinguish it from the rest by touch. In front, the indicator light is not visible, and only in relative darkness on the table under the indicator, when it is lit, can you see a circle of blue light.
On the back panel, behind three white translucent plastic inserts, there are decorative lighting LEDs.
Illumination modes are selected in the setup menu. There are three of them – the backlight is off, lit steadily, blinking. The latter mode can be used if you need to drive crazy the person sitting opposite. The backlight color is still the same corporate one.
All interface connectors and the power connector are located in an open recess on the rear panel and are oriented downward. There are several ventilation grilles on the back panel at the bottom and at the bottom end. There are tiny loudspeakers built into the monitor behind the grilles.
The Cooler Master GM34-CW design of the stand is notable for the fact that the contour of the base follows the contour of the manufacturer’s logo. Made of aluminum alloy, the base, the stand and the screen block holder are anodized and painted in dark gray silver. The cylindrical stand is an all-metal profiled pipe. The stand is of a fixed height, but a spring-supported mechanism allows vertical movement of the assembly to which the screen assembly is attached. This assembly has rollers that slide along the rails inside the rack. As a result, the screen can be set to the desired height with a slight movement of the hand. One hinge in the mounting unit allows the screen unit to be slightly tilted forward from the vertical position, more backward, and the second allows the screen to be rotated to the right and left.
Rubber pill legs are glued to the base of the stand.
The base of the stand allows you to lean on the far edge of the keyboard, and in the space limited by the base, you can fold any office small things. Note that all supporting elements of the stand are metal (steel or aluminum alloy). The construction of the kickstand is quite rigid, but the screen unit is heavy, so it still sways with slight bumps. Cooler Master GM34-CW is very stable on the stand. If necessary, the stand can be detached (or not initially attached) and the display unit can be attached to a VESA-compatible bracket (75 x 75 mm pad, you must use the supplied adapter).
We got the Cooler Master GM34-CW packed in a large, strictly and even gloomily decorated box made of corrugated cardboard. Foam inserts are used inside the box to distribute and protect the contents. You can carry the monitor packed in a box alone, grabbing the plastic handle on top. True, in our case, the handle box has already lost.
Cooler Master GM34-CW have four video inputs – for a pair of HDMI version 2.0 (the inscription 1.4 under one, apparently, is erroneous) and DisplayPort version 1.4. Of these, only DisplayPort supports the input signal with the maximum resolution and frame rate for Cooler Master GM34-CW, as well as operation in the Nvidia G-Sync Compatible mode. Inputs are selected in the menu, in addition, in the absence of a signal at the current input, the automatic selection of the active input is triggered (but not always). The HDMI and DisplayPort inputs are capable of accepting digital audio signals (PCM stereo only), which are output after conversion to analogue to the built-in speakers and via the 3.5mm minijack. An external powered speaker or headphones can be connected to this jack.
The headphone output power was sufficient to provide sufficient volume in 32-ohm headphones with a sensitivity of 92 dB, and even with a small margin. The sound quality in the headphones is poor, since the sound is harsh, only midrange frequencies are reproduced, but at least in the pauses, the noise is not audible.
The quality of the built-in speakers is also poor. Even for the user sitting in front of the monitor, they are quite quiet, there is no low frequencies at all, there is a pronounced “plastic” sound due to the resonances of the body, the stereo effect is weak. Let’s compare the frequency response of Cooler Master GM34-CW with the frequency response of two high-end TVs (pink noise was reproduced, sound pressure levels in 1/3 octaves were recorded by the Octava-110A-Eco sound level meter, 30 dB is approximately the background noise level):
It can be seen that Cooler Master GM34-CW does not have low frequencies, in the middle there are resonance peaks, the range of reproduced high frequencies is narrow.
Menu, controls, additional functions and software
Basically, the Cooler Master GM34-CW menu navigation itself is convenient, the lists are looped, the rate of change of numerical parameters increases with a long press, there is some thoughtfulness, but you can get used to it. Unfortunately, to navigate the menu, you have to use four buttons (rather tight, by the way), which is much less convenient than the five-way joysticks, which are increasingly found in monitors.
If the monitor is working and there is no menu on the screen, then the first time you press any button, the start menu with five icons approximately above the buttons is displayed on the screen.
The functions called in this menu are fixed and cannot be reassigned. The first icon is to call the main menu, the second is to display the sight in the center of the screen, the third is to switch image profiles, the fourth is to enable HDR mode (in fact, nothing changes), the fifth is to turn off the monitor.
The main menu is quite large (white field – the entire display area):
The text in the menu is readable. When adjusting the image, the menu remains on the screen, which slightly interferes with the assessment of the adjustments made. As you navigate through the menus, hints on the current button functions are displayed at the bottom of the menu. The position of the menu on the screen, the transparency of the background and the delay of the automatic exit from the menu are adjusted. From the company’s website, you can download the monitor driver (color correction profile with the installer), as well as the user’s manual.
Of the additional features, there is one “gamer” function: the display in the center of the screen of the sight of four types.
Screen Cooler Master GM34-CW
There are not very many settings, which is expected for a gaming monitor. There are adjustments for brightness (backlight), black level and contrast, a function of dynamic brightness adjustment.
You can correct the gamma curve by choosing preset profiles, as well as color balance by choosing a color profile from two preset ones, or a custom one, in which the color balance can be adjusted by adjusting the intensity of the three primary colors. If a separate setting that adjusts the intensity of the blue component, you can also adjust the hue and saturation.
Additionally, adjustments for contour sharpness and matrix overclocking are available. There are functions of advanced scaling, dynamic adjustment of the gamma curve and black frame insertion.
Combinations of settings are stored in several preset profiles (Picture mode list). When you select most of them, some of the settings become unavailable, and only in the Standard profile you can change everything except the color gamut. The latter is almost always wide and only with an active sRGB profile does the coverage approach the space of the same name.
There are four geometric transformation modes: the image is stretched to fill the screen; enlargement to the borders of the screen while maintaining the original proportions (pixels are considered square); output one-to-one pixel by pixel in the center of the screen and output with an aspect ratio of 4: 3.
There is a function of displaying a picture from two sources at once in the form of a pair of the main window in full screen and a small one in the selected corner of the selected size (out of three). For a small window, we were unable to get a 1: 1 pixel by pixel output. The second version of the demonstration from two sources is to display two images in two (right and left) halves of the screen. In this case, you cannot get point-to-point output, even if you select 1720 × 1440 pixels in the display settings.
In the case of DisplayPort and a professional video card, operation in 10-bit per color mode is supported, but the output to the monitor screen is still in 8-bit per color mode. We run this test using an Nvidia Quadro K600 graphics card and NEC Display Solutions 10 bit Color Depth Demo. This test shows whether programs such as Adobe Photoshop and Adobe Premier Pro that use OpenGL can display a 10-bit color display on a professional video card such as Nvidia Quadro, AMD FirePro, or AMD Radeon Pro.
Cooler Master GM34-CW supports AMD FreeSync technology via DisplayPort and HDMI inputs. For a visual assessment, we used the test utility described in the above article. The inclusion of FreeSync allowed to obtain an image with smooth movement in the frame and without tearing. The range of supported frequencies, which is indicated on the settings panel of the video card, is 48-144 Hz for modes with a frame rate of 144 Hz (connection via DisplayPort) and 48-100 Hz for modes with a frame rate of 100 Hz (for example, when connecting via HDMI). With Nvidia graphics cards, Cooler Master GM34-CW supports G-Sync in G-Sync Compatible mode, but only on DisplayPort input. To check, we used the G-Sync Pendulum Demo utility – the G-Sync mode is turned on, and the effect from turning on is exactly what it should be.
Only when connected to a computer via DisplayPort, resolutions up to 3440 × 1440 are supported at 144 Hz frame rate in 10-bit per color input mode and without loss of color clarity. When connected via HDMI in 3440 × 1440 resolution, you can get a maximum of 100 Hz frame rate.
Judging by the description, Cooler Master GM34-CW meets the requirements of DisplayHDR 400 (however, at the time of publication of the article, it was not on the list of certified ones). One of the compliance criteria is a long-term brightness of at least 320 cd / m² on a white background in full screen and 400 cd / m² when displaying a white rectangle with an area of 10% against a black background, or a short-term increase in brightness to the same value when displaying a white field in full screen after 10 seconds of displaying a black field in full screen. To eliminate the impact of our choice of test applications and images, as well as the combination of monitor settings, we decided to use the official DisplayHDR test tool, which VESA offers to use to verify that display devices meet the certification criteria. When using this program, the variation of conditions is practically excluded, since it is enough to follow the instructions-prompts. In particular, the monitor settings should be reset to their default values, which we did. The result is excellent: a special test gradient showed the presence of 10-bit output with good quality (but the monitor itself copes with this worse than a video card with 8-bit output with dynamic tint mixing). Even on a white field in full screen, the steady-state brightness reaches 480 cd / m². Thus, in terms of maximum brightness and the number of gradations of shades (as well as color gamut), Cooler Master GM34-CW meets the DisplayHDR 400 criteria.
We tested the theater modes using the Sony BDP-S300 Blu-ray player. Work on HDMI was tested. The monitor accepts 576i / p, 480i / p, 720p, 1080i and 1080p signals at 50 and 60 fps. 1080p @ 24fps is supported, but frames in this mode are output with a aspect ratio of 2: 3. In the case of interlaced signals, non-changing areas are displayed frame by frame, moving objects – more often in the fields. The subtle tint gradations for the standard video range vary in both highlights and shadows. The luminance and color fidelity matches the characteristics of the current signal, but with progressive signals, the color fidelity is higher. Interpolation of low resolutions to matrix resolution is performed with good quality.
There is no “crystalline” effect. The matte surface of the matrix allows you to work with comfort in the case of a typical arrangement of the monitor, user and luminaires in the room.
Micrographs of the matrix
A clear image of the pixel structure due to the matte surface cannot be obtained, however, if desired, in the fragment below you can see a typical VA structure in the form of parallel stripes, and also that each subpixel is divided into four equal domains in which the stripes are oriented at different angles (black dots are dust on the camera sensor):
Focusing on the surface of the screen revealed chaotically located micro-defects of the surface, which are responsible for the matte properties themselves:
The grain of these defects is several times smaller than the size of the subpixels, therefore, focusing on microdefects and “jumping” of focus across subpixels when changing the angle of view are weak, because of this, there is no “crystal” effect.
Color rendering quality assessment
The real gamma curve depends on the selected profile in the Gamma list (the values of the indicators of the approximating function are given in parentheses in the captions, in the same place – the coefficient of determination):
Formally, in terms of the exponent value, the real gamma curve is closest to the standard one when Gamma = 2.4 is selected, but the behavior in highlights is better when Gamma = 2.2. Therefore, we further measured the brightness of 256 shades of gray (from 0, 0, 0 to 255, 255, 255) at this value. The graph below shows the increase (not an absolute value!) In brightness between adjacent semitones:
The growth of the increase in brightness on most of the gray scale is more or less uniform, and almost every next shade is brighter than the previous one. There are noticeable deviations in the very shadows and in the highlights. In the shadows, the shade of gray closest to black does not visually differ in brightness from black, and then a couple of shades also do not differ in brightness from the previous ones:
An approximation of the resulting gamma curve gave a score of 2.11, which is below the standard value of 2.2. In this case, the approximating power function in highlights deviates noticeably from the real gamma curve:
An i1Pro 2 spectrophotometer and an Argyll CMS software package (1.5.0) were used to assess the color rendering quality.
Cooler Master GM34-CW original color gamut is much wider than sRGB and is close to DCI-P3:
Selecting an sRGB profile shrinks the coverage to the sRGB borders, but decreases the red saturation more than necessary:
Below is the spectrum for the white field (white line) superimposed on the spectra of red, green and blue fields (lines of the corresponding colors), when there is no explicit gamut correction:
By how narrow the peaks are, it can be assumed that the backlight LEDs use a blue, green and red emitter. However, the description of the monitor mentions “quantum dot technology”, so it is possible that a blue emitter and green and red quantum dots are used.
In the case of adjusting the gamut to sRGB, the components are already largely mixed with each other.
The color balance in the brightest mode (that is, without correction) and even when the sRGB profile is selected is different from the standard, since the color temperature is close to the standard 6500 K, but ΔE is too large. Therefore, we tried to improve it by adjusting the gain of the three primary colors. The graphs below show the color temperature at different parts of the gray scale and the deviation from the blackbody spectrum (parameter ΔE) in the absence of intervention and after manual correction (R = 50, G = 47, B = 50):
The range closest to black can be disregarded, since the color rendition in it is not so important, and the error in measuring the color characteristics is high. Manual override reduced the ΔE value to an acceptable value. However, there is no need for correction for gaming use.
Measurement of uniformity of black and white fields, brightness and power consumption
Cooler Master GM34-CW brightness measurements were carried out at 25 points on the screen, located with a step of 1/6 of the width and height of the screen (screen borders are not included, monitor settings are set to values that provide maximum brightness and contrast while maintaining acceptable image quality). The contrast was calculated as the ratio of the brightness of the fields at the measured points.
White uniformity is very good, and black uniformity, and as a result, contrast is much worse. Contrast for this type of matrices is not the highest by modern standards, but it is still significantly higher than is usually the case with IPS matrices. It is visually visible that the black field is overexposed in some places. The photo below demonstrates this:
The nature of the exposed areas suggests that they appeared mainly due to the deformation of the matrix as a result of its bending. Note that a fairly high contrast reduces the visibility of uneven black field: it is visible only in the dark when black is displayed on the whole screen and after the eyes have slightly adapted to the dark. Also, unevenness is more visible when viewed from a large angle to the screen.
When you turn on the mode with dynamic brightness control, the established contrast formally increases, but not indefinitely, since even on a black field across the entire screen, the backlight does not turn off at all. The graph below shows how the brightness (vertical axis) increases when switching from black (after five seconds of output) to white with dynamic brightness off and on:
It can be seen that in dynamic mode, the backlight brightness increases slowly and nonlinearly to its maximum value. The practical use of this function is zero.
Brightness of the white field in the center of the screen and power consumption from the mains (other settings are set to values that provide maximum image brightness while maintaining acceptable image quality, steady-state value after long-term operation, 144 Hz mode).
In standby mode, the Cooler Master GM34-CW consumes about 0.3W, and in a conditionally off state – 0.2W.
The brightness of the Cooler Master GM34-CW changes precisely by the brightness of the backlight, that is, at the same time, the contrast and the number of distinguishable gradations are preserved. The brightness of the monitor is sufficient for comfortable work even in a brightly lit room, but for conditions of complete darkness, the minimum brightness may seem too high.
At any brightness level, there is no backlight modulation, which eliminates visible screen flickering. For those who are accustomed to recognizing the familiar abbreviation, let’s clarify: there is no PWM. As proof, we present the graphs of the dependence of the brightness (vertical axis) on time (horizontal axis) at various values of the brightness setting:
There is a mode with the insertion of a black frame called MPRT. Dependences of brightness (vertical axis) on time (horizontal axis) for four values of the corresponding parameter:
Clarity in motion does increase (and the manufacturer may indicate an incredible 1 ms with the proviso that this is a characteristic in MPRT mode), but artifacts appear in a dynamic picture, which will be discussed below, and due to flickering with a maximum frequency of 144 Hz, this mode it is recommended to use with caution, as flickering can lead to increased eye fatigue.
The heating of the Cooler Master GM34-CW can be estimated from the images from the infrared camera taken after long-term operation of the monitor at maximum brightness in a room with a temperature of about 24°C.
According to the distribution of heating areas, it can be assumed that the LED backlight bar is located at the bottom of the screen. Rear heating is moderate.
Determining response time and output latency
The response time depends on the value of the setting of the same name, which controls the matrix overclocking. The graph below shows how the on and off times change with the transition black-white-black (“on” and “off” bars), as well as the average total time for transitions between semitones (“GTG” bars) for the four values of this setting:
As the overclocking increases, characteristic bursts of brightness appear on the graphs of some transitions – for example, this is how the graphs look for the transition between the shades of 70% and 90% (the response time settings are shown above the graphs):
Visually, in the case of maximum overclocking, visible artifacts appear. From our point of view, at the penultimate stage of overclocking, the matrix speed is sufficient even for the most dynamic games.
Here are the dependences of the brightness with alternating white and black frames at 144 Hz frame rate for options without overclocking and in the case of maximum overclocking:
It can be seen that even without overclocking, the maximum brightness of the white frame approaches the level of 90% of the white, and the minimum brightness of the black frame drops to the black level. The amplitude of the change in brightness is greater than 80% of the white level. That is, according to this formal criterion, the matrix speed without overclocking is sufficient to display an image with a frame rate of 144 Hz. In the case of overclocking, the amplitude of the brightness change is even higher than 100%.
For a clear idea of what such a speed of the sensor means in practice, what artifacts from overclocking can be and whether the MPRT setting described above increases the clarity in motion, here is a series of images obtained with a moving camera. Such pictures show what a person sees if he follows with his eyes an object moving on the screen. The description of the test is given here, the page with the test itself is here. Note that it is difficult to get good quality pictures in this test in the case of a curved monitor, so the quality is average, but the pictures convey the general trend. We used the recommended settings (motion speed 960 pixels / s), shutter speed 1/15 s, the photos show the refresh rate values, as well as the Response Time / MPRT settings in the form of numbers indicating the effect level.
It can be seen that, all other things being equal, the image clarity increases with the increase in the refresh rate and the degree of overclocking, but even at maximum overclocking, artifacts are not noticeable (at least for this picture). Turning on MPRT increases clarity, but objects in motion are tripled, which reduces the positive effect.
Let’s try to imagine what would happen in the case of a matrix with instantaneous pixel switching. For it, at 60 Hz, an object with a motion speed of 960 pixels / s is blurred by 16 pixels, at 120 Hz by 8 pixels, at 144 Hz by 6.6 (6) pixels. Blurs because the focus of the view moves at the specified speed and the subject is stationary at 1/60, 1/120, or 1/144 of a second. To illustrate this, let’s simulate the blur at 16, 8, and 6.6 (6) pixels:
It can be seen that the clarity of the image, especially after overclocking the matrix, is almost the same as in the case of an ideal matrix.
Duplication / tripping of objects in dynamics when the MPRT function is turned on is explained by the following graph, which shows the dependence of the brightness when alternating between white and black frames at 144Hz frame rate for options with the MPRT function turned off and with the turned on:
It can be seen that the backlight on the black frame turns on when the white brightness has not yet decreased to a minimum. As a result, the pulse of light “reveals” the afterimage, making it more noticeable, hence the parasitic doubling / tripping of objects.
We determined the total output delay from switching the video buffer pages to the start of displaying the image on the screen (recall that it depends on the peculiarities of the Windows OS and the video card, and not only on the monitor). The latency of the image output at 144 Hz (DisplayPort connection) is 6.6 ms, and when you enable G-Sync, the delay increases to 6.9 ms, but the output is many times more stable. In any case, this is a very small delay, it is absolutely not felt when working on a PC, and in games it will not lead to a decrease in performance.
Measurement of viewing angles
To find out how the brightness of the screen Cooler Master GM34-CW changes when you deviate from the perpendicular to the screen, we carried out a series of measurements of the brightness of black, white and shades of gray in the center of the screen in a wide range of angles, tilting the sensor axis in vertical, horizontal and diagonal (from corner to corner for format 16 : 9) directions.
The viewing angles are not very wide in terms of the brightness drop rate. We note approximately the same character of the decrease in brightness when deviating from the perpendicular to the screen in all three directions, while the graphs do not intersect in the entire range of measured angles. With a deviation in the diagonal direction, the brightness of the black field reaches a higher value than in the other two cases. However, the maximum black field brightness is still less than in the case of a typical IPS monitor. Contrast in the range of angles ± 82 ° approaches the 10: 1 mark only with a deviation along the diagonal, but still remains significantly higher than this value.
To quantify the change in color rendition, we carried out colorimetric measurements for white, gray (127, 127, 127), red, green and blue, as well as light red, light green and light blue fields in full screen using a setting similar to that that was used in the previous test. The measurements were carried out in the range of angles from 0 ° (the sensor is directed perpendicular to the screen) to 80 ° with a step of 5 °. The obtained values of the intensities were recalculated in ΔE relative to the measurement of each field at the perpendicular position of the sensor relative to the screen. The results are presented below:
As a reference point, you can select a deviation of 45°, which may be relevant in the case, for example, if the image on the screen is viewed by two people at the same time. The criterion for maintaining the correctness of colors can be considered a value of ΔE less than 3.
It follows from the graphs that when viewed from an angle, at least the primary colors change slightly, but the halftones (especially light blue) change significantly, which is expected for a * VA matrix and is its main drawback.
The Cooler Master GM34-CW is a high-end gaming monitor. It features a large, 1.5m curved, high-definition ultra-wide screen and a neutral design. A decorative design element is the switchable backlight. There is support for refresh rates up to and including 144Hz, AMD FreeSync 2 and Nvidia G-Sync Compatible, and a good HDR implementation. However, despite the obvious focus on gamers, the monitor turned out to be universal, suitable, for example, for comfortable office work, for not very responsible work with graphics, for working in CAD / CAM systems and for watching movies. Further in the lists:
- Low output latency
- Effective adjustable matrix overclocking
- Virtual sight
- Comfortable and adjustable stand
- Flicker-free backlight
- Low intensity blue component
- Picture-in-picture and picture-by-picture modes
- Four video inputs
- VESA platform 75×75mm
- Practical stand design
- No significant flaws