Before I was lucky enough to participate in Auntie Zhang's "Millet TV 3s 65" version of the test. After the publication of the official testimony, it also triggered everyone's heated discussions, but it was rather surprising that this time everyone received acclaim, the comment area did not appear in the expected "war situation." Many of them also questioned Xiaomi's 65-inch and micro-whale's 65-inch which is more worth buying, but the landlord did not come in contact with the micro-whale 65 inches. Not to mention contact, even to see it, so this The problem is really unable to answer everyone. For the purchase of televisions, especially if there are certain requirements on the picture quality and specific functions, I strongly recommend that you bring test videos and pictures, go to the scene, record the problems found, and then go to see another candidate Brand, is not the same problem. Online evaluation articles, because of the evaluation focus, depth, and the reviewer's own limitations, only a certain reference, such as the landlord, television is used to see the US drama based, so do not care about dynamic compensation and smearing issues, Instead, I was always worried about the "dynamic contrast" I couldn't turn off. However, using television to watch the game and value-playing friends may not be able to find the problem for 800 years. First of all, we must affix the link of the previously tested articles. The content of the previous evaluation, in addition to correcting some errors, the same content will not be repeated:
The main purpose of writing this original book is to correct some of the flaws that occurred in previous testimonials. Originally I wanted to add an explanation to Zhang Aunt in the original testimony. However, many value-friends who read the original text may not look at the additional part. Therefore, they should write a separate article.
The first thing to say is the HDMI input function in the previous article that I did not thoroughly evaluate because of my negligence or laziness. At that time, some of the screen problems caused by the previous image quality engine led me to conclude that:
This is now corrected, and this paragraph is completely wrong, because when the HDMI is connected, two display modes will be added to the picture quality parameters, namely "game mode" and "monitor mode". In the "monitor mode", the picture will be drawn. The quality engine does not make any adjustments to the HDMI input screen. So we can use the “monitor mode†when using other devices to connect TVs via HDMI to get controllable colors. It should be noted here that after connecting to HDMI, only the display mode will bypass the host's image quality engine in the two modes (game mode, monitor mode) where there are more image parameters, so switch to display mode or monitor mode in other modes. When switching to other modes, the screen will briefly turn on or off the picture quality engine. This is normal.
Therefore, the landlord can also use the school color meter to measure the performance of the millet TV 3s 65-inch version without the intervention of the picture quality engine. Of course, using a color correction instrument to evaluate a TV is a non-conventional alternative. It has been said before that the entertainment-oriented display device and the working-oriented display device start from the original intention of designing in two directions. One is to allow display. The picture that comes out is more like people, and the other is how to restore the original effect of the picture more accurately. This difference is probably like the difference between the monitor box and the HI-FI box in the speaker world.
The use of entertainment television for work purposes can be said to have been a non-professional business. However, in certain groups, the use of television is not a minority. The specific groups referred to here are basically photography and design industries. Although there are many projectors in use today, first of all, not to mention that a hardware can achieve high-resolution output (4k and 1080P are not at all an order of magnitude for the details of the show), and can basically restore the color of the original projector how much money Even if the projectors are all OK, then how about the corresponding display environment, including shading, curtains, and projection space, and how much does it cost? Believe this fee is also a great expense for a small studio. So many small studios now widely use large-size flat-panel TVs for demonstration purposes. In all the testimonials, there are value-friends who question the accuracy of the screen shots to express the quality of the TV. Therefore, this time testing with spiders can be even more effective. Objectively analyze the screen performance of the millet TV 3s 65-inch version.
The color correction equipment used in this part of the test is Red spider 5 (Spyder5 Elite), which is currently the most common color correction equipment. The calibration analysis software uses the open source DispCal. Everyone may be familiar with the spider's original software, but the original software It can only be said to be simple to use, but the accuracy and quantity of the sampling are slightly insufficient. (The report reports that the size of the colorimetric plate is 1005, which is far higher than the spider's 48. However, the result is that the whole process is very long and the report is about 5 hours. One device). The computer used is RMBP. Millet TV HDMI2.0 is set to on, the color dynamic range is automatic, and the image parameter mode is display mode.
Let's start with the next test, if you only test a TV alone, because the instrument's own error, and many of the value of the color management is not very understanding of the value of the Friends of the full text may feel in the clouds, even after reading the landlord does not know Forced to spend a long time this machine is considered bad. Therefore, in addition to some basic instructions, an additional RMBP15 (Samsung 2012 screen) application will be set up for calibration analysis as a comparison. Of course, not that rmbp's display is good, but the unified test environment. Obviously, comparing a monitor with a TV is obviously unfair, so it will be added to the LeTV S50 as a comparison (all reports will be shared by everyone on the shared disk at the end of the text):
1. Calibration settings
The basic setup is this: the target is a color temperature of 6500K, a Gamma value of 2.2, Profiling uses a sample library of 1136 color patches, and the brightness is not the target, it is the actual measurement value, of course, the brightness of all devices is adjusted to the maximum, although Generally, the monitors we use to adjust the work will set the target brightness to 100 or 120 nits, but this time we measured the display devices for entertainment, so we will not specify them. The test position is the center of the screen.
2. Measurement report settings
Here we use the XXXL verification colorimetric board (video). In the past, we can see that many evaluations only use the datacolor's own monitor analysis. The color accuracy of this piece of color is only 48, and even some people use 12,24 colors. Blocks came out of the report. To tell the truth, such a report is still convincing.
Color gamut testThe basic knowledge of the color gamut has been almost universalized by various value-added friends. The purpose is that the wider the color gamut, the better. Of course, the more expensive the price, the more inevitable it is. The line is better. Here is to explain that the data of this comparison is calibrated data, not the data before the screen calibration. Let's look directly at the test results below:
The figure above is an overview of the Retina Macbook Pro (Samsung Screen, 2012) color correction file. It can be seen that this screen self-test after calibration shows that the mean value of the E color shift is only 0.31, the maximum value is 1.54, and the RMS RMS is also only 0.38, the smaller the value is, the better. (Details about ∆E will be explained later in the chapter on color accuracy.) The Gamut coverage is 86.4% sRGB, 59.7% Adobe RGB and 63% DCI P3.
This is an overview of the millet TV 3S 65-inch color correction file. The average value of this screen calibrated ∆E is 0.41, which is slightly lower than that of Apple. The maximum difference ∆E1.99 is a little larger and the RMS is also higher. Some of this is 0.49. Continuing to look at the color gamut, Xiaomi TV's color gamut coverage is 89.4% of sRGB, 76.5% of Adobe RGB, and 75.9% of DCI P3. Just look at this item. Doesn't it feel that millet TV's color gamut is larger than RMBP? But what everyone needs to know is that even though the screen of RMBP is very good in the notebook, it is not up to the professional and wide color gamut.
Finally, we look at music as the S50, is worthy of the cheapest 50-inch smart TV of the year, and its calibrated ∆E average is 0.82, the maximum even as high as 6.75, RMS corresponding to up to 1.01. The gamut coverage is 92% of sRGB, 69.6% of Adobe RGB, and 74.6% of DCI P3. The difference between color gamut and Xiaomi is not very great.
Everyone likes to compare the maps of the color gamut. Although I don’t think it’s a big deal, I’m still playing with sRGB. From left to right, it’s RMBP, Xiaomi TV, and LeTV S50. (The gray dotted line is the sRGB color gamut range)
Here we have to point out that, in the above picture, you can see that there is a Gamut volume behind the Gamut coverage value. This gamut volume refers to what the display can do. Display colors, such as the sRGB gamut volume of millet TVs, are as high as 123.5%. Theoretically speaking, a range of more than 100% must be able to display all the colors contained in the sRGB color gamut, but it can display a certain color. And two concepts that can accurately display a certain color are completely different. The gamut coverage refers to the fact that under the conditions set by us, with a color temperature of 6500K and a gamma value of 2.2, millet TV can only accurately cover 89.4% of the sRGB color gamut, and you may have doubts about this. The å•¥RMBP sRGB gamut volume is only 91%, but its sRGB color gamut coverage can reach 86.4%? This is because even if you do not calibrate the RMBP screen, its original display state is very close to the calibration target we set. Please see the calibration curves of the following three displays. (The order is still RMBP from left to right, Xiaomi TV, LeTV S50, then the puzzles are all in this order)
This is a RGB tri-color calibration curve. You can see that there is a grey line that is 45 degrees from 0-255. If the RGB tricolor lines completely coincide with this gray line, it means your monitor is perfect, even if Can be used directly without calibration. But of course, the actual situation will not be so idealized, so we see the best performance in the above picture is the RMBP's reality screen, its RG color almost coincides with the gray line, only B color high light output is reduced About 18%. However, the R-color of millet TV only coincides with the gray line, in which the G color is shifted down by 5% in the high-light part, and the B-color high-light output is adjusted by 30%. And LeTV S50 RGB calibration curve, really, I do not want to say. In this section, the slope will directly affect the calibrated color gamut coverage. In addition to the slope, we can see that even after adjustment, the RMBP's three-color curve shape is close to a straight line, while millet has a certain wave curvature. As for LeTV, it completely becomes a big wave. This is to tell us the corresponding output amplitude that RGB needs to adjust when it outputs with different brightness. The closer to the straight line, the closer the color shift to a linear change, the higher the accuracy after calibration. The more twisted lines indicate that the color shift of the monitor is completely irregular. However, the color correction is to take a certain number of color samples to adjust and generate a calibration curve. For the unsampled range, it is simulated according to the calibration values ​​of several sampling points. Therefore, the most direct performance is that the final measurement will be found The color accuracy will be very poor, that is, the maximum value of ∆E will be very high. Of course, the truth is, as long as I calibrate the number of sampling enough, the accuracy of the calibrated color will be more accurate, but the 8bit panel can display 16.7 million kinds of colors, one by one, don't say monitor color calibration instrument Can be reimbursed directly, even if it can be completed, it is estimated that they can wait until the three generations from the primary school. At the same time, we can all find that the three displays have a common problem, that is, B-color output is too high, late a lot of drop, such a direct display of the display is the color blue, that is, the color temperature is too high. I believe that the millet screen is blue, and everyone can observe it through screen shots. Here, by the way, we will entertain you with a gamma correction just for the millet TV. If you remove the calibration settings, I think this TV is simply an invincible move. This is also the practice of many unscrupulous media. By changing the settings, Make the settings closer to panel native data to fake test results:
Just look at this screenshot, do you think that the performance of Xiaomi TV is going to be counter-attack. This is really my measured data, but you must know that the color temperature value set by the school color meter according to the measured value is as high as 15700K.
Color accuracy testThe above part of the color gamut is generated based on the profile generated after calibration, and the following part is the conclusion obtained from the measurement report after calibration.
With regard to color accuracy, the evaluation report of the display monitor will now be affixed with a ∆E form for everyone to see. Some people may even think that the 48 color plates running the original datacolor software are too time-consuming. 12 came to the conclusion. The landlord has always felt that using 48 colors to judge a monitor capable of displaying 16.7 million colors is not enough. So, if you want to play, you can play 1005 colors. The result is as follows:
The first to play is still RMBP:
In the previous profile, the mean value of the 色E color shift of the RMBP was only 0.31 and the maximum was 1.54. In the calibrated measurement report, the average value of ∆E has reached 0.46, and the maximum value has reached 4.66. The report directly marks NOT OK in the conclusion column (below the red character is beyond the general tolerance value). Why? With this deviation, the maximum value has nearly tripled. This is the question of the number of samples that the landlord said. When the number of samples is large enough, the problem will be more obvious and some extreme situations will be manifested. The following is a screenshot of the color sample with ∆E exceeding 2:
Although there is a NOT OK in the above summary, actually there are only 3 kinds of color samples with a ∆E value greater than 2 in the 1005 color sample library. The landlord expressed great satisfaction. Let's take a look at Xiaomi TV:
In the previous profile, the mean value of ∆E was 0.41, and the maximum value was 1.99, which was also updated to an average value of 0.44 and a maximum value of 3.91. If you only look at these two values, will you feel that Xiaomi TV's performance can spike RMBP again? Oh, we continue to screen out color samples with a ∆E value greater than 2:
There are a total of 9 color samples in Xiaomi TV's ∆E value greater than 2, but fortunately, it's more than 3, so the average can't believe it. Finally, we look at the music as S50's results:
The profile results before the Letv S50 were ∆E mean 0.82, maximum 6.75, and the test report corrected to an average of 1.23 and a maximum of 7.14. This result I find it difficult to cause any misunderstanding. It is too much for the 色E to be more than 2 specific color samples. This workload is really... I will cut off the 0001-0044 chart and let everyone feel it. it is good. If you are interested later, you can download the full test report at the end of the article.
In terms of color accuracy, there is no doubt that the RMBP wins.
Relative color temperature testRMBP:
In this figure, green indicates normal, yellow indicates a large degree of deviation, and red indicates a large degree of deviation. As you can see, the RMBP is red except for 2% and 5%, and the others are green. Although its performance is not perfect, it is still very good. (The white point temperature of RMBP measured after calibration is 6520K)
Millet TV:
Millet TV was a bit far off at 2%, 5%, and 7%, and the rest was pretty good. The relative RMBP is slightly worse. (The white point temperature of the millet TV after calibration is 6486K)
LeTV S50:
Landlord to apply, is it right? Can music as the S50 directly out of the list of contrasts ... ..., this relative color temperature map really do not know what to say. (The color temperature of the LeTV S50 after calibration is 6605K)
RMBP:
RMBP's GAMMA is somewhat biased at the end, but overall it is OK.
Millet TV:
Millet's GAMMA and RMBP seem to be better than others.
LeTV S50:
After LeTV considered the above comparison, I think it belongs to a more “wave†and basically it is not accurate!
RGB grayscale testRMBP:
This test is to measure the gray color, when the gray color trend, if a smooth and smooth center is on the 0% line, and only see the blue circle is the best, of course, this is the theoretical value Actually there are deviations. In addition to seeing the blue circle, see red and green, indicating that in this grayscale, there is a certain color cast, this we can download the test report to view, in the test report, you can mouse Moving to each point to observe the RGB grayscale color cast at this point, we only look at one, roughly, RMBP's performance is still good.
Millet TV:
The RGB grayscale performance of the millet TV is also considered good, but overall it is worse than the RMBP.
LeTV S50:
Let's continue to wave, this is really nothing to say, we must pay attention to ah, this is the result after calibration.
Brightness, contrast test First we look at the brightness:
More strangely, the brightness of these three machines is RMBP. The measured brightness of millet TV 3S before uncalibration is about 250cd/m2, which is reduced by nearly 30cd/m2 after calibration. Of course, the last one mentioned in the survey. There are more than 320 measurements in the media. One of the reasons for this may be the instrument error. There is also a different measurement area. Here is an example of the screen uniformity and degree when a millet TV 3 is not calibrated.
As you can see, Xiaomi TV's 3s 65-inch screen uniformity performance is very poor, with the maximum deviation reaching 21%. If you press 250cd/m2 measured by me, then the brightest area of ​​the screen is certainly more than 300cd/m2.
Black point brightness:
This value represents the brightness of the screen when the screen is pure black. The smaller the number, the better. We can see that Xiaomi TV has the best performance from using the SVA screen, but RMBP only A little bit worse.
Contrast:
Still benefiting from the screen characteristics of the SVA screen, contrast testing is still the best for millet TV.
Millet TV's overall performance under the factory default is a high color temperature, and the screen is blue. If you directly use it as a demonstration machine, obviously it is not competent, but after the instrument calibration, we can still be very pleasantly surprised to find its overall performance. Is not inferior to Apple's retina macbook pro, that is, the use of Samsung SVA screen has a very high ability to restore, in the case of uncalibrated high color temperature this pot, I think the main still by Rui Yi / Xiao Mi jointly developed the bottom Side-entry backlight to back. But at the factory, millet did not correct for this situation (of course, the original factory calibration and its own post-correction principle is basically the same, white LED backlighting is unable to adjust the color temperature later, can only adjust the panel's three-color output, Therefore, high-end TVs and displays will use RGB backlight modules more, control the color temperature from the source, and even calibrate the panel's deviation). But for the music in the comparison S50, can only say sorry, completely unexpected did not expect to be so unbearable before the test (access to relevant information found that LeTV S50 screen manufacturer used by TPV, is a TPV AOC and LG's The joint venture company (incidentally, the 1299 AOC 4K monitor that the landlord bought before is much inferior to the millet TV's calibration result), but fortunately this TV has long been discontinued, and even if it is said that it is black, I think it should be nothing. Black is worth it.
Therefore, if you have a color correction device on your hand and you have recognized the colors of the retina macbook pro's original display screen, the calibrated color of the 3s 65-inch version of the millet TV is fully capable of performing normal presentations. Here is a small photograph with limited budget. / Design studio makes a strong recommendation. Of course, as for the quality control and after-sales service of Xiaomi, there seems to be some controversy. The landlord can only say that the millet TVs of the landlord and several friends' families have not had any quality problems from 2s to this 3s (but they It took me 1 or 2 years to use it.
For melon-eating people, everything mentioned in this article is meaningless because you cannot load any color profiles under the Xiaomi television system, let alone have a picture quality engine inside. Help you "optimize". However, if I would like to comment more, I would say that Xiaomi used a first-class LCD panel (except for the backlight module), but the TV made is far from the level of this panel. regret. In the previous comments of Zhong Wenwen, it was also mentioned that in the television industry, Internet brands and traditional audio and video veteran are totally incomprehensible in the depth of understanding of the picture and technology accumulation, because now TV is a very technical content. Low, panel/backlight/driver/smart chip, all have ready-made solutions. Manufacturers only need to make a shell and put all of them into it to light them directly. Even individuals can buy panels and driver boards on Sham Shui Po. The plug is a TV. Nowadays, Internet brands have focused more on the system UI and content, and have neglected to be the basic function and role of a TV. Just like cooking, I used top-quality ingredients, but the focus was indeed on utensils and plates for cooking. For cooking and seasoning, I just cooked it. So here is still a suggestion to Xiaomi, although the slogan “to be the best TV in the world†is very convenient to cry, but in real practice, it is still hoped that we can make more use of the TV itself, otherwise According to the current direction, the possible result can only be “to be the best TV system in the world†or to “assemble the TV with the best accessories in the worldâ€.
Having said so much, since this TV was almost able to be used in the same way as RMBP, this time it would be rogue if it was not on a point-by-point photogram comparison, so the following screen shot time would be reached. For the sake of comparison, it would be more direct and direct. Put the millet TV and RMBP in the same picture, both use the calibrated configuration file, but the above also said so many differences after calibration, so the actual picture is certainly still different, let's see what How much worse:
The order of the pictures is the original picture - Xiaomi TV - RMBP. The color temperature is set at 5500K and the color tone is +14 (this is based on the gray portion of the white vial in the picture) when the captured RAW file is processed. The configuration file used is the camera standard. The screen shot is very subjective and it is easy to be black. Therefore, everyone can look at it. I will not explain it more.
Test report download link
One of the questions that people are still very concerned about in the comments on the last test article is that the HD TV input of the 3S 65-inch version of Xiaomi TV does not support HDR 10, and it is specifically shown that when using PS4 to connect Xiaomi TV 3S, it cannot open HDR. Options. Friends of the value of contact customer service to ask this question, customer service answer is not supported, but I asked the millet engineer, the engineer said that support, whether it feels that the person is not to do the hardware, so it is not very sure. Or the practice of knowing it, asked friends to borrow a PS4 (system has been upgraded to 4.0 +), worried about the original cable problems, specifically Jingdong orders bought a new HDMI 2.0 line. However, after repeated testing, it is impossible to open the HD4 option of the Shenhai 4 . Therefore, it is basically possible to confirm that the millet TV 3S 65 inch is incompatible with the HDR of the PS4. However, this site's chief living @abrams2415 comment reply I said that when PS4 PRO is connected to the millet TV 3S can open the HDR option in the game, so this pot seems to have PS4 to back ... ....
Headband Headphones have good sound field and are comfortable to wear, Bluetooth Headband Headphones can avoid scratching the ear canal; the sound quality of high, medium,and bass are very good, have a good sense of hearing even when playing symphony; The midrange effect of the headphones is thick, pop songs' hearing effect is very contagious, music effect is great.
Advantages:
1: Sound field is good, comfortable to wear.
2: No in-ear, avoid scratching the ear canal.
3: longer time to use, compare to in-ear earphone.
Headband Headphones,Wireless Headband Headphones,Headband Earphones,Bluetooth Headband Headphones
Shenzhen Linx Technology Co., Ltd. , https://www.linxheadphone.com