These are the samples that were taken more cheaply in the previous Samsung A5 single evaluation. The comparison model is the strongest domestic night shot Xshot, and the domestic Xiaomi 2s with the highest inventory; The other sample is the comparison between the highest pixel MX4 made in China and Xiaomi 2s.

If you have it on hand or are ready to buy one of the vivo Xshot/Samsung Galaxy A5/MX4/, I hope this article can help you get a general understanding of the performance of night photography of the machine.

Parameters and description

Vivo Xshot uses Sony IMX214 13 million pixel sensor, with CMOS size of 1/3.06 inch, pixel size of 1.12 µ m, F1.8 aperture with optical anti shake, and 6P lens. Snapdragon 801 integrated ISP (image processor) is used, and the algorithm is provided by Hongruan ArcSoft;
Samsung A5: 13 megapixel camera, parameter unknown. The test machine is BOC Telecom version, and the system is Android 4.4;
Meizu MX4 uses Sony IMX220 20.7 million pixel sensor, CMOS size is 1/2.3 inch, pixel size is 1.2 μ m F2.2 aperture, 5-chip lens, 4-channel ISP, algorithm unknown;
Xiaomi 2s: Sony IMX135 sensor, 13 million pixels, aperture F2.2. The test machine is 32G telecom version, the system is MIUI V5, and the third-party camera app (promote) is used.

Xshot/Xiaomi 2s/Samsung Galaxy A5

The following default is Xshot on the left, Xiaomi 2s in the middle, and Samsung A5 on the right

Central magnification 100%

The three machines all adopt automatic mode. Except that the Xshot has the optical anti shake ability to set the shutter to 1/10s, the other two machines are 1/16s and 1/17s respectively, with the same film yield. The white balance of the Xshot is green, and the Xiaomi 2s and Samsung A5 are red.

Under the outdoor weak light at the roadside, the ISO has been used up for Mi2s and Samsung A5, which are 1665 and 800 respectively. But the picture of Xiaomi 2s is still dark. The highest picture brightness is Xshot, followed by A5 (closer to the naked eye). Except that the glare control of M2 is not good, the other two are good.

In terms of discrimination, Xiaomi 2s is quite old and noisy and the picture is dark, and some dark details will be lost; A5 has an algorithm for sharpening, and the edge outline of the object is clear, even surpassing Xshot in details.

whole

Central amplification

In the weaker light (sidewalk with shade), the ISO of Xshot is also forced to 2700, but the image noise control is still the best. The glass behind the ice-cream cone has the most details, but the text on the ice-cream cone has traces of algorithm smearing (the font is smaller);

The second is Xiaomi 2s with full screen noise; The Samsung A5, which has undergone double processing of sharpening and smearing, has the worst detail performance, but the best appearance, and the brightness is closest to the naked eye.

Xshot is on the left and Samsung A5 is on the right.

The noise control of Xshot is better than that of the previous file, but the algorithm smearing makes the edge of the object blurred, resulting in some loss of detail; The A5 has a powerful picture sharpening, and there are victories and defeats in discrimination.

Original drawing

Central amplification

Central amplification of another group of samples

Left Xshot normal mode, central A5 normal mode, right A5 night view mode

Xiaomi on the left for 2s, A5 normal mode in the center, A5 night view mode on the right

Because this is the sample group of Samsung A5 at that time, the A5 night view mode was specially tested. On the left is the 13 million pixel sample of Xshot and Mi2s, and on the right is the 8 million pixel sample of A5 night view mode (it will also become 8 million in HDR mode).

After the night view mode is turned on, the performance of Samsung A5 has been greatly improved. The brightness of the image is close to Xshot, and the noise control and the Xshot in the normal mode have a win-win situation (the noise control in the bright part is better, but the noise control in the dark part is weaker); However, the reduction of resolution makes the analytical power slightly backward, and details will be lost in places with complex details such as grass and gravel.

Original drawing

Central amplification

The upper side is the general mode and night view mode of Xshot, and the lower side is the general mode and night view mode of A5.

In extremely weak light (even the ISO of Xshot has reached 3900 when shooting city night scenes), the A5 night scene mode has greatly improved the brightness of the picture, and the picture details have slightly exceeded the ordinary Xshot mode.

However, the nightscape mode of Xshot is also surprisingly strong. In addition to fixing the noise, it also retains a large number of picture details, and hangs A5 (but the picture will have a sense of oil painting).

Summary

In terms of night scene performance, the outcome was as expected: Xshot>Samsung A5>Xiaomi 2s. But more unexpectedly, there is no gap in discrimination.

Millet 2s, the oldest and cheapest, is undoubtedly at the bottom of the list. The picture is noisy and has insufficient photosensitivity. The picture is very dark, and the white balance often tends to red and yellow. The only thing worth praising is that there are many details in the picture, which is basically equal to the Xshot and A5 (maybe because the noise has not been smeared by the algorithm, but left a lot of details?);
Samsung A5 performs unexpectedly well. The brightness of the image is the closest to what the naked eye can see. The noise control is commendable, and the impression of the proof is very good. When the night view mode is turned on, the brightness, noise control and detail performance under ultra weak light can be greatly improved, and even the general mode of Xshot can be slightly exceeded. However, in places with relatively strong light (near street lights or neon lights), it is still impossible to get rid of the curse that night scene mode will lose details;
Xshot has IMX214 plus optical anti shake. Even in normal mode, image brightness and noise control can be used to suspend Xiaomi for 2s and A5. However, the performance of discrimination was unexpected. In addition to the ultra weak light (city night view), the detail discrimination failed to open the gap with Samsung A5 and Xiaomi 2s.

Meizu MX4/Xiaomi 2s

There are only two environments tested here, one is ordinary weak light (in front of the store at night and in the KTV lobby light room where the light is weak), and the other is ultra weak light (in the nighttime roadside, KTV room and other environments).

The MX4 uses the automatic mode, while the congenitally deficient Xiaomi 2s uses a third-party procure camera (version 1.6.1.1) to provide normal sharpness and white balance.

Ultraweak light

The size difference between the MX4's 20 million pixels and the Xiaomi 2s' 13 million pixels is as above, and their wide-angle sizes are similar. In the field environment, both machines have exhausted their sensitivity and are under the highest ISO and slowest safety shutter. However, the MX4 (IMX220) on the left side has stronger light sensitivity, and the picture brightness is higher and closer to the naked eye.

The available ISO of the MX4 is lower, the shutter is slower (1/10s), and it is easy to shake and destroy the tablet; The available ISO of Xiaomi 2s is higher, but the safety shutter is fast (1/16s), and the film rate is high but the noise is serious.

The white balance on both sides is floating, and the MX4 is relatively close to reality, while the Xiaomi 2s is still the classic yellow of Xiaomi. In addition, the lens glare of the MX4 is really serious

After zooming in the center, Xiaomi 2s hardly processed the noise. The brightness was low and there were many red and green noise points, which made the phone numbers in the left box difficult to identify, which greatly affected the appearance, but the background details (the right box) were very clear and sharp; The MX4 has a lower ISO and has been smeared with noise. The image is much cleaner, but the edges of objects are blurred and details are lost seriously.

Original picture comparison

Central amplification

Due to the lack of photosensitivity, Xiaomi 2S's picture is dark, and a lot of dark details are lost. The railing and van in the box are almost invisible.

Original drawing

Central amplification

Xiaomi 2s lost the details of the refrigerator in the lower left corner, but the MX4 light metering is slightly bright, and the bright details of the goods on the upper side of the picture are lost.

Sample after central amplification

Original drawing

Central amplification

The MX4 has higher picture brightness and better photo appearance, but the smudging algorithm makes it lose a lot in a more discriminating future.

Weak light

When the ambient light is stronger, there are more lights in front of the store or at the intersection. The Xiaomi 2s, ISO (1665) and shutter speed (1/16s) on the right have not changed, so the brightness of the picture has been greatly improved, which exceeds that of the MX4 and the naked eye.

For the MX4 on the left, the ISO has dropped below 600, and the imaging levels on both sides have reversed. The performance of the MX4 has "recovered" to the normal level, and the discrimination and tolerance of Xiaomi have been suspended for 2s.

Summary

The MX4 performs much better than Xiaomi for 2s under normal low light (shop front, indoor light, etc.); However, in the environment where the light is weaker (beside the road), although the image brightness and noise control of the MX4 are good, the discrimination performance is seriously degraded (even worse than Xiaomi 2s).

summary

The camera of a mobile phone is like a headset. Each mobile phone has its own characteristics in imaging. Corresponding to different earphones, there will be different high, middle and low frequency performances. Under different brightness environments, each machine has its own imaging weaknesses and strengths: under brightness A, this mobile phone may have better imaging, and under brightness B, that mobile phone may have better imaging.

In the two comparisons in this article, even Xshot failed to show a comprehensive advantage in the night scene (the difference in discrimination is not big), and even the Xiaomi 2s three years ago, its discrimination performance in many scenes is not inferior to that of modern models (I really don't know whether this is a good thing or a bad thing).

Combined with the comparison of the previous three photos, here is a brief summary of the night/weak light imaging characteristics of these mobile phones (ranking from high to low):

vivo Xshot： The night scene and even the overall photography level are very balanced. The noise reduction algorithm is powerful and the picture is pure. It is undoubtedly the strongest in China under extremely weak light. The brightness of the picture is beyond the naked eye, and the night scene will appear very pleasing. However, the white balance is not accurate, and the frequent ISO racing results in relatively poor detail performance under normal indoor lighting (but it can still beat most models);
Meizu MX4: The highest domestic pixel does not show the advantage of discrimination in daytime, but the performance of normal indoor lighting can surpass that of Xshot. The white balance, metering and noise control are good, and the brightness and hue of the sample are very close to the naked eye. But the performance of outdoor weak light is weak, even losing 2s to Xiaomi;
Samsung Galaxy A5: It shows that it is far more capable of taking photos than the mid-range computer should be. After turning on the night view mode, it can even match the general mode of Xshot. Even the outdoor weak light can maintain a lot of picture details. The noise control is good, and the picture brightness is sufficient for most of the time;
Moto X： Although it is a new generation of noise king, the picture brightness can be close to Xshot, the picture details are retained, the white balance is accurate, and the photography experience is very pleasant;
Samsung Galaxy S5: The performance of night scene is inferior to that of its own mid-range computer. Although it is supported by the rage algorithm, the performance of indoor lighting is good, but once it goes to the outdoor lighting where the light is weaker, the performance of details becomes weak;
Xiaomi 4: With IMX214 and large aperture, it should have performed better, but poor noise reduction algorithm led to serious loss of machine details, and serious drift of the innate white balance, resulting in more points being reduced;
Xiaomi 2s: There are many noise points, but because there is almost no noise reduction algorithm, the image discrimination is good. However, the photosensitive ability is seriously insufficient in outdoor night scenes, which leads to dark pictures and unattractive images;
Glory 4X: In order to ensure the brightness of the picture, the low light ISO rises quickly, and the indoor light performance is good, but as long as the light is slightly weak, the noise will increase rapidly;
Hongmi Note.

《 Fighting! Xiaomi 2s/Xiaomi 4/Hongmi Note/Glory 4X photo contrast 》

《 The strongest domestic photo contest Meizu MX4/vivo Xshot photo comparison 》

《 Night shot battle Moto X/Xiaomi 4/Samsung Galaxy S5 night shot comparison 》