On September 15, 2020, Samsung released four new CMOS with unit pixel of 0.7 μ m: 108 million pixel HM264 million pixel GW348 million pixel GM5 (for periscope telephoto, 1/2.55 '') and 32 million pixel JD1 (for front, 1/3.14 ''). HM2 and GW3 are destined to be old friends of 1000-2000 yuan mid-range computers in the future.
The 64 megapixel GW3 was first released by the True Self V15 released on January 7, and then used by the vivo S9 and iQOO Z3. GW3 unit pixel is 0.7 μ m, 1/1.97 inch, supports Super PD focusing, and can shoot up to 4K 60fps video.
Today's comparison is to see how GW3 actually performs. To facilitate the control of variables, the selected competitors are iQOO Neo5 (IMX5984800W, 0.8 μ m, 1/2 '', F1.79+OIS) with close sensor size, and iPhone 12 Pro Max (1200W, 1.7 μ m, F1.6, sensor displacement OIS) as the baseline standard.
"Long focus" contrast
The iQOO Z3 and iQOO Neo5 do not have a telephoto lens, but use tactical ultra wide angle lens, so this part is simply omitted.
Long focus contrast, Scene 1
Long focus contrast, scene 1, central magnification
Long focus contrast, Scene 2
Long focus contrast, scene 2, central magnification
Long focus contrast, scene 3
Long focus contrast, scene 3, central magnification
Summary of long focus:
The iPhone 12 Pro Max has a 2.5x zoom, and the iQOO Z3 and iQOO Neo5 have a 2x zoom (in fact, they are digital zoom). In fact, it is a 2.5 times light variable digital trimming of high pixel CMOS. Measured, can't win. However, the iQOO Neo5 digital zoom is more discriminative than the ordinary 2x optical zoom, and there are not many iPhones with 2.5x focal power.
Although they are all 1/2 ", IMX598 is obviously stronger than GW3 in discrimination and tolerance. The advantage of GW3 in the number of pixels is not reflected in the digital zoom.
Ultra wide angle contrast
IQOO Z3 is an 800W ultra wide angle OV8856 of 120 degrees (108 degrees after correction), 1.12 μ m, 1/4 '', F2.2；
IQOO Neo5 is a 120 degree S5K3L61300W, 1.12 μ m, 1/3.1 '', F2.2， With 2.5cm macro function.
Ultra wide angle comparison, daytime scene 1 (comparison between the actual size of the sample and the range of view)
Ultra wide angle contrast, daytime scene 1, central magnification
Ultra wide angle contrast, daytime scene 2, central magnification
Ultra wide angle contrast, daytime scene 3
Ultra wide angle contrast, daytime scene 3, central magnification
Ultra wide angle contrast, light scene 1
Ultra wide angle contrast, light scene 1, central magnification
Ultra wide angle contrast, weak light scene 1
Ultra wide angle contrast, weak light scene 1, central amplification
Ultra wide angle contrast, weak light scene 2
Ultra wide angle contrast, weak light scene 2, central amplification
Ultra wide angle summary:
The iQOO Z3 is a super wide angle with an equivalent focal length of 16mm. The field of vision is relatively small, and 8 million pixels are also very difficult to distinguish. It is tactical.
The iQOO Neo5 has a 15mm focal length of 13 million pixels in ultra wide angle (1.12 μ m), which is significantly better than the iPhone's ultra wide angle in quality, except that some scenes are heavily smeared. It can also be seen that the algorithm difference/gap between iQOO Neo5 and Apple;
However, the super wide angle of the iPhone 12 Pro Max is only better at noise reduction and discrimination than other super wide angles. If Apple's multi frame synthesis algorithm had not remedied tolerance and highlight suppression, I could not have imagined how weak the ultra wide-angle imaging of the 10000 yuan mobile phone would be.
Subjective contrast
Compare the size of the sample with the range of view. The 64 megapixel GW3 of iQOO Z3 has 16 megapixel output in daily use, which is one circle larger than the common 12 megapixel proof.
Day Scene 1
Daytime scene 1, central zoom in
Day Scene 2
Daytime scene 2, central zoom in
Day Scene 3
Daytime scene 3, central zoom in
Indoor scene 1
Indoor scene 1, central magnification
Indoor scene 2
Indoor scene 2, central magnification
Weak light scene 1
Weak light scene 1, central amplification
Weak light scene 2 (iPhone lens shows serious glare ghost, which is harmful and tragic)
Weak light scene 2, central amplification
Weak light scene 3
Weak light scene 3, central amplification
Main summary
Although they are all blue factory algorithms, the iQOO Neo5 white balance is blue and cold, the iQOO Z3 white balance is yellow and magenta, and the night scene algorithm is completely different (the latter is stronger). The iQOO Z3 does not have optical anti shake, and the weak light scene will not have the common reminder of "keep the handheld stable". However, the main camera will produce a 3 second slow shutter, and the ultra wide angle will also have 2 seconds. Fortunately, the film yield is OK.
In terms of hardware specifications, the main camera (1200W, 1.7 μ m, F1.6+sensor displacement OIS) of the iPhone 12 Pro Max is stronger than the IMX598 (4800W, 4in1 is equivalent to 1200W of 1.6 μ m) of the iQOO Neo5, F1.79+OIS）， Not to mention the GW3 of iQOO Z3 (6400W, 4in1 is equivalent to 1600W of 1.4 μ m).
But the actual results are very unexpected. The tolerance and discrimination of the iPhone 12 Pro Max in the daytime are not much different from the iQOO Neo5. At night scenes, there is no gap between iQOO Z3 and iQOO Z3
How does GW3 perform?
To sum up simply, it is still a matter of money and goods. Even though GW3 has 64 million pixels, its ultimate discrimination and tolerance are not as good as the 48 million pixel IMX series. High pixels make it lonely, increasing capacity and computing pressure.
When other variables are controlled, the higher the pixels of CMOS with the same size, the smaller the actual pixel area (the more pixel spacing, the smaller the "practical area"). In addition to the quality gap of the sensor itself, GW3's ultimate analytical power in the daytime is not as good as that of close to size, but the IMX598 with lower pixels is also less tolerant.
The stronger night scene algorithm of iQOO Z3 conceals the weakness of GW3's weak light imaging, and on the contrary, it surpasses iQOO Neo5 (surprise!) in discrimination, which is just the opposite of Xiaomi 11 Ultra.
Because of the interference of the commonly used night view mode algorithm, the advantage of CMOS at the bottom is more obvious in the well lit daytime scene.
Follow our Weibo @ Love Computer
Follow our WeChat official account: playphone
Of course, we also pay attention to our Bilibili account: love computer
