That explains it a bit better I think, thanks for the response. It still doesn't make complete sense to me though, it could be like you said and the technical details are just beyond the average ***** like me.
Is the sensor physically larger because of larger pixels? If so, why would that not let in more light per pixel?
edit: really? i d i o t is blanked out?
Oh boy. Okay, lets see if we can get more technical without sounding more technical. Warpdrived already helped, but I'll try to expand.
Okay a larger pixel, as a sensor in its own right, does "let in" more light, but of course how much is a function restricted by the f-stop and shutter speed of the lense as well as the area of the pixel. But it does not "absorb" that light. If that was really the reason for better low-light pictures, thenn as you can imagine sunlight pictures would just get blown out (over exposed) all to ****.
So what they do, sort of, is make better use of the light they do get. They're digital so of course their output voltage is divided into descreet steps, and then the processing software maps each step or value to a gray level. The number of steps, and the mapping done by the software, determine how much dynamic range you can get in the final image.
Trying to think of a good analogy. How about we consider analog to digital audio conversion. The better the frequency response of the analog mic, and the higher frequency sampling of a better converter, can result in an audio file containing a more accurate representation of the original sound. Even though the values in the file are stepped, there is a wider range of available values and a closer distrubution of them. Okay that analogy breaks down on many levels, but it kinda shows how a larger pixel can improve the dynamic range of the image. It can take the light it gets and divide it into a wider range yet smaller separation of steps.
Hmmm. Can we get even simpler? Imagine a tiny volume dial. It's difficult to finely tune the exact volume because your fingers can't select as many finely divided positions around the dial. A bigger dial allows you to precisely select from a larger set of possible volumes.
Okay thats kind of crappy too, but I hope it helps with a basic understanding. Still, we must qualify some things. A bigger pixel by size alone doesn't make it perform better. You've got to have the f-stop and shutter speed to give the pixel something good to work with, the pixel has to have been designed to take advantage, and the processing software has to do a good job. So bigger pixels on the sensor by themselves don't automatically mean better pictures. I suspect that's why the "don't let HTC fool you" thought was expressed.
Now we get into the screen you view the pictures on as well. The specs of Samsung's S4 suggests that the screen on it has more dynamic range available for display than the ONE has, by virtue of its deeper blacks alone. Regardless of what range of values is in the image file, when you view it all those values are again mapped to the display output. So the screen itself and whatever graphics engine is driving it determine the dynamic range of what you're actually looking at. The light that hit your lense, by the time you're looking at on screen, has undergone several levels of interpretation input and output along the way. What comes into your eyeball at that point may or may not be pretty close to what you might have seen if you had just put your eyeball where your camera lense was instead. The best cameras and displays in the world arent perfect, and probably never will be, although technological improvements have brought us a long way. And cell phone cameras, depending on how picky you want to get, aren't even close.
Most of us aren't that picky. So just decide whether the better zoomability/detail/size or the possibility of a very slightly improved low-light capability is more important to you, consider the other features, and choose your phone accordingly.
Sent from my SPH-D710 using Tapatalk 2