AnonD-436524, 03 Sep 2015Raw mode from that sensor? I mean: would it be possible for a custom camera app to actually sa... moreThe benefits would be that you could decide on a aspect ratio afterwards.
The downside would be, that at the time when you framed your picture, so that, for example that church fit in both in height and width, you would have used the full sensor.
And later when you try to crop it to a aspect ratio that makes sense, it might not fit.
AnonD-375713, 03 Sep 2015Since we are moving towards 4K (and yes I bet it will take a while, at least 5 years more if n... moreWell it's not that easy.
To have any benefits from higher resolution, for video.
First, let's assume that the sensor and optics was actually able to deliver the specified resolution (it's not, but more on that later).
With the compression used in mobiles, with 4K video. Due to the very low bitrate, you would not be able to capture anywhere near 4K resolution. If you upped the 1080p bitrates in mobiles to match, there would not be a big difference, if any at all, in terms of captured details.
So you are not really getting 4K.
That introduces a problem.
If you have a good 1080p video, with good amount of details, most 4K TVs today have good upscaling (might not be true forever, with prices falling, things like upscaling usually takes a hit), so you would end up with a picture that looks good on your TV. Well even if it pretty much just doubles the pixels in height and with, it would look pretty good.
If you however have a 4K video, no upscaling will take place. And if that video doesnt offer much details, well you will see the result on your TV.
If we are using the same sensor, and if it really had could offer the specified resolution, you would have the benefit with the 1080p video that you could zoom in more, and thus focus on details.
Due to the compression (due to low bitrate), 4K in mobiles, actually aren't a good thing, even if you have a 4K TV.
But then we come to the other issue.
With high resolutions, you need really good optics.
At these resolutions, even with DSLRs, you will need a more expensive lens to take advantage of the resolution.
I really can't belive, that Sony (and others), have manage already to minimize optics to be sharp enough for these resolutions.
So if we assume the sensor is actually capable of capturing details that matches the specified resolution, it wont, since the optics aren't sharp enough.
If we are talking video. Well that means that if you zoom in, to pixel level of the sensor, it would not be that much of a benefit, as it doesnt capture that amount of details. That would be the same if the video was 1080p or if it's 4K (but the 1080p would be more zoomed in, making it more visible).
You would increase the blur, compared to using the full sensor, and then downscaling to the resolution of the video resolution you are using.
If we are talking still pictures, you would see this if you zoom in, as well.
Then we have the other issue.
Small pixels, generate more noise than bigger pixels.
Noise also has an impact on captured details, is noise and not captured detail.
There are advances in this field, so pixels can be made smaller and smaller and still perform as well as larger pixels from years ago. So we can get higher resolution, but without having to deal with more noise.
However, using the latest tech, we could keep the resolution from years ago, instead make the pixels bigger, and have even less noise. However, in DSLRs they reached a point, where the noise is so low that in good light, using a lower resolution sensor, means that you will capture less details. You can actually capture a good amoung of details, even in less good light.
So most of the strive goes for higher resolutions these days.
However, those camera still do worse than your own vision, when it comes to seeing things in the dark, so I think they should work harder on lowlight capabilities than they currently are... but resolution is easier to market.
The Sony A7s took a step back in terms of resolution, to offer better lowlight performance, but that one is partly developed for 4K video as well. And a canon has one that is not yet out, that can capture video in really dark enviroments, but only 1080p, but that one actually surpasses our vision when it comes to seeing in the dark, however in good light it falls very very far behind).
However, when it comes to mobiles, even in really good light, they capture less details, then DSLRs with the same resolution, due to noise.
And it gets even worse, when the light no longer is really good. You can't even compare to DSLRs, they are worlds apart.
HTC tried with their 4MP module, that had bigger pixels, to get better lowlight performance.
But I think neither their optics, nor that sensor was the best.
So they ended up with low resolution in daylight, and not outperforming the rest even when the light wasnt that good.
There is no reason to have sensors that have a resolution that surpasses that of the optics they sit behind. And there is no reason to have sensor that produce too much noise.
They might cancell eachother out in really good light, so that the cap for amount of captured detail is the same with both the optics and the noise level of the sensor.
But in low light you would benefit from a sensor with less noise (bigger pixels).
And to reach te best compromise even if that would mean a slight loss in resolution in the best of light, it's worth it.
Bigger pixels usually are better when it comes to dynamic range, as well. Meaning that it will capture more details in the lightest and the darkest part of the pictures. HDR functionality sometimes is even better at that, but that requires that you hold the camera still. And with a sensor with better dynamics, the result from the HDR would be even better.
I think the best compromise right now, for mobiles is somewhere in the 8-12 MP range.
That would still mean that it would be more focused on resolution than on lowlight performance. But you dont want to go to low in resolution, especially not with the high resolution monitors/displays we have today.
But also with the lack of optical zoom, you do want to capture the most important details.
But otherwise I'm a strong believer in the concept of using a high megapixel sensor, behind wide angle lens (or rather ultra wide angle lens).
I like the idea of caputring what your eyes can see (that would include some slight eyemovement, as that is how your vision works, so you would need higher resolution than that of your eyes combined and slightly more wide angle as well).
I do have a K zoom.
I like that the look sets it apart, and I have been asked several of times what it is. And if it takes good pictures.
But also, the zoom makes up for the lack of details. I can still zoom in to capture details. But then I need to take a lot of pictures.
If it had a better sensor, it could have been made with less of a zoom range, and still captured as much details. and then it could have either been thinner, to have had the potential of selling more. Or the optics could have had a bigger aperture, to get better result in low light.
It uses a sony sensor, the Same one as my Z3. Both are way off from capturing 20.7 MP. And even if this new sensor, with smaller pixels, still has the same performance. Imagine how much better it could have been with bigger pixels.
Raw mode from that sensor? I mean: would it be possible for a custom camera app to actually save the entire sensor area pixels (with corners "cut off") and enable later pixel manipulation in app/web/computer? Technically thrilled, though real world benefits may be marginal... :-)
AnonD-398362, 03 Sep 2015I know right. They always do that. Remember when they announced the iPhone 6 and and the apple... moreI realise that, but I was refuting someone's comment that Nokia developed the idea.
AnonD-292897, 03 Sep 2015I bet Apple would market the s**t out of it (of course, as their own huge genuine innovation) ... moreI know right. They always do that. Remember when they announced the iPhone 6 and and the apple watch will house nfc? They acted like they just gave us the keys to economic heaven and we should thank them for making mobile payments possible and safe.
Andrew, 03 Sep 2015Actually this technique has been used in digital cameras for quite some time. It's not super-c... moreIt's not about "innovation", it's about the fact that sony did not marketed this nice feature in his own favor.
AnonD-152638, 03 Sep 2015I still don't get why they had to push the resolution upwards.
Lower resolution, but the sa... moreSince we are moving towards 4K (and yes I bet it will take a while, at least 5 years more if not 10), keep for example an 8 megapixel sensor. You'll be pixel per pixel on 4K, no headroom for any zooming without upscaling. The 5984 pixels width already don't allow much zooming without upscaling in 4K, about 1.5x only and that's it.
If you are taking pictures, of course most of them will be trash but the ones worth keeping, you'll want to keep the best out of it, you'll not want them to look terrible in the future, so it's logical to keep a high enough resolution.
When do people will understand that MP count is not every thing in the quality of pictures..look at the note 5 and lg g4 the best phone cameras to date..i want to se how it goes against the king LG G4 on night time!
It is actually a new Sony 24.8 Megapixel, 1/2.3", 1.12 um Exmor RS multi-aspect sensor, with 192 PDAF points and 2-dimensional contrast detection AF. It is almost similar to Omnivision's IS (http://www.ovt.com/products/sensor.php?id=170), but this new addition to Exmor RS for mobile family is different due to the use of multi-aspect sensor. Though it doesn't have an Optical Steadyshot (OIS), as per Engadget, it uses a closed-loop actuator (which probably made by ON Semiconductor: http://www.onsemi.com/PowerSolutions/product.do?id=LC898212XA), which guides the sensor and the lens for a faster AF and better image stabilization.
I still don't get why they had to push the resolution upwards.
Lower resolution, but the same size sensor, means that they could have potentially produced a sensor with better dynamic range, and better low light performance.
I have been wondering for long, why cameras aren't 16:9. Not only mobile cameras, but many other cameras as well.
I suspect, that there are no real size or cost advantages, to making optics for 16:9 sensors. And that more squarish sensors makes better use of the image circle made by the optics.
Who's said the sensor was a 4:3 a ratio sensor? GSMArena cannot be out smart Sony the actual image sensor maker right? So please stop this speculation
I bet Apple would market the s**t out of it (of course, as their own huge genuine innovation) and take all the time they need to explain it.
Omar Mohamed, 02 Sep 2015Actually this tech is invented by Nokia since N9Actually this technique has been used in digital cameras for quite some time. It's not super-common but it's always well-respected. Google "multi-aspect sensor".