Hello, I was sort of looking forward to Android 14 as I was under the impression that it would introduce a new camera API which would enable third-party camera apps such as G-CAM to fully take advantage of the 50 megapixel sensor. However, it doesn’t seem to be the case that this has happened. I’m not entirely sure why this is; per my testing it looks like I have access to level 3 hardware support for the back camera, but I am still not able to run the full 50 megapixels with any app other than the default phone camera, which looks pretty bad if I’m honest as a result of its heavy image processing. Does anyone else have any opposing experiences? Are there any plans in the works to fix this? Thank you.
I would also accept any method of turning off the image processing on the default camera, as my end goal is to take full-res photos with no processing and I don’t really care which app I use to do it.
I don’t remember where I got this impression. I’m sorry if I made it seem like I thought it was an official statement from Fairphone. I just heard it somewhere as one of the features of Android 14 and I assumed it would be implemented.
I think many people would appreciate this, however full res wasnt made available for the FP4 so far, so I’m a bit pessimistic for the FP5, so best is to contact Fairphone directly. Whatever GCam port you use, is wideangle accesible and working?
Is there actually a 50Mpix output from the camera module? I was under the impression that not and that the camera is logically a 12Mpix system.
Which would be (in my scientific view) the preferable option in most cases:
50Mpix is way beyond the optical limits of physics. Even with a system camera most optics will be limited to a smallest Airy disc that’s larger than a single pixel. With the smaller optics on a smartphone there is no point in having such high resolution since the optical resolution is lower anyhow.
The reason higher pixel count makes sense (aside from marketing) is in postprocessing, e.g., denoising - keyword pixel binning, which can be done in hardware*, so the output from the camera module would already be lower than 50Mpix (typically a quarter, so 12Mpix for a 48Mpix sensor).
I was under the impression that the FP5 module used pixelbinning and that the hardware resolution was not outputted from the module at all.
*Usually preferable, however, not updateable if completely new postproc algorithms come along.
Edit: my opinion is, this was introduced with the FP4 as many people did not understand the pixel binning part and complained a lot when the FP3 48Mpx module was released…
Good to know. Doesn’t really change the fact that pixel binning in hw/fw is likely to be the best solution unless you are an expert in digital image processing.
So most likely just introduced as marketing, since the people I know who can make the best out of raw output are definitely not using smartphones
Though I haven’t been able to test it properly, I think my personal preference is to have the large resolution photos and to simply take the camera noise, as I feel it appears more natural than many denoising artifacts.
