camera phone photo, test upload

[stuck]

Wondering what the quality of this shot is like.

Ken

[stuck]

Hmm, better than I anticipated!

[HansV]

Very good. Did you buy a smartphone?

[stuck]

I acquired a smartphone back in the summer, as a birthday present from myself and my sons. My eldest son sourced it but even though it was brand new, the USB-C port was faulty. Although a replacement was obtained within a couple of weeks, it was delivered to my son's house in London. There it sat until he next came to visit, at Christmas.

I've yet to fully the switch to the new handset, I've not yet ported my number across. I'm in no rush

Ken

[ChrisGreaves]

...I'm in no rush

Me neither.
I much prefer the leisured pace of email ....
Cheers, Chris

[BobH]

Ken, can you share what smartphone you used to make the photograph? It is, as usual, an excellent one!

[stuck]

Samsung S10e.

I'm still investigating the camera settings. This image is probably the only one worth sharing so far.

[HansV]

Your photo has the same kind of cold winter light as this one from this morning (phone picture too)

Winter.jpg

[stuck]

Yes, we woke up to a thin dusting of powdery snow this morning. We weren't expecting that. The only bits that melted were those that had full sun on them.

Nice photo!

Ken

[stuck]

Another test. This is a second shot of the same scene, taken about a minute before the one above.

It differs from that one in that this is the output from non-Samsung camera software. Software that saves the actual RAW data from the hardware camera sensor (rather than automatically processing it into JPEG format, which is what the Samsung camera software does) and then I've 'developed' that RAW file using Affinity Photo.

Any comments?

Ken

[ChrisGreaves]

Software that saves the actual RAW data from the hardware camera sensor (rather than automatically processing it into JPEG format, ...

Hi Ken, and @Graeme: Is this close to what Graeme's system does?

At its simplest Graeme's setup captures data and then he processes it (I believe) using a standalone application, to obtain a JPEG image which he publishes here.
Thanks, Chris

[HansV]

I have moved this thread to Scuttlebutt; it would have been deleted automatically in Test Area after a while.

The differences are subtle. The colours look a fraction more natural in the processed RAW image, while the JPG looks more immediately attractive (the processing done by camera software has a tendency to do that).

[stuck]

I'll leave Graeme to give a specific answer about his workflow but ahead of that...

All digital images start with light falling on a sensor, which is an array of light sensitive photo sites, aka pixels. Some of the pixels are red light sensitive, some are green sensitive, others blue. The more light that lands on each pixel, the greater the number recorded / captured by that pixel. Thus the output from the sensor, the so-called RAW data, is just an array of numbers representing how much R, G or B landed on each pixel. Those numbers have to be processed to turn them into something that can be displayed on screen as an image.

The vast majority of digital cameras do that conversion for you, they have their own built in RAW conversion algorithms and they save the converted data in JPEG format. You have no control over how the RAW to JPEG conversion is done. You can of course edit the final JPEG file, to try and adjust it to your taste and in most cases that will be all you need.

But... photo geeks like to think they can do better than the built-in automatic RAW to JPEG algorithm and so they like to start with the RAW data from the sensor and use software like Abode Camera Raw (which is part of both Photoshop and Lightroom), or the 'Develop' persona in Affinity Photo or DxO PhotoLab or any one of many other applications to set their own parameters for the RAW conversion. They will also almost always save their conversion in TIFF, since that is a lossless format.

Now read this tutorial:
https://www.cambridgeincolour.com/tutor ... format.htm

As for me...

The physical camera in my new smartphone, by default, is controlled by the Samsung camera app that comes with the Android OS on the handset. That app does have some settings that influence the automatic RAW to JPEG conversion but it only ever gives you a JPEG file, there is no option to get a copy of the RAW data the camera app used to create that JPEG file.

Being a photo geek it didn't take me long to search for, download and install and alternative camera app, one that as well as automatically converting the RAW data to JPEG also saves a copy of the RAW data, in a format called DNG. Affinity Photo can handle such DNG files and so that's what I used to generate the second test image in this topic.

I was able to adjust the basic crop, 'fix' what appeared to be unnatural curved distortions, control the contrast, control how saturated the colours are, etc. I think I ended up with a more natural looking version of the actual scene than that produced automatically by the default camera app. Of course, it took me time to all that and in the grand scheme of things the effort to benefit ratio is probably the wrong way round but I'm retired, it's freezing cold outside and I had fun, in the warm, inside doing it.

Ken

[stuck]

I have moved this thread to Scuttlebutt; it would have been deleted automatically in Test Area after a while.

The differences are subtle. The colours look a fraction more natural in the processed RAW image, while the JPG looks more immediately attractive (the processing done by camera software has a tendency to do that).

Hans, I didn't envisage the topic growing like this, which is why I posted it in Test Area.

Yes, I think it's the slightly boosted saturation that gives the camera JPEG the more immediate impact. I may also have over done the distortion correction, which has altered the crop.

Ken

[stuck]

Here are two more images. Both taken using the non-Samsung camera app. The first is the default JPEG produced by that app, the other is my processing of the RAW data. The differences are clear but perhaps not surprising, I was shooting into the light and the snow on the frozen pond was very bright, any automatic algorithm is going to struggle with non-standard lighting conditions like these.

With more tweaking I could probably improve my version further.

Ken

[HansV]

That's a very pronounced difference!

[Graeme]

I'll leave Graeme to give a specific answer about his workflow but ahead of that...

All digital images start with light falling on a sensor, which is an array of light sensitive photo sites, aka pixels. Some of the pixels are red light sensitive, some are green sensitive, others blue. The more light that lands on each pixel, the greater the number recorded / captured by that pixel. Thus the output from the sensor, the so-called RAW data, is just an array of numbers representing how much R, G or B landed on each pixel. Those numbers have to be processed to turn them into something that can be displayed on screen as an image.

Excellent quality for a <250kb jpg.

Not much to add except to say that all camera pixels irrespective of the camera are monochrome. All they do is convert photons into electrons using the photoelectric effect and the camera hardware counts each electron from each pixel for each exposure. Colour cameras have a grid of colour filters over the pixels called a Bayer Matrix. My astro camera has a filter wheel in front of it so I take a Red image, through the red filter, then Green then Blue. They're all grey scale images until they're processed and combined into an RGB image.

Graeme

[RonH]

My Google Pixel phone has RAW option though I rarely use it, sticking to .jpg images. My lady decides what images she wants to print for her albums. I adjust these (if necessary) using Adobe Camera RAW which also allows a lot of tweaks on JPEG images.

I bulk save the adjusted images in the Adobe format .dng, then resize for 13/15x10cm image printing with conversion back to .jpg. Then off to the online printer who prints in Germany and trucks images to our local camera shop.

Our image needs these days are fairly basic and this methodology works for us, though likely somewhat 'unapproved'

[ChrisGreaves]

The vast majority of digital cameras do that conversion for you, ...

Thank you Ken. Someone (you? Graeme?) told me all that about a year ago, so now I remember. I am a poor studemt.

I am a simple-user, so my digital cameras do the raw data number-crunching for me to deliver a printable, or at any rate, viewable image.

On a parallel track I note that the human brain, including mine, does a similar thing with photons, although it does not harbour the raw data in binary.

There remains for me the mystery of two related images in the human brain:
(1) The "image" conjured up in real-time by the sudden view of a dangerous tiger in my vicinity
(2) The "image" that re-surfaces in my brain later that night as I struggle to go to sleep in my tent.

I suspect that these images have different sources, but they are equally vivid, to me.
Cheers, Chris

[stuck]

...I am a simple-user, so my digital cameras do the raw data number-crunching for me to deliver a printable, or at any rate, viewable image...

No shame in that, if it works for you that's all that matters.

Ken