Color channels, 8, 12, 14, 16 bits

Quote:

---

Originally Posted by MacMahon

To all intents and purposes we think of the RAW data with which the computer works as consisting of number triples as long as we're working within an RGB model.

---

Pardon me, Tim, if exclude myself from "we". Bayer data is, by definition, four-channel and I can not think of four as three, sorry.

Quote:

---

Originally Posted by GrumpyDiver

Ted - no confusion what so ever. There may be twice as many green photosites as there are red or blue in a Bayer array, and this is certainly the way the RAW data is saved. Add the demosaicing step to turn the data into an image, we are dealing with three colour channels with each colour made up of a red, green and blue value.

---

Hello Manfred,

We must be talking at cross-purposes. I was only talking about raw file data when mentioning four channels. As we both know, a de-mosaiced (i.e. converted) file is no longer a raw file and, BTW, . . . thank you for clarifying what colors comprise an RGB color :rolleyes:

Quote:

---

I can't seem to recall any abiity to manipulate two separate green channels in post....

---

Well, DCraw can influence the four-channel values by inputting R,G,B,G so as to set white balance. RawDigger can output TIFFs of any one of the 4 channels for further manipulation by anything. I suppose it all depends what "in post" means....

Quote:

---

Originally Posted by xpatUSA

Pardon me, Tim, if exclude myself from "we". Bayer data is, by definition, four-channel and I can not think of four as three, sorry.



Hello Manfred,

We must be talking at cross-purposes. I was only talking about raw file data when mentioning four channels. As we both know, a de-mosaiced (i.e. converted) file is no longer a raw file and, BTW, . . . thank you for clarifying what colors comprise an RGB color :rolleyes:



Well, DCraw can influence the four-channel values by inputting R,G,B,G so as to set white balance. RawDigger can output TIFFs of any one of the 4 channels for further manipulation by anything. I suppose it all depends what "in post" means....

---

Hi Ted

I understand what you are saying about the information you can get from a Bayer array. I'm sure that the RGGB data from groups of four pixels can be output for any amount of 'manipulation'. But does anyone manipulate four component 'bayer data' (for want of a better term) in photographic image processing? I guess its possible, but I haven't heard of it.

I also appreciate the semantic about whether a demosaiced file should still be called a RAW file. Good question.

Nevertheless I'm with Manfred here.

The point of demosaicing that 'bayer data' is to produce an RGB triple, per pixel. That is, to assign a 'colour' value to the pixel that is more than just red, or green, or blue so that a high quality image can potentially be constructed from all the 'coloured' pixels. The demosaicing algorithms account for the values of all neighbouring green-filtered pixels when calculating the green number for a pixel. (See here, and here.

AFAIK it's the demosaiced data, RGB triples per pixel, that Raw processors such as ACR and Lightroom work with, albeit having been scaled to a particular colour space.

Interesting stuff. I wish knowing about it would make me better able to hold a camera steady!

Tim

Dave, Manfred, Ted, Tim; you were great.
Thank you very much for your help and explanation.
I really appreciate it.
I will read them again tonight to understand them more.

Quote:

---

Originally Posted by Macmahon

Hi Ted

I also appreciate the semantic about whether a demosaiced file should still be called a RAW file. Good question.

Tim

---

Tim, the "semantic" was not a question - it was a statement, which you can either agree with or refute.

I do understand your and Manfred's points of view; they are simply not mine. Let's agree to disagree. No doubt Hafedh will form his own view, perhaps concluding that a raw file is made of 3-1/2-lets :D

Thank you for the education and links about de-mosaicing which I'm sure will be of interest to Hafedh.

Quote:

---

Originally Posted by xpatUSA

Tim, the "semantic" was not a question - it was a statement, which you can either agree with or refute.

---

Hi again Ted

Forgive me. I did not intend to be smart or dismissive. I meant only that I have no language for the 'limbo' state of a file after it has undergone 'raw conversion' (aka demosaicing etc) into an image representation, but not yet saved out into some other file format. I've always loosely thought of it as a raw file, but it is clearly not, pixel for pixel, the same data that was captured by the sensor.

My language is deficient on this. It's a semantic issue for me. My bad, as they say.

Cheers

Tim

Quote:

---

Originally Posted by Macmahon

. . . I have no language for the 'limbo' state of a file after it has undergone 'raw conversion' (aka demosaicing etc) into an image representation, but not yet saved out into some other file format. I've always loosely thought of it as a raw file, but it is clearly not, pixel for pixel, the same data that was captured by the sensor.

---

No problem, Tim,

Here's how I understand the 'limbo' state. When a raw file is opened for the first time it is converted, including de-mosaicing, into a 'working file' in RAM on your computer. A file in RAM is indeed in 'limbo'. Usually the file is 16-bit RGB, although I've read that Adobe uses 15-bit, believe it or not. The color space for this as yet un-saved file is usually a wide gamut color space such as ProPhoto (which used to be called Kodak ROMM). However what you see on the screen is a transformed version using your screen's profile (usually sRGB, Adobe RGB for posers ;-). Now you play around with the image, editing, cropping, sharpening. All this work is applied to the as-yet un-saved file which, being still in RAM, is still the 'working file'. However, after doing all that, the working file must be 'saved as' something and this is the point where you choose the file type and bit depth, the color space, etc. If you use ACR, the raw file remains as it was but with an added 'sidecar' file which is kind of like a record of all your editing while in ACR. Other editors work differently. RawTherapee saves a sidecar file for any file you edit, not just raw files. Just to be different, Sigma Photo Pro embeds your settings in the raw file itself, should you select that as an option.

Then, and only when you 'save as', does an actual file appear on your hard drive in RGB triplet form as a jpeg, tiff, png, gif, etc. - whatever you chose. Before that, your work was at the mercy of your computer or sheep eating through the mains cable.

Hope this helps . .

Significant addition Ted. things are clearer to me now.
thank you all for your significant output that enriches my information.

I took a workshop from him a few months after that. That experience changed my whole approach to photography. At that workshop in Yosemite in 1973 I decided I wanted to try and see if I could pursue this for myself, and I'm still trying.

Quote:

---

Originally Posted by kashifalijugnoo

I took a workshop from him a few months after that. That experience changed my whole approach to photography. At that workshop in Yosemite in 1973 I decided I wanted to try and see if I could pursue this for myself, and I'm still trying.

---

You took a workshop with whom? and what was it about?

Hafedh - this member has been posting rambling and unconnected comments on a number of different threads today. I wouldn't expect an answer.

Quote:

---

Originally Posted by GrumpyDiver

Hafedh - this member has been posting rambling and unconnected comments on a number of different threads today. I wouldn't expect an answer.

---

Thank you Manfred. it seems so.

Starting from a state of confusion I have read throughout this whole thread and have to thank you all for an "enlightenment". Really excellent stuff and many thanks to the OP.