ACR Adobe Camera Raw Idea

antibiotic

:dunnoHello all,
I was experimenting with Adobe Camera Raw,, and I opened my Canon 5D MarkII a 21mp, yet in camera raw at the bottom, i changed the setting to a 50mp and both have close results, a little softer on the 50 size.

Also should I import my raw files in at 96 or 200 resolution, if I set it at 96, then I also get a bigger print, then I up the resolution in Photoshop CS6, they both look great, yet I also get a bigger print to work with and crop.:barb

Richard

I don't really understand what setting you changed. The number of pixels is fixed by the sensor and you can't change the raw pixels. You can resample the image to make it larger (or smaller) in PS if you need to for some reason, but this will not improve the image quality, ever. It will only degrade it to a greater or lesser extent, which may be appropriate if you're printing something huge that will be viewed at a distance.

Resolution is strictly an output issue. Your image will have the pixels it has regardless of what import or working resolution you set. Resolution determines how much space the pixels occupy on the output medium.

antibiotic

Adobe Camera Raw
when you open a raw picture the ARC pre-corrections appear,, done at the bottom there is a high-lighted bar,, go into that and try changing the camera pixel size like I have a 21mp canon, i made it a 50mp,, and it came out very nice.

ziggy53

While it's true that all Bayer imaging chips have a fixed number of photosites, none are complete for either chrominance (color) or luminance information because of the color filter over each and every photosite. Bayer chips generally have an R-G-B-G filter sequence and every photosite must be interpolated during the demosaicing function of the image's RAW processing, either in-camera or in post-processing.

Since every photosite must be interpolated anyway, in order to produce pixels (picture elements), it actually does make some sense to also interpolate for the number of pixels (but mostly if the RAW processor does the upres during demosaicing). The advantages include:

A much greater harvest of pixels for further noise reduction and sharpening algorithms, which simply gives greater control over halos, etc. (I greatly prefer to defer noise reduction for the post-RAW phase of image processing for critical work.) Specifically, High-Pass and Deconvolution sharpening can both get grainy rather quickly, so adding more pixels beforehand makes a lot of sense.

A demonstrable improvement over image rotation and straightening operations, even though Adobe does use sub-pixel rendering for those operations. Stair-stepping and aliasing issues mostly go away with the increase in pixels.

Using multiple captures of the same subject and then combining the images in post for resolution and detail increase and/or noise reduction also improves with RAW upres first, IMO.

BTW, I commented on this very issue previously :

Using ACR 5.6 to interpolate/upres crazy high, Feb-11-2010

I have been using the process of RAW interpolation for size ever since (but not on every image, just those which I feel may need it).

Sam

I forgot all about this!

Sam

AceCo55

ziggy53 wrote: »

While it's true that all Bayer imaging chips have a fixed number of photosites, none are complete for either chrominance (color) or luminance information because of the color filter over each and every photosite. Bayer chips generally have an R-G-B-G filter sequence and every photosite must be interpolated during the demosaicing function of the image's RAW processing, either in-camera or in post-processing.

Since every photosite must be interpolated anyway, in order to produce pixels (picture elements), it actually does make some sense to also interpolate for the number of pixels (but mostly if the RAW processor does the upres during demosaicing). The advantages include:

A much greater harvest of pixels for further noise reduction and sharpening algorithms, which simply gives greater control over halos, etc. (I greatly prefer to defer noise reduction for the post-RAW phase of image processing for critical work.) Specifically, High-Pass and Deconvolution sharpening can both get grainy rather quickly, so adding more pixels beforehand makes a lot of sense.

A demonstrable improvement over image rotation and straightening operations, even though Adobe does use sub-pixel rendering for those operations. Stair-stepping and aliasing issues mostly go away with the increase in pixels.

Using multiple captures of the same subject and then combining the images in post for resolution and detail increase and/or noise reduction also improves with RAW upres first, IMO.

BTW, I commented on this very issue previously :

Using ACR 5.6 to interpolate/upres crazy high, Feb-11-2010

I have been using the process of RAW interpolation for size ever since (but not on every image, just those which I feel may need it).

You continually amaze me at the breadth and depth of your knowledge!!!

Great info.