Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
03-09-2016, 06:52 AM   #1
New Member




Join Date: Jan 2015
Posts: 7
Channel bleedthrough in PDCU5 ?

Hi,

I was always wondering how images PP'ed either in Adobe CameraRaw or PDCU could look so different in terms of colors (especially reds), no matter what settings you use. Here I am showing the most bizzare case I have so far experienced. I am shooting to DNG, sRGB with my K-3 II. In PP I am using default parameters in both apps with all corrections turned off.

First image is the original DNG developped in camera.
https://www.pentaxforums.com/forums/attachments/32-digital-processing-software-printing/301924d1457530280-channel-bleedthrough-pdcu5-orig.jpg

The second is a crop of CameraRaw 6.1 developped image
https://www.pentaxforums.com/forums/attachments/32-digital-processing-software-printing/301925d1457530280-channel-bleedthrough-pdcu5-cameraraw.jpg

The third is the very same image processed in PDCU 5.4.2
https://www.pentaxforums.com/forums/attachments/32-digital-processing-software-printing/301926d1457530280-channel-bleedthrough-pdcu5-pdcu.jpg

As you can clearly see in the PDCU image, the originally exclusively orange LED turns red-white, as blue signal sneaks in the highlights (check the histogram).


Now, it gets even weirder. You know, that there are 2 ways of making jpg from your DNG in PDCU, the "save as" and the "save as with image processing" option. I don't really see the difference as obviously the raw sensor data has to be processed somehow in order to scale down to from 14 bits to 8 bits per channel. But OK, the SA does it "it's own way" whatever that is (camera setting?, PDCU default?) , whereas SAWIP takes into account the parameters you have changed in the Laboratory mode. Now, the funny thing is that the LED stays white-red on the images processed via Laboratory even though I turned off all corrections and set everything to default, but the simple "save as" gives much better (and more orange) color of the LED. The last three images are only blue channel from jpgs developped in: 1 - PDCU Laboratory (all default), 2 - PDCU save as, 3 - CameraRaw (all default).
https://www.pentaxforums.com/forums/attachments/32-digital-processing-software-printing/301927d1457530280-channel-bleedthrough-pdcu5-blue.jpg

And now the questions
Where the bloody hell did the blue come from?
How and why is there a difference between "save as" and "laboratory".

All in all it makes me feel that the very well known "Pentax Red Overdrive" issue is not a problem of the camera sensor or exposure as some say, but rather of imperfect image processing.
Can somebody comment on the difference in handling the raw file in PDCU and Adobe applications? Perhaps, it's been solved in another thread already?
I'd like to keep PDCU as PP option, especially for pixel-shift images as they are misread or not recognized by CameraRaw (at least in 6.1). But the color abuse in PDCU prevent me from using it more.

Attached Images
View Picture EXIF
PENTAX K-3 II  Photo       
03-10-2016, 04:20 AM   #2
Senior Member




Join Date: Feb 2013
Posts: 159
Hi mooobr,

I didn't know about that notorious "Pentax Red Overdrive"... Is it similar as the "Canon skin tone"?

I think it's only a matter of white balance / RGB mix. The Red, Green and Blue signals are present (in the 14 bit raw file). PDCU puts more blue into the highlights of the 8 bit file. But indeed, the behavior of PDCU is strange. Maybe it's a deliberate choice for the Japanese market?
03-10-2016, 07:48 AM   #3
New Member




Join Date: Jan 2015
Posts: 7
Original Poster
Hi Azergoth,

you of course made a good point with the WB. As changing WB really makes a difference in color appreciation, that was one of the first things I tried to mess around. But see for yourself... In the range 3000 - 7000K no blue in CameraRaw, but a prominent blue peak in all of PDCU processed images (I am aware of the fine color WB tune, but it also made no difference).

Well, I thought that there were many people around reporting issues with reds in pentax... At least I noticed myself, then did some googling and found a number of forum links like this at no time - https://www.pentaxforums.com/forums/115-pentax-k-5/244792-whats-up-color-red.html
Apparently, sometimes when you take a pic of a prominently red object, in-camera and PDCU processing can lead to a red punch where most detail is lost. I had some of this myself. If I find some examples, I'll post later. I never had this problem, however, while using CameraRaw. On the other hand, certain images come out much more lively out of PDCU in a way I struggle to achieve elsewhere. I was wondering if it was color space management, or something else?
Attached Images
 
03-10-2016, 08:25 AM   #4
Senior Member




Join Date: Feb 2013
Posts: 159
Very interesting indeed...

I would say it's a deliberate choice, as you find the same results in in camera jpegs and in PDCU. Maybe it's their way to add a certain "Pentax touch"? I heard many people tried to reproduce te "Canon touch" in raw PP...

It would maybe interesting to analyse the files with RAWdigger. I think there is a trial version of their software...

03-10-2016, 02:13 PM   #5
New Member




Join Date: Jan 2015
Posts: 7
Original Poster
I just found a perfect instance of this phenomenon in my bathroom.

It's probably no question what is now PDCU (no corrections, natural profile, all settings "middle") and what Adobe (all defaults, WB slightly adjusted to match PDCU) processing...

I know this very much depends on the actual monitor, calibration and stuff, but it is the upper left corner, that takes the gold for documenting the color. On the other hand the left looks like it was already washed 50 times, whereas the right almost new. So perhaps, this vivid red effect might be desirable at times.

That said, I wish there was an easy way from left to right and vice versa. I never (completely) matched the look of an image processed in the other software. :/

(I tried to match the WB using the tiles. There is still a boundary visible, but you know.... it's pretty close.)
Attached Images
View Picture EXIF
PENTAX K-3 II  Photo 
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
bleeding, camera, channel, channel bleedthrough, channels, default, dng, image, images, laboratory, pdcu
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
K-5 with PDCU4 vs. PDCU5 k0og Pentax K-5 1 04-30-2015 03:19 PM
Lens prices and our channel strategy in the U.S. - Ned Responses LFLee Pentax News and Rumors 716 06-18-2012 08:38 AM
RF-602 like: 16 channel 2.4Ghz & AAA in both transmitter and Receiver Russell-Evans Pentax Camera and Field Accessories 24 12-17-2010 05:17 PM
Landscape Patches of rain in the English Channel juu Photo Critique 14 11-12-2010 01:59 AM
8 bits per channel -vs- 16 bits per channel Ed in GA Digital Processing, Software, and Printing 5 03-25-2008 09:02 AM



All times are GMT -7. The time now is 10:43 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top