[todd]

Got around to my first PS test with the K1, which required taking my first trip into DCU. I think it would have helped to have the image filename appended with PS or something when shooting PS'd in Raw only, so we know which image(s) are PS'd.

I am normally a raw only shooter (because jpegs can be extracted so easily when I need/want them) so thinking my workaround will be to try to remember to turn on Raw+ (or assign one of the users for PS with Raw+...) so I have jpegs only with the PS'd shots to help me identify things...

I did see some artifacts in the jpeg processed file from my test...

[Oakland Rob]

Got around to my first PS test with the K1, which required taking my first trip into DCU. I think it would have helped to have the image filename appended with PS or something when shooting PS'd in Raw only, so we know which image(s) are PS'd.

I am normally a raw only shooter (because jpegs can be extracted so easily when I need/want them) so thinking my workaround will be to try to remember to turn on Raw+ (or assign one of the users for PS with Raw+...) so I have jpegs only with the PS'd shots to help me identify things...

I did see some artifacts in the jpeg processed file from my test...

What software do you use?

PS's are pretty easy to ID...they are the ones that take up half your drive space

There is some ID'ing info in exif, but aside from using exiftool it's kinda inaccessible in most applications. But in most you can do a sort or view size, then label or rate or keyword or whatever. I have a saved stationary pad I use in Photo Mechanic, but you could use a template in Bridge, or a preset in Lr (although it's harder in Lr to get at sizes).

---------- Post added 05-11-16 at 05:27 PM ----------

HI Penfred,
I'm copying and pasting from a response I left in another thread out of laziness, but I hope it helps you. I was curious about the same thing. Thinking the camera combines the 4 images, and then using say Adobe Camera Raw to process the one DNG file would give the full benefits of PS Res. It appears, at least right now to get the full effect one needs to run the dng through DCU 5 and click on a pixel shift option.

Here's my other response:

Somewhat formal test ... enough for me anyway. Mirror up. Motion Correction off . Tripod . Remote Release.

I was thinking/hoping I would be able to see the full benefit of PS resolution by processing through Adobe Camera Raw (up to date) . When I compared a ps res. image to a non ps shift res. image there was only a small difference (marginal..10% ?) between the two dng files . But, tonight I decided to install DCU5 and run the ps res. dng through it to see what difference it would make, and it is the difference that I was expecting...significant. It comes pretty close to a 645z file of the same scene, also shot under the same conditions. PS resolution is really impressive.

By the way the DCU 5 interface is quite busy, though given a few minutes I was able to minimally figure out how to process the raw. I did not see a way to combine 4 images from the one dng file, but if you shoot with PS res. on and then tell DCU5 that it is a PS image it re-processes the file and results in the more detailed image.

Hope that may help someones wondering about the processing. I was questioning the same thing.

See https://www.pentaxforums.com/forums/172-pentax-k-3/311865-pixel-shift-finicky.html

In DCU, at least with a K-3II, you have to use a special sharpening option to get optimal results. And crank it up. Ditto in Lr/ACR; crank up the sharpening a bunch. You can separate out the frames with dcraw ps variant if you wanna explore more. Or separate out a frame to use in masking. dcrawps can also mark areas affected by motion.

[todd]

What software do you use?

rawtherapee and gimp are the main apps (on windows).

PS's are pretty easy to ID...they are the ones that take up half your drive space

haha yeah.. could do it that way but a little slower to have to switch the folder from thumbnail view to detail view..

In DCU, at least with a K-3II, you have to use a special sharpening option to get optimal results.

Which one? On my tests yesterday I used deconvolution in rawtherapee on the PS'd tif's from DCU and it looked really nice.

dcraw ps

thanks for mentioning this. might have to try it out at some point.

What software do you use?

In DCU, at least with a K-3II, you have to use a special sharpening option to get optimal results. And crank it up. Ditto in Lr/ACR; crank up the sharpening a bunch. You can separate out the frames with dcraw ps variant if you wanna explore more. Or separate out a frame to use in masking. dcrawps can also mark areas affected by motion.

Where do I find this special sharpening?

[Oakland Rob]

Where do I find this special sharpening?

It's a drop down menu in the Laboratory module under sharpening. The regular one shows an "S" and the one for PS shows an "FS."

Rawtherapee and GIMP are not good tools for organization, so you'd need to use some ingestion tool to mark or maybe rename the PS images. XnviewMP is free and could do it.

It's a drop down menu in the Laboratory module under sharpening. The regular one shows an "S" and the one for PS shows an "FS."

Rawtherapee and GIMP are not good tools for organization, so you'd need to use some ingestion tool to mark or maybe rename the PS images. XnviewMP is free and could do it.

I found this "manual" Pentax K-3 II Review - Processing Pixel Shifted RAWs | PentaxForums.com Reviews

[HopelessTogger]

Lloyd Chambers . . .

diglloyd: Pentax K1 Super Resolution Mode has Color Problems

Lots of posts for the K-1. I think he likes it . . .

diglloyd - Blog

[Franc]

Lloyd Chambers . . .

diglloyd: Pentax K1 Super Resolution Mode has Color Problems

use of ACR instead of the Pentax software?

[cali92rs]

So, the basic gist of this thread that in order to get acceptable results (i.e. without the nasty artifacts present in the K-3II whenever there is the slightest bit of motion) is that you have to use DCU when processing your K-1 PS files?

[todd]

use of ACR instead of the Pentax software?

I had the same thought...

Rawtherapee and GIMP are not good tools for organization, so you'd need to use some ingestion tool to mark or maybe rename the PS images. XnviewMP is free and could do it.

Yeah I've been trying to get them to add at least some tagging features to RT but probably not going to happen... I use Xnview for some plugins but can't see myself adapting to it as an organizer. Maybe I need to upgrade to the MP version and have a poke around though, so thanks...

use DCU

In my initial tests the jpegs weren't bad at all... The few artifacts I saw in the jpeg I think were small and from light movement through the trees on a windowsill in the frame... But those same artifacts were gone in the DCU version... But then when I looked around the image, there were some spots where the jpeg version looked cleaner/sharper... So I think I'm going to be shooting Raw+Jpeg whenever I do these and continue to compare them...

[HopelessTogger]

use of ACR instead of the Pentax software?

Dunno? I've had no probs using LR CC.



https://www.flickr.com/photos/56053365@N07/26895489102/in/dateposted-public/lightbox/

[Rondec]

I use the DCU to generate a TIFF file and then do anything further to that file in Lightroom. It isn't that you have to really torque up the sharpening either, but more that there is just more detail/texture/color depth to the images that have PS versus ones that don't.

This is about worst case scenario with it -- a situation where it was a really windy day and I used too slow a shutter speed. I see motion blur/artifact in the tree on the right, although the rest of the image looks pretty clean to me:



I posted this elsewhere, but this is a comparison with PS off:



And PS on:



Full size are available on Flickr. I didn't bump up the sharpening on either image, but when I zoom in, I can just see the texture on the paper and everything looks crisper with PS than without it. That said, I think non-pixel shift images are adequate for 98 percent of imaging needs.

[infoomatic]

I have not managed to get a good PS picture so far ... OK in the center, but weirdly artifacts on the buildings on the left hand side (OOC)

http://robert.pixelbert.com/temp/IMGP0704.JPG

[HopelessTogger]

I have not managed to get a good PS picture so far ... OK in the center, but weirdly artifacts on the buildings on the left hand side (OOC)

http://robert.pixelbert.com/temp/IMGP0704.JPG

I think Pixel Shift is only good for truly static subjects in controlled (mostly indoor) conditions. But that's good enough. For product, macro and other still life it's superb. For landscape, there's just too much movement regardless.

Horses for courses. Everything has its limitations.

[Oakland Rob]

I think Pixel Shift is only good for truly static subjects in controlled (mostly indoor) conditions. But that's good enough. For product, macro and other still life it's superb. For landscape, there's just too much movement regardless.

Horses for courses. Everything has its limitations.

I agree. I think any motion control sacrifices too much. But motion can spoil even regular shots. For landscape, its another tool in the box. If you can isolate the movement traditional post work with masking would seem the best way to go. Especially considering that in only some cases can the viewer of the photo discern the difference (eg big prints).