[Oakland Rob]

Back to PS....

f course you'd rather use a photo merge, what other choice do you have? The other day kengoh (the amazing) suggested a trick he uses with birds, he capture the bird with high ISO, waits until he bird is gone, takes another identical low noise image at low ISO. masks the bird out and then pastes it into the low ISO background. You could do the same with pixel shift, I'm guessing with spectacular results.

Maybe. If the subject is the bird, THAT's where you want the detail. Not necessarily in the background. Some wildlife photographers try to use as small an aperture as they can get away with with some subjects so they get decent DOF for the subject,and then use the same technique to blur up the background, or darken it. And it's still faster to use denoise tools with masks or brushes, especially if you wanna soften up the background a bit anyway.

But if the real subject is say the mountain behind, and you still need to catch the bird, it does work well. I'd still prefer improving detail with a better lens, since it's more versatile, but PS is fun to use on some outside stuff within the limitations of the technology.

I saw an image someone did with a K-3II in PS with a shutter speed like 2-3 seconds with moving water, and it didn't have artifacts. I can't say for sure that it didn't do what I think K-3II's do with movement and JPEG generation in the camera, which is chuck out whole frames with movement. The comparison with non-PS shot isn't conclusive because of different settings on other parameters.

[BrianR]

No, it wasn't (see below)..

Sorry? I thought you were saying a flash wouldn't help freeze motion with PS as the subject has to stay immobile in between each shot, and this is what you meant by "identical" (no subject movement in each of the frames). No?

[Dave L]

Sorry? I thought you were saying a flash wouldn't help freeze motion with PS as the subject has to stay immobile in between each shot, and this is what you meant by "identical" (no subject movement in each of the frames). No?

No movement of the subject or parts of it (e.g. rustling leaves) between the 4 successive frames used to compile a PS image. Thought that would have been clear, as it's the basis of PS.

[BrianR]

No movement of the subject or parts of it (e.g. rustling leaves) between the 4 successive frames used to compile a PS image. Thought that would have been clear, as it's the basis of PS.

I've been agreeing with this from the start, but never mind.

[Dave L]

I've been agreeing with this from the start, but never mind.

We've somehow drifted off into speaking at cross purposes I think, so yes, we should stop.

[stevebrot]

I'd rather do a photomerge. Takes less time in field an you can include moving subjects. Don't always need a tripod.

Similar, but not the same.


Steve

[Sliver-Surfer]

Similar, but not the same.


Steve

Your right not the same but multiple merged shots with a tele can give you a super resolution wide angle image. The limitation would be the image always ends up wider than the lens you are using.

[Oakland Rob]

Your right not the same but multiple merged shots with a tele can give you a super resolution wide angle image. The limitation would be the image always ends up wider than the lens you are using.

There's a good description of a similar technique here: A Practical Guide to Creating Superresolution Photos with Photoshop

Not ideal, but it actually works better with some motion (and many more frames) so is MORE useful handheld or at least not perfectly rigid.

And some have been experimenting with a combination of one PS shot, and one regular RAW. That allow you to have control over what parts of the movement get masked out.

[Zen4Life]

From a post related to a report from CP+ 2016 :

Pixel shift is really good. I was able to take pictures of the exhibition floor handheld, with the camera applying the pixel shift resolution enhancement to the parts that weren’t moving, while detecting the people and using the image from a single shot for those parts of the image. The rep confirmed that the SR is still active while the pixel shift is working. He said that it should be good for handheld shooting in a lot of situations as long as you keep to a relatively high shutter speed.

Read more at: https://www.pentaxforums.com/forums/16-pentax-news-rumors/315224-my-report-cp...#ixzz41ZmRCbnB

I will probably move back to Pentax if and when PS works on moving water (plus leaves, grass, trees, etc.). Preferably in a next gen K-3...

[Rondec]

Your right not the same but multiple merged shots with a tele can give you a super resolution wide angle image. The limitation would be the image always ends up wider than the lens you are using.

I think the whole point of pixel shift is not to give super-resolution, but to increase color depth and dynamic range -- hopefully in a fairly easy to use way that doesn't require a lot of post processing. I think the K-1 will be a step closer to that point than the K3 II. Once you reach the point where you are doing a bunch of layers and masks, it then becomes a choice as to how much you are willing to do. This should be automated and be usable with some subject movement.

But I don't think the point is to generate 80 or 160 megapixel images. It is to produce better 36 megapixel images.

[normhead]

I think the whole point of pixel shift is not to give super-resolution, but to increase color depth and dynamic range -- hopefully in a fairly easy to use way that doesn't require a lot of post processing. I think the K-1 will be a step closer to that point than the K3 II. Once you reach the point where you are doing a bunch of layers and masks, it then becomes a choice as to how much you are willing to do. This should be automated and be usable with some subject movement.

But I don't think the point is to generate 80 or 160 megapixel images. It is to produce better 36 megapixel images.

Exactly... there's a lot being said here that just indicates a lot don't really understand what PS does. There's a lot of discussion about what you can do. Well here's something else you can do. You can turn your camera horizontally shoot a pano then stick the two together and you don't need an FF at all.

The point is, I'm not going to do all that stuff. I am not going to come home from a canoe trip, take the 1000 images I've taken probably 50 of which might be pixel shifted, and combine 200 images to get 50 after processing the other 950. Talk about masochisitic. You are going to do all that work, so you don't "need" pixel shift, and I'm just going to press the button, and look at my images when I get home like I always do. And you are pretending that this is in some way equivalent? Really?

It's amazing how when something looks amazing done by some camera company that isn't your favourite, you can come up with all kinds workarounds. IN this case they will probably ad hours to your work flow. It's all about how you want to spend your life. For me, I never once put two APS-c images together to approximate an FF image. And I will never put together 4 images to try and emulate what pixel shift does. I did that kind of stuff with Photoshop creating my own HDRs years ago, and I'm not interested. I'm interested in things I might actually do. There are always pictures to take, being behind the camera interests me more than being in front of a computer. Pixel shift I will use. These makeshift solutions, I don't want them. Been there, done that. Ease of use is more important than saving spending more money for camera that doesn't have PS and wasting your life.

[stevebrot]

I think the whole point of pixel shift is not to give super-resolution, but to increase color depth and dynamic range...

...But I don't think the point is to generate 80 or 160 megapixel images. It is to produce better 36 megapixel images.

Exactly... there's a lot being said here that just indicates a lot don't really understand what PS does

Add my voice here as well. What pixel shift gives you is something closer to "truth in data" as compared to a Bayer-interpolated merge. The result is higher detail capture for a given number of pixels as well as more accurate microcontrast and color tonality.


Steve

[Winder]

Add my voice here as well. What pixel shift gives you is something closer to "truth in data" as compared to a Bayer-interpolated merge. The result is higher detail capture for a given number of pixels as well as more accurate microcontrast and color tonality.


Steve

Exactly. It is simply a way a eliminating the color and detail loss/artifacts generated by the Bayer filter. This basically gives you the resolution of a Foveon sensor without all of the color/ISO issues that Foveon has.

[Rorschach]

Exactly... there's a lot being said here that just indicates a lot don't really understand what PS does. There's a lot of discussion about what you can do. Well here's something else you can do. You can turn your camera horizontally shoot a pano stick then two together and you don't need an FF at all.

The point is, I'm not going to do all that stuff. I am not going to come home from a canoe trip, take the 1000 images I've taken probably 50 of which might be pixel shifted, and combine 200 images to get 50. You are going to do all that work, so you don't "need" pixel shift, and I'm just going to press the button, and look at my images when I get home like I always do. And you are pretending that this is in some way equivalent? Really?

It's amazing how when something looks amazing done by some camera company that isn't your favourite, you can come up with all kinds workarounds. IN this case they will probably ad hours to your work flow. It's all about how you want to spend your life. For me, I never once put two APS-c images together to approximate an FF image. And I will never put together 4 images to try and emulate what pixel shift does. I did that kind of stuff with Photoshop creating my own HDRs years ago, and I'm not interested. I'm interested in things I might actually do. There are always pictures to take, being behind the camera interests me more than being in front of a computer. Pixel shift I will use. These makeshift solutions, I don't want them. Been there, done that. Ease of use is more important than saving spending more money for camera that doesn't have PS and wasting you life.

I am totally with you here, bro I do not have the interest, the energy or the time to spend hours after hours at the computer after the fact. What comes out of the camera, that's it, in my case.

Then again, I am no pro and have zero aspirations to win photo contests. Just a keen hobbyist who has a day job that is already demanding and challenging enough. So yes, I am one of them despicable jpg people

[clackers]

So yes, I am one of them despicable jpg people

The 'J' word?

Wash your mouth out with soap, young man!