Pages: [1] 2   Go Down

Author Topic: New Article - Sony a7r III Pixel Shift  (Read 7275 times)

Kevin Raber

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1339
  • Kevin Raber
    • Kevin Raber
New Article - Sony a7r III Pixel Shift
« on: November 20, 2017, 11:23:44 pm »

I just published an article on the home page of LuLa about the NEW Sony a7r III pixel Shift feature.  The article contains files to view at 100% plus to download.  Check it out
Logged
Kevin Raber
kwr@rabereyes.com
kevin@photopxl.com
rockhopperworkshops.com
photopxl.com

mlewis

  • Full Member
  • ***
  • Offline Offline
  • Posts: 118
Re: New Article - Sony a7r III Pixel Shift
« Reply #1 on: November 21, 2017, 07:58:52 am »

The pixel shifted image of the car had slight ghosting on the LH headlight compared to non pixel shifted Sony shot.  The camera must not have been quite stable enough.  The pixel shifted engine was slightly sharper compared to non pixel shifted Sony shot.  I couldn't see a difference with the red truck between the Sony shots.
Logged

Kevin Raber

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1339
  • Kevin Raber
    • Kevin Raber
Re: New Article - Sony a7r III Pixel Shift
« Reply #2 on: November 21, 2017, 09:15:03 am »

On the red truck, the logo shows the most difference.  On edges as well as the inside of the yellow logo.  The camera was well locked down but if someone walked in front of the chrome it changed the refection. You can see in the grasses too a bit of ghosting.  This is definitely a feature that should be used in a good lighting condition such as a studio or interior.
Logged
Kevin Raber
kwr@rabereyes.com
kevin@photopxl.com
rockhopperworkshops.com
photopxl.com

jsachs

  • Newbie
  • *
  • Offline Offline
  • Posts: 1
Re: New Article - Sony a7r III Pixel Shift
« Reply #3 on: November 21, 2017, 10:38:35 am »

I don't think this test really shows what pixel shift can do because you need to start with really sharp images at the individual pixel level to get the best results.

There are three reasons you are not getting images as sharp as possible:

1) Diffraction - to get a circle of confusion equal to the pixel spacing on the A7RII sensor, you need to shoot around F/3.5 or less.

2) Depth of Field - depth of field using a circle of confusion equal to the pixel spacing is extremely narrow, especially at F/3.5.

3) Motion blur - the tiniest bit of subject or camera motion can blur images at the single pixel level. This may not be an issue shooting cars with the camera on a heavy tripod, but trees, grass and water will be an issue except on dead calm days.

In my opinion, the best subjects for pixel shift technology are photographing flat objects using a copy stand or some architectural subjects with limited depth of field. I don't think pixel shift will be terribly useful for most landscape photography.

Jonathan Sachs
Digital Light & Color
Logged

Kevin Raber

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1339
  • Kevin Raber
    • Kevin Raber
Re: New Article - Sony a7r III Pixel Shift
« Reply #4 on: November 21, 2017, 11:34:43 am »

Jonathan . . . you are right and wrong.   This was done with the Sony guys next to me.  We the camera was locked on a very sturdy tripod.  It didn't move and when the Phase One was used the seismograph showed no movement. F/8 and f/11 with G-Master lenses are not going to have huge or any diffraction.  Even at f/16 they will be fine.  There is differences that can be seen at 200% and 300% thus the reason I provided files for download.  In the end you still get a 42 MP file.  And if you look where I mentioned you can see differences.

Maybe flat field and copy stand will show a difference.  I can only work with what I had.  The car wasn't moving and the tripod wasn't moving.

I'll certainly test this further when I get my own camera.
Logged
Kevin Raber
kwr@rabereyes.com
kevin@photopxl.com
rockhopperworkshops.com
photopxl.com

GrahamBy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1813
    • Some of my photos
Re: New Article - Sony a7r III Pixel Shift
« Reply #5 on: November 21, 2017, 12:23:48 pm »

So if I understand it, from a single image you get 42MPx of luminance information, but roughly one third (1/2 for green, 1/4 for blue or red) of that for the colour info because of the Bayer filter: the final image colours are reconstructed in the raw converter but with necessarily lower spatial resolution.

http://www.strollswithmydog.com/bayer-cfa-effect-on-sharpness/

With 4 images, you still get 42MPx of luminance info, but you now get full resolution for the colour components.

In short, if you shoot B&W there is no advantage, but there may be some improvement on sharp edges between different colors... if everything stays perfectly still.

In a way, we've gone back to three-plate colour, as described by James Clerk Maxwell in 1855...  :D

http://www.clerkmaxwellfoundation.org/html/first_colour_photographic_image.html
Logged

adias

  • Guest
Re: New Article - Sony a7r III Pixel Shift
« Reply #6 on: November 21, 2017, 12:30:08 pm »

Take away: The Phase One Trichromatic is very good. Nothing beats resolution.
Logged

jmlphotography

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 63
Re: New Article - Sony a7r III Pixel Shift
« Reply #7 on: November 21, 2017, 01:22:37 pm »

It seems from what I've read that Olympus has a much better implementation of pixel shift technology.  Would you agree and whether yes or no, perhaps comment a bit on this?
Logged

uscholdm

  • Newbie
  • *
  • Offline Offline
  • Posts: 4
Re: New Article - Sony a7r III Pixel Shift
« Reply #8 on: November 21, 2017, 03:42:05 pm »

I just published an article on the home page of LuLa about the NEW Sony a7r III pixel Shift feature.  The article contains files to view at 100% plus to download.  Check it out

This is  very good article, except it is incredibly clunky and painful to do side by side comparisons. It would be wonderful if you could give us side by side images. If you take 30-60 min, then you save us many 100s of minutes and more importantly, more people will get the real value of this post.

Logged

Kevin Raber

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1339
  • Kevin Raber
    • Kevin Raber
Re: New Article - Sony a7r III Pixel Shift
« Reply #9 on: November 21, 2017, 04:03:19 pm »

I provided links so you can download the same files used in the article and do your own side by sides in Photoshop or whatever program you like.

Did you see the links?
Logged
Kevin Raber
kwr@rabereyes.com
kevin@photopxl.com
rockhopperworkshops.com
photopxl.com

Majohnson

  • Newbie
  • *
  • Offline Offline
  • Posts: 34
Re: New Article - Sony a7r III Pixel Shift
« Reply #10 on: November 22, 2017, 04:12:31 am »

Thanks Kevin for the comparison.

I can definitely see the improvement on the car with the pixel shift and the difference appears quite significant to my eye when pixel peeping.
I am looking forward to trying this out on some food and product shots when I get the camera to test.

Mark

Logged

Jeff

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 82
Re: New Article - Sony a7r III Pixel Shift
« Reply #11 on: November 22, 2017, 10:04:00 am »

With no knowledge / understanding of the technology it must be quite clever in shifting the sensor by a single pixel as I guess the size of a pixel is quite small.

Also, as Phase One have a special relationship with Sony Cameras & Capture One Pro so one wonders where this is going in the future as C1 cannot merge raw files.
For example, will you have to do the merging in Sony software then import into C1 for processing or are you totally reliant on Imaging Edge for all of the processing ?
Logged

SteveB

  • Newbie
  • *
  • Offline Offline
  • Posts: 6
Re: New Article - Sony a7r III Pixel Shift
« Reply #12 on: November 22, 2017, 01:21:57 pm »

Interesting comparison Kevin.  This is intriguing technology but I strongly suspect it will Not provide increased resolution in the making of prints (we'll see ...) unless you print them to trendy humongous size, so I don't find this technology all that useful for yours truly.  File resolution has gotten so good that you can buy almost any camera these days and get very good if not excellent print files.  My little Olympus 4/3 system can make almost as good prints technically as my Sony A7rii and Canon 5DMk3 systems to about 24" x 20" print size. These types of features are of more of interest to technologists and camera suppliers than folks wanting to make compelling pictures.  After a career of 35 years in engineering, I still stay abreast of camera technology (a proud geek) but I'm much more interested in capturing compelling images than good ol' pixel peeping.  Oops, sorry for getting off tangent ... Cheers, Steve     
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: New Article - Sony a7r III Pixel Shift
« Reply #13 on: November 22, 2017, 02:03:56 pm »

Jonathan . . . you are right and wrong.   This was done with the Sony guys next to me.  We the camera was locked on a very sturdy tripod.  It didn't move and when the Phase One was used the seismograph showed no movement. F/8 and f/11 with G-Master lenses are not going to have huge or any diffraction.  Even at f/16 they will be fine.  There is differences that can be seen at 200% and 300% thus the reason I provided files for download.  In the end you still get a 42 MP file.  And if you look where I mentioned you can see differences.

Maybe flat field and copy stand will show a difference.  I can only work with what I had.  The car wasn't moving and the tripod wasn't moving

Kevin,

In what ways is Jonathan right and wrong? The pixel pitch of the A7riii is 4.51 microns, and the diagonal of the pixel is 6.38 microns. At f/8 and f/11 the Airy disc diameters are 10.3 and 14.2 microns respectively and there will be additional defocus blur at object distances other than the focus distances. Add aberrations and camera shake to that, and the blur circle will be considerably greater than that of the pixel size. To optimize the blur circle to the pixel, you should really not shoot at apertures smaller than f/5.6.

Perhaps you don't know that Jonathan is quite accomplished and his opinion should be taken seriously. He was the author of the popular DOS spreadsheet Lotus 1-2-3 and also of an excellent depth of field calculator that takes diffraction into account (DoF). He also wrote a photo editing program that is highly recommended by Norman Koren (Picture Window Pro).

Regards,

Bill

 
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: New Article - Sony a7r III Pixel Shift
« Reply #14 on: November 22, 2017, 03:37:26 pm »

Hi Kevin,

Good lenses perform best at around f/5.6. Diffraction definitively plays a role at f/8. This was shot at 4.8 microns with a something like 25 years old Minolta 100/2.8 Macro lens.

f/5.6f/11f/16

The loss of sharpness at f/11 is not catastrophic, but clearly noticable.

The pixel shift will not increase image detail, after all, it is still a 42 MP image. But interpolation of colours on a RGBG image causes loss of fine detail contrast. So, resolution will remain the same, but visual acuity will improve.

For a proper comparison, the Sony should be shot at f/8 and the Phase One at f/12, that would give the same DoF.

Note how much wider DoF is for the A7rIII in the enclosed image. Phase one image downsized to A7rIII image size.

Best regards
Erik


Jonathan . . . you are right and wrong.   This was done with the Sony guys next to me.  We the camera was locked on a very sturdy tripod.  It didn't move and when the Phase One was used the seismograph showed no movement. F/8 and f/11 with G-Master lenses are not going to have huge or any diffraction.  Even at f/16 they will be fine.  There is differences that can be seen at 200% and 300% thus the reason I provided files for download.  In the end you still get a 42 MP file.  And if you look where I mentioned you can see differences.

Maybe flat field and copy stand will show a difference.  I can only work with what I had.  The car wasn't moving and the tripod wasn't moving.

I'll certainly test this further when I get my own camera.
« Last Edit: November 22, 2017, 04:02:51 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

Kevin Raber

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1339
  • Kevin Raber
    • Kevin Raber
Re: New Article - Sony a7r III Pixel Shift
« Reply #15 on: November 22, 2017, 04:06:09 pm »

In real-world picture taking using optimal f/stops is not always possible.  I'll never argue with science but when needed you use the f/stop that delivers the DOF needed to get the shot.  And, I made it very clear in the article that the result of the Pixel Shift is a perceived improvement in image quality.  The test I did was typical of conditions many photographers would use Pixel Shift.  While I love the Physics and Science of photography I am a photographer and thus sometimes I have to use not so optimal f/stops like f/22 god forbid.  As many know I am a Capture One user and there are tools in C1 that help minimize many of the issues of diffraction and other lens aberrations.  Bottom line is that in the end, most people will never know the difference between an image shot at f/5.6 or f/11 or even f/22.  Unless you have something to compare to and in most cases, the results will be hard to see.  For me, it is getting the best image possible using the tools I have.  And, yes many times when I need to leave the sweet spot of my lens I say to myself - ouch, this is going to hurt but at least I got a photograph and usually, I can make the best of it.

Either way, Pixel Shift by Sony and the others is pretty damn cool especially when you consider the science and technology that it takes to make it happen.

For me, I don't let physics and science get in the way of my art.
Logged
Kevin Raber
kwr@rabereyes.com
kevin@photopxl.com
rockhopperworkshops.com
photopxl.com

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: New Article - Sony a7r III Pixel Shift
« Reply #16 on: November 23, 2017, 01:21:04 am »

Hi Kevin,

If you look at the blue car shot, the front of the car has a lot of fake detail on the pixel shift exposure. Some of that is probably coming from overly aggressive processing. The plain exposure with the A7rII is much cleaner. The 100 MP back really plays it's resolution advantage. See the first attachment.

If DoF mattered for artistics reasons, the 100 MP Phase One camera would need to be stopped down to f/16 to match DoF, if you compare the rear tire it is kept in focus on the A7rIII but it is totally out of focus on the Phase One shot. See second attachment. That part of the pixel shift exposure is totally unusable BTW. See attachment #2.

In the above images the Phase one image was downsized to match the resolution of the A7rII.

The third attachment really shows the advantage of the larger sensor. There are processing differences, of course, the Phase One image has a higher contrast tone curve, quite aggressive sharpening combined with quiet aggressive noise suppression, fairly typical of Capture One default processing.

As the pixel shift mode does not increase resolution, it's main benefit is reduction of colour aliasing and removing the colour interpolation used in the demosaic algorithm. The point Jonathan Sachs tried to make is that using smallish apertures removes the main advantage of the pixel shift. Stopping down to f/11 adds quiet enough blur to eliminate most of the aliasing on the A7rII. So no benefits will remain. The any benefit we see is grittyness from oversharpening in the Sony software plus a generous amount of artefacts.

Some of us are interested of both the art and the technology behind, it can be a learning experience to all.

Best regards
Erik





In real-world picture taking using optimal f/stops is not always possible.  I'll never argue with science but when needed you use the f/stop that delivers the DOF needed to get the shot.  And, I made it very clear in the article that the result of the Pixel Shift is a perceived improvement in image quality.  The test I did was typical of conditions many photographers would use Pixel Shift.  While I love the Physics and Science of photography I am a photographer and thus sometimes I have to use not so optimal f/stops like f/22 god forbid.  As many know I am a Capture One user and there are tools in C1 that help minimize many of the issues of diffraction and other lens aberrations.  Bottom line is that in the end, most people will never know the difference between an image shot at f/5.6 or f/11 or even f/22.  Unless you have something to compare to and in most cases, the results will be hard to see.  For me, it is getting the best image possible using the tools I have.  And, yes many times when I need to leave the sweet spot of my lens I say to myself - ouch, this is going to hurt but at least I got a photograph and usually, I can make the best of it.

Either way, Pixel Shift by Sony and the others is pretty damn cool especially when you consider the science and technology that it takes to make it happen.

For me, I don't let physics and science get in the way of my art.
« Last Edit: November 23, 2017, 01:24:09 am by ErikKaffehr »
Logged
Erik Kaffehr
 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: New Article - Sony a7r III Pixel Shift
« Reply #17 on: November 23, 2017, 01:31:53 am »

Hi Jonathan,

Welcome to the forum. I used to be a Picture Window user before I converted to MacOS. (*)

I also enjoyed quite a few of the articles you published about colour spaces.

Please, keep on posting!

Best regards
Erik

(*) Well, I am back on Windows. Needed to replace the Macs and Macs with specs I needed were way to expensive.



I don't think this test really shows what pixel shift can do because you need to start with really sharp images at the individual pixel level to get the best results.

There are three reasons you are not getting images as sharp as possible:

1) Diffraction - to get a circle of confusion equal to the pixel spacing on the A7RII sensor, you need to shoot around F/3.5 or less.

2) Depth of Field - depth of field using a circle of confusion equal to the pixel spacing is extremely narrow, especially at F/3.5.

3) Motion blur - the tiniest bit of subject or camera motion can blur images at the single pixel level. This may not be an issue shooting cars with the camera on a heavy tripod, but trees, grass and water will be an issue except on dead calm days.

In my opinion, the best subjects for pixel shift technology are photographing flat objects using a copy stand or some architectural subjects with limited depth of field. I don't think pixel shift will be terribly useful for most landscape photography.

Jonathan Sachs
Digital Light & Color
Logged
Erik Kaffehr
 

Kevin Raber

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1339
  • Kevin Raber
    • Kevin Raber
Re: New Article - Sony a7r III Pixel Shift
« Reply #18 on: November 23, 2017, 07:44:46 am »

Erik and Jonathan
Thanks for your input and further explanation.  I included the Phase One image just because I had the camera with me and thought it would be interesting to compare. I should have shot at a smaller f/stop great DOF on Phase One. Frankly, I didn't see a lot of advantage with Pixel Shift.  The way you guys explained it one would hardly ever use it.  It does reduce artifacts, stair stepping in diagonals in an image if you really pixel peep.  The feature is there if photographers want to use it.  Many photographers aren't going to be aware of much what you discuss regarding f/stops and diffraction. 

Pixel shifting aside the Sony a7r III for $3100.00 delivers a beautiful file.  I have now shot a lot with the camera.  Although we were told we were shooting with production cameras the firmware on mine was .91.  I ordered the camera and when it is received I'll share a lot more images in another article.  The camera is a big step in features even though it has the same chip as the a7r II.  I shot 1500 plus images and was still above half on battery power.  The image Stabilization really works great.  I was shooting handheld at 1/8th of a sec. with good results. Dynamic range is also pretty impressive. 

The thing with Sony is you know it will only get better from here.

Thanks to both of you for good posts here.
Logged
Kevin Raber
kwr@rabereyes.com
kevin@photopxl.com
rockhopperworkshops.com
photopxl.com

Farmer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2848
Re: New Article - Sony a7r III Pixel Shift
« Reply #19 on: November 23, 2017, 01:40:20 pm »

I would have thought that pixel shift is something that you could use to create an image to be used as a layer.  Take a "main" shot, then take a pixel shifted shot.  Where the pixel shifted shot has more details or improved image quality and so on, you can use that and where it's an issue because of motion blur or areas that were out of focus and what not, you use the "main" shot.

I'm sure there are other techniques, too, where this could be useful - something not unlike bracketing and then blending.
Logged
Phil Brown
Pages: [1] 2   Go Up