DPR Forum

Welcome to the Friendly Aisles!
DPRF is a photography forum with people from all over the world freely sharing their knowledge and love of photography. Everybody is welcome, from beginners to the experienced professional. Whether it is Medium Format, fullframe, APS-C, MFT or smaller formats. Digital or film. DPRF is a forum for everybody and for every format.
Enjoy this modern, easy to use software. Look also at our Reviews & Gallery!

Pixel Shift Comparison

MEDISN

DPRF-Patron Gold
Finally had a minute to use the new pixel shift feature on the R5. I shot the same static scene with SL2, OM1 and R5 all using pixel shift. A couple of observations.
1. The R5 is relatively quick to combine the shots - the entire process took about 7 seconds.
2. Highlights can cause trouble so if there are important highlights to protect, consider dialing in some negative exposure compensation.
3. Overall balance of noise and sharpening is pretty good for a JPG.
4. The resulting JPG is 24576 x 16384 and 112MB.
5. The camera really does need to be completely still. I used my phone to trip the shutter after 15s delay. I don't see a way to add a delay in camera.
6. Long exposures are limited by R5's 0.5s electronic shutter.

Here is the full view:

L1000707.jpg


Here is the cropped view normalized to the SL2 (180MP) which I feel hides some of the JPG color noise and jagged edges best. Really, anywhere between 100-200MP looks great but I feel the "400MP" is a stretch.

SL2 v R5 v OM1 at 180MP.jpg


^SL2 on the left, R5 center, OM1 right

The OM1 outputs 80MP files from its 20MP sensor so it was enlarged to approximately same subject size. Normally I use MFT pixel shift at 25-50MP. Likewise SL2 at around 100-150MP. I would probably use the R5 around the same final result.

There is less latitude to work with the JPG obviously so getting exposure right in camera is a must. If I attempt to lift the shadows on the right side of the image, the JPG does not hold up as well as the 14-bit DNG from the SL2.

Shadows +100.jpg


Not bad by any means but something to be aware of.

I tried some shots outdoors but with the breeze they were unusable - jaggies everywhere there is slight motion. The Panasonic and Leica cameras with pixel shift have a motion compensation option which does a pretty good job of freezing motion but not without risk of strange artifacts left behind. Olympus cameras leave a slight blur behind with some jaggies on high contrast edges.

Ultimately, I am thrilled to see more manufacturers offering computational features like this, especially straight out of camera.
 
wow!

Thank you for this great comparison!

112MB is a lot. Strangely, the Olympus image in the crop-comparison looks the best to me on image #2. Is the quality of the lenses so different?
 
wow!

Thank you for this great comparison!

112MB is a lot. Strangely, the Olympus image in the crop-comparison looks the best to me on image #2. Is the quality of the lenses so different?
It is a big file but I think I can resize for transfer in camera. For example, my default Olympus export is 80MP raw + 25MP JPG because the pixel shift 25MP file looks amazing and has small footprint.

SR vs HHHR 25JPG.jpg


As for lenses there is some difference yes. The SL2 had the Panny 24-105 (good but not great), the R5 had the 14-35/4 (better) and the OM1 had the 10-25/1.7 at f/2 (great) to match the FoV and DoF of the other two. The micro contrast on the PL lens is truly excellent.
 
It is a big file but I think I can resize for transfer in camera. For example, my default Olympus export is 80MP raw + 25MP JPG because the pixel shift 25MP file looks amazing and has small footprint.

View attachment 2861

As for lenses there is some difference yes. The SL2 had the Panny 24-105 (good but not great), the R5 had the 14-35/4 (better) and the OM1 had the 10-25/1.7 at f/2 (great) to match the FoV and DoF of the other two. The micro contrast on the PL lens is truly excellent.

Both images are at 100% view. I do not see a difference between the two on my screen. What am I doing wrong? Or is the Jpeg on the right the pixelshift version saved as Jpeg directly in the camera?

The Panasonic-Leica Lumix 10-25/1.7 seems to be a masterpiece. I only hear good things about it. Unfortunately, it is big and heavy. I personally prefer the small lenses in my MFT system. Like the 9/1.7.
 
Both images are at 100% view. I do not see a difference between the two on my screen. What am I doing wrong? Or is the Jpeg on the right the pixelshift version saved as Jpeg directly in the camera?
Really? Try downloading and opening these in your favorite image viewer. The difference in detail in the water droplets, the pollen lying on the flower pedal. The lack of noise in the green background. All readily obvious to my eyes.

Flower in Rain-1.jpg


Flower in Rain-3.jpg


The Panasonic-Leica Lumix 10-25/1.7 seems to be a masterpiece. I only hear good things about it. Unfortunately, it is big and heavy. I personally prefer the small lenses in my MFT system. Like the 9/1.7.

It is a gem as is its brother the 25-50. I have many of the systems small lenses as well but different jobs require different tools and MFT provides ample choices. That is one of my chief complaints about RF system (back on topic :LOL: ) is the lack of high quality, compact options. All the great lenses are large and expensive.
 
Thank you for sharing this result. Your images are pretty good for the R5 and competitive with the others when shooting indoors at products. I still don't know why I really need more the 45MP of stationary objects. Some people do I am sure but just not yet for me. I will keep it in mind for future use though.
 
Back
Top