Another pixel shift, but also focus stacked.

Bollygum

Bollygum

Messages
439
Name
Steve
Edit My Images
No
It's a bit long winded to process, but the result is a bit special. After putting the images together (20 imges, 5 layers and 4 pixel shift images per layer, all at f16 on an A7R3) there was no need for any sharpening or other photoshop processing except crop and resize. I don't know about anyone else, but I think the colour is amazing - and that's exactly as the camera registered it.
P.S. The fungus is Cyptotrama aspartum which is quite a common fungus around here, though this was a particularly good specimen.


DSC00721_PSMS-helicon.jpg
 
Now that is amazing.
Now your talking gobbledygook with this (20 images, 5 layers and 4 pixel shift images per layer) Got no idea what it mean but I understand it took a superb photo :)
 
Superb.
I too do not have a clue about the technical bit that helped to put this together. But i do know how much i like it.
 
Thanks Boogie. I'm not sure that the gobble is all that important in a 4k photo
 
Very nice work Steve
 
There's a very good reason for that (why not use a "better" aperture) - and that is - time and probability of error. I take most of my photos in the field which us usually rainforest. It is a challenging environment. I have found that it is only worth taking extra photos and/or taking extra time if I can get noticeably better results. Using "better" apertures does not produce noticeably better results and it adds significantly to both the time taken and the chances of errors that would make the final image useless. These errors include - changes of light, wind, tripod moves, errors in focusing, etc. By using a "worse" aperture, I can use the focus ring to focus, whereas I need to use a focus rail for "better" apertures. Pixel shift produces far more difference than using a wider aperture, so I will use it sometimes, but only for that exceptional photo. One of my current projects is to produce some very high resolution prints. The prints will be of fungi macros. To do this I intend to use my Sony A7R3 using Pixel Shift, with a Canon TS-E 135mm macro lens. I will use the shift function of the TS-E lens to produce multiple images that can easily be combined into a panorama. I will also use focus stacking and pixel shift. The final image will be about 100MP but with Pixel Shift it will be equivalent to significantly more than that. I should be able to print an image that is 2000x1350mm (60inx40in) in very high resolution.
As an example of where I will make compromises in order to increase my chances of a good photo, I will often use ISO200 or 400 or even 800 if the light means I would have to use exposures of gt 3 secs. I have found that the chances of an error increase dramatically, the more time is taken in completing the picture.

The final file is 42MP both before and after pixel shift.
 
Nice composition,but it doesent look in focus to me in certain areas?

I’ve no idea how to do what you did,but can’t you achieve the same with a tripod and hyperfocal focussing?
 
futurelegend said:
Nice composition,but it doesent look in focus to me in certain areas?

I’ve no idea how to do what you did,but can’t you achieve the same with a tripod and hyperfocal focussing?
Click to expand...
I agree, it's not perfect in all areas - the fungi on the left are a bit soft. That is probably because it was my first time using pixel shift and I was still learning. It was shot with a tripod and no, you can't get the dof with a single shot.
 
Bollygum said:
There's a very good reason for that (why not use a "better" aperture) - and that is - time and probability of error.
Click to expand...
I certainly understand that. But are you saying that you don't see the effects of diffraction at f/16, or that it is less of a detriment than other factors using wider apertures introduce (i.e. stacking errors/movement)?

I was curious as the file size in MB, not MP... I would guess with edit layers it would be pushing towards a gigabyte.
 
sk66 said:
I certainly understand that. But are you saying that you don't see the effects of diffraction at f/16, or that it is less of a detriment than other factors using wider apertures introduce (i.e. stacking errors/movement)?

I was curious as the file size in MB, not MP... I would guess with edit layers it would be pushing towards a gigabyte.
Click to expand...
The diffraction effect is small compared to other effects. The file size isn't more than a GB as you don't keep all the interim layers in the file. The largest file will be the merge of 3 tiff images at the end, which is 3 X 120MB. The saved file will be less than that
 
That looks very promising.

You used four pixel shift images per layer. Does that mean that each layer is a 2 x 2 panorama constructed from 4 pixel-shift images? My mind is a bit boggled as to the physical mechanics of achieving that (keeping the focus plane the same for each set of four). I think I must have misunderstood something here. :)

Perhaps each layer is a pixel shift composite image constructed from four pixel-shift captures of the same part of the scene.
 
GardenersHelper said:
That looks very promising.

You used four pixel shift images per layer. Does that mean that each layer is a 2 x 2 panorama constructed from 4 pixel-shift images? My mind is a bit boggled as to the physical mechanics of achieving that (keeping the focus plane the same for each set of four). I think I must have misunderstood something here. :)

Perhaps each layer is a pixel shift composite image constructed from four pixel-shift captures of the same part of the scene.
Click to expand...
The camera does it all by moving the sensor by 1 pixel increments. I couldn't possibly control that myself..
 
Bollygum said:
The camera does it all by moving the sensor by 1 pixel increments. I couldn't possibly control that myself..
Click to expand...
TBH, I don't think the shift has to be that precise/controlled at all... it might require a couple extra images if it was done more randomly (i.e. camera shake).
I might have to experiment with this...
 
This is a very nice image but I have no idea of the size of the subjects.
Can you please give me some indication?
 
You must log in or register to reply here.
Back
Top