E-M1 III Focus Stacking II, IQ bad Compared to PS

Gary from Seattle said:

Bassam Guy said:

Sdlv said:

You left out some details. Did you shoot with a tripod or handheld? And what was the stacking setup (number and differential) you used?

. . . Steven

Click to expand...

Handheld. Just eight shots.

Click to expand...

That could lead to a problem because the focal depth of the flower is significant.

Click to expand...

Pedagydusz said:

By the majority of us, that is!

Photo Stack is not miraculous. It is a marvellous, but difficult technique, full of pitfalls. First of all, the parameters can be wrong: the number of shots, combined with the f value and the difference in between can easily result in areas, or strips of unfocused subject, which are a total killer of the shot. But this is more or less apparent.

But worse is the non-apparent difficulty. If the camera and lens is not held very steadily, things can go wrong. They can, of course, go totally wrong - but that is a blessing, because you are notified that there was a failure, a stacked image is not produced, but the individual frames are still available, as JPEGs or ORFs, to be processed afterwards, preferably in Helicon Focus or Zerene.

But they can also be apparently ok, no failure reported but the stacked image sucks. That happens when there was excessive wiggle of the photographer-camera-lens, and one or more frame just is not in the right place, and the stacking process includes some oof areas that totally ruin the image! It was, I suspect, the OP case. Fortunately, the individual frames are still available, in this case, and everything can be made good.

But to me that is not correct: if the shot is not 100% successful, a failure should be reported immediately, not by observing the image later. Perhaps a firmware update will improve this...

Note that camera motion in-between frames is much easier than it is thought. A special note to the fact the IBIS, however excellent it is, does not stop that. To my understanding, IBIS (and OIS, and synchronous function) correct camera motion while the frame is taken, not to my knowledge efficiently afterwards, not even when pressing half-way the shutter apparently freezes the image.

So, I think that Focus Stacking can be wonderful, but if we are not sure of total immobility we should brace ourselves, or find a support to steady the camera-lens during the process. Even a tree trunk, a fence, a door knob, a camera case, a boot...

Click to expand...

Digital **** said:

My E-M5 II can do in camera focus stacking but I don't have any of the Pro lenses need for that feature to work.

Click to expand...

Digital **** said:

However, I do a lot of focus bracketing and then stack the images in PS. I've never noticed any problem with sharpness. I agree that PS is slow particularly on my old computers.

****

Click to expand...

cba_melbourne said:

Bassam Guy said:

Thanks but my question is about SOOC vs PS, not about camera movement, number of shots, aperture, etc. PS does a much better job than the camera using the very same images.

Click to expand...

The answer to that is trivial.

A PC simply has way more processing power and fast memory available than a camera. The stacking software can use much more complex algorithms to stack.

Whereas the camera has to do the best it can with very limited resources, and provide a (any) result in short time.

In camera stacking is grossly overrated by marketing. It almost only serves the purpose of convincing consumers to buy the few Olympus/OM lenses that are "compatible" with focus stacking. The results are sub-par and only useful for two things: to assess if the bracketing worked or needs to be repeated, or for those that for some reason or another need to immediately upload the stacked results to show someone on the web.

Click to expand...

Pedagydusz said:

I am afraid I didn't make myself clear. The technique is not "sub-par". If the images are perfectly aligned, it works like a treat - provided that 15 images are enough, which is not always the case. I have on several occasions used dozens of frames for stacking to my liking (afterwards). But if 15 frames or less are enough, and the camera is very steady, the result is excellent.

If the answer to any of these conditions is "no" - i.e., 15 frames are not enough or the camera moved during the take - then, as you say the little computer in the camera is not enough. No problem, the frames are there, go to a "big computer"!

However, on many, many circumstances, stacking works, and looks like a miracle! It just is not an universal miracle, but when it works it is as good as any stacking done some other way.

And I have seen faults in stacking done in many other ways, with different software, with halos and ghosting, and oof areas! In fact, I know photographers who swear they will never use it. But photography is full of pitfalls - noise, lack of dynamic range, lens aberrations, diffraction, metamerism, ... And we go on doing it, fighting one problem at a time - and enjoying it, and being proud of our resources and skills!

Click to expand...

cba_melbourne said:

Pedagydusz said:

I am afraid I didn't make myself clear. The technique is not "sub-par". If the images are perfectly aligned, it works like a treat - provided that 15 images are enough, which is not always the case. I have on several occasions used dozens of frames for stacking to my liking (afterwards). But if 15 frames or less are enough, and the camera is very steady, the result is excellent.

If the answer to any of these conditions is "no" - i.e., 15 frames are not enough or the camera moved during the take - then, as you say the little computer in the camera is not enough. No problem, the frames are there, go to a "big computer"!

However, on many, many circumstances, stacking works, and looks like a miracle! It just is not an universal miracle, but when it works it is as good as any stacking done some other way.

And I have seen faults in stacking done in many other ways, with different software, with halos and ghosting, and oof areas! In fact, I know photographers who swear they will never use it. But photography is full of pitfalls - noise, lack of dynamic range, lens aberrations, diffraction, metamerism, ... And we go on doing it, fighting one problem at a time - and enjoying it, and being proud of our resources and skills!

Click to expand...

My answer remains the same. A PC based stacking software can not only use more elaborate algorithms. It can do additional steps to the stacking itself, like are necessary for merging imagines of slightly different size, or perspective as the camera ever so slightly moves between the shots. Especially if handheld bracketing. There are lots of things that can be done, that the in-camera firmware just cannot do for lack of processing power and storage space and the resulting time to execute.

Another way to explain would be: there is a reason a PC workstation cannot run on a tiny 10 Watt/hour camera battery.

Click to expand...

Bassam Guy said:

Gary from Seattle said:

Bassam Guy said:

Sdlv said:

You left out some details. Did you shoot with a tripod or handheld? And what was the stacking setup (number and differential) you used?

. . . Steven

Click to expand...

Handheld. Just eight shots.

Click to expand...

That could lead to a problem because the focal depth of the flower is significant.

Click to expand...

Thanks, but that doesn't explain the disparity between PS and SOOC - and the ORFs have all areas of interest well focused.

Click to expand...

Gary from Seattle said:

Gary from Seattle said:

[...] Instead short comings lie in some ghosting (likely motion or change of lens perspective on largely OOF elements of an individual image), or largely failed stacks where something moved during the shooting of the individual images.

Click to expand...

Click to expand...

Pedagydusz said:

Gary from Seattle said:

Pedagydusz said:

[...] Instead short comings lie in some ghosting (likely motion or change of lens perspective on largely OOF elements of an individual image), or largely failed stacks where something moved during the shooting of the individual images.

Click to expand...

Click to expand...

Gary, when something moves during the shooting but not much, strange things can happen and some frames can be ignored from the final result, without any information. I have seen this, have even "forced" this to happen.

Click to expand...

Pedagydusz said:

It is not common, though, but happens.

Click to expand...

Pedagydusz said:

Gary from Seattle said:

Pedagydusz said:

[...] Instead short comings lie in some ghosting (likely motion or change of lens perspective on largely OOF elements of an individual image), or largely failed stacks where something moved during the shooting of the individual images.

Click to expand...

Click to expand...

Gary, when something moves during the shooting but not much, strange things can happen and some frames can be ignored from the final result, without any information. I have seen this, have even "forced" this to happen.

It is not common, though, but happens.

Click to expand...