R6 Focus Stacking bracketing to overexposed images

JustUs7 wrote:

If you used manual with fixed ISO and the exif data show’s consistency in the values, then isn’t the only possibility that the light changed?

In theory, yes, that's why this is so curious, because that definitely isn't what happened.

1. I took 7 stacks then and there and each showed the same behavior: First shot correctly exposed, rest overblown, although settings are (seem to be) the same.

2. It's taken with the Electronic shutter: that means 20fps, light would have needed to change for exactly 1/20s and then change back seven times in a row. That's not what happened there

So yeah, it seems that the camera actually changes some setting but it doesn't register on the EXIF data, because the additional light has to come from somewhere.

Since firmware 1.8 was pulled, I suppose it has something to do with that.

Blue-Shift wrote:

JustUs7 wrote:

If you used manual with fixed ISO and the exif data show’s consistency in the values, then isn’t the only possibility that the light changed?

In theory, yes, that's why this is so curious, because that definitely isn't what happened.

1. I took 7 stacks then and there and each showed the same behavior: First shot correctly exposed, rest overblown, although settings are (seem to be) the same.

I suspect that the aperture is returned to f/2.8 after the first shot. Maybe do a trial an look into the lens what happens to the aperture. And then forget about the mess of taking stacks, they are not worth the hassle, especially if parts of the subject overlap within the stack, that will always lead to horrible artifacts that no software can remedy because they are down to the physical facts of focusing "through" foreground elements which will be depicted unsharp but larger as they are in the stack layer that is sharp...

charlyw64 wrote:

I suspect that the aperture is returned to f/2.8 after the first shot. Maybe do a trial an look into the lens what happens to the aperture. And then forget about the mess of taking stacks, they are not worth the hassle, especially if parts of the subject overlap within the stack, that will always lead to horrible artifacts that no software can remedy because they are down to the physical facts of focusing "through" foreground elements which will be depicted unsharp but larger as they are in the stack layer that is sharp...

Good call Charly,

I tried reducing the exposure by 2x in Lightroom and now the pics are identical, the formerly bright one with reduced DOF. So this is what is happening here.

Which is weird for 2 reasons.

1. It shouldn't happen in the first place

2. If it does, it should register as F2.8 in the EXIF data. But that says F5.6 still.

Well, I assume it will be removed when Canon releases the fixed firmware. We'll see.

Also, about the stacking: I found it to be practical for some cases. Sometimes, it works quite well without too much of a hassle if one isn't too perfectionist. Also, it can be good just to have a range of shots to choose from, so I like the functionality.

Especially since the high fps of the R6 in electronic shutter can do it very efficiently, the 40fps of the R6 Mk2 no doubt increasing this even more.

Do you have 'exposure smoothing option'? Like here:

https://cam.start.canon/vi/C004/manual/html/UG-03_Shooting-1_0290.html

New firmware update is also available.

Polcastrol wrote:

Do you have 'exposure smoothing option'? Like here:

https://cam.start.canon/vi/C004/manual/html/UG-03_Shooting-1_0290.html

New firmware update is also available.

Two good points, thank you! Exposure smoothing is available for R6 MK1. I never tried it, although  I doubt it is the explanation here. I know  that effective aperture gets darker with closer focusing, but it hardly can explain that difference.

Ah, and thanks for the headsup. Installing the new firmware now

Blue-Shift wrote:

charlyw64 wrote:

I suspect that the aperture is returned to f/2.8 after the first shot. Maybe do a trial an look into the lens what happens to the aperture. And then forget about the mess of taking stacks, they are not worth the hassle, especially if parts of the subject overlap within the stack, that will always lead to horrible artifacts that no software can remedy because they are down to the physical facts of focusing "through" foreground elements which will be depicted unsharp but larger as they are in the stack layer that is sharp...

Good call Charly,

I tried reducing the exposure by 2x in Lightroom and now the pics are identical, the formerly bright one with reduced DOF. So this is what is happening here.

Which is weird for 2 reasons.

1. It shouldn't happen in the first place

Indeed this is strange. In burst mode the aperture on the R7 stops down at the beginning of the burst and then stays put.

2. If it does, it should register as F2.8 in the EXIF data. But that says F5.6 still.

I think this is a protocol error that crept in (or was always present but has gone unnoticed)...

Well, I assume it will be removed when Canon releases the fixed firmware. We'll see.

I assume so too...

Also, about the stacking: I found it to be practical for some cases. Sometimes, it works quite well without too much of a hassle if one isn't too perfectionist. Also, it can be good just to have a range of shots to choose from, so I like the functionality.

The latter (focus bracketing) could well be an option for some circumstances. But I usually focus manually and focus a smidge beyond what should be the closest area in focus - because DOF extends in both directions from the plane of focus and so I don't waste any of the DOF for thin air between the subject and the camera.

Blue-Shift wrote:

Polcastrol wrote:

Do you have 'exposure smoothing option'? Like here:

https://cam.start.canon/vi/C004/manual/html/UG-03_Shooting-1_0290.html

New firmware update is also available.

Two good points, thank you! Exposure smoothing is available for R6 MK1. I never tried it, although I doubt it is the explanation here. I know that effective aperture gets darker with closer focusing, but it hardly can explain that difference.

Ah, and thanks for the headsup. Installing the new firmware now

I seem to recall someone posting a really nice looking in-camera focus stank here, although I can't recall who; but I'm pretty sure that it was on this R forum. Maybe do a search for it. If the details in it aren't specific enough, at least you might be able to ask the poster about his/her experience and settings.

John Photo wrote:

Blue-Shift wrote:

Polcastrol wrote:

Do you have 'exposure smoothing option'? Like here:

https://cam.start.canon/vi/C004/manual/html/UG-03_Shooting-1_0290.html

New firmware update is also available.

Two good points, thank you! Exposure smoothing is available for R6 MK1. I never tried it, although I doubt it is the explanation here. I know that effective aperture gets darker with closer focusing, but it hardly can explain that difference.

Ah, and thanks for the headsup. Installing the new firmware now

I seem to recall someone posting a really nice looking in-camera focus stank here, although I can't recall who; but I'm pretty sure that it was on this R forum. Maybe do a search for it. If the details in it aren't specific enough, at least you might be able to ask the poster about his/her experience and settings.

The camera can't perform wonders. If more distanced details are overlapped by elements closer physics will prevent capture of areas around the closer elements. Those areas will be obscured no matter which step size you set. So if you photograph flowers, insects or anything else that has overlapping elements within the stack you will have to spend hours carefully masking these element surrounding areas and invent (content aware fill with fitting subject areas) those. The only ever decent stacks I have ever seen were printed in a French nature photography magazine and the photographer said he expects to spend between 12 and 24 hours of straight editing on a single image to remove those un-capturable areas...

The camera will probably be better than Zerene or Helicon focus though - as it will have access to the phase difference because of the dual pixel design of current camera sensors - thus instead of having to calculate which layer contains which subject areas by means of frequent analysis it can use the phase difference, that is more precise and less error prone. But it can't remedy the above mentioned problems.

Other problems that I regularly encounter when presented with stacks is a too sharp drop of sharp captured area to the background (that can be an artistic choice so open for discussion if you like it or not) and a lack of depth indications within the stacked area. Often you can not determine which subject area belongs where - like which antenna is oriented towards the camera and which is pointing away from it, you will have to work hard with your lighting setup to remedy that deficiency...