Helicon Soft Forum

I think I can see *what* is happening in the stack, but I'd appreciate some suggestions on how best to work around it.

In the stack (that's uploaded to your ftp server in a folder with this username) it appears that the foreground section of the ring in particular is being used when out of focus. This is using the lens for shifting focal plane, not a rail; as such, the size changes quite a bit between the start & end of the stack.

Oh, and the program reports a difference of around 15% lighting, but given this is using continuous lights and everything set to manual, I think that's another manifestation of the focus breathing.

It looks like the only way to cope with this problem is to reduce aperture from current f5 to something like f8 or f11. Blurred foreground masks background details and the program is not able to get missing information.
You can use retouching brush to clone from source images but even this will not solve the problem completely.

Stas Yatsenko wrote:It looks like the only way to cope with this problem is to reduce aperture from current f5 to something like f8 or f11.

Right, I shall give that a try, thanks. Sweet spot on this lens is supposed to be 5.6 according to some sites (I read it on the internet so it must be true!). Nikkor 105mm FYI.

Blurred foreground masks background details and the program is not able to get missing information.
You can use retouching brush to clone from source images but even this will not solve the problem completely.

Yeah - there's nothing there to clone *from* (same as the algorithm doesn't have anything to work with) so the only way I could think of doing that was something like Photoshop's smart brush. AKA "making it up".

Quite like the Android app for building the stacks; how feasible would it be to be able to set a destination path on a network drive for the images rather than on the local device?

That's made a fair difference, although there's still some evidence of the problem. Shouldn't be too bad to take it out manually though.

Do you want to use Android application and save to network drive? Windows and Mac versions can handle network paths.

I have been playing around for a couple of nights between working and have found a similar affect. This is a stack of 50 shot over a 2cm distance using some close up filter on an 18-55 lense, f5, calculated DOF of 0.48mm.

I used a home built macro rail (built from an old DVD RW) to move the bee. I will try at a larger f-stop (maybe 8?). However I was thinking that if i was to set the macro rail to a smaller step, 0.25 for argments sake, would this go any way to resolving this sort of "stuttering" around the edges

I would recommend to explore source images more thoroughly. If source images have enough information to reconstruct perfect image, then it is a question of processing parameters and maybe using clone brush from source images.

If you see that some details are not available, because foreground object is out of focus and masks them, then I would recommend to cope with the program by setting smaller aperture.

I have been doing a little more experimenting and have found that, areas that either reflect more light or dont reflect as much light as other parts of the image seem to cause some of this. Taking shots in a darker area with a makeshift lense reflector (I think thats what they are called, the shinny thing monuted to lenses, mine was a bit if white card with whole cut in the middle for the lense to stick through), gave better results

Thanks for the tip!