Combine z vs Helicon Focus

I am curious about the relative performance and ease of use of the "combine z" and "Helicon Focus" image processing software products. (These software packages improve a photo's depth of field by processing multiple images.) I have been using the Helicon Focus software for about a year with satisfactory results. I would like to hear from micro digital photographers who have tried both software packages.

Hello Tom,

I am using Helicon sisncea few years, in a version which also allows "retouching", this means to correct artefacts manually. I have tested Combine Z only a little. What I can see is that Helicon seems much faster and als is not limited to JPEG formats, it als supports raw formats.

On the other hand Combine Z is for free and ok for most cases.


Volker

Hi,

Helicon is somewhat easier to use but as far as I'm concerned the two tools are equally powerfull although some say the algorythms in Helicon are more powerfull.
I ran several comparative tests using the same frames with both programms, there was hardly any noticeable difference between the two results.
I allways use CombineZ by the way...

Cheers

Harjo

There is a Combine Z forum @ [forum.pbase.com]

Bill G

the "official" CZ forum is here:
[tech.groups.yahoo.com]

here:
tech.groups.yahoo.com/group/combinez/

I've decided to go back to using CombineZ5 instead of the latest versions of CombineZM. The stacking algorithm of Z5 yields much better depth of field than ZM, especially deeper into the stack. Has anyone else noticed a change in stacking results with CombineZM?

Thanks!

Steve Stuart

Steve, I tried the latest (December) version of ZM and it gave me much larger halo artifacts than the previous version I had (I think of July '07). I tried ironing out the kinks with Hadley but despite his help, couldn't make it work properly. I think the July '07 version of ZM is superior to Z5, if only because it's easier to use (does stacks in both directions, for example). I have not noticed the problem regarding depth that you mention.

Tony

Helicon Focus has a new version out. It has a mode A and mode B option for stacking. I have run a couple of stacks using the mode B and found that it does a better job on the out of focus background and has less of a halo problem. It make two passes through the stack.

Doug

Alan Hadley has posted a new version of CombineZM that fixed a typo in his code. It works slightly better than the late December version, but not as well as CombineZ5!

How can I get a copy of the July 07 Combine ZM. I had it, but replaced it long ago with the newer versions.

Thanks!

steve, send an email to tpeterso@sympatico.ca and I can probably mail you the version I have.

tony

Hello,

i updated to Helicon 4.4 and tested the new B stacking method. This gives much better results for halos.

Volker

This isn't exactly about CombineZ vs Helicon but is appropriate to the thread......in addition to the halo problem - which can't be avoided when a close object sits in front of a distant background - I sometimes have a problem with CombineZM not producing a well-focussed image of the terminations of objects that project outward, particularly if they are low-contrast. It's especially acute for deep images with many pictures in the stack. I got frustrated with poor results from a gemmy weloganite micro, taken in 50 images, so I tried re-stacking this into 5 sets of 10 each, then combining the 5 resulting images (all tiffs to avoid jpeg "round-off" error). Voila! Much better results! A pair of images for comparison have been attached (FOV is about 1 cm). The only penalty, aside from the extra work involved, seems to be perhaps excessive sharpness in finely detailed, high-contrast areas.

I wonder if the best algorithm for these programs wouldn't be, instead of sequentially stacking the images, to stack adjacent pairs, then pairs of pairs, etc. until only 1 image is left. Alan, are you there?

Tony

FROM SPAIN. Could somebody send me a manual of combine z?

@Tony,
That's what I do too when the sequence involves many frames, stack f.i. 4 groups of 8 frames and then stack the 4 resulting frames instead of stacking the whole 32 frames in one sequence.
The reason why this works better is probably due to the fact that the program discriminates between the sharp, sharper or sharpest areas, obviously there's less to discriminate between 8 frames worth of sharpness compared to 32..

@Steve,
I also still use the Z5 version, I compared the results between Z5 and ZM but will stick to Z5 for the moment (I compared the standard stacking sequence as well as customised stacking sequences)

Cheers

Harjo

Harjo et al - Alan was kind enough to provide me with a macro that stacks 2 at a time, then 2 of the 2s, etc. till it's down to one. It worked fine but even with the high filter set at (1000,949) things were still oversharpened...I might try it again at (1000,980) or something. mManwhile, he also provided me with an experimental macro called pstack (for pyramid), it's quite different. It seems to split frames into rgb components) and works very well in reasonable time, also works in batch mode, my only complaint is things may not be quite sharp enough, or the contrast is a little low - can't say for sure because my test images are rather uniformly of pale minerals, anyway. Suggest you ask him for it but you'll have to use the latest version of CZM, it requires additional code.

keep stacking,

Tony

Hi Tony,

I did a stack like that (2 by 2 then 2 by 2 etc) the other day.
The result can be quite good but some problems arouse with an object lacking high contrast.
In that case this method seems to exaggerate the problems encountered when stacking tools have to deal with very low contrast areas....

Cheers

Harjo (keeping up stacking :-) )