Questions about Stackshot -- observations: Helicon / Zerene

[pittendrigh]

I have Helicon Remote, Helicon Focus and Zerene Stacker.

Helicon Remote is tethering software that provides setting closest and farthest focus points plus automated focusing increments in between. The software electronically twists the lens focus ring rather than moving the camera on a rail. The concept is convenient but the driver for my camera (Nikon D7000) has too many bugs for a good review.

Helicon Remote can automatically pipe a stack sequence to Helicon Focus, but Helicon Focus only knows how to deal with JPEG.

I have been using Helicon Remote on a Mac. Then copy the images (via scp on a home network) to a linux box where I use UFRAW to do a preliminary image edit on the first raw image. Then I jump out of the UFRAW GUI and use a terminal window command telling UFRAW to convert all RAW images in the sequence to JPEG, using the same edits manually performed once, on the first image in the sequence. My setup sometimes (often) leaves a partial 1 or 2 pixel black border around the edges of the RAW image which totally confuses any subsequent stacking software, so I also use an automated ImageMagick script to crop 4 pixels away from all JPEG edges before sending it to Zerene Stacker.

Then I import the cropped JPEG stack into Zerene.

So far so good. But I want an alternative to Helicon Remote. What about Stackshot by Cognisys? I've been sifting through posts mentioning Stackshot but still cannot figure it all out.

Stackshot moves the camera rather than twisting the focus ring. Is one method better for extreme macro? Helicon claims camera moving is NOT so good for extreme macro because it exaggerates perspective changes. Is that true?

Can I abandon the view finder and the camera's LCD screen and do everything from a tethered computer with Stackshot?

[TheLostVertex]

Rik wrote a page coving this topic here, http://zerenesystems.com/cms/stacker/do ... versusrail

What sort of magnification and optics you are using may influence the method of focusing you choose to use. For lower magnification the perspective change of using a rail can be pretty noticeable. For very high magnification the change can be imperceptible. If you give information about your subjects and optics people maybe able to give you more definite recommendations.

[pittendrigh]

Rik wrote a page coving this topic here, http://zerenesystems.com/cms/stacker/do ... versusrail

What sort of magnification and optics you are using may influence the method of focusing you choose to use. For lower magnification the perspective change of using a rail can be pretty noticeable. For very high magnification the change can be imperceptible. If you give information about your subjects and optics people maybe able to give you more definite recommendations.

Interesting. Thank you. I have a Nikon D7000. Someday a full frame replacement. I use a Nikor 105mm macro lens as is or with up to 6" inches of extension tubes. I also have a bellows, which--when combined with the tubes--magnifies a lot. I use a microscope lamp to illuminate the subject during focusing adjustments.

Typically use "soft box" lights with daylight compact flourescents. But I have umbrella strobes too. And the incandescent microscope lights.

So, sometimes I do minimal magnification (mostly of insects). Sometimes I do a lot. Trout flies too.

=============
The link above is helpful (about rail vs focus ring). What about the Cognisys Stackshot Rail? Can it be operated from a tethered computer screen? I suppose I should just call the salesman for that one. I was hoping for a less-biased review of Stackshot.

[mawyatt]

I use the Stackshot focus rail connected via USB to a MacBook Pro running Zerene stacking software. Zerene has the ability to directly control Stackshot. The Stackshot controller has an output to trigger your camera, you need a special cable which Congisys offers.

This works well for my use; Nikon 105mm Macro with extension tubes and Mitutoyo microscope objectives on a Nikon 70-200mm @200mm or reversed Raynox as 200mm "tube lens".

[rjlittlefield]

Hello! I'm the fellow who wrote the ring-versus-rail page. I'm also the author of Zerene Stacker, and a great fan of StackShot. I probably count as an extremely biased observer. But partly to counter that, I also tend to do a pretty thorough job of analyzing and documenting whatever I claim to understand.

Helicon claims camera moving is NOT so good for extreme macro because it exaggerates perspective changes. Is that true?

I'm sure hoping this statement lost something in summary or translation, because no, as stated here it's not true.

The big drawbacks to moving the camera apply at low magnification, not high. That's because it's with low magnification that you get large changes in perspective due to high ratios of distance_moved versus distance_to_subject. For what I would consider to be "extreme macro" applications, looking at subjects of say 5 mm field width and smaller, I don't think I have ever seen degradation that I would attribute to changes in perspective.

As an example of the image quality that you can get with StackShot and with relatively inexpensive optics (a 10X objective costing less than $100, operated at 5X on a 100 mm tube lens), see HERE.

You can also find a recipe for use of the StackShot with Zerene Stacker on a Nikon D5000 at http://zerenesystems.com/cms/stacker/do ... 2_software .

but the driver for my camera (Nikon D7000) has too many bugs for a good review.

I am sorry to hear this, except that it makes a good example of why I don't write software that attempts to talk with cameras directly. There are lots of opportunities for things to go wrong, and I am not psychologically suited to releasing software that is not reliable. Instead, I defer to other people to do camera control where that is appropriate.

In the case of camera control for Nikon cameras on Mac, you might consider taking a look at qDslrDashboard (http://dslrdashboard.info/). I have no experience with Mac + Nikon D7000, so for all I know it may be useless.. But I have had decent results with qDslrDashboard and D5000 on a couple of platforms, so I think it is worth a shot.

I hope this helps!

--Rik

[pittendrigh]

RE> "Helicon claims camera moving is NOT so good for extreme macro because it exaggerates perspective changes. Is that true?" .............

I might have the above wrong. Helicon Remote has great promise but many bugs. At least for my camera (Nikon D7000).

I'm a retired software developer, C-programmer, Java and database guy. So I have a pretty good idea about how difficult it is to make a GUI front end that is supposed to work for dozens of different cameras. I also see lots of complaints from user's of Nikon's own in-house remote camera control software. There are just too many if and or maybe software/hardware exceptions to catch.

The Helicon product support guys are great. But I am looking for an alternative. I like the idea of Zerene Stacker running Stackshot directly. I did pay for Zerene Stacker and I like it a lot. Thank you for making a great product.

Also, the ability to batch pre-process RAW files (with UFRAW) BEFORE sending them Zerene is critical for me. Unless........can Zerene handle RAW files directly? Oouch. I just assumed it couldn't but now that I think about it I never actually looked into that.

I'll start saving my pennies. I want a Stackshot controller and rail now. Thank you.

[rjlittlefield]

Unless........can Zerene handle RAW files directly? Oouch. I just assumed it couldn't but now that I think about it I never actually looked into that.

No, it cannot. Raw conversion is another one of those things that other people do really well so I defer to them. My standard recommendation (from the FAQs) is "For highest quality, we recommend converting raw files to 16-bit TIFF using your favorite raw converter and whatever settings make it work the best. After stacking the 16-bit TIFFs, tell Zerene Stacker to save its output also as 16-bit TIFF."

--Rik

[pittendrigh]

Just an Apr 6 2015 update on Helicon Remote

Earlier I reported Helicon Remote had too many bugs for a good review, at least for me when used with a Nikon D7000, running from a MacBook Pro laptop

I just downloaded a "new release" and much has been fixed. Still a few issues. But improvement is a good sign. They're working on it.

If this software ran without issues it would be very good.


=============
Also, earlier in this thread rjlittlefield recommended (to me) converting Nikon RAW files (I use UFRAW in batch mode, after manually setting up the first image in a sequence) to tif format instead of jpeg. That was a very good suggestion. I see better results already.

[henrikfoto]

I have been experimenting with stackshot a lot and I have had a lot of issues with the perspectice change from moving the hole camera. I leads to a softer image and errors in the image.

For me the camranger with autosteps within the camera is much better.

But the very best way to do it is to fix the camera-lens and just move the camera on the back of the setup with stackshot. This way the perspective
changes are cero and you will get a perfect picture every time.
This way you will not need any smoothing in the stacking-program and
the sharpness will be perfect.

[ChrisR]

That's an interesting observation, Henrik.

The magnification still changes if you change the sensor to lens distance (assuming a "simple" lens), so there's still an alignment issue, and/or introduction of a false change in perspective in the stacked image.

I mean here that the background details should look smaller than foreground. If you turn alignments off, that will be influenced by the magnification change when focus is changed.

The entrance pupil of the lens is moving relative to the plane of focus within the subject.

If the sensor-lens-subject is in perfect alignment I don't think there's a problem.
If the camera (sensor + lens) is twisted as it's moved towards the subject then yes, perspective does change laterally not just longitudinally. I've often done that, and found it annoying, but not nailed a sharpness problem due to that.


Could you show one of the examples of where you have had a problem with perspective change?

[rjlittlefield]

At the risk of sounding redundant, the issue that Henrik mentions is exactly the one discussed at http://zerenesystems.com/cms/stacker/do ... versusrail, which has been previously linked in this discussion.

That page is titled "Is it better to use a focus rail or the ring on my lens?" The opening sentence says "The short answer is that it’s usually better to use the ring." A deliberately extreme example is provided.

I have not seen Henrik's examples, but I have no doubt they exist.

In fact I wrote the web page mentioned above because at some point I started seeing a disturbing number of examples come in as Zerene Stacker support requests. Paraphrasing one of those from memory:

Please see the attached photograph. It shows a necklace draped on a modeling torso. I used a long-axis StackShot rail to move in fine steps from front to back, and I can see in the individual shots that every part of the necklace has been captured sharply in some source frame. But in the Zerene Stacker output, the necklace is not sharp. What's going wrong?

The answer, of course, is that in this situation it's a mistake to change focus by moving the lens+camera. Doing it that way means that the entrance pupil moves a lot. Using the focus ring is better because then the entrance pupil moves much less, potentially not at all. Or as Henrik says, one can keep the lens in one place and adjust focus by changing the rear extension. That guarantees that the entrance pupil does not move.

Unfortunately this whole question of "what's the best way to focus" is a matter of tradeoffs. Larger subjects will usually give better results from focusing "by ring". But there's a large range of situations -- all of them involving small subjects -- for which focusing "by rail", moving the camera and lens together, is simple and is equally or even more effective. That's the proper regime for StackShot, microscope focus blocks, and so on. Focusing by adjusting rear extension is excellent in most cases if you have the optics and mechanics to do it, but even that method runs into problems with deep stacks at high magnification, where rails and focus blocks are great.

--Rik