Finally getting closer - first test shots

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Planapo
Posts: 1581
Joined: Tue Nov 07, 2006 2:33 am
Location: Germany, in the United States of Europe

Finally getting closer - first test shots

Post by Planapo »

I´ve finally found time to mount and screw everything together for my stacking set-up, and for the first time have seriously used the 400D since it had arrived in January.

Although I am not able to run a stack at the moment as I first have to have more RAM installed for computing power, I have started some test shots to see whether the lighting and stability of the set-up will be good enough for stacking and how the lenses perform. But I find that with my set-up I have stable, straight and smooth control of motion of the focus in well controllable and reproducible intervalls along the whole depth of the specimen.

So here is one of the test shots with the 2.8/50 EL-Nikkor. Length of extension from rear end of lens to flange at about 240 mm. Not cropped. Recorded as RAW and tranformed to jpg with the Canon software. Then downsized and improved with editing software. Aperture: Well, I must have subconsciously thought of the convenient Exif.., but of course, with bellows and reversed lens...#-o probably it was about half stopped down.
Shutter speed was 4 seconds. Mirror lock-up.

Tungsten-halogen lighting from a cold light source via two fibre light guides (Lighting was not as I had planned since the transformer for my other cold light source gave up the ghost before I did even start shooting and has to be replaced).

Light diffused with hemispherical plastic diffuser. Funny side note: When strolling through the Ikea store in early January I came across the left-overs of christmas lightings and tree decorations. Catching a glimpse of white, transparent plastic balls I immediately thought of Charlie on the other side of the globe and loaded my trolley... :D


Image


Now my questions concerning this test shot: What do you think could be improve? Should I control for flare with a DIYed lens shade or an inner aperture in the light path?

To me the shot looks rather sharp where the focus lies. e. g. on the mandibles. So vibration doesn´t seem to be a problem?
And although the extension with 240 mm is quite long, the lens performs still quite satisfyingly, or what would you say?
Am I ready for stacking?


(Some info for those interested in the ant photographed. It´s a worker of Formica polyctena, our most abundant red wood ant. As it´s just a test shot I haven´t showered, blow-dried and combed her properly :wink:, hence some dust flakes and fibres.
But the dark 'pimples' that can be seen on the antennae, the right eye and between the latter and the ocelli and on the pronotum (OOF above the head) are not dirt specks but a parasitic fungus that grows on the ant. It has already grown on her while she was still alive, thus it´s not a kind of mold due to poor storage conditions.)

--Betty

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Betty,

Yes, you are ready to stack. :smt023

240 mm extension with a 50 mm lens gives about 4-5X magnification. You should shoot a mm scale to get precise, but 4-5X is good enough for thinking.

If you have things screwed down well, then vibration is almost certainly not a problem with 4 seconds exposure and mirror lock-up. With my setup, 1-2 seconds exposure with 1.5 seconds lockup delay works fine even at 20X.

Your image looks sharp rescaled to 800 pixels, but it could easily be fuzzy at actual pixels. Be sure to run a test sequence to find optimum aperture. It's probably around f/5.6 marked, maybe even wider.

Illumination looks excellent for scientific purposes -- very even and diffused. Be sure to shoot a photographic gray card for color balance. The posted image looks a bit warm to me.

I have no experience with that lens to know whether a lens shade would help. With some of my lenses, adding a shade makes a huge improvement, but with others it does nothing. It's not much trouble to cut one from a piece of black paper, so best to run the experiment.

Very interesting story about the fungus -- thanks for that detail!

--Rik

DaveW
Posts: 1702
Joined: Fri Aug 04, 2006 4:29 am
Location: Nottingham, UK

Post by DaveW »

Rik just said what I was going to say.

If your background was white it is now a creamy colour, so if you adjust the colour balance in either camera or post processing to get the background white your insect colours should come right. You could use a gray card, but I think that would set exposure rather than balance the lighting. Assessing the correct tone of gray would be harder for me than getting something looking white since it is easier to judge the exact colour cast on white rather than gray. I think it will be easier simply to get the background white in camera or post processing if that was it's actual colour?

DaveW

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

I often use a white index card (note card) as a quick-and-dirty substitute for an official photo gray card. But that's only after quite a bit of testing to know that it works OK for the brand of index cards that I buy.

The issue is that "white" papers are surprisingly variable in what they actually reflect. Gray/white cards that are manufactured for photographic purposes have very flat spectra, so that imaging the card does actually give you an integrated sample of the lighting, and nothing else. If you do the same exercise with a half-dozen different types of "white" paper, you're likely to end up with a half-dozen different color calibrations.

The variability is not so bad with incandescent lighting, but it can get pretty serious with daylight or fluorescent bulbs where those nice bright blue dyes added to the paper have a chance to fluoresce too.

--Rik

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

OK Betty let's get stacking!

If you can "batch" resize the images for a stack (and save them under a different name!) you should be able to do a few even if your computer is not up to working with large files. HF seems pretty good in that respect. 800-1000 pixels in the long dimension should be fine.

This looks good. Don't worry about the amount of extension. This 50mm enlarging lens is likely "optimally" designed to make an 8x10 inch print from a 35mm negative. So about an 8-10X enlargement. I suspect you could add a lot of bellows and still get a great image. (... but it becomes physically impractical to do so in most cases). The aperture you use is important however. You can easily do some tests. As a rough guide I would suggest the following as the smallest aperture to use (highest marked f-number) at certain magnifications.

f8: 2X
f5.6: 3-4X
f4: 5-7X
f2.8: 9-10X

If the lens is really good your results can always look a little sharper at slightly larger apertures than suggested above. But if you close the lens down more than indicated you will most definitely start to notice a significant reduction in resolution, regardless of the lens quality.

Your message made me laugh a couple of times. Occasionally I will still absent-mindedly think about checking the EXIF data to see what I used for a particular bellows or microscope shot! :smt017 #-o

And I saw the same Christmas lights at Ikea before Christmas and nearly bought a string of them. (If it had been after Christmas and they were on sale I probably would have). I'd have had enough little diffusion domes to last the rest of my life.

Planapo
Posts: 1581
Joined: Tue Nov 07, 2006 2:33 am
Location: Germany, in the United States of Europe

Post by Planapo »

Rik, Dave, Charles, thanks guys for your helpful answers!

Thanks for the aperture values and for your feedback, Charlie. I am happy to hear that I´ve made you laugh several times, and I wasn´t aware that Ikea has spread thus far!:D

I have a question concerning my lighting. I use two cold light sources of the same model type with 150 W halogen light bulbs. Both can be regulated electronically, but the newer version has a built-in diaphragm which allows for regulating the light intensity after the light is set electronically to a certain light temperature [K]. The light temperature is shown on a dispay.
But the older one can solely be regulated electronically, and I don´t see the light temperature as it has not the display like the newer version. I only know that it´s 3200 K when set on max. output. Now, when using both of them, can I mix light of different light temperatures? Can this be compensated for, when later editing the RAW image? Or should I rather avoid such a kind of light mix for documentary shooting?

--Betty

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Planapo wrote:Now, when using both of them, can I mix light of different light temperatures? Can this be compensated for, when later editing the RAW image? Or should I rather avoid such a kind of light mix for documentary shooting?
For documentary purposes, avoid this like the plague!

If the two sources were uniformly mixed, then you could fix most of the issues by adjusting color balance, either in the camera with custom white balance or in post-processing.

But that won't happen.

Instead, different parts of your subject will be lit by different mixes, and even in theory there's no way to tease apart differences in the lighting from differences in subject coloration.

This topic is a perennial favorite among photographers of living spaces, who encounter unavoidable mixed lighting with say daylight, incandescent, and fluorescents all in the same scene. There are several standard tricks for dealing with it, but they are all troublesome, subjective, and amount to "here's how to make the result look less bad".

--Rik

augusthouse
Posts: 1195
Joined: Sat Sep 16, 2006 1:39 am
Location: New South Wales Australia

Post by augusthouse »

Betty,
Are you using single light guides on each of your units?
Using a dual (or triple) light guide on 1 unit might be worth considering.


Rik wrote:
This topic is a perennial favorite among photographers of living spaces, who encounter unavoidable mixed lighting with say daylight, incandescent, and fluorescents all in the same scene. There are several standard tricks for dealing with it, but they are all troublesome, subjective, and amount to "here's how to make the result look less bad".
Rik, what about something like ExpoDisc. Would this be helpful in such situations?

Craig
To use a classic quote from 'Antz' - "I almost know exactly what I'm doing!"

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Rik, what about something like ExpoDisc. Would this be helpful in such situations?
No. Either the ExpoDisc or a standard gray card will give accurate calibration of averaged illumination color at a single point. The problem is that other points in the scene have other illumination colors. You can spread a multitude of gray cards around the scene and accurately determine the illumination colors at all of those points too, but then what? You still have a multitude of different colors to deal with.

Even worse, the whole concept of "averaged illumination color" is flawed when working with 3D subjects. Suppose you illuminate with a yellowish light from one side and a bluish light from the other side. On average, the light is neutral, and either the ExpoDisc or a gray card will tell you that. But nonetheless, if you stick a uniform gray ball in the scene, it will appear yellowish on one side and bluish on the other because the shape of the ball makes it reflect more of the light from one side than the other. In this situation, there's no way to make the ball appear gray without knowing what it is supposed to be, and fudging the image to match -- with a different fudge value at every point!

Sorry, but for 3D subjects the only practical way to get accurate color at all points is to use the same color for all illumination.
Using a dual (or triple) light guide on 1 unit might be worth considering.
Yes, this works great -- cheap and effective.

--Rik

Planapo
Posts: 1581
Joined: Tue Nov 07, 2006 2:33 am
Location: Germany, in the United States of Europe

Post by Planapo »

Rik and Craig, thanks for your input.

Yes, one of my light sources is equipped with a dual light guide. But I wanted to use the other for some nifty backlighting.

Well, I could set them both on 3200 K and then regulate the intensity of the newer one by its diaphragm whereas 'muffling' the intensity of the older one with neutral gray filters or some material like tissue layers etc., couldn´t I ?

--Betty

lauriek
Posts: 2402
Joined: Sun Nov 25, 2007 6:57 am
Location: South East UK
Contact:

Post by lauriek »

Betty, theoretically that should work, but you need to watch that your material used for dimming is a neutral colour. As Rik pointed out earlier in this thread, most things which appear pure white are actually not pure!

Neutral density filters would work well as these should obviously be made specifically to not introduce colour casts.

I would do some experimentation though with a simple test subject, and a simple lighting setup, one light unit on left, one on right, and then in raw processing, select a point on either side of your subject (ie two points, each mainly lit by each of your light rigs) to select as a basis for white point, and see how swapping between the two points affects the colour of the picture...

It occurs to me that if we had a grey sphere made of grey card material, that theoretically a computer could maybe use a shot of this to correct mixed lighting, if it could build a 3d model of the subject...

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

Rik and Craig... some of the most complicated shots I've ever worked on were large format (4x5") interiors of commercial buildings. Not only was there a huge range of light intensities, but often different areas in the same picture illuminated by very different light sources.... high wattage incandescent bulbs for one section, low wattage incandescent in another, fluorescent for another (or perhaps in a product display case)... and maybe even some halide type somewhere to make it still more fun. These shots were a real challenge. Sometimes the best results were obtained by getting access to all the lighting switches and circuit controls. Then, using multiple exposures on the same piece of film, each source would be turned on by itself, and separate exposures were made with a "gel pack" of filters to correct color for that source. While doing this, different exposure times would be used to balance the intensities. Lot's of initial Polaroid testing!

Betty... use the "dual-pipe" one for the overall lighting and the other for your backlighting. The color temperature for the backlighting is less critical, although I would run it at 3200k. It's easy to physically move a back light closer or farther to vary it's intensity on the subject.

For stacked shots it's best to avoid "AWB" (auto white balance) and set it for "tungsten"... or better yet do a few test shots and determine a manually set color temperature setting with perhaps a little additional color bias correction as well if needed (if your diffusion material imparts a little of it's own color). This will give jpg's that are usable right away (raw files would be unaffected).

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Charles Krebs wrote:Sometimes the best results were obtained by getting access to all the lighting switches and circuit controls. Then, using multiple exposures on the same piece of film, each source would be turned on by itself, and separate exposures were made with a "gel pack" of filters to correct color for that source.
Wow -- that is definitely the most disciplined approach I've ever heard for handling that problem! Also the most elegant -- since from the standpoint of the film, each light source got independently balanced to be "neutral", or more likely, to be whatever you needed to give the right effect. I imagine it would look pretty unnatural if tungsten-lit display cases got balanced to be just the same as daylight coming through the windows.

Betty, the trick that I usually use for white-balancing behind pingpong balls & similar setups, is to get everything set the way I want, including illumination levels to give appropriate exposure times, then at the last moment set white balance by inserting a strip of index card just in front of the subject. There are some setups where that doesn't work because I physically can't reach in front of the subject. In those cases, if I care a lot about color balance, I'll do a rough balance on the outside of the diffuser, shoot the stacks, then remove the subject, insert a white card, create an adjustment layer as needed to finish balancing the card, and finally add that same layer to the other stacks. It's not perfect, but it's a lot better than what I can get by just trying to reproduce the dimmer setting on my illuminator.

--Rik

Post Reply Previous topicNext topic