Butterfly wing detail

[Beatsy]

Working on improving my lighting technique now I've got the macro rig. Quite pleased with tonights effort. It's a small section of a butterfly wing plus a close crop from the same image to show the detail in the scales.

I included an image from earlier experimenting too. The left side used standard diffused light, the right side also used diffused but made more directional with a smaller diffuser and bits of card shading out some of the incident and bounce light. Both sides of the image are from the exact same spot on the wing. I was really surprised at the huge changes to be had from the slightest little adjustments. Having a lot of fun playing with this...

All shots taken using a 50x mitty on 135mm tube lens for 33.75x onto the sensor.

[mikeatnip]

Looks good to me. At least I'll be happy if I ever achieve something equal!

[rjlittlefield]

Looking good!

I was really surprised at the huge changes to be had from the slightest little adjustments. Having a lot of fun playing with this...

As I think about the problem, lighting these small subjects should be just the same as lighting large subjects that have the same shape --- except for three additional wrinkles.

The first wrinkle is that with small subjects our light sources have to be quite a lot wider angle to avoid an assortment of strange artifacts.

The second wrinkle is that small subjects have a bigger "exclusion zone" of angles where we can't put lights because they get blocked by the camera or lens. As a result we're often using more oblique lighting than we would probably choose with a larger subject. I think that turns into higher sensitivity to light position.

The third wrinkle is that even if we're using continuous illumination with live view, we can't tell what the scene is going to look like because we can only see one little slice of it!

It would be really cool if we could see stacked results in real time, even in low res preview quality. I'm pretty confident we'll get there, but I have no idea how long it will take.

--Rik

[Beatsy]

Wrinkle 4: hard-edged shadows and absence of (unwanted) specular highlights are mutually exclusive.

As a related aside - have you noticed that shrinking the size of the diffused light source has a similar visual effect to closing down the condenser aperture on a microscope - especially when you get close to a point source?

Conversely, increasing the size of the source (diffusing more) reduces the contrast of small, low-contrast features even more - also similar to the effect of opening a condenser aperture wider.

I don't *think* resolution is actually affected by light source size/diffusion in this case, but it sure looks like it is...

[fotoopa]

Very interesting test. Your result is clearly positive.

It has also kept me busy for several days. Hence the realization of my controllable LED flash in power and light direction. With 10 LEDs I have many opportunities. The direction I have divided in 40 parts under software control. The LEDs always remain in the same physical position. The individual LED power (time) is defined by a profile (FPGA). The profile determined, the selectivity of your lighting.

This corresponds to your test. The great advantage of my setup is reliability. The LEDs are always in the same initial state relative to the object. The preset values can be used for other images. My experience is awesome. After some experimenting, I have a default profile that works great. I just have to choose a direction out of the 40 positions via a potentiometer. This is already done in preview mode. Then, I make a few test shots for the final adjustment. Then the stacking of the images begins.

Thanks for your nice demo.

Frans.

[Beatsy]

Thanks Frans. Yes - I've been reading the posts about your flash setup with some envy. I want one! It must certainly help with consistent results. Very "clinical". Having said that - I'm content to keep tinkering with my separate lights, diffusers and bits of card too - simply for the novel "happy accidents" that emerge from time to time

[Olympusman]

Very dramatic lighting. I think you've got something going there.

Mike

[rjlittlefield]

Wrinke 4: hard-edged shadows and absence of (unwanted) specular highlights are mutually exclusive

As a related aside - have you noticed that shrinking the size of the diffused light source has a similar visual effect to closing down the condenser aperture on a microscope - especially when you get close to a point source?

Conversely, increasing the size of the source (diffusing more) reduces the contrast of small, low-contrast features even more - also similar to the effect of opening a condenser aperture wider.

I don't *think* resolution is actually affected by light source size/diffusion in this case, but it sure looks like it is...

I'm thinking that it actually is. There is a concept of "utilized aperture" (the term coined by Chris S.), in which properties of the light source and the subject affect how much of the lens aperture is actually used in the process of imaging a point on the subject. For many subjects, making the light source smaller will also reduce the utilized aperture, which should have all the same effects as closing down a condenser aperture.

There is some discussion of this effect in False color and detail/DOF from shiny metal, though offhand I'm not sure which of the 7 pages of that thread will be most relevant.

--Rik

[ChrisR]

Balance of several things...?

increasing the size of the source (diffusing more) reduces the contrast of small, low-contrast features even more - also similar to the effect of opening a condenser aperture wider.

Increasing the diffusion reduces apparent sharpness available from lighting relief - which is often seen as contrast, but only if there's a physical structure.
Reducing diffusion (starts to ) mean the utilised aperture reduction may matter, rather like the combined NA of objective & condenser, which can reduce resolution (sharpness) - and therefore contrast,
but you also get less out-of-focus haze which can help with contrast.

Some of the effects, such as where a highlight will appear to throw off a visible flare of light like a comet's tail, I don't think we have a handle on.

[Beatsy]

Thanks for the comments. The "shiny metal" thread was interesting reading and covered a lot of the effects I've been seeing.

Regarding the comet-tail highlights, they are a right royal pain when stacking diatom images, particularly in DF or DIC. I recently discovered that turning off *all* alignment options in Zerene significantly reduces their impact in the final image, but it's still best to adjust the lighting to avoid them in the inputs where possible.

[ChrisR]

They are odd. If you use oblique top lighting, it's like the light bouncing off the subject is illuminating dust in the atmosphere. You can focus on it perfectly easily.

[rjlittlefield]

They are odd. If you use oblique top lighting, it's like the light bouncing off the subject is illuminating dust in the atmosphere. You can focus on it perfectly easily.

I would be interested to see an animation of that effect. Based on just the words, I'm wondering if you're seeing an interference pattern that got formed in the space just above the surface, as a result of operating in that "partially coherent" regime.

--Rik

[GaryB]

I love the lighting. Gives a feeling of real depth and texture missing from most shots of this type. You get a good sense of how the wings and scales really sit with the shadows.

Very well done.

[abpho]

Incredible work Beatsy. The lighting is superb.

What size of sensor does your camera have?

[BugEZ]

Excellent images and interesting discussion. Thanks for sharing!

Keith