Proof of the pudding...or.....new diffuser works

[sonyalpha]

I took all of these shots using my new diffuser with one sheet of opaque plastic over the open end:

I did need to lighten each shot a little....then I used the unsharp mask for posting here:

The hairy legged fly is a new one to me....it looked quite aggressive...as if it would bite???......any ideas on ID please?

The Greenfly Aphid shot is cropped to get rid of unwanted OOF foliage...........I like the way it is looking down:

comments and constructive criticism welcome

sonyalpha

[Eric F]

Lovely shots SA; your new diffuser is working great! I especially like #2. The fly is a dung fly, Scathophaga stercoraria, and it is a predator.

Eric

[Harold Gough]

Very pleasing!

Harold

[Peter M. Macdonald]

The new diffuser seems to be working well. I like the fly shots. I always think that these are some of our most attractive flies - the Genus is Scathophaga and the species looks like stercoraria. I was watching a pair of these maiting and then egg laying on one of the byproducts of dog ownership only a couple of hours ago!

As you will see from the generic name, they eat dung, at least in the immature stages.

Regards,

Peter

[sonyalpha]

Thank you for the kind comments folks, here is a link to further information on the subject:

http://www.haworth-village.org.uk/natur ... sp?pic=151

Another excellent insect ID site:

http://www.bugsandweeds.co.uk/flies%20p3.html



Also.....I am wondering if that is one of its victims behind it in shot #3? which is why it was sitting contentedly on top of a bush in a fair-cool breeze:

That reminds me............conditions for hand held focussing are made even more difficult for me at this particular shrub because it is circa 6ft (2m) from a busy main road.......every time a truck passes the bush is blown wildly back and forth:

sonyalpha

[gmazza]

Like the second photo, the profile include interesting detail. Not sure about the diffuser effeiciency, would prefer see more shoots, different subjects.

[LordV]

LOvely captures SA.
Re the diffuser, assuming the flash was the major light source, you should be able to avoid the hexagonal type reflection off the fly's eye with the right amount of diffusion.
Exposure control and diffusion should be separate issues unless you are really blocking most of the light from the flashgun.
Brian v.

[sonyalpha]

LOvely captures SA.
Re the diffuser, assuming the flash was the major light source, you should be able to avoid the hexagonal type reflection off the fly's eye with the right amount of diffusion.
Exposure control and diffusion should be separate issues unless you are really blocking most of the light from the flashgun.
Brian v.

Thank you Brian,

The shooting conditions were fast moving clouds, a stiff breeze plus the frequent passing of heavy trucks blowing the bush about wildly ....the flash (inside the cereal container diffuser) was diffused further with a sheet of opaque folder plastic:

The aim being (as you know) to soften the harshness of highlights caused by the flash.................are you suggesting that the hexagonal eye highlights are still too harsh?

I thought they were about just right in emphasising the shape of the eye...........is it the flash that has caused the hexagonal shape then?

I will experiment further with a kitchen-towel diffuser as suggested by you earlier:

Still learning.....onwards and upwards:

sonyalpha

[LordV]

LOvely captures SA.
Re the diffuser, assuming the flash was the major light source, you should be able to avoid the hexagonal type reflection off the fly's eye with the right amount of diffusion.
Exposure control and diffusion should be separate issues unless you are really blocking most of the light from the flashgun.
Brian v.

Thank you Brian,

The shooting conditions were fast moving clouds, a stiff breeze plus the frequent passing of heavy trucks blowing the bush about wildly ....the flash (inside the cereal container diffuser) was diffused further with a sheet of opaque folder plastic:

The aim being (as you know) to soften the harshness of highlights caused by the flash.................are you suggesting that the hexagonal eye highlights are still too harsh?

I thought they were about just right in emphasising the shape of the eye...........is it the flash that has caused the hexagonal shape then?

I will experiment further with a kitchen-tower diffuser as suggested by you earlier:

Still learning.....onwards and upwards:

sonyalpha

SA - of course it's a matter of personal taste but yes that shape in the eye is caused by the flash (unless there was bright sun present also). I try to avoid these shapes in the eye, prefering a smooth light gradient which more diffusion should give you. The bright reflection is also influenced by how big the diffuser is relative to the fly (ie size and distance away).
I've always compared flash shots to natural light shots taken under bright direct sun and taken under complete cloud cover. Bright sun gives high contrast and hot spots. Complete cloud cover gives low contrast normally with no hot spots. With flash diffusion you are trying to get somewhere between these two extremes. The more diffusion the more you move towards the cloudy day lighting. It is a matter of personal taste and how much light loss you can afford. It's also dependant on the magnification you are using. The higher the mag the more diffused the flash tend to be as the ratio of apparent diffuser size to subject increases with magnification

Example shot i took yesterday below.

Brian v.

[rjlittlefield]

SA - of course it's a matter of personal taste but yes that shape in the eye is caused by the flash (unless there was bright sun present also). I try to avoid these shapes in the eye, prefering a smooth light gradient which more diffusion should give you. The bright reflection is also influenced by how big the diffuser is relative to the fly (ie size and distance away).

To expand on this point...

What makes a hard edge on direct reflections is having a hard edge on the light in the first place. To get the softest highlight, the subject must see a soft edge in the light coming from the diffuser.

This implies that the diffuser should not be evenly bright across its whole surface, but instead should be dark on its edges, with smooth gradients between the edges and the more central bright areas. If the diffuser is evenly bright across its whole surface, then there is a hard edge where the diffuser stops.

Bryan, I can't remember -- have you ever shown us a picture of your diffuser alone, say imaged in a chrome mirror ball and exposed so that the diffuser appears in shades of gray?

--Rik

[LordV]

SA - of course it's a matter of personal taste but yes that shape in the eye is caused by the flash (unless there was bright sun present also). I try to avoid these shapes in the eye, prefering a smooth light gradient which more diffusion should give you. The bright reflection is also influenced by how big the diffuser is relative to the fly (ie size and distance away).

To expand on this point...

What makes a hard edge on direct reflections is having a hard edge on the light in the first place. To get the softest highlight, the subject must see a soft edge in the light coming from the diffuser.

This implies that the diffuser should not be evenly bright across its whole surface, but instead should be dark on its edges, with smooth gradients between the edges and the more central bright areas. If the diffuser is evenly bright across its whole surface, then there is a hard edge where the diffuser stops.

Bryan, I can't remember -- have you ever shown us a picture of your diffuser alone, say imaged in a chrome mirror ball and exposed so that the diffuser appears in shades of gray?

--Rik

Rik - Not sure i agree with the explanation. I may well be wrong but my assumption was that diffusers work by acting as secondary light scatterers, so the light beam will naturally be softer at it's edges and this is not necessarily the same thing as how evenly lit the diffuser appears to be when looking at it ?

As an example in the shot set below which I hope is fairly self explanatory, bright spot spreads as more and more diffusion is applied and becomes less blown in the center. I would have expected the light eveneness across the diffusion material would have been the same in all cases as all I'm doing is changing the diffusion material on the face of the diffuser ?

[EDIT] Just did a reflection underexposed picture of the diffuser in action (excuse the dirt) - because of the light spreader film I use on the head, it's fairly even vertically but not horizontally.

Brian v.

[Craig Gerard]

Here is an image I took that shows the result of two light sources, one hard (sunlight) and one soft, but not soft enough (diffused flash), both striking the face of a dragonfly (this was hand-held, the motion blur was introduced by me - I think?).

The flash diffuser in this case was one of those plastic, over priced, commercial, rectangular speedlight softbox products.

http://www.photomacrography.net/forum/v ... php?t=8806

Craig

[rjlittlefield]

SonyAlpha, I apologize for diverting your thread into a technical discussion. I thought about splitting the thread, but couldn't see how to do that cleanly. No doubt we'll talk about this stuff later in the technical forums.

Brian, thanks for the constructive skepticism. I don't have time to set up a good experiment/demo, so let me see if I can do something useful with words.

my assumption was that diffusers work by acting as secondary light scatterers, so the light beam will naturally be softer at it's edges and this is not necessarily the same thing as how evenly lit the diffuser appears to be when looking at it

I'm not sure what you mean by "looking at it". The only thing that affects the image is how the illumination appears when the subject looks at it. Appearances from other viewpoints may be quite different, but those viewpoints don't matter.

Assuming that the back side of the diffuser is evenly lit, then what the subject sees depends on the size and quality of the diffuser. If the material is weakly diffusing, then the subject will see bright light from the direction of the source, and progressively dimmer light from other directions. If the material is strongly diffusing, then the subject will see almost even illumination clear to the edge of the diffuser, at which point the break between diffuser and unflashed background will create a hard edge.

See the second panel HERE, and imagine that instead of a clear plastic box and a fiber bundle, I had used a flashtube and an opaque snoot. Then the left side with weakly diffusing material would look quite similar: a bright spot in the direction of the flashtube, surrounded by dim diffuse light. But the right side with strongly diffusing material would look quite different: it would be an evenly lit rectangle cutting to black at the edge of the snoot.

Now suppose the subject has a smoothly curved and somewhat shiny surface that produces direct reflections. The direct reflection of the diffuser is simply a distorted view of the diffuser, as seen from the standpoint of the subject.

If the diffusion material is weak, then that reflection has a small bright center that's likely to be captured as a blown highlight. If the diffusion material is strong, then the reflection will be larger and proportionally less bright (assuming constant total exposure). If the diffuser is made larger, at some point it becomes sufficiently less bright that the direct reflection no longer blows out. This point is a critical threshold: diffusers that are effectively smaller will produce blown direct reflections; diffusers that are effectively larger will not. (For illustration, see the tomatoes HERE. In image 1, the highlights are blown; in images 2 and 3 they are not.)

So, size and uniformity are the critical parameters for determining whether the highlight will blow out.

But what about the highlight's edge? Again, the direct reflection from a smoothly curved shiny surface is just a distorted view of the diffuser. If the diffuser is uniformly bright clear to its edge and then stops, the direct reflection will do the same thing. In other words, lighting with hard edges produces highlights with hard edges -- at least when reflected from a smoothly curved shiny surface.

On the other hand, if the diffuser has a soft edge, darkening gradually before it ends, then the direct reflection will do that too. Lighting with soft edges produces highlights with soft edges. The "ideal" situation would be a diffuser with a center area that is large and uniform, combined with edges that fall gradually to dark. The diffuser you show fits this pattern pretty well in the horizontal direction, not so much in the vertical.

I'm pretty confident about the preceding analysis for subjects with smooth curves -- the extreme example would be shooting a shiny sphere.

Where things get trickier is with complex fine structure like arrays of ommatidia on insect eyes. Quite frankly I'm not sure what all goes on there.

I think the surface of each ommatidium is basically a "bit of shiny sphere with a polygonal boundary".

If we were close enough to clearly resolve detail within each ommatidium, I bet we would see a clear though warped reflection of the diffuser. Certainly that is what we see in high mag shots of spider eyes and fruit fly eyes. If we want those tiny reflections to have soft edges, then the light coming from the diffuser has to have soft edges too.

But when we move back so far that we can't resolve detail of the individual reflections, something more complicated happens. Each ommatidium reflects to the camera light coming from a big chunk of space. Assuming I'm right about that "bit of shiny sphere with a polygonal boundary", then the chunk of space will be also be a warped polygon.

Let me think about this... Each ommatidium is reflecting light from a polygonal chunk of space, and what the camera gets is pretty much the average brightness of that chunk. Putting on my mathematics hat, this sounds like a 'convolution' operator. It's like what happens to OOF background in regular pictures. What you see there is a convolution of the background with a "blur polygon" that is the shape of the lens aperture. If there is an isolated bright spot in the background, what you see in the image is a not-nearly-so-bright blur polygon, with hard edges.

By analogy, with the array of ommatidia I guess what we're seeing is a convolution of what I'll call the ommitidia's "reflection aperture" against whatever light sources sit inside that aperture.

Ah, that actually makes testable predictions. If the light source is narrow angle, like the sun, then the highlight should have the same overall shape as the reflection aperture. That would be hexagonal if the arrangement of ommatidia is hexagonal, or four-sided if the arrangement of ommatidia is more rectangular (a less common case).

On the other hand, if the light source is wide angle, like a nearby large diffuser, than the highlight will have a much mushier shape with a soft edge. This would be analogous to large features in OOF background, which tend to just look blurred (in contrast to point sources, which turn into hard-edged images of the aperture). In that case, the degree of blurring depends mostly on the size and shape of the aperture, not the hardness of the background features. To get a smooth background ("good bokeh"), what you want is a large and round aperture. By analogy, what you want to get a smooth reflection from an array of ommatidia is a large and round diffuser.

OK, I think you've convinced me (or I've convinced myself) that we're both right. When the subject is smooth and shiny, I claim that what's needed to get a soft-edged reflection is a soft-edged light source. But when the subject is an insect eye, then maybe what matters more is that the light source be large and round.

Unfortunately, after all this it's still unclear what's going on in SonyAlpha's images. The highlight in his fly's eye is quadrilateral. By itself, that does not let us distinguish between two cases: A) the reflection aperture is small and any shape and the light source is quadrilateral, and B) the reflection aperture is large and quadrilateral, and the light source is small and any shape. In case A, what we're seeing would be mainly information about the light source, while in case B, it's mainly information about the eye. In either case, the highlight could be made larger and softer by increasing the size of the light source, making it round instead of polygonal, and softening the source's edges.

I don't know if this is helping anybody else, but I'm finding the mental effort valuable. Thanks for the stimulation; apologies for the intrusion...

--Rik

[LordV]

Thanks Rik for the thoughts.
Thinking about this after I did think there must be a combination of truths in there somewhere.
I suspect the odd bright normally 4 to 6 sided ring effect you get is a combination effect of a nearly parallel light beam with an edge hitting the eye where some odd effect goes on with the eye structure and produces a straight sided bright figure. I think adding diffusion both softens the physical light edge and scatters the light more making it a lot less parallel.
I was just a bit worried as I was fairly certain that you can have a diffuser with perfectly even illumination that does not produce that effect on insect eyes.

SA - hopefully you might find the diffuser materials comparison useful- the main point being that many of the plastics unless they look quite white do not actually diffuse the light that much (eg plastic milk bottle) whilst the more fibrous type paper materials do. In virtually all the tests I've done I seem to need a material that gives about 2 stops light loss to get the diffusion I prefer. So for example I have found that whilst a single styrofoam sheet does not give me enough diffusion, using 3 sheets of it did.
One other point is that your shooting style can affect the diffuser material choice. Because I often shoot fairly rapid sequences for focus stacking, I need to be able to take multiple shots without the flash recycling. So whilst I might like to use even heavier diffusion with associated greater light loss, I can't because I need the multiple shot capability (unfortunately I can't use an external battery pack with my 430Ex flash guns).

Brian V.

[sonyalpha]

Please don't apologise for your incredibly interesting and informative discussion on diffusers and light diffusion.........In fact I am delighted that my posts initiate responses of this calibre:

They are just what I need to encourage me to experiment further with the use of paper towel and other tissues ...............not to the depth and extremes that both of you might ....but just enough to improve my macro shots of flies especially:

The beauty of my ex cereal container flash reflector is that the wide-flat surface is ideal for quick changes and thickness's of diffusing materials....all of which are curled upwards over the open end of the container:

Question:

I only have a cheap SUNPAK PF 30x flash unit..............if for a potential stack.......will it re-charge quickly enough with the camera set in burst Sports Mode?

I have only used Sports Mode for my birds-in-flight series...............without the flash though:

The weather forecast for later today is sunny and warm after a very misty start.............it might persuade a fly or two to venture out so that I can see a tissue covered diffuser rids my sots of the hexagonal eye highlights:

I will post results in this thread:

sonyalpha