Speedlight's, Umbrellas, Coverage and Light Loss

[TMR Design]

I've been working on beefing up and refining my location lighting kit and in the process I decided to do some measurebating to demonstrate how different umbrellas and modifiers performed in terms of evenness of coverage and light loss.

Obviously, a Speedlight that is forward firing with no modifiers will produce much more light than using a modifier, so even though I'm going to reference light loss with modifiers in relation to output with no modifier, the real comparison should be from one modifier to another.

The modifiers tested were a 60" Photogenic Eclipse umbrella, 46" Photek Softlighter II and Photoflex 24" x 32" LiteDome softbox. The 60" umbrella was tested as a reflective bounce (with black backing in place) and shoot through, the 46" Softlighter was tested as reflective bounce (with black backing in place) , shoot through and as a brolly-like softbox with the front diffusion material. The 24" x 32" softbox was tested with both inner and outer diffusion panels in place.

Each of these modifiers and various configurations were tested with a single flash and with dual flash. The Speedlight's used for the tests were Nikon SB-800's. Readings for light loss were all taken at 4 feet from the light source at ISO400 with the flashes set to 1/4 power. The flash zoom setting was set to 24mm and then at 17mm with the wide diffusion panel down. I always allowed for full recycle and stabilization before taking another reading and the Speedlight's were powered by an Al Jacobs Black Box so AA batteries were eliminated and the flashes were never affected by diminished power.

Reference readings are as follows:

Single Flash (24mm): f/16 + .4
Single Flash (17mm wide panel): f/11 + .7
Dual Flash (24): f/22 + .1
Dual Flash (17mm wide panel): f/16 + .3

The images with the zoom setting at 24mm were all shot with the umbrella shaft extended as far as it could go from the flash head for maximum coverage, and the images at 17mm with the wide panel were done with the shaft pulled closer so as not to have spill outside the edges of the umbrellas. Using the wide panel not only increased coverage but greatly reduced or nearly eliminated the hot spot, depending on the modifier and configuration.

The blue triangles at the top left of each image show light loss relative to bare bulb output. Some of the results were expected and some were not. You can see that in every instance, using the wide panel improved coverage, sometimes in a significant way and sometimes only slightly. If you're using AA batteries to power your Speedlight's then you're probably better off sticking to 24mm for the zoom setting but if you've got external battery power and battery life or recycle times are not a concern then using the wide panel makes a great deal of sense. In many cases you simply can't get even coverage, or full coverage for that matter without using the panel and if you're shooting groups or full length shots the wide panel will really improve your ability to get those shots without the rapid falloff in the modifier or the loss of light from top to bottom (or left to right) of your subject(s).

Make what you will of the data and use it to your advantage when determining how you're going to use your umbrellas and modifiers, and how you'll set up your flash(es).

Additional tests with the Diffusion Dome (Sto-Fen) can be found in post #15.

[simply.blue]

Thank you for your effort!
I have a quick question if you don't mind. Do you think we'd have similar results with the smaller 30-something umbrellas?

[TMR Design]

Thank you for your effort!
I have a quick question if you don't mind. Do you think we'd have similar results with the smaller 30-something umbrellas?

My pleasure.

Similar, but different :D

Obviously with a smaller umbrella you'll get greater coverage at the 24mm setting and in some cases the hot spot may actually fill the umbrella, making it more efficient for the size but using a smaller umbrella means a smaller apparent light source, and that means that it won't produce the same soft light without getting it much closer to the subject.

That's why in most cases people select umbrellas starting at about 43" and up.

[agedbriar]

Thank you Robert, a great help, especially to someone who hasn't got a flash meter.

The small difference in light output between reflective and shoot-through umbrellas is really someting I didn't expect.

[Moppie]

Wow, some measurebating I can make use of :)

Cheers Robert, I am always amazed at, and apreciative of the amount of work you put into these things.

[photagraph]

Wow, thanks so much for this info. I've always wondered myself if flipping down the wide panel would make a difference or not.


bw!

[Lotto]

Nice work, Robert! thanks for sharing the results.

From looking at your shots, I can't say the bounced Eclipse provides more even lighting than what I though it would. Is it because of the speedlight? or just the pictures don't show it.

[TMR Design]

The small difference in light output between reflective and shoot-through umbrellas is really someting I didn't expect.

Exactly!! Most people think that using an umbrella as a shoot through is some terrible waste of light and far less efficient that using it as a reflective bounce. It really isn't. The real difference is the spill outside the edge of the umbrella and behind it.

[TMR Design]

Wow, some measurebating I can make use of :)

Cheers Robert, I am always amazed at, and apreciative of the amount of work you put into these things.

My pleasure. As I began the tests I realized that many would probably be interested and benefit so it seemed appropriate to compile and present the data and images in the forum.

[TMR Design]

Wow, thanks so much for this info. I've always wondered myself if flipping down the wide panel would make a difference or not.


bw!

As I said in the original post, I think that most of the time we use Speedlight's we're conditioned to thinking about battery life and recycle times and do everything we can so as not to waste light and power and keep those recycle times as fast as possible.

With AA batteries we have to think that way or we'd be changing batteries constantly.

[TMR Design]

Nice work, Robert! thanks for sharing the results.

From looking at your shots, I can't say the bounced Eclipse provides more even lighting than what I though it would. Is it because of the speedlight? or just the pictures don't show it.

Thanks Lotto.

Personally, I was surprised and a bit disappointed in the results of the Photogenic Eclipse umbrella. While Photogenic has hidden the umbrella spokes and created a more pleasing catch light, their design also creates that unusual clover leaf pattern and doesn't have the evenness in coverage of the standard umbrella. I suspect that the 60" Softlighter would look more like the 46" Softlighter than the Eclipse.

The images really do give a clear picture of what is happening.

I'm going to be doing another set of tests with strobes in the next week or so.

[SYS]

Fantastic job of presenting the results, Robert! I'm always appreciative of your time and effort you so willingly invest in providing highly useful information to us forum members.

One question... I'm wondering whether the results would be about the same if you had tested with the Sto-fen on the flash instead of the wide panel. My thinking is that it would benefit those users of flash that is not equipped with the wide panel if the results also show a more even distribution of light. In fact, I've seen Scott Kelby doing exactly that in one of his videos. I'm guessing that it's for the same reasons why he's using the Sto-fen in firing the Speedlite flash into the umbrella.

[Curtis N]

Nice post, Robert.

[TMR Design]

Fantastic job of presenting the results, Robert! I'm always appreciative of your time and effort you so willingly invest in providing highly useful information to us forum members.

One question... I'm wondering whether the results would be about the same if you had tested with the Sto-fen on the flash instead of the wide panel. My thinking is that it would benefit those users of flash that is not equipped with the wide panel if the results also show a more even distribution of light. In fact, I've seen Scott Kelby doing exactly that in one of his videos. I'm guessing that it's for the same reasons why he's using the Sto-fen in firing the Speedlite flash into the umbrella.

Thanks Soo Young.

Coming right up :D

Nice post, Robert.

Thank you Curtis

[TMR Design]

I just conducted the same tests with the Nikon Diffusion Dome (essentially the same as a Sto-Fen). I tested with single and dual flash just as before. The only modifier I did not test in this series was the 24" x 32" softbox.

On the Nikon SB-800, once you attach the Diffusion Dome the flash automatically changes the zoom setting to 14mm.

The average loss from the 17mm wide panel to the 14mm setting with the diffusion dome is .3 stop, ranging from .1 to .6 in some instances.

If you're using flash with a wide panel I would use it rather than the diffusion dome with the exception of the Softlighter. Due to the fact that the flash is much further inside the umbrella of the Softlighter, the diffusion dome really seems to give a greater spread because light is coming out from the sides and not just a greater angle of coverage from the front.

If your flash does not have a wide panel and you can afford the loss of power, coupled with decreased battery life, then without question the diffusion dome (or Sto-Fen) is the way to go.

Thanks to Soo Young for making the suggestion.

[PacAce]

Hi, Robert. Looking at the images from your test results, I couldn't help but notice that the BG of the images of the shoot-throughs are gray while the BG of the other tests are black. What that tells me is that the light reflected backwards is being "wasted" and, in fact, can contribute to the final shot as an uncontrolled light spill. Although this may not be practical for the Photogenic umbrella (I'm not familiar with it), if you employ the diffuser with the Photek umbrella even though it's being used as a shoot-through, you can harness another 1/3 stop of light to light up the subject and if you also place the black backing over it, you can prevent any light making it through the diffuser turning into "stray" light. That should improve the "light loss" you are registering for the Photek umbrellas.

[TMR Design]

Hi, Robert. Looking at the images from your test results, I couldn't help but notice that the BG of the images of the shoot-throughs are gray while the BG of the other tests are black. What that tells me is that the light reflected backwards is being "wasted" and, in fact, can contribute to the final shot as an uncontrolled light spill. Although this may not be practical for the Photogenic umbrella (I'm not familiar with it), if you employ the diffuser with the Photek umbrella even though it's being used as a shoot-through, you can harness another 1/3 stop of light to light up the subject and if you also place the black backing over it, you can prevent any light making it through the diffuser turning into "stray" light. That should improve the "light loss" you are registering for the Photek umbrellas.

You can't use the black backing on the Photek umbrella in any way other than over the umbrella. It's not made to be used if you're using the umbrella as a shoot through.

I'm not sweating over the light loss Leo. This is just to demonstrate coverage and light loss when the modifiers are used in these configurations. Personally, when I use an umbrella as a shoot through I'm not concerned with spill out the back nor am I concerned with light loss. I understand the concept of wasted light but in the scheme of things you don't really lose much compared to using the umbrella as a reflective bounce and when powering my flashes from an external battery pack I don't even worry about light loss.

That's just me. :D

[PacAce]

You can't use the black backing on the Photek umbrella in any way other than over the umbrella. It's not made to be used if you're using the umbrella as a shoot through.

I'm not sweating over the light loss Leo. This is just to demonstrate coverage and light loss when the modifiers are used in these configurations. Personally, when I use an umbrella as a shoot through I'm not concerned with spill out the back nor am I concerned with light loss. I understand the concept of wasted light but in the scheme of things you don't really lose much compared to using the umbrella as a reflective bounce and when powering my flashes from an external battery pack I don't even worry about light loss.

That's just me. :D
I hear you, Robert. I was more concerned about the uncontrolled light spill than the waste itself. But if you have dark or black walls (or a large studio space), then light spill wouldn't be an issue. For someone like me with a small make-shift studio in the basement with light colored walls, controlling of light spills would be more of an issue. :)

[TMR Design]

I hear you, Robert. I was more concerned about the uncontrolled light spill than the waste itself. But if you have dark or black walls (or a large studio space), then light spill wouldn't be an issue. For someone like me with a small make-shift studio in the basement with light colored walls, controlling of light spills would be more of an issue. :)

I absolutely agree Leo. I don't typically use umbrellas in the studio. My umbrellas are only used on location and mostly outdoors so any spill that comes out the back doesn't affect the shot whatsoever. Generally speaking I only use shoot through umbrellas for groups of full length in situations where I don't have another modifiers available. I still prefer the 46" Softlighter over the Softlighter minus the black backing and diffuser. I like the light quality and when feathering, either from left to right or top to bottom I can do couples or a small group, or full length. Before I got my Elinchrom 53" Octa I used the 46" Softlighter in the studio for tons of full length and I loved it.

[WayneF]

Each of these modifiers and various configurations were tested with a single flash and with dual flash. The Speedlight's used for the tests were Nikon SB-800's. Readings for light loss were all taken at 4 feet from the light source at ISO400 with the flashes set to 1/4 power. The flash zoom setting was set to 24mm and then at 17mm with the wide diffusion panel down.



FWIW, I get rather different actual results. I see no way shoot through is within 0.2 EV of reflected. Perhaps your softlighter might be, with the additional rear fabric panel to cover it either way? But which would not account for the Photogenic.

I could not assign any meaning to your specifications of "4 feet to light source", where 1/4 power gives f/16 4/10. Assuming you refer to umbrellas, that is totally mindboggling. Your SB-800 outputs much more power than mine if that is true, which I doubt. ISO would equate our metering, but mine is full power and you said 1/4 power.

With my SB-800, Manual mode, full power 1:1, at 24mm zoom and ISO 200.

I measured a white 45" umbrella at a careful 48 inches from fabric (and about 6 inches off center). I meter f/11 2/10 reflected, and f/8 5/10 shoot through. That is 0.7 EV difference.

My Photogenic Eclipse 60 inch measures f11 2/10 reflected, and f/8 3/10 shoot through (funny looking thing as shoot through). The shaft of the larger umbrealla is 5 or 6 inches longer (so 24mm works out well for both), but the 48 inches metered distance was carefully adjusted to be the same. This is 0.9 EV difference.

Of course, shoot though is used much closer than reflected, so the intensity loss is not so important... except that 2/3 of the light goes out the rear somewhere. :)

My "light source" would the flash, and is 24 or 30 inches (shaft length) greater than the 48" to the fabric that I measured.

[PacAce]

FWIW, I get rather different actual results. I see no way shoot through is within 0.2 EV of reflected. Perhaps your softlighter might be, with the additional rear fabric panel to cover it either way? But which would not account for the Photogenic.

I could not assign any meaning to your specifications of "4 feet to light source", where 1/4 power gives f/16 4/10. Assuming you refer to umbrellas, that is totally mindboggling. Your SB-800 outputs much more power than mine if that is true, which I doubt. ISO would equate our metering, but mine is full power and you said 1/4 power.

With my SB-800, Manual mode, full power 1:1, at 24mm zoom and ISO 200.

I measured a white 45" umbrella at a careful 48 inches from fabric (and about 6 inches off center). I meter f/11 2/10 reflected, and f/8 5/10 shoot through. That is 0.7 EV difference.

My Photogenic Eclipse 60 inch measures f11 2/10 reflected, and f/8 3/10 shoot through (funny looking thing as shoot through). The shaft of the larger umbrealla is 5 or 6 inches longer (so 24mm works out well for both), but the 48 inches metered distance was carefully adjusted to be the same. This is 0.9 EV difference.

Of course, shoot though is used much closer than reflected, so the intensity loss is not so important... except that 2/3 of the light goes out the rear somewhere. :)

My "light source" would the flash, and is 24 or 30 inches (shaft length) greater than the 48" to the fabric that I measured.

When I did my testing, with the "average" surfaces of the umbrella in shoot-through or relfected the same distance from the subject (the camera in this case), there was a 2/3 stop difference, with the advantage going to the reflected umbrella.

When I used the diffuser to bounce back more light towards the subject in the shoot-through mode, the difference was only 1/3 stop.

Just as an FYI, the umbrella I used is the 45" Softlighter II.

[dave_p]

Rob,

Slightly off topic, but I haven't seen threads that talk about mounting two flashes on one stand. How is this done? Can you provide some close-up shots of how that mounting looks and perhaps links to products? I can post this as a stand-alone thread if would be better that way. Thanks!

[TMR Design]

FWIW, I get rather different actual results. I see no way shoot through is within 0.2 EV of reflected. Perhaps your softlighter might be, with the additional rear fabric panel to cover it either way? But which would not account for the Photogenic.

I could not assign any meaning to your specifications of "4 feet to light source", where 1/4 power gives f/16 4/10. Assuming you refer to umbrellas, that is totally mindboggling. Your SB-800 outputs much more power than mine if that is true, which I doubt. ISO would equate our metering, but mine is full power and you said 1/4 power.

With my SB-800, Manual mode, full power 1:1, at 24mm zoom and ISO 200.

I measured a white 45" umbrella at a careful 48 inches from fabric (and about 6 inches off center). I meter f/11 2/10 reflected, and f/8 5/10 shoot through. That is 0.7 EV difference.

My Photogenic Eclipse 60 inch measures f11 2/10 reflected, and f/8 3/10 shoot through (funny looking thing as shoot through). The shaft of the larger umbrealla is 5 or 6 inches longer (so 24mm works out well for both), but the 48 inches metered distance was carefully adjusted to be the same. This is 0.9 EV difference.

Of course, shoot though is used much closer than reflected, so the intensity loss is not so important... except that 2/3 of the light goes out the rear somewhere. :)

My "light source" would the flash, and is 24 or 30 inches (shaft length) greater than the 48" to the fabric that I measured.

Hi Wayne,

I can understand that you may not get the same results. I never said I measured distance in a precise way. Knowing that an inch either way would not affect a reading at that distance, I did the best I could to keep my distance , plus or minus an inch. If you doubt my findings that's fine and if you find it hard to believe that's fine too. I gain nothing by fudging numbers.

The conditions for my tests are as stated, so to say "perhaps your softlighter might be, with the additional rear fabric panel to cover it" is fine but that is not the case and not how I arrived at my numbers. Each test was done exactly as I've indicated.

You say you could not assign any meaning to my specifications. Where is it you're going with that? What type of meaning are you looking for? Whether or not that is what you get with your SB-800 has no bearing on the results. Nowhere in my testing or the subject of this thread did I indicate this is about output. The tests are about light loss, which is very different.

You then go on to say "assuming you refer to umbrellas...". No, I never said I'm referring to umbrellas. I needed a point of reference so I knew how much light was lost. It wouldn't have mattered what reading I got at what distance at what power level at what ISO as long as I measured the same way with each modifier to reflect the difference between my reference and the modifier being tested.

The reference was done bare bulb, flash head facing forward, on the same vertical and horizontal axis as the dome of the light meter and settings were as I've documented. Every test that followed was done in the same way to give me the amount of light loss from the reference.

I also don't know whether you and I are defining and measuring from the same place when we say "distance from the light source". The source of illumination refers to the place that the light is emitted from the modifier, not the flash. If I measure 4 feet from a shoot through umbrella I'm measuring from the most forward part of the umbrella... the peak, if you will. If I'm measuring a reflective bounce umbrella I measure 4 feet from the leading edge of the umbrella.

Umbrella shafts are of different lengths and we all use different hardware to position and place the flashes within a modifier. My flashes are aligned to the axis of the umbrella shaft and point directly towards the center of the umbrella. Most umbrella adapters, unless custom made do not create that alignment. They usually place the flash head at least a few inches above the axis and that will assuredly affect the reading you get from a modifier if you're taking readings at the center. If you took a standard umbrella adapter or one that holds two flashes with the heads aligned horizontally you would also get a very different image that would not appear centered within the umbrella. There would be an offset with the hot spot off center. In the case of my custom bracket the flash heads are about 4 inches further forward than they might be with a standard adapter.

There are many variables that could yield different results but the images and numbers I've published represent my tests as accurately as possible. I'm sure that 10 people could do the same tests and get different numbers.

Regardless of the variances that could occur, the tests and my findings will give people an idea of what happens when they use modifiers in different configurations as well as using their flash units at the different zoom settings with the wide panel or diffusion dome. Splitting hairs in this scenario won't change the results enough to make a difference.

[TMR Design]

Rob,

Slightly off topic, but I haven't seen threads that talk about mounting two flashes on one stand. How is this done? Can you provide some close-up shots of how that mounting looks and perhaps links to products? I can post this as a stand-alone thread if would be better that way. Thanks!

HI Dave,

I had someone else ask me this and sometime later this evening I'll post images of the bracket I fabricated to give me the alignment and placement with either single or dual flash. Unfortunately I've never come across a product that does just that.

[dave_p]

Awesome, thanks! More things to spend money on!

[WayneF]

You say you could not assign any meaning to my specifications. Where is it you're going with that? What type of meaning are you looking for? Whether or not that is what you get with your SB-800 has no bearing on the results. Nowhere in my testing or the subject of this thread did I indicate this is about output. The tests are about light loss, which is very different.


I simply made the assumption that if you are posting test results meaningful to anyone else, then anyone else ought to be able to duplicate your results. I could not duplicate it, and it was also contrary to my prior "common knowledge" of what the answer should be.

The only one detail I cared about or speak about is that shoot through is simply not within 0.2 EV stop of reflected, not from any standard umbrella. I do exclude the softlighter from my statement, as its extra fabric panel surely does have similar losses out either side, which I dont know about it, but you said 60 inch Photogenic Eclipse, and I know it is not an exception. You said 0.2EV and I say 0.9EV. This is a difference.


You then go on to say "assuming you refer to umbrellas...". No, I never said I'm referring to umbrellas. I needed a point of reference so I knew how much light was lost. It wouldn't have mattered what reading I got at what distance at what power level at what ISO as long as I measured the same way with each modifier to reflect the difference between my reference and the modifier being tested.

The reference was done bare bulb, flash head facing forward, on the same vertical and horizontal axis as the dome of the light meter and settings were as I've documented. Every test that followed was done in the same way to give me the amount of light loss from the reference.


Ah so, a reference then. This probably was origin of your "light source" comment too. I did not grasp why direct flash mattered if you were measuring umbrellas. I mean, you could meter it reflected from a pink Cadillac too, but so what if you are measuring umbrellas? :) No big deal here, I just did not understand.


I also don't know whether you and I are defining and measuring from the same place when we say "distance from the light source". The source of illumination refers to the place that the light is emitted from the modifier, not the flash. If I measure 4 feet from a shoot through umbrella I'm measuring from the most forward part of the umbrella... the peak, if you will. If I'm measuring a reflective bounce umbrella I measure 4 feet from the leading edge of the umbrella.


Thanks, I better understand now. I measured from my assumed maximum "working area", the flat fabric near center shaft (trailing for reflected, leading if shoot through), but otherwise I did the same thing. Perhaps the softlighter extra fabic panel was responsible again here. These few inches of measuring difference would seem to make your reflected slightly weaker - more distant than it should be (according to my method). Your case could be 15% farther (fabric) than my case, which is only about 11% different from the light sources ISL point of view. This measuring difference would help to equalize reflected/shoot-through in your case, but I suspect that is not enough to equalize it with my numbers however.

The flash tube is the "light source". The flash tube emits the light, there is no light without the flash tube. Inverse square law works from the actual light source (flash tube). We will become very confused if we attempt to measure ISL from the fabric. But yes, nevertheless, the fabric is a more convenient measuring point, and I also measured metered intensity 48 inches from the fabric in all cases.

I don't care about any of that here. I just cannot understand how only 0.2 EV difference is possible in any conventional umbrella shoot through and reflected (I do exclude the softlighter having fabric panels on both sides, as an unknown to me).

[TMR Design]

Sorry, I disagree Wayne. The source of illumination is not the flash head. What about when it's firing into a reflective umbrella. We measure the source of illumination from the light that is output by the modifier.

If you do it differently that's cool but I don't know anyone else that measure it that way.

[WayneF]

Sorry, I disagree Wayne. The source of illumination is not the flash head. What about when it's firing into a reflective umbrella. We measure the source of illumination from the light that is output by the modifier.



You must have to shut the closet door to keep the light inside at night when you store your umbrellas? My umbrellas don't do that. :)

To each his own, and it is amazing that it is an issue, but the inverse square law only works one way, and it is quite easy to verify... Just connect a steel measuring tape to the light stand, and stretch it out on the floor in front of the light. Measure the offset from that stand connection (end of tape) to the flash tube, which ever way that is (reflected, shoot-through), and add this offset to all measurements (to account for the longer light path). Actually it is much easier to simply attach that offset measurement at the light stand pole (use adhesive tape), then it is automatic, and the tape reads directly.

Then hold a string with a little plumbbob weight in the hand with the flash meter (string pointing down to precise measurement on tape - it also holds a constant height above floor if you adjust string length to center of the light), and meter at various distances along the tape, including up as close to fabric as you can get, and including at 10 or 15 feet from the stand. Plot this out. Look at it... think about it.

Here is one I did with a softbox, but an umbrella does the same thing (yes, I've done it, but I dont find it now). This is true so long as you measure from the actual source of the light.

http://www.scantips.com/g2/b400-softbox.gif

Some will always argue (of course) that room reflections can affect things, but 1) use a larger room, keep away from walls, and 2) any reflection is a longer path and a much weaker light, and 3) if we get the right answer, we don't have to explain any problems. :) Or go outside at night if you have to...

Try to explain this way how the fabric is the source of any light. What wattage is your fabric anyway? LOL

But I dont even care about that, it is simply how things work. I just want to know how to make my shoot though umbrellas meter within 0.2 EV of the reflected value?

[PacAce]

You must have to shut the closet door to keep the light inside at night when you store your umbrellas? My umbrellas don't do that. :)

To each his own, and it is amazing that it is an issue, but the inverse square law only works one way, and it is quite easy to verify... Just connect a steel measuring tape to the light stand, and stretch it out on the floor in front of the light. Measure the offset from that stand connection (end of tape) to the flash tube, which ever way that is (reflected, shoot-through), and add this offset to all measurements (to account for the longer light path). Actually it is much easier to simply attach that offset measurement at the light stand pole (use adhesive tape), then it is automatic, and the tape reads directly.

Then hold a string with a little plumbbob weight in the hand with the flash meter (string pointing down to precise measurement on tape - it also holds a constant height above floor if you adjust string length to center of the light), and meter at various distances along the tape, including up as close to fabric as you can get, and including at 10 or 15 feet from the stand. Plot this out. Look at it... think about it.

Here is one I did with a softbox, but an umbrella does the same thing (yes, I've done it, but I dont find it now). This is true so long as you measure from the actual source of the light.

http://www.scantips.com/g2/b400-softbox.gif

Some will always argue (of course) that room reflections can affect things, but 1) use a larger room, keep away from walls, and 2) any reflection is a longer path and a much weaker light, and 3) if we get the right answer, we don't have to explain any problems. :) Or go outside at night if you have to...

Try to explain this way how the fabric is the source of any light. What wattage is your fabric anyway? LOL

But I dont even care about that, it is simply how things work. I just want to know how to make my shoot though umbrellas meter within 0.2 EV of the reflected value?
The inverse square law applies only to point light sources. A large softbox is not a point light source at close distances. The intensity of the light from an infinitely sized softbox would be the same no matter how far from the light source you measure it. However, since practical softboxes are not infinite in size, they would behave more like a linear light source whose light intensity varies inversely with the distance (note, I did not say "the square of the distance") from the light source. In other words, at close distances from the softbox, the light intensity will decrease by one stop for every doubling of the distance.

Take a look at your reading at 14" from the softbox. It's f/8 0.8. And at 28" it is f/5.6 0.7. That's a difference of 1 stop. Now, double the distance again and check the reading. You have a reading at 55" which is close enough and that's f/4 0.4. Again, close to a stop difference.

Of course, as you start getting farther and farther away from the softbox, it starts to behave more and more like a point light source and then the inverse square law starts applying to the distant softbox.

[WayneF]

Take a look at your reading at 14" from the softbox. It's f/8 0.8. And at 28" it is f/5.6 0.7. That's a difference of 1 stop. Now, double the distance again and check the reading. You have a reading at 55" which is close enough and that's f/4 0.4. Again, almost stop difference.




Right, that is my point too... measuring from the fabric is doomed to fail. The fabric does not emit light. Turn off the flash tube, and all goes dark.

Double distance should be TWO stops (according to ISL). Instead from fabric, it is 0.8 and 1.1 stops at each "double".

And of course, it is TWO stops at double distance if we measure from the source of the light, as shown above. QED

[PacAce]

Right, that is my point too... measuring from the fabric is doomed to fail. The fabric does not emit light. Turn off the flash tube, and all goes dark.

Double distance should be TWO stops (according to ISL). Instead from fabric, it is 0.8 and 1.1 stops at each "double".

And of course, it is TWO stops at double distance if we measure from the source of the light, as shown above. QED

What you really want to do is measure from the apparent light source. That does not necessarily mean that it's the light source that's actually providing the light. The fabric of a softbox is the apparent light source, as is the surface of an umbrella.

[PacAce]

Right, that is my point too... measuring from the fabric is doomed to fail. The fabric does not emit light. Turn off the flash tube, and all goes dark.

Double distance should be TWO stops (according to ISL). Instead from fabric, it is 0.8 and 1.1 stops at each "double".

And of course, it is TWO stops at double distance if we measure from the source of the light, as shown above. QED

BTW, I already mentioned that a softbox does not necessarily follow the inverse square law so doubling the distance does not necessarily have to result in a two stop light loss.

[WayneF]

What you really want to do is measure from the apparent light source. That does not necessarily mean that it's the light source that's actually providing the light. The fabric of a softbox is the apparent light source, as is the surface of an umbrella.


Maybe what you want to do, but certainly not what I want to do. Because, the inverse square law simply does not work from the fabric, because it simply is not the source of the light. We are both arguing that fabric does not work.

What does work is to measure from the flash tube, regardless if in softbox or umbrella. Just measure around the obstacle. It is exactly the same if you bounce off the ceiling, we still have to compute the total path from the flash (NOT just from the ceiling). In the old days of non-automatic #5 glass flash bulbs, we did exactly that computation, because that is the way it works. Bounce is no different from the reflected umbrella. The ceiling is also not the source of the light.

I have posted the numbers that show this, and you could simply take ten minutes to quickly setup with a steel tape and a string, and confirm it, and then you will see it, and then we can end all this. :)

[Bearmann]

I am hesitant to get involved here, as I am no expert. However, Leo has already pointed out that a flashgun is not a point source of light, so the inverse square law does not apply. Also note that a white ceiling is not a mirror, but both reflects and diffuses light. It's not just a simple matter of measuring the distance to the ceiling from the light source and adding the distance from the ceiling to the subject as I understand it.

Robert, thanks for going to all this trouble to both measure and post your results. I appreciate it very much!

[TMR Design]

Robert, thanks for going to all this trouble to both measure and post your results. I appreciate it very much!

My pleasure Barry.

[SYS]

I just conducted the same tests with the Nikon Diffusion Dome (essentially the same as a Sto-Fen). I tested with single and dual flash just as before. The only modifier I did not test in this series was the 24" x 32" softbox.

On the Nikon SB-800, once you attach the Diffusion Dome the flash automatically changes the zoom setting to 14mm.

The average loss from the 17mm wide panel to the 14mm setting with the diffusion dome is .3 stop, ranging from .1 to .6 in some instances.

If you're using flash with a wide panel I would use it rather than the diffusion dome with the exception of the Softlighter. Due to the fact that the flash is much further inside the umbrella of the Softlighter, the diffusion dome really seems to give a greater spread because light is coming out from the sides and not just a greater angle of coverage from the front.

If your flash does not have a wide panel and you can afford the loss of power, coupled with decreased battery life, then without question the diffusion dome (or Sto-Fen) is the way to go.

Thanks to Soo Young for making the suggestion.

This is very helpful. I've always been curious about this, and now I have a solid test-based information thanks to you. Much appreciated your effort, Robert! :D

[PacAce]

Maybe what you want to do, but certainly not what I want to do. Because, the inverse square law simply does not work from the fabric, because it simply is not the source of the light. We are both arguing that fabric does not work.

What does work is to measure from the flash tube, regardless if in softbox or umbrella. Just measure around the obstacle. It is exactly the same if you bounce off the ceiling, we still have to compute the total path from the flash (NOT just from the ceiling). In the old days of non-automatic #5 glass flash bulbs, we did exactly that computation, because that is the way it works. Bounce is no different from the reflected umbrella. The ceiling is also not the source of the light.

I have posted the numbers that show this, and you could simply take ten minutes to quickly setup with a steel tape and a string, and confirm it, and then you will see it, and then we can end all this. :)
Wayne, if you are the same WayneF I think you are, then I respect your knowledge in the area of lighting but in this case, we're obviously not seeing eye to eye. So we'll just have to agree that we disagree. :)

But before I end this discussion, I would like to pose a rhetorical question (meaning you don't have to answer it): If it really doesn't matter whether light is being bounced, for example, off a ceiling, then why is it that the distance scale of a flash, like on the SB-900 or the 580EX, is disabled when the head is in the bounce position. Shouldn't the inverse-square law still apply, as you claim, even when the flash is being bounced? ;)

[WayneF]

I am hesitant to get involved here, as I am no expert. However, Leo has already pointed out that a flashgun is not a point source of light, so the inverse square law does not apply.


Of course, Inverse Square Law obviously does apply to flashguns. Inverse Square Law does in fact give correct approximations about metered intensity expectations at distance from the softbox. Not of course if trying to measure from the softbox fabric, but it does apply if measuring from the flash tube source, as my chart demonstrated, and as it would take anyone ten minutes to set it up and verify that it did apply. Of course it obviously applies. 2x distance results in 1/4 intensity of the flash, within 0.1 stop. Seems plenty close enough. The results are about perfect, and as expected.

How would you say it did not apply? What did it get wrong?

If I can meter f/11 at 2 feet, f/5.6 at 4 feet, and f/2.8 at 8 feet from flash tube, what would you call that relationship? A lucky guess? I call it observing the expected ISL, perfectly.

I only checked a few, but here are 14,800 links about how photographers do consider that the ISL does apply to their flash.
http://www.google.com/search?hl=en&q=flash+%22inverse+square+law%22&btnG=Search

If your flash unit contains a distance range scale as mine does, then as you change aperture, it is calculating new maximum range based on inverse square law. Flash Guide Number is based solely on inverse square law and aperture number. There are no other factors for computing flash expectations. If you claim ISL does not apply to flash tubes, then I fear our photography world is in danger of collapse. :)

How does your theory explain those facts?


When the distance reaches about five diameters from source diameter, then common radiators can be reasonably considered a point source which is in fact subject to the inverse square law... Flash tubes are tiny, so outside the softbox is probably twenty times diameter.

[WayneF]

But before I end this discussion, I would like to pose a rhetorical question (meaning you don't have to answer it): If it really doesn't matter whether light is being bounced, for example, off a ceiling, then why is it that the distance scale of a flash, like on the SB-900 or the 580EX, is disabled when the head is in the bounce position. Shouldn't the inverse-square law still apply, as you claim, even when the flash is being bounced? ;)


Technically, its distance scale is about subject distance, but it has no clue then how high the ceiling is, etc. It cannot compute to show this maximum subject distance. The new flash models will have a new button to enter ceiling height. :) Of course, today it is only important to this scale, not to the metered exposure. But I can promise that old timers who were used to #5 flash bulbs (before days of thyristor automation) have a pretty good eyeball guess about it. Those today whining about TTL being a bit off ain't seen nothin'. :)

I do not claim no discontinuity exists at the bounce surface, or at the softbox diffusion surface. But these are mere obstacles, a little step function, and Inverse Square Law is in fact STILL an excellent approximationl It is all we have to go on, and it still works well.

Disagreement is fine, we have no need to believe the same thing, but of course, I am concerned you may be missing out. :) If you spend ten minutes with the tape measure and string, you too can believe.

[PacAce]

Technically, its distance scale is about subject distance, but it has no clue then how high the ceiling is, etc. It cannot compute to show this maximum subject distance. The new flash models will have a new button to enter ceiling height. :) Of course, today it is only important to this scale, not to the metered exposure. But I can promise that old timers who were used to #5 flash bulbs (before days of thyristor automation) have a pretty good eyeball guess about it. Those today whining about TTL being a bit off ain't seen nothin'. :)

I do not claim no discontinuity exists at the bounce surface, or at the softbox diffusion surface. But these are mere obstacles, a little step function, and Inverse Square Law is in fact STILL an excellent approximationl It is all we have to go on, and it still works well.

Disagreement is fine, we have no need to believe the same thing, but of course, I am concerned you may be missing out. :) If you spend ten minutes with the tape measure and string, you too can believe.
And I have the same concern for you, Wayne. :p :lol:

Actually, I did spend ten minutes with a tape measure and what I discovered is that we are both right. Turns out there is more than one way to skin the cat. I think my way is more intuitive (it doesn't require you to dissect the cat and get to the heart first, i.e. figure out what the flash to fabric distance is inside a softbox) as you only need to measure from the surface of the fabric, assuming that you aren't bull-headed about the need to apply ISL (which doesn't apply to a non-point apparent light source). ;) :D

[WayneF]

Actually, I did spend ten minutes with a tape measure and what I discovered is that we are both right. Turns out there is more than one way to skin the cat. I think my way is more intuitive (it doesn't require you to dissect the cat and get to the heart first, i.e. figure out what the flash to fabric distance is inside a softbox) as you only need to measure from the surface of the fabric, assuming that you aren't bull-headed about the need to apply ISL (which doesn't apply to a non-point apparent light source). ;) :D


Yeah, life is hard sometimes. :)

Forgive me, I must have missed "your way". I remember you said:


Take a look at your reading at 14" from the softbox. It's f/8 0.8. And at 28" it is f/5.6 0.7. That's a difference of 1 stop. Now, double the distance again and check the reading. You have a reading at 55" which is close enough and that's f/4 0.4. Again, close to a stop difference.


I probably misunderstand, but I am not aware of seeing anything else. Was that your method? One stop each time distance doubles? What do you call this new law? It seems to creep pretty bad. I do that with histograms, each 20% down from the right side is one stop. Easier to remember than right, but halfway so, close enough at the time.

Yes, it is convenient to measure distance from the fabric, and we do think that way about "closeness" and shadows, which is about relative size of the fabric. But the fabric is not the source of the light, no matter how hard we try.

[TMR Design]

How about taking the math and physics lesson to a new thread? Thanks for turning my thread into a petty pissing contest.

Good night gentlemen.

[Moonjo]

Thanks, Robert. I completely missed post #15. Little surprised at the results but happy to see the actual data. I've been considering getting the 60" Softliter II instead of just a 60" umbrella, but I was concerned whether or not my speedlites would provide enough power. I feel more confident after reviewing your data.

[PacAce]

How about taking the math and physics lesson to a new thread? Thanks for turning my thread into a petty pissing contest.

Good night gentlemen.

Sorry, Robert. I was just trying to defend your position of metering from the front of the softbox/umbrella and got carried away. I'm done now.

[TMR Design]

Sorry, Robert. I was just trying to defend your position of metering from the front of the softbox/umbrella and got carried away. I'm done now.

No Problem Leo. I always appreciate the effort. :cool:

[cortes]

A very informative test Robert. I'd been shooting at 24mm, but never thought to put down the diffuser. I tried the test with my 43" Westcott shoot through. I could never get dark areas similar to yours. Maybe the 43" is just right for 24mm. Anyway, do you recall your shutter speed and aperture?

Another question is what was the orientation of your SB800? I'm using the Strobist set up where the head is angled down 90 degrees. The umbrella is angled up so the flash in centered on the umberlla. I'm not positive from you picture, but yours looks like it may be laying flat. Maybe it's as small as it is because of the distance. When I set the aperture to 1/250 sec, f/16 (high as I could go on my 85mm), and the flash power to 1/4 and 24mm, I would get blown out highlights in the top half of the umbrella.

[TMR Design]

Hi Curt,

The images were intentionally shot slightly underexposed so that the light didn't just blow out the whole umbrella. That wouldn't really show the pattern and coverage. Any of those images could have shown even coverage if I cranked up the power and overexposed.

Regarding the orientation and mount of the Speedlight(s).. I have another thread floating around here with my bracket showing the flashed mounted and their position with respect to the umbrella shaft and I know I promised some other forum members those images but I've been really busy and will get that posted as soon as I can for you to see.

The bracket I fabricated has the flash(es) laying within 1/4" of the umbrella shaft, pointing straight in to the center of the umbrella. Every commercially available umbrella adapter I've ever seen places the flash head several inches above the umbrella shaft and direct the light above the center of the umbrella.

I didn't include those images in my tests but if you were to conduct some tests and take test shots you would see the hot spot shifted upward and much more uneven coverage due to that misalignment.

[siniquezu]

good info. will read later

[WayneF]

Maybe the 43" is just right for 24mm.


Yes, 24mm zoom is about as good as you can do with any umbrella.

The 24mm zoom of course covers an angle slightly greater than a 24mm lens should cover (speaking of full frame sensor coverage). The lenses have angular specs, and Nikon speedlight specs also include this angle - they say 78x60 degrees at 24mm zoom. A 24mm lens will be slightly smaller, maybe 4 or 5 degrees smaller.

Umbrellas have curved tops, and their spec size is the size of that curved fabric, measured over the curved top, so to speak. So, a 45 inch umbrella typically measures about 40 inches straight across, but it depends on how deep or how flat they might be, they vary slightly. A 45 inch umbrella has an effective shaft length of about 24 inches at full length, but you can measure from flash head to fabric on yours.

Trig says the angle to fill the umbrella is:

2 arc tan (opposite radius / adjacent shaft length) = degrees.

for 45", 2 arc tan (20" radius / 24" shaft) = 79.6 degrees.

for the 60" Eclipse, 2 arc tan (24" / 29") = 79.2 degrees.

Reflectors like Alienbees are 80 degrees too.

The larger 60 inchers have longer shafts, and the smaller 32 inchers have shorter shafts, so that they all work the same with the lights. This will work out to be about 80 degrees for about any umbrella. So 24mm zoom will be about as good as you can do with speedlights, af full shaft length.

Only reason to use the diffuser would be if you think your flashes coverage is not even, but your flashes coverage is even. Otherwise, it just robs power. If you use like 35mm zoom, or put the light too close on shaft, you just get a hot spot in the center, not that harmful to shoot-through, but for reflected, the metered light at the subject will decrease somewhat, not increase. I am speaking of umbrellas. The softlighter is exception in many ways, one is that flash must be extremly close to its fabric, no concept of a coverage angle.

The way to examine for actual spill around the reflected umbrella is to take a flash picture of a near wall behind the umbrella. At full shaft length, 24mm shows no spill, 17mm shows a slight trace of spill.

[Moonjo]

I believe this is the thread that Robert is referring to for his speedlite alignment to the umbrellas.

[TMR Design]

I believe this is the thread that Robert is referring to for his speedlite alignment to the umbrellas.

Thanks for posting that for me Moonjo.

[cortes]

The 24mm zoom of course covers an angle slightly greater than a 24mm lens should cover (speaking of full frame sensor coverage). The lenses have angular specs, and Nikon speedlight specs also include this angle - they say 78x60 degrees at 24mm zoom. A 24mm lens will be slightly smaller, maybe 4 or 5 degrees smaller.

Umbrellas have curved tops, and their spec size is the size of that curved fabric, measured over the curved top, so to speak. So, a 45 inch umbrella typically measures about 40 inches straight across, but it depends on how deep or how flat they might be, they vary slightly. A 45 inch umbrella has an effective shaft length of about 24 inches at full length, but you can measure from flash head to fabric on yours.

Trig says the angle to fill the umbrella is:

2 arc tan (opposite radius / adjacent shaft length) = degrees.

for 45", 2 arc tan (20" radius / 24" shaft) = 79.6 degrees.

for the 60" Eclipse, 2 arc tan (24" / 29") = 79.2 degrees.



Nothing like a little trig to get me whip out my old pre-caculus book to research. Thanks for the informative response.

[cortes]

Hi Curt,

The bracket I fabricated has the flash(es) laying within 1/4" of the umbrella shaft, pointing straight in to the center of the umbrella. Every commercially available umbrella adapter I've ever seen places the flash head several inches above the umbrella shaft and direct the light above the center of the umbrella.

I followed the link and saw your set up. Animated pictures! Robert, you've out done yourself. My rough calculation has the bulb of your SB800 about 4.5" farther into the umbrella than mine. Maybe that's why I can fill mine at 24mm. Your set up has the advantage of doing a better job of filling the umbrella evenly. It makes me appreciate the placement of the umbrella hole on Elinchrom strobes.

Just did a few measurements. The outside edge of my Westcott 43" describes a circle 34" in diameter. It's 16.5" from the end of the shaft to its intersection with center of the circle above. Subtract 2.5" to the location of the flash bulb. That would be 14/34 if I followed WayneF correctly and I just burnt out on trying to calculate the angle.

[WayneF]

Just did a few measurements. The outside edge of my Westcott 43" describes a circle 34" in diameter. It's 16.5" from the end of the shaft to its intersection with center of the circle above. Subtract 2.5" to the location of the flash bulb. That would be 14/34 if I followed WayneF correctly and I just burnt out on trying to calculate the angle.


16.5 - 2.5 = 14 inches? I may not understand your description. That seems an incredibly short shaft?

I think it is described as a "compact" umbrella, but that shaft would not even extend outside the fabric when it is collapsed. It must be several inches shy of extending out into view. Is that right? Most umbrellas extend out several inches. Might the shaft have some type of telescoping extension on it?

opposite radius / adjacent shaft

From those numbers, 34" diameter is 17" radius.

2 arc tan (17 / 14.75) = 98 degrees

2 arc tan (17 / 14) = 101 degrees

So a 24mm zoomed flash should be around 78x60 degrees, and seems like it could never fill that short shaft umbrella.

Nikon says their wide angle adapter or diffusion dome gives more like 120x110 degrees.

[m3rdpwr]

I don't know how I missed this post.

Thank you very much for your continued contributions...

-Mario

[GH_ATL_GA]

Robert, many thanks for your efforts in performing these tests & posting the summary. The resulting discussion has also been informative & entertaining.

A couple of months ago I did some simple tests for my own purposes to examine the 580's emission pattern at various zoom settings. I thought some of the results from those tests might be of interest to folks following this thread.

The attachment below shows a summary example of patterns obtained when the 580 was fired at a surface 3 feet distant & set at 1/64th power with zoom settings of 14, 24, & 105mm. The plots were generated using imageJ, a very powerful & free (public domain) image analysis application ( http://rsb.info.nih.gov/ij/.)

The top row of the grid contains the source images, the middle row XY plots with the intensity scale at 0 - 255 for all 3 plots, & the bottom row 'surface' plots with the intensity scale optimized (neglecting the specular reflection region) as appropriate for each image

A larger version of the attachment along with images & plots for all 580EX zoom settings can be found at http://www.pbase.com/gh_beauphoto/580ex_samples . (An animated gif located in the gallery will play if the display size is set to 'original.') Other than the flash zoom setting, the parameters for all images were identical & chosen to avoid saturation in any image within the test set.

There's nothing new here, just an alternate way of 'seeing' things.

[TMR Design]

Robert, many thanks for your efforts in performing these tests & posting the summary. The resulting discussion has also been informative & entertaining.

A couple of months ago I did some simple tests for my own purposes to examine the 580's emission pattern at various zoom settings. I thought some of the results from those tests might be of interest to folks following this thread.

The attachment below shows a summary example of patterns obtained when the 580 was fired at a surface 3 feet distant & set at 1/64th power with zoom settings of 14, 24, & 105mm. The plots were generated using imageJ, a very powerful & free (public domain) image analysis application ( http://rsb.info.nih.gov/ij/.)

The top row of the grid contains the source images, the middle row XY plots with the intensity scale at 0 - 255 for all 3 plots, & the bottom row 'surface' plots with the intensity scale optimized (neglecting the specular reflection region) as appropriate for each image

A larger version of the attachment along with images & plots for all 580EX zoom settings can be found at http://www.pbase.com/gh_beauphoto/580ex_samples . (An animated gif located in the gallery will play if the display size is set to 'original.') Other than the flash zoom setting, the parameters for all images were identical & chosen to avoid saturation in any image within the test set.

There's nothing new here, just an alternate way of 'seeing' things.

Very interesting and pretty cool. Thanks for posting these images.

[cortes]

16.5 - 2.5 = 14 inches? I may not understand your description. That seems an incredibly short shaft?

It's a Westcott 2001 collapsible. A picture is worth a thousand words. I'm subtracting 2.5" since the flash head is rotated 90 degrees into the umbrella. It's a pretty rough guess.

[WayneF]

It's a Westcott 2001 collapsible. A picture is worth a thousand words. I'm subtracting 2.5" since the flash head is rotated 90 degrees into the umbrella. It's a pretty rough guess.


Ah so, thank you Curt, I understand now. You have corrected my error, and convinced me that your way is the correct measurement. That would compute 101 degrees wide. The speedlight 24mm zoom spec is more like 78x60 degrees, none too much for sure, but it would be a bit wider than that, in that it would be a gradual taper, and not a sharp cutoff there. Plus due to the curvature, I suspect bottom line is that the remote field pattern with of an umbrella must be measured to actually know anything. :)

[GH_ATL_GA]

Very interesting and pretty cool. Thanks for posting these images.

Robert,

Glad to do it & happy that you found it interesting. I've found imageJ to be a great measurbation aid.

[GH_ATL_GA]

It's a Westcott 2001 collapsible. A picture is worth a thousand words. I'm subtracting 2.5" since the flash head is rotated 90 degrees into the umbrella. It's a pretty rough guess.

Hi Curt,

I have the Westcott 2001 collapsible & made some tests with it a couple of months ago. I pulled the images up tonight & used the imageJ application to generate these plots.

The images & plots show the results of firing a Canon 580EX at 1/64th power into the Westcott with it mounted at absolute maximum extension. The flash was mounted on a typical umbrella tilting type bracket, i.e. the flash is vertical with the head at appx 90 degrees to the control stage.

The images are small to meet the posting requirements, making it difficult to read the scales (the intensity scale is 0 -255 for all plots.) I can post larger sizes & the individual images & plots online should anyone be interested.

[m3rdpwr]

Your 3rd and 6th row of graphics looks like a Bundt Cake... :)

Now I want some... hehe

-Mario

Hi Curt,

I have the Westcott 2001 collapsible & made some tests with it a couple of months ago. I pulled the images up tonight & used the imageJ application to generate these plots.

The images & plots show the results of firing a Canon 580EX at 1/64th power into the Westcott with it mounted at absolute maximum extension. The flash was mounted on a typical umbrella tilting type bracket, i.e. the flash is vertical with the head at appx 90 degrees to the control stage.

The images are small to meet the posting requirements, making it difficult to read the scales (the intensity scale is 0 -255 for all plots.) I can post larger sizes & the individual images & plots online should anyone be interested.

[GH_ATL_GA]

Your 3rd and 6th row of graphics looks like a Bundt Cake... :)

Now I want some... hehe

-Mario
Hmmmm. Hadn't noticed that, but it might explain my near constant case of the munchies...:)

[TMR Design]

I thought we were looking at the thermal scans that Predator sees.

[form]

The original tests seem genuinely useful for future reference; I'll remember them. However I think the shorter shaft of smaller umbrellas will impact the spread, so I don't think you'll necessarily get greater % of coverage in a smaller umbrella because the distance between flash and umbrella will also be shorter.

[M Powered]

Very informative thanks Rob ;)

[TMR Design]

Very informative thanks Rob ;)

Always a pleasure Kesler. :D

[EOS mE]

Your 3rd and 6th row of graphics looks like a Bundt Cake... :)

Now I want some... hehe

-Mario

Bundt cake.. lol.. that reminds me of that movie "My big fat greek wedding" when the parents met and the greek family was given a bundt cake. :lol:

[cortes]

Hi Curt,

I have the Westcott 2001 collapsible & made some tests with it a couple of months ago. I pulled the images up tonight & used the imageJ application to generate these plots.

Thanks GH_ATL_GA,

From the images, it looks like I can go down to 35mm if I wanted. I can also go with a larger umbrella. Although, given what WanyeF said about them having longer shafts, it boils down to the power of the flash. I'll have to play around a bit more myself.

[GH_ATL_GA]

Thanks GH_ATL_GA,

From the images, it looks like I can go down to 35mm if I wanted. I can also go with a larger umbrella. Although, given what WanyeF said about them having longer shafts, it boils down to the power of the flash. I'll have to play around a bit more myself.


Hi Curt,

With the 2001 extended as far as possible on the bracket you're still getting pretty good use of the umbrella face at 35mm without exacerbating the hot spot problem too badly.

As Robert mentioned in an early post, 43" & larger umbrellas are more commonly used; you can always choke up on the shaft if you want a smaller source. You can make a 45 inch umbrella mimic a 30 inch umbrella but you can't make a 30 inch fully mimic a 45.:) The only real disadvantages of the larger umbrellas, at least that I know of, are the cost, the space required, & the greater 'sail' they present if shooting outdoors.

George

[adebartolo]

Robert,

Thx for this post and the and the illustration.

Tony

[silvex]

great work like always robert!

[adebartolo]

Brilliant piece of work Robert. As usual there is always a lot to learn from your posts.

[TMR Design]

great work like always robert!

Brilliant piece of work Robert. As usual there is always a lot to learn from your posts.

Thanks Ed and Tony. Glad you found it interesting and/or useful.

[Bob Hartman]

Thank you for a professional and helpful test. You must excuse my lack of knowledge, but I want to test my own equipment using yours as a guide. To do that I need to know exactly how you measured the distances.
1. For example, it is obvious for the 'bare bulb' tests you measured from the face of the bare bulb to the light meter?
2. For the shoot thru umbrella tests you measured from the portion of the umbrella that is the closest to the light meter, and then to the meter? Generally shoot thru results in a softer light than reflected because the umbrella is closer (bigger) at 4 feet compared to a reflected umbrella
at 5 to 6 feet?
3. For the reflected umbrella, at what distance was the center of the umbrella (the effective light source) to the light meter?
4. Or still discussing using the reflected umbrella, did you measure from the bare bulb directly to
the light meter? Or How?
Any information you can provide will be greatly appreciated. Thanks again.

Bob Hartman