A Test: Flash Exposure Compensation w/ 9 Examples

Scott_Quier · June 23, 2007

I've been trying to get fill flash, in an outdoor setting, working for me and have been having some trouble nailing it down. On another board, I read that, on a sunny day, you wanted to set EC to -1 and FEC to +1 and move those numbers to zero as cloud cover increased. But, I've never seen any "proof", with pictures, of this actually working and, on the surface of it, it seemed counter-intuitive. So, I figured I would run a test and see what I got.

The following are a series of 9 images, taken at 1 stop intervals of EC and FEC. My model (my loving wife!) is standing in the solid shade of a tree on a cloudless day, approx 2:15PM EDT (local). The background is the side of a neighbor's house.

The numbers are EC and FEC, in that order

0/-1

0/0

0/+1

-1/-1

-1/0

-1/+1

+1/-1

+1/0

+1/+1

All were captured using:

Canon 30D
Av priority (f/4.0)
EF-S 17-44 at approx 45mm
ISO: 100
Shutter speed varied to maintain "desired" exposure
Direct light from 580EX II on bracket and positioned above lens

The full sized images can be seen in this gallery (link), should anyone be that interested

pathfinder · June 23, 2007

Very nice study Scott.

The use of fill flash is such an easy way to improve images, especially outdoors in sunlight or shade, that the more we know about its subtleties, the better photographers we can be.

In theory, shooting in Av mode, an EOS camera will choose the correct exposure for the background, and illuminate the forground with flash properly. In theory, if the metering is done accuarately, you should not need to dial in EC or FEC.

But.........there's always a but:D

What light metering mode was your camera in? Probably, Evaluative, and your lovely wife is shaded but standing in front of mixed sunlight and shade. In this particualar series of images, I do tend to prefer the shot with the -1 EC and the +1 FEC. This brightens up the subject and darkens the background which is a classic lighting technique to keep the viewer's eye on the subject. The results may vary a quite bit here just by moving the camera around due to the nature of Evaluative metering with a reflective reading camera light meter, and the large differences in lighting strength behind the subject.

Why not just shoot in Manual Mode with the flash in ETTL?? This will let you have total control over the balance of ambient versus flash lighting. (Remember that in bright sunlight, you may need to use High Speed Synch to allow shutter speeds faster than 1/200th - without HSS the camera will not allow you to adjust to a faster shutter speed.) Aperture will control flash lighting amount, and shutter speed will control ambient lighting. For more flash, open your aperture, for more ambient, lengthen your shutter speed.

Generally, when I use fill flash in Av, I tend to shoot without any EC, and use FEC at - 1/3 or -2/3 because I want the flash to be invisable. In your image with -1 EC and +1 FEC, the added light might be apparent to the knowledgeable eye. Very attractive, but apparent.

With +1 FEC, you MIGHT get over exposure or blown highlights also - think of the eyes of a groom in a black tuxedo where the meter wants to lighten the black to gray for instance.

It all finally comes down to your goals, I believe. As a general rule, I prefer shooting in Manual Mode with flash. However, I also shoot in Av mode with Fill Flash, from time to time, as it is faster and more easier to use in situations where the light is changing from shot to shot.

Once again, I commend you for your very interesting post.

Antonio Correia · June 24, 2007

Scott.
Good morning. This Fill flash business interesses me very much.
Your experiments were very interesting and instrutive.

Today, Sunday, I can't write here much because I have to go and shoot kite surf for my son and friends, but allow me to suggest a photo in P mode, flash straight on with compensation of (-) 1 1/3, and Speed sync.

I suppose it will work. The background with all it's light, will make the camera to command the flash to work on Fill flash. I have a strong feeling it does work.

I hope I will be able to make an experiment like yours one of these days.

thumb

Art Scott · June 24, 2007

Great Study....now I have to translate into the lnaguage of non canon camera users.....:D for me to understand....AV (audio - visual correct).....:D

dunno

I do see that the - +1 is the best of the bunch.

Scott_Quier · June 25, 2007

pathfinder wrote:

What light metering mode was your camera in? Probably, Evaluative, and your lovely wife is shaded but standing in front of mixed sunlight and shade. In this particualar series of images, I do tend to prefer the shot with the -1 EC and the +1 FEC. This brightens up the subject and darkens the background which is a classic lighting technique to keep the viewer's eye on the subject. The results may vary a quite bit here just by moving the camera around due to the nature of Evaluative metering with a reflective reading camera light meter, and the large differences in lighting strength behind the subject.

Why not just shoot in Manual Mode with the flash in ETTL?? This will let you have total control over the balance of ambient versus flash lighting. (Remember that in bright sunlight, you may need to use High Speed Synch to allow shutter speeds faster than 1/200th - without HSS the camera will not allow you to adjust to a faster shutter speed.) Aperture will control flash lighting amount, and shutter speed will control ambient lighting. For more flash, open your aperture, for more ambient, lengthen your shutter speed.

With +1 FEC, you MIGHT get over exposure or blown highlights also - think of the eyes of a groom in a black tuxedo where the meter wants to lighten the black to gray for instance.

Once again, I commend you for your very interesting post.

Thanks PF for your input and you did nail the light metering mode dead on. Yes, evaluative mode.

I was trying to confirm something I had read somewhere else about how to get a decent exposure from a setting such as this. This experiment/study seemed to support what I had read. Then, when I'm out with the Bride on Saturday, couldn't get consistant results.

So, this afternoon I will attempt some experiments using manual exposure and ETTL and see what what gets me. I'm hoping it's as simple as:

Set the aperature to what I need to blur the background while still maintaining sufficient DOF
Set the shutter speed high enough to bring the background exposure down where I need it. I'll be in a position where I can use an incident light meter as well as the reflective meter of the camera to meter the light on the background; it'll be interesting to see what the differences are between these
Set the flash to ETTL and HSS - understanding that HSS will decrease the total effective flash power (TANSTAAFL)
Move the subject into place
Press the shutter
Chimp and, if necessary, adjust FEC to taste

I'll post some more samples here once this has been done - later this afternoon.

In the meantime, if anyone has a better idea on how to approach this, please feel free to chime in.

pathfinder · June 25, 2007

I think this is a very worthwhile and educational effort for us all here, with you doing the heavy lifting, Scott.

I predict you will find that you prefer shooting in Manual Mode in ETTL as I do. Not that I don't use Av with fill flash - I do - it is easy, convenient, and quick to set up and do. That's why I do it too.

But Manual Mode with ETTL leaves more control of ambient vs flash in the shooter's hands, and this is better in the long run.

BenA2 · June 25, 2007

Great thread. I'm happy to see an increasing interest in flash here on Dgrin since it's where I'm focusing my learning efforts right now.

I did some similar fill shooting this weekend with an XT and a 430 EX. I was shooting in the middle of the day in a covered patio so had bright, daylight backgrounds with people shaded by the patio "roof". I was bouncing the flash off the mostly white ceiling. I shot in Manual mode with the exposure -1 stop from the meter reading (evaluative) of the daylight background. I was going for low DOF, so had a circular polorizer on board to get down to sync speed at f/2.8.

I started shooting in E-TTL with -2/3 FEC and, over several shots, ended up pushing it up to +1 FEC until I was happy. I was really surprised by this, because I usually find E-TTL tends to overexpose faces. So, while it's not what I expected, I'm at least pleased to see it matches Scott's results.

I'm wondering, could the polarizer have affected the E-TTL metering?

I'd post pics, but my wife has the CF card with her camera right now a couple hundred miles away. I'll get them up on Wednesday.

LiquidAir · June 25, 2007

Here are my guesses as to why you need those settings:

EC -1: Generally when the Canon metering system thinks there is mixed sun and shade it overexposes the sunny areas to recover some detail in the shade. You are getting that service from the flash, so would rather properly expose the sunlit areas.

FEC +1: In typical sunlight exposure situations, Canon assumes the sun is the main light and the flash is filling the shadows. As a result, the metering sustem underexposes the flash a bit to retain some shadow contrast. In this shot, your subject is entirely in the shade so the flash is actually the main light light rather than fill. FEC +1 brings the flash power up to an appropriate level for a main light.

Typically in lighting situations like the one you have there I dial some flavor sunny 16 in manual mode. That gets the sunlit part of the background properly exposed. Then I dial FEC anywhere from 0 to +1 depending on how much I think I can get away with without the shot looking "flashy". Honestly, mixed sun and shade like that is one of my least favorite shooting conditions because it takes a lot of fiddling to get anything that looks reasonable. If I can find full shade, I'll put the flash on a stand with an umbrella, dial the flash to run +1 to the ambient, and things get much better. Another trick is to gel the flash slightly blue and set a custom WB for the gelled flash. That will warm up your shadows.

Scott_Quier · June 26, 2007

This was a test of a number of different things:

Does my light meter "correctly" measure incident light. Or, put another way, does my camera agree with the light meter?
Can I manually set the camera to "correctly" expose the background?
To what do I set the FEC to "correctly" expose the subject with fill light?
Because I run into this situation a lot - having to shoot the bride and/or groom outside and them not wanting to stand in the sun and full coverage shade being nearly impossible to find, can I actually do this in a meaningful and repeatable fashion? Is this something that I can use on the job?

The sun was out (d'oh) and the sky was nearly clear (a little water haze). Should conform to the sunny-16 rule. My light meter said

ISO: 100
f/16
SS: 1/90

OK, that's close enough. This shot is at 100:16:1/80. It's plain that we could have used a bit more light (1/3 stop maybe?). But close enough that post processing would be simple.

Here's the set up without fill flash (100:f/5.6:1/800)

+2 FEC

+1 FEC - Very little difference between this and +2 FEC. I believe this fact to be attributable to the limits of the flash. As powerful as the 580EX II is, it might just not be able to throw +2FEC worth of light under these conditions. Remember, this is at HSS of 1/800, more than 3 times the x-sync speed of the shutter.

0 FEC - Here we see a real difference and an improvement. It is clear that there is still too much flash though.

-1 FEC - OK, we've just gone too low on the flash. As, just as PF posted above, it looks like the sweet spot will be somewhere in the neighborhood of -1/3 to -2/3 stop FEC.

-2 FEC - OK, by now we know we are well beyond the point of "correct" exposure. But, this is included for completeness.

Conclusions:

The light meter and the camera are in close agreement, but not spot on. To "expose to the right" I need about 1/2 more stop of light than admitted to by the light meter. This is good to know for future reference. I had seen this before when working with my AB800 stobes, but attributed it to pilot error.
The light meter closely agreed with the sunny-16 rule. I think the camera sees it more like the sunny-14 rule.:D
Manual settings - yep, I can do this.
Just like PF said, FEC works well somewhere between -1/3 and -2/3 stop.
Can I I do this quickly enough to make the process usable on the job. OH YEAH! Take a quick reading, dial it in, offset the exposure for the variance between the light meter and the camera and let 'er rip!
These Fill-Flash results are not good enough to post in the "Assignment #37: Fill Flash" thread, but it would'nt take a lot of work to get it there, maybe one or two more exposures.

I hope this has been useful for others. I know that I have learned a bunch from this quick exercise.

Antonio Correia · June 26, 2007

Scott.
Nice experience.

Good work, very useful.

What is curious is that I have myself made a similar experience today but I erased all the pictures because I thought it was pointless.

However, the results were similar.
The difference was that you gave a far better explanation of the subject than I would because you do know much more and speak better English.

Anyway, I would like to try again tomorrow and post my attempt. Can I ?

pathfinder · June 26, 2007

I think this last post was a great demonstration of the utility of fill flash and the accuracy of Sunny 16

I like Incident meters too for their superb accuracy. The meter in our cameras always seems to vary a little more from shot to shot.

Your comment, Scott, about the limitations imposed by High Speed Synch are very apt. I frequently shoot at ISO 400 indoors with a 580ex bounced off the ceiling and with a foamie diffuser throwing light forward. This works very nicely and High Speed Synch allows me to choose the shutter speed whereever I choose to put it. At these settings, the 580ex has plenty of horsepower to spare. But....

I also use an MT-14ex in Manual Mode in High Speed Synch out doors for shooting macro, and I notice a very large difference in the power of the MT-14ex when used in High Speed Synch and without High Speed Synch. I can change the FEC from +1 2/3 to -2/3 by switching from High Speed Synch to shooting without High Speed Synch in ETTL II mode. This seems to be the cost of light when using HSS in the MT-14ex. I suspect there is a similar cost lightwise with the 580ex or the 430ex. Maybe not this large, but it is there.

This is just one of those things to be aware of - if you seem to lack enough pop with your EOS Speedlight in High Speed Synch, turn off the HSS, and dial you shutter back to 1/200th, adjust your aperture accordingly, and see if that does not improve the situation light wise.

Again, very good discussion of the use of the 580ex outdoors.

Yes, Antonio, I am sure Scott has no objections to your joing the party

Scott_Quier · June 26, 2007

Antonio Correia wrote:

Scott.
Nice experience. Good work, very useful.
What is curious is that I have myself made a similar experience today but I erased all the pictures because I thought it was pointless.

However, the results were similar.
The difference was that you gave a far better explanation of the subject than I would because you do know much more and speak better English.

Anyway, I would like to try again tomorrow and post my attempt. Can I ?

Do I know more than you - NO. I've seen some of your work. For example, your fill flash items posted in the "Assignment #37: Fill Flash" thread are wonderfully done. I might speak/write better English, but it's the work that really shows!

As for posting your examples? Please do so. We all can only benefit from your additions to the thread. I don't own this thread, I only created it! Like PF said, "Join the party - PLEASE!"

Antonio Correia · June 26, 2007

Scott_Quier wrote:

Do I know more than you - NO. I've seen some of your work. For example, your fill flash items posted in the "Assignment #37: Fill Flash" thread are wonderfully done. I might speak/write better English, but it's the work that really shows!

As for posting your examples? Please do so. We all can only benefit from your additions to the thread. I don't own this thread, I only created it! Like PF said, "Join the party - PLEASE!"

Scott_Quier · June 26, 2007

pathfinder wrote:

I think this last post was a great demonstration of the utility of fill flash and the accuracy of Sunny 16 I like Incident meters too for their superb accuracy. The meter in our cameras always seems to very a little more from shot to shot.

I hadn't noticed a variance in the light meter, but then I wasn't looking for it either. Hmm, another experiment!

pathfinder wrote:

Your comment, Scott, about the limitations imposed by High Speed Synch are very apt. I frequently shoot at ISO 400 indoors with a 580ex bounced off the ceiling and with a foamie diffuser throwing light forward. This works very nicely and High Speed Synch allows me to choose the shutter speed whereever I choose to put it. At these settings, the 580ex has plenty of horsepower to spare. But....

I also use an MT-14ex in Manual Mode in High Speed Synch out doors for shooting macro, and I notice a very large difference in the power of the MT-14ex when used in High Speed Synch and without High Speed Synch. I can change the FEC from +1 2/3 to -2/3 by switching from High Speed Synch to shooting without High Speed Synch in ETTL II mode. This seems to be the cost of light when using HSS in the MT-14ex. I suspect there is a similar cost lightwise with the 580ex or the 430ex. Maybe not this large, but it is there.

This is just one of those things to be aware of - if you seem to lack enough pop with your EOS Speedlight in High Speed Synch, turn off the HSS, and dial you shutter back to 1/200th, adjust your aperture accordingly, and see if that does not improve the situation light wise.

That would be another interesting experiment - would one get more power back from the flash as a result of turning off the HSS than you would need to overcome the stopping down of the aperature?

pathfinder wrote:

Again, very good discussion of the use of the 580ex outdoors.

Yes, Antonio, I am sure Scott has no objections to your joing the party

LiquidAir · June 26, 2007

Here is a quick rundown on HSS and to use it:

Remember, once the shutter speed is faster than the sync speed, the shutter is a slit that sweeps the sensor. As you increase the shutter speed, you change the width of the slit but the sweep time stays the same. That is the key to understanding how HSS works. Lets take my 5D as example. The 5D has a sync speed of 1/200s and a maximum shutter speed 1/8000s which is 40 times faster. So, when my shutter speed is set 1/8000 second the shutter covers only 1/40 of the frame. For the flash to light the entire frame it must pop at least 40 times in the 1/200 sweep. In practice I am sure it is much faster so the multiple images of the shutter slit blend nicely.

A key thing to understand about HSS it behaves more like a continuous light than a flash because unlike normal sync, when using HSS the shutter speed affects the flash exposure. The flash puts out a fixed amount of power in 1/200 of a second and the size of the shutter slit controls how much of that light hits the sensor in a similar way to the aperture. So, then HSS does not help you overpower ambient light because faster shutter speeds reduce the apparent brightness of the flash by the same amount they reduce the ambitent light. If the problem you are trying to solve is to overpower the abient light, you cannot do better than putting the flash on full power and setting the shutter to the maximum sync speed. HSS actually slightly reduces the effective power of your flash because it is not 100% efficient.

There are only really two common applications for HSS:

1. You need higher that the flash sync speed to stop motion.
2. You want to open the aperture further than the sync speed allows to control DoF.

If your particular situation doesn't fall into either of these categories, you are likely better off using standard sync.

pathfinder · June 26, 2007

LiquidAir wrote:

Here is a quick rundown on HSS and to use it:

Remember, once the shutter speed is faster than the sync speed, the shutter is a slit that sweeps the sensor. As you increase the shutter speed, you change the width of the slit but the sweep time stays the same. That is the key to understanding how HSS works. Lets take my 5D as example. The 5D has a sync speed of 1/200s and a maximum shutter speed 1/8000s which is 40 times faster. So, when my shutter speed is set 1/8000 second the shutter covers only 1/40 of the frame. For the flash to light the entire frame it must pop at least 40 times in the 1/200 sweep. In practice I am sure it is much faster so the multiple images of the shutter slit blend nicely.

A key thing to understand about HSS it behaves more like a continuous light than a flash because unlike normal sync, when using HSS the shutter speed affects the flash exposure. The flash puts out a fixed amount of power in 1/200 of a second and the size of the shutter slit controls how much of that light hits the sensor in a similar way to the aperture. So, then HSS does not help you overpower ambient light because faster shutter speeds reduce the apparent brightness of the flash by the same amount they reduce the ambitent light. If the problem you are trying to solve is to overpower the abient light, you cannot do better than putting the flash on full power and setting the shutter to the maximum sync speed. HSS actually slightly reduces the effective power of your flash because it is not 100% efficient.

There are only really two common applications for HSS:

1. You need higher that the flash sync speed to stop motion.
2. You want to open the aperture further than the sync speed allows to control DoF.

If your particular situation doesn't fall into either of these categories, you are likely better off using standard sync.

Good description of High Speed Synch - which makes a strobe act like a Focal Plane flash bulb - remember those??

There is one other reason that I frequently use High Speed Synch for, though.

When shooting macro in the wild in a garden, the backgrounds are frequently a jumble of leaves, twigs, dirt, fences, etc.
I like to shoot at smaller apertures for good DOF, and then use enough flash and a short enough shutter speed, so that I overpower ambient light, and drive the background to black. For this I frequently need a shutter speed of 1/500th or so in daylight. Ergo, High Speed Synch, like here. This does cost 1 -3 stops of light compared to non-HSS, as I alluded to in my previous post. But then my backgrounds would not be driven to black.

ISO 200 f10.0 1/400th Manual Mode Exposure bias 0 compulsory flash fired - the shutter speed was determined by trial and error until the background was dark enough. Shot in full shaded daylight against the white side of a barn.

LiquidAir · June 26, 2007

pathfinder wrote:

ISO 200 f10.0 1/400th Manual Mode Exposure bias 0 compulsory flash fired - the shutter speed was determined by trial and error until the background was dark enough. Shot in full shaded daylight against the white side of a barn.

My point is that you could have taken the same shot in standard sync at ISO200 f/15 1/200 and the background would have been just at dark becase the ratio of the sunlight to the flash power is the same. The service HSS is providing on this shot is letting you open the aperture up to f/10 to reduce diffraction effects.

pathfinder · June 26, 2007

That is true, and a valuable thing it is

BenA2 · June 28, 2007

BenA2 wrote:

I'd post pics, but my wife has the CF card with her camera right now a couple hundred miles away. I'll get them up on Wednesday.

When I went to look at my pics, I was surprised to see how they looked on a propper monitor vs. how they looked on the camera LCD display. I actually don't see that much of a difference between -2/3 FEC and +1 FEC. The difference seemed much more dramatic in camera. I do notice, however, that the further away the subject was, the more pronounced the difference. I also notice that they all may be a tad bit underexposed, even those at +1 FEC. Fortunately, that's all easily fixed in RAW conversion.

These were all taken in manual mode, shot in RAW, and converted by LR 1.1 with default settings except for White Balance adjustment (no exposure/tone adjustments). Exposures were ISO 100, 1/200s, f/2.8 (except for one shot at 0 EV, 1/100s).

Background Metered to -1 EV, -2/3 FEC (10-15 ft)

Background Metered to -1 EV, -2/3 FEC (4-6 ft)

Background Metered to 0 EV, +1 FEC (10-15 ft)

Background Metered to 0 EV, +1 FEC (4-6 ft)

Scott_Quier · June 28, 2007

BenA2 wrote:

When I went to look at my pics, I was surprised to see how they looked on a propper monitor vs. how they looked on the camera LCD display. I actually don't see that much of a difference between -2/3 FEC and +1 FEC. The difference seemed much more dramatic in camera. I do notice, however, that the further away the subject was, the more pronounced the difference. I also notice that they all may be a tad bit underexposed, even those at +1 FEC. Fortunately, that's all easily fixed in RAW conversion.

These were all taken in manual mode, shot in RAW, and converted by LR 1.1 with default settings except for White Balance adjustment (no exposure/tone adjustments). Exposures were ISO 100, 1/200s, f/2.8 (except for one shot at 0 EV, 1/100s).

Background Metered to -1 EV, -2/3 FEC (10-15 ft)

Background Metered to 0 EV, +1 FEC (10-15 ft)

The exposure on all but the first one looks very good to me. Your #4 might be a touch over exposed, but not by much.

The background not withstanding, I see a significant difference in the quality and quantity of light illuminating your subject in these two exposures.

BenA2 · June 28, 2007

Scott_Quier wrote:

The background not withstanding, I see a significant difference in the quality and quantity of light illuminating your subject in these two exposures.

Yeah, these are the two that lead me to point out that the difference was more notable at longer camera-to-subject distances. Those shot at closer range... not so much.

jdryan3 · June 28, 2007

BenA2 wrote:

When I went to look at my pics, I was surprised to see how they looked on a propper monitor vs. how they looked on the camera LCD display. I actually don't see that much of a difference between -2/3 FEC and +1 FEC. The difference seemed much more dramatic in camera.

I think this is why most people will tell you to use the LCD image to look for compositional issues and the histogram for exposure. As Pathfinder mentioned earlier, the type of metering used makes a huge difference. Given the DOF I like and my typical framing, I prefer Center-Weight. Somtimes I do Partial. Rarely do I do Evaluative.

If I am doing bracketed shots, I do Partial or Spot. I primarily shoot Manual, but do use Aperature Priority often enough (used to use it all the time). I also know that my 5D often requires me to 'overexpose' by 2/3 - I almost do it by default, but in my manual settings rather than my EC settings.

cmason · June 28, 2007

Great stuff Scott. PopPhoto did this very thing years ago, and I keep this 'cheat sheet' in my camera bag, and refer to it often in these situations. I typically meter for subject, then use -1 2/3 on my flash...that seems to be my default, and then I work it from there.

http://www.popphoto.com/assets/download/821200311318.pdf

Antonio Correia · June 28, 2007

cmason wrote:

Great stuff Scott. PopPhoto did this very thing years ago, and I keep this 'cheat sheet' in my camera bag, and refer to it often in these situations. I typically meter for subject, then use -1 2/3 on my flash...that seems to be my default, and then I work it from there.

http://www.popphoto.com/assets/download/821200311318.pdf

I have been busy these days and I had no time to shoot an experience and post it here.

But I use, as you do, FEC ( - ) 1 1/3 the way you do with good results.
For me, Fill flash should be so soft that we can hardly notice it !

And thank you for that link !!!

BenA2 · June 28, 2007

jdryan3 wrote:

I think this is why most people will tell you to use the LCD image to look for compositional issues and the histogram for exposure. As Pathfinder mentioned earlier, the type of metering used makes a huge difference. Given the DOF I like and my typical framing, I prefer Center-Weight. Somtimes I do Partial. Rarely do I do Evaluative.

If I am doing bracketed shots, I do Partial or Spot. I primarily shoot Manual, but do use Aperature Priority often enough (used to use it all the time). I also know that my 5D often requires me to 'overexpose' by 2/3 - I almost do it by default, but in my manual settings rather than my EC settings.

Sorry, maybe I should have included more from my previous post to better explain that the real issue being discussed here is flash exposure compensation (FEC) in E-TTL mode. I'm perfectly happy with how my background exposures come out and, yes, I use the histogram for those. But, the histogram is not useful for gaging flash exposure on peoples' faces, because they do not take up enough of the frame to distinguish their tonal range in the histogram (unless you completely blow them against a dark background). That's what we're supposed to be able to rely on E-TTL for with Canon speedlites, but this example just illustrates my experience with how variable its results can be with the hope I might learn some why to make them more consistent.

Antonio Correia · July 4, 2007

Scott good morning.
I saw this and I remembered you and all the experiences and real work with the flashes.
Have a look, please.
May be you have already seen it ...

I like the results. Don't you ?

Scott_Quier · July 4, 2007

Antonio Correia wrote:

Scott good morning.
I saw this and I remembered you and all the experiences and real work with the flashes.
Have a look, please.
May be you have already seen it ...
I like the results. Don't you ?

No, I had not seen this before. I took a quick glance at it and it looks interesting. I'll take a better read of it later, when I have more time.

Thanks,

rwells · July 6, 2007

Two additional things:

1. 30D in Av mode will only shoot flash as "fill flash"
-- C.fn #03 forces the x-sync speed to 1/250 --

2. 580EX mounted on 30D --- FEC (physically set on the flash)
over-rides any set via the camera.

You can use Av mode with E-TTL flash and FEC, but your going to be chasing the exposure, never knowing exactly what it's going to do.

For that reason, I now use camera in manual mode (set however I need for ambient exposure) - flash in E-TTL and dial-in any FEC that I deem needed. (although in the above mode, the exposure is almost allways correct)

YMMV, but I did extensive reseach and testing to come to the above info.

Scott_Quier · July 7, 2007

rwells wrote:

Two additional things:

1. 30D in Av mode will only shoot flash as "fill flash"
-- C.fn #03 forces the x-sync speed to 1/250 --

2. 580EX mounted on 30D --- FEC (physically set on the flash)
over-rides any set via the camera.

You can use Av mode with E-TTL flash and FEC, but your going to be chasing the exposure, never knowing exactly what it's going to do.

For that reason, I now use camera in manual mode (set however I need for ambient exposure) - flash in E-TTL and dial-in any FEC that I deem needed. (although in the above mode, the exposure is almost allways correct)

YMMV, but I did extensive reseach and testing to come to the above info.

I agree about the manual mode thing. I did a portrait session with a young lady, her mother, and her grandmother about 10 days ago. Never had the camera out of manual. Set the FEC on camera. Worked very well (just looking at the raw files in ACR). As and when I have them processed, I'll post a couple.

But, yeah, the more I work with the 30D and flash, the more I like shooting manual mode. I am discovering that the computer in the camera can be pretty stupid sometimes.

Antonio Correia · July 7, 2007

I have now the ST-E2.
I have used it here for the first time.

In all the pictures I used the same speed and aperture, which has been measured for the ambient light, in the back.
I suspect that when I use fill flash with day light it blueishes (can I say this for tuning blue ? ) the light from the hole picture.
I also felt that I will like to use the ST-E2 in the future for the relative freedom it gives to the flash.

Thank you for watching.

Scott_Quier · July 7, 2007

I went to your site so get a better look at the larger size of your photos. The examples of on-camera and off-camera fill provide a very strong argument for why off-camera flash is so favored! Thanks for the this.

A Test: Flash Exposure Compensation w/ 9 Examples

Comments