Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
02-11-2009, 09:10 AM   #16
Forum Member
polur101's Avatar

Join Date: Jun 2008
Location: Buffalo, NY
Posts: 58
Highly reflective surface.

QuoteOriginally posted by audiobomber Quote
Could someone please tell me why this didn't work? The K20D was set the same way for both photos: Aperture priority, Auto ISO, +2 EV, 55-300mm at 300mm, F8.

Flash on camera, Sigma ST 530 Super, P-TTL mode.



Flash off



I need a longer lens
I had a similar situation. Volunteered to be the photographer for my son's swim team. Action shots using K20D. Water reflected a ton of light, making P-TTL a bad choice. An inherent drawback of the technology, not unique to Pentax. Marc Sabatella explains it well. The way I went around this. Have camera in manual mode. Hit green button to get an idea of adequate shutter and apperture. If the shutter speed is below 100, raise it and lower apperture appropriately to compensate. Keep the flash in manual mode. Use flash power of 1/1 to 1/4 depending on how far the subject is. Worked about 90% of the times. Cheers.

02-11-2009, 10:50 AM   #17
Veteran Member
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,685
QuoteOriginally posted by audiobomber Quote
I did use +2 EV with the flash. I agree with you, the photo should have turned out, at least it would have had I used a Canon. It seems that Pentax only accomodates fill flash in Hyper-Program mode.
I don't think that's quite true, although I admit I didn't follow that whole aspect of the discussion regarding how the Canon system differs from Pentax. But even with Pentax, this would have worked perfectly had you simply chosen an ISO high enough for the ambient light to have had any effect. Or chosen a larger aperture.

I suppose you are saying that even in Av mode, Canon *would* have adjusted the auto-ISO or shutter speed after the pre-flash. That is, once it discovered the flash wasn't reaching, it would have changed the exposure settings it already settled on to allow the ambient light to correctly expose the scene. And maybe it would have - I've never used a Canon flash system. So in that sense, it might be more automatic in situations like this. Although I suspect there would be other situations where you could fool it, too.

But really, I think fill flash *should* require a little thought - you should be in control of the ratio of ambient light to flash. At least, it seems you should want to be if you're considering using fill flash, as only you are aware of the specific aspects of the scene that require it's use to get the effect you want.

What I would have done is use M mode, no auto ISO, so I am in control of everything. Of course I'm already shooting that way all the time, but even if I shot Av or P mode and used Auto ISO otherwise, I'd switch to manual operation for this. I'd think through what is the minimum ISO/aperture combination for the distance I am shooting and pick one that would allow the flash to reach. Then meter the scene - with the flash turned off - using Green button. Then slow down the shutter a stop or two because I am dealing with snow (or meter off something neutral in the first place). This would yield the shutter speed that would correctly meter the scene using ambient light. If I think the shutter speed is too slow to be usable, I increase ISO or aperture and speed up the shutter to compensate. Then, I decide how much fill I want, and speed up the shutter to underexpose the shot by that amount. Then switch on the flash and take the shot.

Sound complicated? Sure, it's not point and shoot. But I know I'm less likely to be surprised by the camera's decisions, because the only decisions I am letting the camera make are ones I know it can handle (eg, choosing a shutter speed to expose the scene correctly given the available light, and choosing a flash output to expose the scene correctly given the current exposure settings and available light).

BTW, I'm also not convinced fill flash would actually work too well here even had you been in range - any amount of light sufficient to brighten the birds would also have blown out the snow. Fill flash works to help even out uneven lighting, but it can't do anything about objects that are inherently different in brightness. I don't remember if anyone has pointed that out yet. That's actually a bigger problem here - the *best* you could have hoped for was a shot that looked identical to the one without flash, or one in which the snow was exposed *exactly* as if you had simply dialed in more exposure compensation (or done the equivalent in M mode, or metered off a grey card).
02-13-2009, 06:10 AM   #18
Veteran Member
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,806
Original Poster
QuoteOriginally posted by Marc Sabatella Quote
I don't think that's quite true, although I admit I didn't follow that whole aspect of the discussion regarding how the Canon system differs from Pentax. But even with Pentax, this would have worked perfectly had you simply chosen an ISO high enough for the ambient light to have had any effect. Or chosen a larger aperture.
The K20D was set for Auto ISO. It should have raised the ISO itself if there was insufficient light. The Canon system would have found sufficient light in a daylight scene. I have Canon A720. I calll it a mini-dslr because it has all the controls of a dslr. The little Canon exposed the scene properly with its on-board flash, yet you all are telling me the Pentax couldn't expose properly because the big Sigma flash didn't reach 25 feet. I accepted this answer, but it still smells funny IMO.

QuoteQuote:
But really, I think fill flash *should* require a little thought - you should be in control of the ratio of ambient light to flash. At least, it seems you should want to be if you're considering using fill flash, as only you are aware of the specific aspects of the scene that require it's use to get the effect you want.
I disagree, and I still can't swallow that the Sigma doesn't have enough power to light a daylight scene at the limit of its Guide Number rating. Did the light hit a brick wall at the rated limit of 23 feet?

QuoteQuote:
What I would have done is use M mode, no auto ISO, so I am in control of everything.
I'm not interested in manual mode with flash. A camera with P-TTL should be able to figure out a proper exposure. I sent my photos to Sigma and here's the response:

Thank you for choosing Sigma products.
There should be an upgrade available for this flash model to fix the problem. You would have to send it to Gentec International since you are located in Canada.

Gentec International


QuoteQuote:
BTW, I'm also not convinced fill flash would actually work too well here even had you been in range - any amount of light sufficient to brighten the birds would also have blown out the snow. Fill flash works to help even out uneven lighting, but it can't do anything about objects that are inherently different in brightness.
Using fill flash is a very common technique in wildlife photography. The idea is to expose the subject, and if the sky, snow or water is overexposed, oh well. In fact this scene did expose correctly at +1EV and ISO 200, so the DR was within the camera's capability.

Sorry for the late response, I'm having a brutal month. February is crunch time at work and we just finished buying and selling houses, so I've been trying to stay off the internet.
02-13-2009, 03:01 PM   #19
Veteran Member
Marc Sabatella's Avatar

Join Date: Sep 2006
Location: Denver, CO
Photos: Gallery
Posts: 10,685
QuoteOriginally posted by audiobomber Quote
The K20D was set for Auto ISO. It should have raised the ISO itself if there was insufficient light.
Except that it simply does not *know* there is going to be insufficient light at the moment it set the ISO, which was before the pre-flash. It thinks the flash *will* provide enough light, because it doesn't know the scene is actually too far away.

I've gotten curious about this to do some experimentation to be sure what we are saying is not off-base. It isn't, and it is quite easy to prove. I do hope you read this, because this is some new information I have discovered and I think it will finally convince you. Feel free to follow along with your camera - you won't even need your Sigma to reproduce my results.

I took my K200D, put it in Av mode, set aperture to f/8, set auto ISO to range from 100-800, and set my zoom to 200mm. The shutter speed selected by the camera always read 1/180, because the algorithm always seems to choose something around 1/focal_length. I scanned about my living room and watched the ISO change in the viewfinder as one might expect, showing that Auto ISO isn't totally disabled with flash (which was something I was otherwise beginning to suspect).

Now, here's the interesting part. When I went outside and pointed at the house down the street, then *regardless* of what aperture I chose, the camera went with ISO 100. That's right, even at f/32, the camera wanted to go with ISO 100. That's what was displayed in the viewfinder, and that's what it used when I took the picture.

Think that ended up being fast enough? No, *of course not*. I got a *completely black* frame at f/32, because the light outside wasn't *nearly* bright enough for a good exposure at ISO 100, f/32, and 1/180 shutter speed, and my little popup flash isn't *nearly* strong enough to light up the house across the street at that ISO and aperture. At f/8, I got an underexposed picture. Try it yourself and see that I'm not making this up.

Now, *if* system were adjusting exposure parameters based on the actual light measured during the pre-flash, it would of course have noticed it had screwed up big time, and either raised ISO, lower the shutter speed, or both. But it did no such thing. It stubbornly held to the shutter speed and ISO it naively chose when it thought maybe the subject would turn out to be close enough for the flash to reach.

This is completely consistent with what we've been saying all along: the flash system just isn't as smart as you want it to be. It's not a defect in the camera or flash - it's just how it is. The camera simply does *not* adjust any exposure parameter a result of the pre-flash. Whatever exposure parameters it is displaying in the viewfinder when you press the shutter, that's what it is going to use. The *only* thing it will change based on pre-flash is the flash output. No "fix" to your Sigma flash is going to change the fact that the camera is still going to choose ISO 100 for you when shooting birds outside, no matter how far away they are and no matter what aperture you select.

Now, you may be wondering, why *did* the system choose ISO 100? That's a great question, and I don't know the answer. No matter what I tried, when the flash was popped up, it *always* chose ISO 100 any time there was a "decent" amount of light (eg, an amount of light that might hav resulted in an ISO 100 setting if I were in P mode). Indoors, it might choose ISO 200, 400, or 800 as it thought appropriate, but outdoors it *always* chose ISO 100. Apparently the algorithm measures the total amount of light and stops right there - it doesn't take into consideration shutter speed or aperture. Maybe it assumes the flash is infinitely powerful - I suppose it has no way of know just how powerful the flash is (after all you might have wireless slaves attached, too). But then, why go ahead and bump up the ISO indoors? I don't know; it's kind of a mystery, really. On the surface, though, it seems a pretty dumb algorithm.

But this validates what I said earlier - it's another reason why I would never trust the camera to make the right decisions on matters like this where I have no insight whatsoever into what it is "thinking". I know it works pretty darned well as a center-weighted meter and can go a good job of setting shutter speed to my selected aperture and ISO or putting out the right amount of flash for my selected shutter speed, aperture, and ISO. So I'm perfectly OK with letting the camera make those decisions. Anything beyond that, move over - I'm taking control.

QuoteQuote:
I have Canon A720. I calll it a mini-dslr because it has all the controls of a dslr.
OK, but does it also do TTL flash metering using a pre-flash system? I kind of doubt it. I'm not saying it isn't a solvable problem given a design change to the system. I'm just saying it is not one that it is reasonable to expect the current Pentax pTTL system to get right.

QuoteQuote:
The little Canon exposed the scene properly with its on-board flash
Nonsense. The Canon exposed the scene using available light. Try setting ISO, shutter speed, and aperture as in the Pentax and see if the flash can really make up the 2.5 stop difference.

QuoteQuote:
yet you all are telling me the Pentax couldn't expose properly because the big Sigma flash didn't reach 25 feet. I accepted this answer, but it still smells funny IMO.
Well, which part of it do you disagree with? The fact that the guide number of the Sigma flash is 175 in feet at ISO 100? The fact that 175/8 is less than 25? The fact that dividing the guide number by aperture is the correct way to calculate the effective range of the flash? Or the fact that the amount by which you exceeded the effective range was sufficient to yield some underxposure? I can only assume the last of these, since the first three are absolutely incontrovertible.

Since we don't know the *exact* distance involved, or just how accurate Sigma is in stating their guide number, the only way to truly gauge this is to actually perform the experimented I suggested: set f/8, 1/180, and ISO 100 yourself in M mode (to remove the camera from the equation), force the Sigma to fire full power (assuming it has such a control), and see what you get. I don't find it hard at all to believe that exceeding the published specs by a few feet would result in underexposure. But again, if you test it, we can put the issue to rest.

QuoteQuote:
I disagree, and I still can't swallow that the Sigma doesn't have enough power to light a daylight scene at the limit of its Guide Number rating. Did the light hit a brick wall at the rated limit of 23 feet?
No, it falls off with the reciprocal of the square of the distance, just as light always does, yielding to more an more underexposure the farther out of range you get.

Can we go with your original estimate of 30 feet? Dividing 175 by 30 yields 5.8. That's pretty close to 5.6. Meaning that you should expect to be able to do no better than one stop of underexposure with the flash. And that's one stop away from "correct" exposure - you actually wanted two stops of *over*exposure. Meaning, it's three stop below what you wanted. Available light has already been demonstrated to be 2.5 stops under your desired exposure at the settings chosen by the camera.

So no, it shouldn't seem the least bit surprising that light from the flash (which is 3 stops too weak to give you the results you wanted at the select exposure settings), even combined with daylight (which is 2.5 stops too weak at those settings) wold not have been sufficient.

But again, you needn't take my word for any of this - it is trivially simple to test if you can force the flash to fire full power. If you find it does indeed do better at full power than it did here, then I think you need to consider the possibility that reflections *did* play a role in that particular case, fooling the camera into dialing down the flash power just enough to make it that much less powerful.

QuoteQuote:
I'm not interested in manual mode with flash. A camera with P-TTL should be able to figure out a proper exposure.
In some cases, sure. but not in others. That's just how it is. When you want to do something tricky, like combine available light with flash at specific apertures - especially apertures that force the flash to the limits of its range - you're going to have to accept the need for some manual control, at least with the pentax pTLL system.

QuoteQuote:
[i]Thank you for choosing Sigma products.
There should be an upgrade available for this flash model to fix the problem. You would have to send it to Gentec International since you are located in Canada.
With all due respect, it's prett obvious they did not do as careful an analysis of the situation as we are doing here. That's essentially a form letter. They know of some osort of problem that mayor may not have anything to do with the situation, so the are offering to fix the problem without actually having any idea how it will affect your situation.

QuoteQuote:
Using fill flash is a very common technique in wildlife photography. The idea is to expose the subject, and if the sky, snow or water is overexposed, oh well.
I realize that. I just want to make sure you understood fill flash wasn't going to the thing that many people otherwise use fill flash for - to even out the lighting. Even had the flash been powerful enough to expose the birds at that distance, it would have done so no differently (except for *direction* of light, of course) then just adding yet more exposure compensation.

02-13-2009, 06:47 PM   #20
Veteran Member
jeffkrol's Avatar

Join Date: Sep 2006
Location: Wisconsin USA
Photos: Gallery
Posts: 8,434
Best guess... again

QuoteOriginally posted by jeffkrol Quote
As to Sigma flashes they are reverse engineered and can suffer malfunctions on newer bodies that they are incorrectly chipped for.

(message too short filler)

A bit more "thinkin"....


photo 1 was at 1/160 f8 iso 100 mean is 109
photo 2 was at 1/200 f8 iso 640 mean is 212

http://www.libraw.org/articles/zone-v-in-digital.html

Theory predicts:
the neutral surface should render 101 RGB if exposed according to the spot-meter; next stop is 138 RGB, then comes 189 RGB,
difference between the 2 is almost exactly the +2. I find this unlikely if the camera was just flubbing the flash.

Last edited by jeffkrol; 02-15-2009 at 09:43 AM.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, dslr, flash, photography
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Is Flash Sync Speed limited by the body or flash? Reportage Troubleshooting and Beginner Help 16 09-16-2010 06:18 AM
Ghosting - Due to UV? ManhattanProject Pentax SLR Lens Discussion 5 08-11-2010 03:00 PM
Pentax is due for a new body soon... jct us101 Pentax News and Rumors 90 07-31-2010 02:28 PM
When is The Next Lens Roadmap Due Out ? wll Pentax News and Rumors 1 05-22-2009 03:04 AM
Due in 2.5 weeks KrisK10D Post Your Photos! 9 04-19-2008 12:02 PM



All times are GMT -7. The time now is 06:55 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top