The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Exposure problems

P

Paul66

Guest
I don't understand what you all mean by what metering technique I use,
 
P

Paul66

Guest
If your talking about the dome in or out, I have always metered with dome in, I have never had an issue before.

Also Bob I did as you ask and it worked fine, just confused me even more ha ha
Also I have the Mamiya 645DF/ 28DM and 80mm LS
 

cunim

Well-known member
If your talking about the dome in or out, I have always metered with dome in, I have never had an issue before.

Also Bob I did as you ask and it worked fine, just confused me even more ha ha
Also I have the Mamiya 645DF/ 28DM and 80mm LS
Was the underexposure consistent? I noted a similar problem (X or M, DF, IQ back, LS80) using a hot shoe flash for fill, with a profoto air transmitter synced from the back. Images tended to be dark, but varied from what I expected to dark. Reducing exposure time made it worse so it was some timing issue between the hot shoe pulse and the output pulse.

I think moving the sync to the DF sync port helped but, in the end, I just got rid of the on camera flash. The air transmitter works fine on its own.
 

johnnygoesdigital

New member
This might be odd, but check under the dome to make sure there's no dust on the meters measuring sensor. It's possible to have dust influence your meter. Also, make sure your drawing the correct amount of power from your lights, and that the elements are intact. Try taking readings with the dome up, facing the light to be measured and then at the lens. The camera's reflective meter should be fairly accurate - high contrast situations can sometimes fool a meter. In those situations I use the spot setting on the meter. With digital, you can keep changing the camera settings until it looks good, so try a combination of settings using the histogram, to match your desired result.

I just looked at you posted photo, and it seems your lights might actually be too far away. Look at the lack of contrast or light falloff in the scene overall. Try moving your lights closer to your model. Remove and modifiers such as a soft box and baffles too, to experiment with output.
 
P

Paul66

Guest
I tried a friends light meter and it's dead on, not sure what happened with mine but I guess electronics just fail.
Thanks for all the advise and thanks for not bashing my ignorance for tech terms, I am self taught and learned trial and error
Doug are you sating the skyport isn't really compatible with the Mamiya?
 

goesbang

Member
Paul,

You don't need to buy a new light-meter, it would be much more precise and easy to use the histogram. It couldn't be more precise.

Thierry
Thiery is absolutely right here.

Meters need to be calibrated to a standard, which assumes shutters, apertures etc also conform to said standard, blah, blah, blah. Frankly it's all crap when the only thing that matters is how much light is arriving at your sensor. Thankfully, your individual sensor's sensitivity is a constant, unlike the batch variation that our comrades at both Kodak and Fuji inflicted on us.
I am a working pro who, in the days of film, carried a meter bag with a Minolta flashmeterIV, a Minolta Colormeter, a Minolta spotmeterF and the groundglass probe. With the advent of backs with histograms, I sold all my meters and have never looked back. Shoot a test frame, check the histogram and adjust from there. This is WAY faster and more accurate than any meter.
If you are a picky SOB as I can sometimes be, shoot tethered. Then you can read the brightness levels down to single pixels to your hearts content.
I shoot mostly with an Alpa STC and I have the most accurate meter ever made - the histogram on my IQ180.....
Frankly, I don't understand why anyone who doesn't shoot film needs a meter.

Stress less, shoot more!

Cheers,
 

SergeiR

New member
I don't understand what you all mean by what metering technique I use,
No, i didn't mean dome and stuff.. I mean if you were metering reflected light - what zone you were using , and if you were metering incident - how you aim meter..

But no matter, if you said that your friend's meter worked fine.
 

SergeiR

New member
. Shoot a test frame, check the histogram and adjust from there. This is WAY faster and more accurate than any meter.
Sorry. It is not faster nor more accurate. I been big fan of "test with histogram" for a while too, after having to deal with old and clunky minolta's and some old soviet meters.. And then i went back to real flash meters from Sekonic and never ever looked back.

Histograms doesn't work b/c it not neutral enough, b/c you always have to remember what kind of scene's main tone is & etc.

Then again. Different folks. Different strokes.
 

SergeiR

New member
Sergi, I point the meter at each light that I am metering.
Ah. So Collins school, like myself.. Not Lane&Co school.. :) I vouch for the funky electronics... Have seen this happening with friend's Minolta (ambient dead on, flash is one stop under), but never seen it with Sekonics..
 

SergeiR

New member
what do you mean with this?
Typically histogram will display values according to current settings of contrast. Also per-channel histogram goes differently and calculated differently in camera/back from software (exception - i never compared new IQ series, and i never compared Phase One backs).

Digital sensor do have linear response, unlike film, which is curved. Histogram on most cameras (i am sure there are exceptions, however i never seen it) represents curved response, thus displaying info that is not neutral to settings. Frank Doorhoff suggested at some point contrast -2, which is getting S-curve more flattened, thus approximating what is in RAW a bit better.

Another interesting problem is channel sensitivity (which is included into the final histogram). E.g - if i shoot Leaf in friend's studio with Alien Bees (older ones) - they do have really really weird thing going in red channel. No matter how i fight it - i always end up with pretty terrible results for skin - it will look ok on back, but once i am processing it with C1 (or any other RAW software, i have few for different occasions) - goodbye skin texture, red channel goes out of control. Same stuff shot on ZD back - don't see it at all. Both backs with Elinchrom lights - never had a single issue.

Granted - i have never ever taken electronic parts apart. And i don't have latest backs, which might have some cool new features.
But i would much rather trust my flash meter and not waste shutter counts / batteries with camera, making scene ready before subject arrives on the set, than having to deal with whole thing on the fly. Plus, having light exactly where i want it and not "oh its within one stop, i will later fix it" - is much better. Not to mention that whole "i will later fix it" never truly yields good results as good.

Or if you meant what i meant by main tone? Its just a whole zone thing. Shooting black cat on the snow, shooting white cat on black paper, shooting gray cat on gray paper... That kind of stuff. Having to remember to compensate and do whole zone placing b/c of reflected metering for histogram is not much of a picnic either :) (not like there is much choice for landscape of course, but at least for portraits there is)

Just me. I learned to not trust histograms much, walking softly and carry around big stick once again..
 

Thierry

New member
Sergei,

Typically histogram will display values according to current settings of contrast. Also per-channel histogram goes differently and calculated differently in camera/back from software (exception - i never compared new IQ series, and i never compared Phase One backs).

Digital sensor do have linear response, unlike film, which is curved. Histogram on most cameras (i am sure there are exceptions, however i never seen it) represents curved response, thus displaying info that is not neutral to settings. Frank Doorhoff suggested at some point contrast -2, which is getting S-curve more flattened, thus approximating what is in RAW a bit better.
That is up to you, how you set the curve, linear, s-shaped, flatter in the shadows, steeper in the highlights, or vice-versa, etc ..., resp. with more or less contrast (this is simply equal to changing the developing process with film). The same can be obtained with film, in some ways, e.g. by changing the process, thus changing the contrast.
It doesn't change anything concerning the precision of using the histogram as a light-meter, once you have the curve you want to use.

Another interesting problem is channel sensitivity (which is included into the final histogram). E.g - if i shoot Leaf in friend's studio with Alien Bees (older ones) - they do have really really weird thing going in red channel. No matter how i fight it - i always end up with pretty terrible results for skin - it will look ok on back, but once i am processing it with C1 (or any other RAW software, i have few for different occasions) - goodbye skin texture, red channel goes out of control. Same stuff shot on ZD back - don't see it at all. Both backs with Elinchrom lights - never had a single issue.
I don't see in which way this has influence on the light-meter and how precise it is. Film has never had equal response or sensitivity in the different color "channels" = layers (greens in B&W, reds and blue in tranies or color negs, all that was such a headache).
With digital you can "easily" correct such responsiveness of the sensor by creating your own camera profile. But in any case, it is better to use a lighting system which is not giving such problems, being it for film or for digital mediums. There are many brands giving close to perfect light.

Granted - i have never ever taken electronic parts apart. And i don't have latest backs, which might have some cool new features.
But i would much rather trust my flash meter and not waste shutter counts / batteries with camera, making scene ready before subject arrives on the set, than having to deal with whole thing on the fly. Plus, having light exactly where i want it and not "oh its within one stop, i will later fix it" - is much better. Not to mention that whole "i will later fix it" never truly yields good results as good.
You misunderstood what I meant when saying the histogram is the best possible light-meter: it is not meant to say that one has then the possibility to "repair" or fix light afterwards when it is within +/- x f stops. It is exactly the contrary, to let you set the light in the subjects, based on your decision (e.g. get that particular shadows part at - x under-exposure, the 1/4 tones at - y, the mid-tones at ..., the 3/4 tones at +..., the highlights at + ..., the first shadows with details at - ..., the last high-lights with details at + ...). You are absolutely free, but obviously this by changing the light in your subject, not by tweaking the file afterwards with the software (though this is still possible to a certain amount, more so with MF digital files, and better so when the exposure is perfect).
That was what I meant, and in the contrary, this method allows to get better files with less needed tweaks afterwards.

Actually, I was running "Black & White - Zone Systems" workshops, at a time were digital was at its beginning: we (me and the participants) would have loved to be able to use such a light-meter to get the correct exposure/zones in our subject.
And in addition to all this, the histogram light-metering allows to measure the light exactly where it falls, on the image plane (TTL light-metering), with no headaches because of exposure compensations to be used (filters, bellows extension/magnification, etc ...).

Or if you meant what i meant by main tone? Its just a whole zone thing. Shooting black cat on the snow, shooting white cat on black paper, shooting gray cat on gray paper... That kind of stuff. Having to remember to compensate and do whole zone placing b/c of reflected metering for histogram is not much of a picnic either :) (not like there is much choice for landscape of course, but at least for portraits there is)

Just me. I learned to not trust histograms much, walking softly and carry around big stick once again..
See my explanations above, Sergei. It is exactly that, the point: zone system metering. A white cat on a white background, or a black panther in a tunnel entry, or in the snow, no matter what, the zone system allows you to set the light exactly to be able to distinguish details where you want them to be. And for this the histogram is the best tool.
Believe me, I have used all types of light-meters, mainly Minoltas, for nearly 20 years, I and the vast majority of the photographers I have met found the histogram to be a "gift from God".
 

SergeiR

New member
Its too hard to do quotes on ipad, so i just will type in order.

- you do not really have control to fine level of getting same response as in raw data , without extensive tweaking of everything, and then having to explain to people how the hell images are bleak and odd when displayed at first. Plus it requires me as photographer to deal with each lens and camera combo individually. Add iso into that. Doing this on the fly is painful. Yes, i can load some backs with profiles. Notice how i am keep saying that for some it might work. But not all people use latest and greatest :)
- adjusting channel sensitivity on bayers sensor is hell of all diseases. If we had foveon structure, or true b&w - yes. Would be possible. But normalizing space when you got to interpolate from rggb or any other weird pattern in sensor array will get you. And nope, raw with strong color shifts is not recoverable in full ( extreme example is shooting one end of k scale, while white balance is set to same end), which is another practical proof of fact that if you got sensor with finite number of states, and offsetting it beyound capacity - register overflow is inevitable and information loss is right there.

I am not arguing about usability of zone system ( not fan of it either, though) or histogram as extra walking stick. But i have found that life is way easier when i have lights (or scene) done with flash meter ( or worked with same light system long enough so you can wild guess exposure within 1/3-2/3 ev). I do have enough issues with controlling people within scene, to run around and keep adjusting lights while subjects holding poses :) but its me.

Calibrated flashmeter i am talking about, like 758dr. Not any.

And its not about film :) film is whole other subject, imho.
 

Shashin

Well-known member
I always find these conversations funny. Member X says I do this and it works for me, and a whole bunch of members come on and say you should not do that because we do it a different way and our way is better. Which begs the question, better for whom? And then when you look at Sergei's work, it is kind of hard to criticize his method.
 

Thierry

New member
It was never meant as a criticism of the method used, nor is it my intention to force into anything, but it is rather an open discussion between photographers in which one tries to give and exchange information, the way this forum works and should be for.

My apology if you felt so, Sergei.

Best regards
Thierry

I always find these conversations funny. Member X says I do this and it works for me, and a whole bunch of members come on and say you should not do that because we do it a different way and our way is better. Which begs the question, better for whom? And then when you look at Sergei's work, it is kind of hard to criticize his method.
 
Last edited:
P

Paul66

Guest
I'm kinda enjoying the discussion since I see it as we are al still learning no matter how advanced one is.
 
J

jeffacme

Guest
I tried a friends light meter and it's dead on, not sure what happened with mine but I guess electronics just fail.
Thanks for all the advise and thanks for not bashing my ignorance for tech terms, I am self taught and learned trial and error
Paul,

So much of the old school theory is vanishing. Developing a Mental/Experiential warning system is key to consistent results and spotting trouble before it affects the outcome of a shoot.

In pure strobe situations it all comes down to simple math. Output level A+distance B=f8 Changing the output setting from 6.0 to 7.0 or moving the light from 5.6' to 4'=f11. The reason many Graybeard film shooters never use flash meters is this understanding of how output and distance relate to exposure. This warning system also would have told you that something was up either with your meter or strobe output. Metering several lights could have eliminated the output as the culprit and called into question the accuracy of the meter.

While distance to subject affects output Johnny is correct in pointing out it also affects the quality of the light on the subject and should be considered.

As many have said the histogram is king but I would go a step further and say that it is always used to confirm what your M/E warning system is already telling you.

If you want total control over of every pixel it is important to become an accomplished practitioner of curves. Any cast can be removed and perfect, at least to your liking, color balance and contrast will always be in your control. Read everything you can by Dan Margulis and experiment with CMYK retouching especially for skin tones. It is easy once you have the chops to balance any skin tone when you can break the red channel down to it's component parts Y and M.
 
Top