exposure metering: slide film vs digital

[Ken Fong]

I was wondering if anyone can explain the following. Sorry, I can't post samples right now, but I will try to describe what I am experiencing...

Using a film camera, polarizer, Velvia 50, and a handheld spot meter (set for ISO 50), I am practicing a method of metering sky and clouds. I meter the value off of the brightest portion of the cloud, then add 2 stops to place it in zone VII, plus I bracket +1 for insurance. The results show that placing the bright portion of the cloud into zone VIII (eight) looks about right with no blown out highlights introduced. So I concluded that I can continue this metering method (place in zone VII with +2 stops, then bracket +1 and +2 for insurance.)

Now getting to the question...
I set up the DSLR with the polarizer and ISO 100. I use the same method as above (handheld spot meter off the brightest portion of the cloud and add 2 stops to place it in zone VII. I made sure I adjusted the handheld spotmeter to ISO 100.) The results are blown out clouds...I won't even think about putting it in zone VIII. So why does the method work for slide film, but blows out the results for digital? Is it related to film reciprocity or user error?
I didn't think reciprocity is a factor at these speeds:
f/8 at 1/30 and 1/15 looked fine for Velvia 50
f/8 at 1/30 thru 1/80 blew out the highlights, started to look normal at 1/100.

(Shots were taken on different days, the only thing that is constant is the metering method described above.)

thanks in advance,
Ken

[ScottE]

I have no idea why the exposures would be so different between Velvia and a DSLR. If anything the DSLR should have a greater exposure range than Velvia which is a high contrast range and can easily blow out the highlights.

I also don't know why you would bother with a hand held light meter when you have a histogram available on your camera. Why not just meter with the camera, take a test shot and then look at the histogram. If you want white whites the histogram graph should be to the right of the range, but not over. If it is not, adjust your exposure and take another shot until you get it right. This can't be any slower than metering by hand.

In any case, it doesn't cost anything to bracet with digital and if you shoot RAW you can fine tune exposure when you get back to the computer.

Could you explain what you hope to gain by using a hand meter?

Scott

[PhotosGuy]

Film & digital worked differently when I ran some tests. Take a look at:
http://photography-on-the.net/forum/showthread.php?t=54281
&
http://photography-on-the.net/forum/showthread.php?t=58677

[Ken Fong]

I also don't know why you would bother with a hand held light meter when you have a histogram available on your camera. Why not just meter with the camera, take a test shot and then look at the histogram. If you want white whites the histogram graph should be to the right of the range, but not over. If it is not, adjust your exposure and take another shot until you get it right. This can't be any slower than metering by hand.

In any case, it doesn't cost anything to bracet with digital and if you shoot RAW you can fine tune exposure when you get back to the computer.

Could you explain what you hope to gain by using a hand meter?

Scott

Thanks Scott. I am actually trying to gain experience with the hand held so I can take better slide photos (MF and 4x5). I do prefer using the histogram for the DSLR, but I am just curious why the handheld method is not quite translating over. It would be nice to have the DSLR act as my proofer. I might just end up keeping the two metering techniques separate.

[kawter2]

MAN I HAVE TO SIT DOWN AND READ THOSE TWO ARTICLES

oops sorry about shouting

[J Rabin]

I was wondering if anyone can explain the following. So why does the method work for slide film, but blows out the results for digital?
OK Ken. I hope this becomes an interesting thread!
What I want to know is HOW DO YOU KNOW THE VELVIA'S NOT BLOWN OUT? For example, when I scan chromes, the Nikon 5000 scanner auto exposure reads the slide, and automatically moves blown out pixels (i.e., 255) lower, between 245 and 253. Looking at the slide for loss of highlight detail is different than reading it post scan. Converesly, with the Canon .CR2 RAWs amazing ability to recover about +/- 2 stops of highlight detail in the RAW converter, "blown out" needs qualifying in comparison with looking at .jpgs, which are contrast range "challenged:" a more limited, different beast entirely.
Let's here more on how you are setting this up. J.

[Ken Fong]

OK Ken. I hope this becomes an interesting thread!
What I want to know is HOW DO YOU KNOW THE VELVIA'S NOT BLOWN OUT? For example, when I scan chromes, the Nikon 5000 scanner auto exposure reads the slide, and automatically moves blown out pixels (i.e., 255) lower, between 245 and 253. Looking at the slide for loss of highlight detail is different than reading it post scan. Converesly, with the Canon .CR2 RAWs amazing ability to recover about +/- 2 stops of highlight detail in the RAW converter, "blown out" needs qualifying in comparison with looking at .jpgs, which are contrast range "challenged:" a more limited, different beast entirely.
Let's here more on how you are setting this up. J.

Thanks J. Okay, I'll need pre-empt my questioning until I can get some scans done. I use the term 'blow out' to refer to loss of detail in the clouds...where it turns pure white. I've been only judging the quality of the slide thru the light table and loupe. I didn't realize looking at jpgs is quite different than looking at slides with a loupe...perhaps you've touched on the answer already.

[PacAce]

Ken, have you tried checking the hand held meter reading of a mid-tone gray to see what you get on the camera? Maybe the ISO of the camera (or the meter) is off by a fraction of a stop or more between the meter and the camera.

[Ken Fong]

Ken, have you tried checking the hand held meter reading of a mid-tone gray to see what you get on the camera? Maybe the ISO of the camera (or the meter) is off by a fraction of a stop or more between the meter and the camera.

Thanks PacAce...I will (and should) do some sort of a handheld to DSLR meter calibration check per your recommendation. This last experiment however used handheld readings only for both cameras (did not use the DSLR meter). The meter method seemed to work well for slide, but did not 'look' well for jpg. I thought I had adjusted for the ISO difference, but there might be some other factor I overlooked (e.g. exposure comp was left on or something like that). I hope to set up a more controlled comparison and post in late June if I am still experiencing the discrepancy. I really would like to be using the DSLR as the proofer for the slide photography...but I think it's safer to keep the metering methods separate for now, and bracket. (I also have to sit down and read the threads presented by PhotosGuy.)

[ScottE]

I take it you are shooting a 4x5 view camera for film.

When you are using a polarizer filter how do you calculate the compensation for the loss of light due to polarization? The method you describe of taking a reading off the brightest part of the clouds and then adding two stops will generally give you an exposure that will make the brightest part of the clouds white when not using a polarizer. Depending on the orientation of the polarizer you would have to add another zero to two stops to compensate for the polarizer.

When I shoot 35 mm or digital I use a similar technique for winter landscapes, but I use the through the lens meter on the camera so the compensation for a polarizer is already metered.

I am just curious about how you make this correction using a hand held meter. I haven't used a meter since the 70's when I got my first SLR with auto exposure. If I remember correctly with the meter I had you had to hold it in the light you were going to shoot and take a reading there since it measured incident light. You apparently have a more modern meter that takes spot readings of reflected light, similar to the TTL reading in an SLR camera.

Scott

[Ken Fong]

I take it you are shooting a 4x5 view camera for film.

When you are using a polarizer filter how do you calculate the compensation for the loss of light due to polarization? The method you describe of taking a reading off the brightest part of the clouds and then adding two stops will generally give you an exposure that will make the brightest part of the clouds white when not using a polarizer. Depending on the orientation of the polarizer you would have to add another zero to two stops to compensate for the polarizer.

When I shoot 35 mm or digital I use a similar technique for winter landscapes, but I use the through the lens meter on the camera so the compensation for a polarizer is already metered.

I am just curious about how you make this correction using a hand held meter. I haven't used a meter since the 70's when I got my first SLR with auto exposure. If I remember correctly with the meter I had you had to hold it in the light you were going to shoot and take a reading there since it measured incident light. You apparently have a more modern meter that takes spot readings of reflected light, similar to the TTL reading in an SLR camera.

Scott
I am lazy. Since I tend to be forgetful about compensations..I placed a 30.5mm (circular) polarizer on the handheld also, while the (non-AF) camera had a linear polarizer...both seem to have the same filter factor. There might be slight variance between the two, but I figured this was close enough...I'm also careful to make sure the polarizer on the camera/handheld are in the same orientation. It's a Sekonic spot meter...it can take incidental readings like you described, but everyting talked about here was with the spot. I ended up buying one because the 10D -20D only had a 'partial' meter (which is fine for most situations) and it was seriously needed for the 4x5 work.

Metering off the brightest portion of the cloud (with polarizer), then adding 2 stops yields a nice dramatic cloud - good depth and detail in the highlights; however, the overall exposure looks underexposed, which is problematic for a high contrast photo.
Adding 3 stops yields a slightly less dramatic cloud...but still nice, and the overall exposure benefits from the the extra stop...these is a lesser chance of the shadows blocking up. I would probably end up going with the +3 shot.
Can anyone confirm if this general practice of placing clouds (bright portion) in zone VIII (and bracket upward) is practical? (The reason I am metering off clouds is because clouds are important in my compositions and I'm using slide film.) BTW, I'm referring to cumulus clouds.

Yes, the 4x5 photography is all film (so far, all Velvia)...can't afford anything near a digital back.

[PhotosGuy]

A rule of thumb for film exposure was expose for the shadows & develop for the highlights. It's a very personal thing & you'll eventually find a method that works for you. Since I tend to be forgetful about compensations..I placed a 30.5mm (circular) polarizer on the handheld also Well, if you can remember that you made this change, find out how much the CP affects your exposure. Probably 2X overall. Then just adjust the meters ISO setting down to compensate for it.

[Ken Fong]

A rule of thumb for film exposure was expose for the shadows & develop for the highlights. It's a very personal thing & you'll eventually find a method that works for you. Well, if you can remember that you made this change, find out how much the CP affects your exposure. Probably 2X overall. Then just adjust the meters ISO setting down to compensate for it.

Thanks Photosguy. That rule works well for print film and I guess I should be doing that for dSLR metering also; is it the opposite for slide film?: "expose for the mid to highlights, then let the shadows take care of themselves". If I had to choose between a blown-out highlight or a blocked-up shadow, I'm supposed to choose the blocked-up shadow? RAW can recover those highlights fairly well, but a blown-out highlight in slide film is supposed to have no recoverable data.
I'm slowly learning more about the limiting characteristics of Velvia.

[PhotosGuy]

That rule works well for print film and I guess I should be doing that for dSLR metering also; is it the opposite for slide film?
I didn't usually have time to "measurebate" when I shot slides & was a bracketing fool! Here's something on that from a previous post:
Relating to transparency film, there are (were?) other valid reasons for bracketing:
1. I once worked with a stereotypical German photographer. Verrrry precise in his calculations. He came back from a 2-day shoot with 3 rolls of 36. Unfortunately, on the trip out, something glitched the diaphragm on his primary lens & it was stuck on wide open. The client was not amused, & who could disagree with him? After all costs for me, transportation, lodging (except for the time I billed $12.00 total for TWO nights in the "Hotel"!), etc are added up, film is the smallest consideration. So I used it - typically 60+ rolls of 36 a day. (I'll put a typical "workflow" at the bottom.)

2. Bracketing will produce "The" correct exposure.
3. Bracketing gives extra close exposures that fit nicely into your Sample files.
4. Bracketing gives extra close exposures that the art directors & clients can fingerprint & scratch.
5. Bracketing gives extra close exposures when the inevitable dust speck embeds itself onto the "money Shot".
6. Bracketing allows one situation to be spread out over several rolls. See "workflow" for why this is important.

Workflow:
Before you go, get all the film for that job in the same batch number, hopefully from the same vendor. We had freezers filled with it. (Pro films were 'aged' to be at their peak right now!). Shoot & process one roll in each body to be sure everything's OK before you go. When you get there, check everything out.

At the location:
Situation 1: First primary subject - seven stop bracket using f-stops. First variation on the primary subject - 5-stop bracket using shutter speeds. Next variation - 5-stop bracket using another body. (OK, so I'm insecure, but I've seen it all happen. How about the time in Italy when the town photo processor was asked to make 8X10" contact sheets of B&W film we'd shot over the weekend to identify locations we'd be using that week? He used his lovely deckel edge trimmer to cut out every shot & gave us 105 1" X 1-1/2" prints... without edge numbers.)

Situation 2: Second primary subject - repeat as above. Maybe drop down to 3-stop brackets if I shoot a lot of variations & my level of confidence that I've got it is up.

After the shoot:
NEVER ship more than 20% of the film out in one batch.
NEVER process more than 20% of the film in one batch.
NEVER process more than ½ the rolls of a variation in one batch.

Bottom line, bracketing was just carrying CYA to it's logical conclusion… the client doesn't care that "the dog ate your pics". If it's even slightly under your control, then you had better have controlled it! Now, with histograms, a Gig of mem cards, & back-up film SLRs, things can be more relaxed. Sigh!

[Ken Fong]

Thanks again Photosguy...sounds like bracketing is the ultimate insurance and it doesn't cost that much more. I'll be bringing about 20 rolls of 220 film (Velvia/Provia) to China. I need to get my confidence up so I can bracket less, but I will definitely bracket more when in doubt. I'll miss the autobracket feature of the dSLRs (wife is using that one).

[PhotosGuy]

Thanks again Photosguy...sounds like bracketing is the ultimate insurance and it doesn't cost that much more. Just remember that I didn't have to pay for the film & processing! ;-)

[rdenney]

Using a film camera, polarizer, Velvia 50, and a handheld spot meter (set for ISO 50), I am practicing a method of metering sky and clouds. I meter the value off of the brightest portion of the cloud, then add 2 stops to place it in zone VII, plus I bracket +1 for insurance. The results show that placing the bright portion of the cloud into zone VIII (eight) looks about right with no blown out highlights introduced. So I concluded that I can continue this metering method (place in zone VII with +2 stops, then bracket +1 and +2 for insurance.)

Now getting to the question...
I set up the DSLR with the polarizer and ISO 100. I use the same method as above (handheld spot meter off the brightest portion of the cloud and add 2 stops to place it in zone VII. I made sure I adjusted the handheld spotmeter to ISO 100.) The results are blown out clouds...I won't even think about putting it in zone VIII. So why does the method work for slide film, but blows out the results for digital? Is it related to film reciprocity or user error?


You have found something that I have found, though there is remarkably little written about it. I think the ISO speeds are off when compared to most films. I routinely set my 10D to underexpose by 1 or 1.5 stops. Digital is like slides in that once the highlights are blown out they are gone forever. I'd generally rather lose shadow detail in both cases, but digital will hide tonality in shadows much better than slide film, and these can be pulled up later.

But I also check the histograms, using that to determine that the light values in the scene are within the data range of the image, rather than trying to evaluate it solely on how it looks on the LCD, which to me is not terribly effective for that.

With my normal correction, I would provide 2 or 2.5 stops less exposure with the DSLR at ISO100 than with ISO50 Velvia.

Remember that the DSLR has a wider dynamic range than slide film, and ISO film speeds are designed for middle gray. Thus, if you align the exposure based on ISO, the highlights will be a stop or two brighter and the shadows a stop or two darker than what you would get with Velvia. Theoretically, you can pull them back in processing, but, as I said before, a blown highlight is gone forever. If you set an exposure based on the high zones (VIII or more), you have to compensate.

I learned this on the copy stand. Setting exposure based on ISO and an incident flash meter resulting in images with white paper backgrounds that could not be displayed with tonality without a lot of processing. 1.5 stops of downward compensation solved the problem.

Rick "who keeps a close eye on the histograms" Denney

[Ken Fong]

Thanks Rick. I tried it again (the method of metering a bright highlight off a cloud, then adding 2 stops to place it in zone VII)...I transferred the results to the 20D (again, not using the 20D meter)...ISO's were set to 100. The result was not blown out this time, but still did not show the 'depth' I was used to seeing on slides; however, the histogram was healthy. Then, I changed the angle of the 20D's LCD and saw more depth (tonal differences)! (Turning the brightness down did not help.) I think what you are saying about some sort of underexposure is still valid however, because the "zone VII" cloud in the 20D still looked brighter than the "zone VIII" cloud (placing cloud highlight in zone VIII) in Velvia...but then again, I am not using the best devices to view these things, so I will have to wait until they are scanned later. It's possible that the "zone VIII" cloud really is blown out but doesn't appear so when viewed on a light table. I think the lesson from this round is that I should not rely on the 20D's LCD image...just use the histogram.

[PhotosGuy]

I think the ISO speeds are off when compared to most films. Then there were variations in shutter speed, metering variations, reciprocity variations, processing/replenishing variations. etc... I SO love digital! ;-)

[rdenney]

Then there were variations in shutter speed, metering variations, reciprocity variations, processing/replenishing variations. etc... I SO love digital! ;-)

And don't forget the monstrously S-shaped characteristic curve.

Most people don't even know what to do with the linearity of digital.

Rick "who likes being able to decide at print time where that S needs to be" Denney

[kellmeister]

I'm waiting on the arrival of my 20d which will be my first digital. This thread has been helpful as I shoot alot of slide film. I've seen alot of digital photos posted that have blown out highlights, which has got me scared regarding the dynamic range. I hope some of my film knowledge will help me out in this.

[PhotosGuy]

I've seen alot of digital photos posted that have blown out highlights, which has got me scared regarding the dynamic range. Learn to get the most out of RAW & how to post process different "exposures" derived from a RAW file & you'll be surprised at how much detail you can retain. I do like the film look a lot better, but find that digital has too many +s going for it to ignore the advantages.