# negative flashing technique

18 replies to this topic

### #1 pramod varma

pramod varma

New

• Basic Members
• 5 posts
• Cinematographer
• India

Posted 28 April 2009 - 04:09 AM

hi,
i would be grateful if found the answer how to co-relate the exposure and the effect of density on the negative while flashing? how to determine a 10%flash or 20%flash negative.
thanks..
• 0

### #2 Dominic Case

Dominic Case
• Basic Members
• 1357 posts
• Other
• Sydney Australia

Posted 28 April 2009 - 07:33 AM

how to determine a 10%flash or 20%flash negative

10% or 20% is a misleading term. Although it is taken to mean 10% or 20% of the maximum density of the negative above d-min, it is often simply measured as a density of 0.10 or 0.20 above d-min in each colour (which is about one third less than the strict percentage measure).

Either way, you need to do a test to get the precise result. But if you want to calculate a starting point for your exposure, look at it this way:

Shooting a 18% grey card at normal exposure produces a density of about 0.70 above d-min in each colour. Now, look at the characteristic curve for the stock in question, (I chose 5212 for this exercise) and find the point on (say) the green curve which is 0.70 above d-min. Follow the line back to a point where it is just 0.20 above d-min. You have moved by four squares on the chart, which is 0.80 logE. That is 2 2/3 stops (roughly). So to get that density you need to shoot an 18% grey card 2 2/3 stops underexposed.

For 0.10 density flashing, you'd need to be about one stop less than that.
• 0

### #3 Paul Bruening

Paul Bruening

(deceased)

• Sustaining Members
• 2858 posts
• Producer
• Oxford, Mississippi

Posted 28 April 2009 - 08:40 AM

THANK YOU DOMINIC!

I was going to shoot a test, flashed roll in 1 stop increments from normal to 8 below and try to determine some form of subject charting. The existing terminology (%'s) applied to film and lab factors. But, what did that all mean for some guy cranking an aperture ring? Your explanation and numbers really help make sense of flashing for all of the shooters here. On behalf of all of us, thanks again.
• 0

### #4 K Borowski

K Borowski
• Basic Members
• 3905 posts
• Camera Operator
• I.A.T.S.E. Local # 600 Eastern Region

Posted 28 April 2009 - 08:49 AM

Yes Paul, it is confusing when trying to correlate log numbers (based on powers of 10) with arithmetic exposure (doubling of aperture area). They don't really line up neatly.

I wouldn't recommend that YOU play around with flashing though, Paul, because your "base fog level" is already dangerously high :-p

Labs really ought to just put it in terms of stops of flash. They now how to convert. We math-stupid cinematographers don't!
• 0

### #5 David Mullen ASC

David Mullen ASC
• Sustaining Members
• 18832 posts
• Cinematographer
• Los Angeles

Posted 28 April 2009 - 10:53 AM

All that matters is how it looks -- no one really knows what an abstract "10%" flash looks like anyway. That's why you test, so that whatever the labelling system used, you can pick the strength of effect you want.

Just note that a flash is more visible in shots with more blacks and shadows in the frame, which is why a light flash may be too strong for a night shot but too subtle for a day shot.
• 0

### #6 K Borowski

K Borowski
• Basic Members
• 3905 posts
• Camera Operator
• I.A.T.S.E. Local # 600 Eastern Region

Posted 28 April 2009 - 11:08 AM

All that matters is how it looks -- no one really knows what an abstract "10%" flash looks like anyway. That's why you test, so that whatever the labelling system used, you can pick the strength of effect you want.

I disagree David. Understanding, scientifically, how something works eliminates a lot of the pain-staking trial-and-error, and guesswork of having to go out and figure out what works optimally.

Now, that in mind, I am not saying that understanding the stop equivalent of a flash eliminates the need for testing altogether, rather, it just gives you a good starting point.

Look at it this way: Is having an ISO speed rating on films (which is often higher than the film's true speed actually is) helpful as opposed to, say, the way film used to be sold, without any speed rating whastoever?

Then you'd have to literally test at every possible speed until you could work out the correlation for yourself, and that would be with every emulsion batch.

Ultimately, understanding the science doesn't eliminate the need to test, but it sure does speed up the testing process when you have an understanding of the science behind it. . .
• 0

### #7 Karel Bata

Karel Bata
• Basic Members
• 487 posts
• Director
• London - a rather posh bit

Posted 28 April 2009 - 01:31 PM

Just how accurate can you expect the speed printed on the can to be? I would have thought that knowing the speed is particularly important here and you need to do a clip test of some sort to ascertain that.

I've never flashed, but I have used Freddie Francis' Varicon (a step on from his lightflex) which does much the same thing, except you can see the difference as you're shooting in the viewfinder and dial in degrees of it. http://web.mac.com/m...7EAF52708E.html and http://www.cinematog...ges/VARICON.HTM

Quite strangely this pic shows it upside down, as this way up it would fall out (unless you gaffer taped it in!)

You can also slip gels into the Varicon to color your shadows. Here's a still with some orange in. Without it half that bookcase would be black. saves a lot of trouble rigging fill lights, particluarly in awkward corners. You have to keep scrupulous notes as you go along, otherwise you'll have difficulty matching shots later. Let's see if this uploads...

Ah!

Freddie Francis won Best Cinematography Oscar using it on Glory. Such a nice man too. Very much missed.
• 0

### #8 David Mullen ASC

David Mullen ASC
• Sustaining Members
• 18832 posts
• Cinematographer
• Los Angeles

Posted 28 April 2009 - 02:28 PM

I disagree David. Understanding, scientifically, how something works eliminates a lot of the pain-staking trial-and-error, and guesswork of having to go out and figure out what works optimally.

I've flashed three features, so I know something about the process, and I can tell you that it is a lot more subjective than you think since the effect is modified by scene content.

You can say "I'm going to flash 10%" based on the scientific definition provided by Dominic... but until you shoot tests and see what "10%" looks like, you're talking out of your ass. And even after you shoot lots of tests, it still creates some variation in effect on the shooting day depending on shot contrast and content.

It's not hard to understand "scientifically" how a flash works. It's a very simple photographic concept and technique. But you are presumably using it for a visual effect on real world subjects, hence why testing and judging the results by eye becomes a lot more important and practical than simply knowing the science behind it.

Besides, my main point is that a "10% flash" is rather meaningless because flash effects vary by lab and by process. A "10% flash" is not going to look the same when done on a Panaflasher versus doing it at Deluxe versus doing it at Technicolor. Hence why you have to test and see the results to figure out what level to use.

Not only do different labs and processes define a percentage of flash differently, but what film stock you use, and how you rate it, and how you process it, and how you expose it create further variations. If you flash by "10%" but you accidentally underexpose by 1/3 of a stop, then that percentage is now different relative to the subject's density.

Honestly, testing and judging the results is the only way to determine the degree of flashing you want to use, and even then, you can't test for every variation possible. Trying flashing a scene with smoke in the air and figuring out what percentage to use as you go to longer focal lengths. It's completely a guess at that point.
• 0

### #9 K Borowski

K Borowski
• Basic Members
• 3905 posts
• Camera Operator
• I.A.T.S.E. Local # 600 Eastern Region

Posted 28 April 2009 - 02:52 PM

I've flashed three features, so I know something about the process, and I can tell you that it is a lot more subjective than you think since the effect is modified by scene content.

You can say "I'm going to flash 10%" based on the scientific definition provided by Dominic... but until you shoot tests and see what "10%" looks like, you're talking out of your ass. And even after you shoot lots of tests, it still creates some variation in effect on the shooting day depending on shot contrast and content.

It sounds as if you are taking this as a personal insult. That wasn't my intent. I've never even shot a project on 35mm, so obvioulsy you have more experience than I do in this arena.

Also, I've never flashed, because I honestly wouldn't want to risk using this dangerous, unpredictable process.

Then again, I always try to shoot clean and accomplish this type of thing in an intermediate stage.

What I'm saying is that, understanding a starting point is easier than having to test every conceivable combination.

You have to do the flashing yourself now, usually, because a lot of labs don't want to be responsible for the results being bad.

I know that flashing immediately before the subject exposure looks completely different than flashing immediately after, too, because the first exposure "takes precedent" over the second.
• 0

### #10 David Mullen ASC

David Mullen ASC
• Sustaining Members
• 18832 posts
• Cinematographer
• Los Angeles

Posted 28 April 2009 - 03:25 PM

I know that flashing immediately before the subject exposure looks completely different than flashing immediately after, too, because the first exposure "takes precedent" over the second.

Not sure if that is really true -- the Panaflasher when back-mounted on a Panaflex is pre-flashing (I believe) while when top-mounted is post-flashing, but the effect is the same IF you test to determine the levels needed to match -- because the distance to the film is different I think. The instructions for the Panaflasher have a compensation factor when top versus back mounting. But once you match the level, the effect looks the same either way.

Whereas a Varicon on the lens is flashing simultaneously with the exposure. Now that seems to have a slightly different affect on contrast for some reason. The added light coming in along with the picture's light seems to be more effective at lifting up detail in the shadows than normal flashing does.

See Mark Wood's article on flashing:
http://web.mac.com/m...7EAF52708E.html
• 0

### #11 Karel Bata

Karel Bata
• Basic Members
• 487 posts
• Director
• London - a rather posh bit

Posted 28 April 2009 - 04:55 PM

Whereas a Varicon on the lens is flashing simultaneously with the exposure. Now that seems to have a slightly different affect on contrast for some reason. The added light coming in along with the picture's light seems to be more effective at lifting up detail in the shadows than normal flashing does.

Could be because you can actually see the effect you're more inlined to creep the level up till it looks good. There's no '10%' or whatever on a Varicon - you just turn the knob up until you get what you want. I would imagine with flashing you'd be more inclined to err on the side of caution. Freddie Francis took it to the limit on Dune, and the subject suited it, but he was much more subtle on Glory. I like the fact that you can switch colors for different scenes too.

Not sure if technically it's really flashing though, because it's a bit like a fog filter.

Here's the Varicon on Dune used lightly (well realatively so!)

And here it's turned up quite a lot!

Be hard to achieve those different effects with varying levels of conventional flashing on a film, I would have thought. With a Varicon you just dial it in! Anyway, I loved it, but like a lot of other neat tricks we took the trouble to learn, it's largely been made redundant by DI...
• 0

### #12 David Mullen ASC

David Mullen ASC
• Sustaining Members
• 18832 posts
• Cinematographer
• Los Angeles

Posted 28 April 2009 - 05:30 PM

OK, tell me how you were able to get an HD frame grab?

Any good way to grab frames off of a Blu-Ray disc? For PC and for Macs?
• 0

### #13 K Borowski

K Borowski
• Basic Members
• 3905 posts
• Camera Operator
• I.A.T.S.E. Local # 600 Eastern Region

Posted 28 April 2009 - 05:42 PM

OK, tell me how you were able to get an HD frame grab?

I am similarly jealous.

Best be careful posting Dune grabs here too. Phil is a HUGE Dune fan; you're going to turn him on posting in HD like that
• 0

### #14 Karel Bata

Karel Bata
• Basic Members
• 487 posts
• Director
• London - a rather posh bit

Posted 28 April 2009 - 06:03 PM

All I did was Google Dune. A clue is in the bottom right: hdmovies.co.nz

But since you ask, for frame grabs for a PC I use captureur: http://www.download3...eur-i42022.html or play back in Quicktime and export to a still. No idea if either would work for HD.

Quicktime saves as a pict file. To convert graphics files quickly I use Picaview: http://asia.cnet.com...9434070s,00.htm right click, convert.
• 0

### #15 David Mullen ASC

David Mullen ASC
• Sustaining Members
• 18832 posts
• Cinematographer
• Los Angeles

Posted 28 April 2009 - 06:05 PM

All I did was Google Dune. A clue is in the bottom right: hdmovies.co.nz

But since you ask, for frame grabs for a PC I use captureur: http://www.download3...eur-i42022.html or play back in Quicktime and export to a still. No idea if either would work for HD.

Quicktime saves as a pict file. To convert graphics files quickly I use Picaview: http://asia.cnet.com...9434070s,00.htm right click, convert.

So what was your approach for those frames from "Dune"?
• 0

### #16 Karel Bata

Karel Bata
• Basic Members
• 487 posts
• Director
• London - a rather posh bit

Posted 28 April 2009 - 06:11 PM

I should have checked that Captureur link before posting it. Here's a better one: http://www.brotherso...ureur-6197.html

So what was your approach for those frames from "Dune"?

http://hdmovies.co.nz/

Edited by Karel Bata, 28 April 2009 - 06:14 PM.

• 0

### #17 pramod varma

pramod varma

New

• Basic Members
• 5 posts
• Cinematographer
• India

Posted 29 April 2009 - 01:26 AM

10% or 20% is a misleading term. Although it is taken to mean 10% or 20% of the maximum density of the negative above d-min, it is often simply measured as a density of 0.10 or 0.20 above d-min in each colour (which is about one third less than the strict percentage measure).

Either way, you need to do a test to get the precise result. But if you want to calculate a starting point for your exposure, look at it this way:

Shooting a 18% grey card at normal exposure produces a density of about 0.70 above d-min in each colour. Now, look at the characteristic curve for the stock in question, (I chose 5212 for this exercise) and find the point on (say) the green curve which is 0.70 above d-min. Follow the line back to a point where it is just 0.20 above d-min. You have moved by four squares on the chart, which is 0.80 logE. That is 2 2/3 stops (roughly). So to get that density you need to shoot an 18% grey card 2 2/3 stops underexposed.

For 0.10 density flashing, you'd need to be about one stop less than that.

THANKS DOMINIC!!
I am going to shoot my tests based on the statical data you have explained.I am glad to have met u here.Thanks once again for your support!
• 0

### #18 pramod varma

pramod varma

New

• Basic Members
• 5 posts
• Cinematographer
• India

Posted 29 April 2009 - 01:45 AM

I've flashed three features, so I know something about the process, and I can tell you that it is a lot more subjective than you think since the effect is modified by scene content.

You can say "I'm going to flash 10%" based on the scientific definition provided by Dominic... but until you shoot tests and see what "10%" looks like, you're talking out of your ass. And even after you shoot lots of tests, it still creates some variation in effect on the shooting day depending on shot contrast and content.

It's not hard to understand "scientifically" how a flash works. It's a very simple photographic concept and technique. But you are presumably using it for a visual effect on real world subjects, hence why testing and judging the results by eye becomes a lot more important and practical than simply knowing the science behind it.

Besides, my main point is that a "10% flash" is rather meaningless because flash effects vary by lab and by process. A "10% flash" is not going to look the same when done on a Panaflasher versus doing it at Deluxe versus doing it at Technicolor. Hence why you have to test and see the results to figure out what level to use.

Not only do different labs and processes define a percentage of flash differently, but what film stock you use, and how you rate it, and how you process it, and how you expose it create further variations. If you flash by "10%" but you accidentally underexpose by 1/3 of a stop, then that percentage is now different relative to the subject's density.

Honestly, testing and judging the results is the only way to determine the degree of flashing you want to use, and even then, you can't test for every variation possible. Trying flashing a scene with smoke in the air and figuring out what percentage to use as you go to longer focal lengths. It's completely a guess at that point.

Hello sir,
I always wondered if i could get your views on this topic and i think you have provided the best possible PRACTICAL way to arrive at a decision.Thanks!
I have seen two of your features"NORHTFORK" & "ASTRONAUT FARMER".Truly inspiring works.
• 0

### #19 K Borowski

K Borowski
• Basic Members
• 3905 posts
• Camera Operator
• I.A.T.S.E. Local # 600 Eastern Region

Posted 29 April 2009 - 08:10 AM

I always wondered if i could get your views on this topic and i think you have provided the best possible PRACTICAL way to arrive at a decision.Thanks!

I don't see why one approach has to be mutually exclusive of the other.

Nor do I recall either myself or Dominic saying that one should just follow the statistical, densitometric response that has been observed in flashed stock and just go off and take it on faith that your stock is going to behave the same way.

I think knowing the science gives you a better idea of what is still "inside the ballpark" so you don't have to do as *much* testing.

So, Pramod, you still need to test. . .
• 0