PDA

View Full Version : Lens flare as "contrast contol" ?



Ken Lee
19-Mar-2015, 11:21
My question is spread across 2 posts.

Given that a sheet of film can hold a lot of tonal information, I wonder if people have been right to describe flare-prone lenses as having less contrast.

Here I have started with a 16-bit image, added "flare", then removed it.

http://www.kenleegallery.com/images/forum/16-bit original.png
16-bit original

http://www.kenleegallery.com/images/forum/16-bit adjusted.png
16-bit adjusted: flare added

http://www.kenleegallery.com/images/forum/16-bit re-adjusted.png
16-bit re-adjusted: flare removed

We added flare (which pushed the low values up and compressed all the values together) and then removed it: There may be gaps in the resulting histogram, but they can't easily be detected.

Ken Lee
19-Mar-2015, 11:22
Here I started with an 8-bit version of the same image, added "flare", then removed it.

http://www.kenleegallery.com/images/forum/8-bit adjusted.png
8-bit adjusted: flare added

http://www.kenleegallery.com/images/forum/8-bit re-adjusted.png
8-bit re-adjusted: flare removed

Because the image is 8-bits only there are gaps in the final histogram - but on an 8-bit monitor the resulting image looks the same as the original.

If we stick with analog process or 16-bit digital capture, can a flare-prone lens serve as a valid method of controlling contrast ?

Kirk Gittings
19-Mar-2015, 11:40
From snippets I have picked up over the years (can't tell you where), many practitioners around the time when multi-coatings were invented and implemented preferred the older single coated lenses for just this reason. The newer lenses were just too contrasty.

Old-N-Feeble
19-Mar-2015, 11:53
I would like to see a test of an uncoated (or any) lens known to have lots of flair vs. a multicoated lens of the same focal length and known to have extremely low flair. Then I'd like to see the differences in prints after contrast correction via fully analog wet process.

Taija71A
19-Mar-2015, 11:55
... If we stick with analog process or 16-bit digital capture, can a flare-prone lens serve as a valid method of controlling contrast ?
__

Yes!

What you have tried to 'simulate' Ken, would be exactly the same...
As adding the proper amount of 'Pre-Exposure' of your Film (*for 'Contrast Control').
--
In an 'Analog Only' workflow... This is most certainly a very, valid option.
One of the best 'treatises' on this subject... Was discussed in the Book:

"Way Beyond Monochrome" by Ralph Lambrecht.

http://www.waybeyondmonochrome.com/WBM2/TOC_files/PreExposureEd2.pdf

--
Best regards,

-Tim.
_________

Drew Wiley
19-Mar-2015, 12:19
I'm having that same schizophrenic debate in my own head right now. This time of year I sometimes covet soft open shadows rather than hard ones, and prefer
the older lens look. Guess I could pack both 360's for my Saturday walk, but my ankle sprain hasn't completely healed, so I'll need to compromise pack weight somewhat. I pampered the ankle last weekend and just shot the Nikon. Developed those negs last nite; and there are sure a couple of shots I'l like to go back and
re-shoot with the 8x10.

Drew Wiley
19-Mar-2015, 12:22
Pre-exposure or "flashing" is a rather poor option in my opinion because it muddies the microtonality in the shadows. I certainly understand it. Too blunt an instrument for me. I'd rather do more subtle controls via film development tweaks, careful lens choice, or perhaps unsharp masking. In other words, I like to have
my cake and eat it too.

Kirk Gittings
19-Mar-2015, 12:24
Pre-exposure or "flashing" is a rather poor option in my opinion because it muddies the microtonality in the shadows. I certainly understand it. Too blunt an instrument for me. I'd rather do more subtle controls via film development tweaks, careful lens choice, or perhaps unsharp masking. In other words, I like to have
my cake and eat it too.

Me too. Back when I was learning the ZS I tested that technique rather extensively and no matter how I did it I never liked the "look" of it.

Taija71A
19-Mar-2015, 12:29
... Then I'd like to see the differences in prints after contrast correction via fully analog wet process.
__

I was just wondering...

How do you propose to do this corresponding 'Contrast Correction' in the Darkroom.
Just curious...

Thank-you!
--
Best regards,

-Tim.
_________

Ken Lee
19-Mar-2015, 12:39
Here are some numbers I am pulling out of my hat. Are these right ?

Let a subject start out at 100% contrast. A lens with low flare delivers 85%. A lens with a lot of flare delivers... 75% ?

In other words, how much are we talking about anyway ?

In my digital "test" I tossed out the lower 50% of the tonal scale and got enough of it back again in the 16-bit version: that's a lot of loss and recovery. Actually, it wasn't tossed out: that would be under-exposure. Instead it was compressed up into higher 50% of the histogram.

Under darkroom conditions, I presume the same would be possible, even better perhaps since film and paper are analog, or so continuous that they appear analog.

On the other hand, if we were using a poor quality film - or a very contrasty film - or a film/developer/paper combination with limited contrast controls - then it might not be possible to spread the tones back out after such... data compression.

Ken Lee
19-Mar-2015, 12:44
How do you propose to do this corresponding 'Contrast Correction' in the Darkroom.

If we include film exposure and development along with choice of paper and paper developer, we can spread the tones out.

We already do this, since no lenses give us as much contrast as the subject starts out with. It's just a matter of degree.

In BTZS testing, a correction is made for flare. In the end it affects what we consider our effective film speed, developing time etc.

Perhaps a BTZS adept could explain what happens when we allow "a much greater" correction for flare. I presume we lose more film speed and develop longer. Sounds like contrast control to me. :rolleyes:

Old-N-Feeble
19-Mar-2015, 12:44
__

I was just wondering...

How do you propose to do this corresponding 'Contrast Correction' in the Darkroom.
Just curious...

Thank-you!
--
Best regards,

-Tim.
_________

Adjustments to film exposure and development techniques and perhaps different choice of paper contrast. A lens with tons of flair would require decreased exposure on film and increased development time for the final print to more closely resemble an image shot with a lens with very little flair.

Taija71A
19-Mar-2015, 12:45
Pre-exposure or "flashing" is a rather poor option in my opinion because it muddies the microtonality in the shadows. I certainly understand it. Too blunt an instrument for me. I'd rather do more subtle controls via film development tweaks, careful lens choice, or perhaps unsharp masking. In other words, I like to have
my cake and eat it too.


Me too. Back when I was learning the ZS I tested that technique rather extensively and no matter how I did it I never liked the "look" of it.

__

Like all other Techniques...
Pre-Exposure or Flashing can be as subtle or blunt -- As you 'choose' to make it.

If you find that it is not 'subtle' enough...
Then 'usually' the problem is... That you have added too much Pre-Exposure.
--
'Cake is Good'... But, we all know what happens if you eat 'way' too much cake! :D
--
Also, please remember...

That, there is nothing to say that you have to 'Pre-Expose' -- Your whole Sheet of Film.
It too can be done 'selectively' (*To a certain degree)...
--
Best regards,

-Tim.
_________

Taija71A
19-Mar-2015, 12:47
Adjustments to film exposure and development techniques and perhaps different choice of paper contrast.

Contrary to popular belief...

I don't believe that any of the techniques that you have cited...
Will have the 'exact' same effect.

*They of course are all Excellent tools... To have in your 'Reproitore'. :)

Jac@stafford.net
19-Mar-2015, 12:51
I'm having that same schizophrenic debate in my own head right now. This time of year I sometimes covet soft open shadows rather than hard ones, and prefer
the older lens look. Guess I could pack both 360's for my Saturday walk

Edison/Mazda base flash bulb units with slaves. At least two, more is better. It can flash-fill a backlit forest. :)

.

Jac@stafford.net
19-Mar-2015, 12:58
Back to lens flare for a moment.

In the earlier days of video we struggled with blooming. You know what that is. Tiffen made, and still offers 'Contrast Filters' (http://www.tiffen.com/contrast_filters.htm) intended to lower contrast. These are not color-contrast filters. They are neutral.

I have also used them with MF film photography at dreaded high-noon light and they did the trick.

So, possibly a lens with the Tiffen filter(s) shown by the link above can be considered.
.

IanG
19-Mar-2015, 13:12
I would like to see a test of an uncoated (or any) lens known to have lots of flair vs. a multicoated lens of the same focal length and known to have extremely low flair. Then I'd like to see the differences in prints after contrast correction via fully analog wet process.

I have images made with an un-coated CZJ 135mm f4.5 Tessar and a 135mm f5.6 Symmar S, I wouldn't say that the Tessar had a lot of flare but there's very definite differences in shadow details and micro-contrast in darkroom prints there's a loss of detail with the Tessar.

At some stage soon I plan to test a few of my lenses side by side, I'm just finishing my new darkroom at the moment.

Ian

Taija71A
19-Mar-2015, 14:34
... If we stick with an analog process or 16-bit digital capture,
can a flare-prone lens serve as a valid method of controlling contrast ?
__

Upon further consideration... I am going to 'refine' my answer slightly. :)
__

Quick Answer:

If we stick with an analog process... Can a flare-prone lens serve as a valid method of controlling contrast ?

Yes.
--

If we stick with 16-bit digital' capture... Can a flare-prone lens serve as a valid method of controlling contrast ?

Valid -- Yes.

Required -- *No.

*In today's 'Digital Age'...

I believe that it 'perhaps' has now become superfluous to our needs...
And can be handled easier (and just as well?)... In Post-production.
--
Best regards,

-Tim.

P.S.: Very sorry Ken...
That I introduced a different topic (*Pre-Exposure of Film) into your Thread.

I was just trying to be helpful...
________

Mark Sampson
19-Mar-2015, 14:53
This sort of question was addressed quite well by Adams in one of his 1950s books, probably 'The Camera'. (Sadly my copy is in storage far away, so i'm working by memory here.) Ansel, of course, was concerned with showing the differences between uncoated lenses and single-coated ones, multi-caoting was 20 years in the future then, but the lessons still apply.

Peter Gomena
19-Mar-2015, 15:24
How much flare do you really want to introduce? A few years ago, I tested my LF lenses and my old Canon FD lenses using a homemade "black box" in open sunlight. The box was about 18" square, painted black on the inside and out, with a 4" hole in the front that had a black cardboard lens shade that shielded it on 3 sides. My LF lenses (80s Schneider multicoated) registered .03 density over film base + fog. My Canon lenses gave .05 over fb+f, both measurements using a calibrated densitometer. How much flare would an uncoated lens produce in comparison? I have no idea, but I doubt it would be much more than one stop more than the coated lenses produced in "normal" lighting conditions. Yes, it might equate to a nice bit of fill light overall. It still would muddy up the mid and mid-high tones a little and compress the highlights a bit.

Mark Sawyer
19-Mar-2015, 16:06
Here are some numbers I am pulling out of my hat. Are these right ?

Let a subject start out at 100% contrast. A lens with low flare delivers 85%. A lens with a lot of flare delivers... 75% ?

In other words, how much are we talking about anyway ?

Exact numbers are pretty loose, but here goes... Each uncoated air-glass surface scatters 4 to 8% of the light passing through. The front element reflects it back out into the world, and the rear surface only reflects what's bouncing back from the film plane, so I discount those two surfaces. So an old uncoated Dagor with two inner air-glass surfaces is losing 8-16% of the light, about half to flare and the other half absorbed by the barrel walls. So you're talking 4 to 8% flare vs. clean image making light. (BTW, two inner air-glass surfaces is the minimum with compound lens designs, so the Dagor was very popular because of its high contrast.)

Split the Dagor into an "air-spaced Dagor" (aka Plasmat), and there are now 6 inner surfaces, so 24 to 48% of light is lost, with 12 to 24% being flare. That's why the Plasmat was never popular until coatings came in.

Early coatings reduced flare to 1 to 2% per air-glass surface. Modern multi-coatings are claimed to get it down to 0.1%.

The flare acts like pre-flashing the film. It affects the shadows most because the flare is evenly distributed, while little image-making light is hitting the shadow areas, so the flare can overwhelm it. If you adjust your processing for uncoated lenses, coated lenses will seem harsh, and if you adjust your processing for modern lenses, uncoated lenses will seem flat or mushy.

Taija71A
19-Mar-2015, 17:19
Since, there seems to be quite a bit of confusion about this topic.
Perhaps, the following may be of interest to some.

Pre-Exposure of course... Doesn't have to be placed as high as Zone II.
(This of course is just an example).

Ansel Adams -- The Negative (Pre-Exposure) pp. 119 - 124.

https://manualesdecine.files.wordpress.com/2012/05/the-negative-ansel-adams-series-no-2.pdf

Regards, -Tim.

sanking
19-Mar-2015, 20:25
....
In BTZS testing, a correction is made for flare. In the end it affects what we consider our effective film speed, developing time etc.

Perhaps a BTZS adept could explain what happens when we allow "a much greater" correction for flare. I presume we lose more film speed and develop longer. Sounds like contrast control to me. :rolleyes:

My presumption is same as yours. We adjust for increasing flare by increasing time of development, which results in higher average gradient. This procedure does control overall contrast, i.e. macro contrast. However, the adjustment in macro contrast does not entirely restore the acutance (micro contrast) that is lost by lens flare. You may recover some of the loss with a digital work flow with sharpening, but lens flare results in some loss of image quality that is impossible to recover.

Sandy

Doremus Scudder
20-Mar-2015, 03:23
Lens flare, since it is non-image light and evenly-distributed, is exactly the same as an intentional overall flashing, or pre-exposure of the film. There's really no way it can be different. These both affect the low values for the most part, flattening out the toe portion of the curve and reducing separation in the area most affected while at the same time allowing more detail to be recorded.

The end result is not much different than using a film with a long toe vs one with a steep one (e.g., TXP vs TMY). Getting more separation in the shadow with either flare or a long-toed film is often just a matter of exposing more to get the desired shadow values up onto the straighter portion of the curve. I suspect this is why a lot of photographers in the past rated their film a bit slower. On the other hand, as proponents of long-toed films, uncoated lenses and flashing realize, the trading of separation in the lowest values for more detail and softer shadows can be quite gratifying.

I routinely manipulate TXP this way, choosing a full exposure in cases where I want more separation in the shadows and less when I want the look the long toe of this film delivers.

Pre-flashing can be done rather easily if one has accurate shutters, and delivers the same effect as lens flare. An evenly illuminated grey card held close to the lens (so it's out-of-focus) and an exposure anywhere from Zone II down to Zone 1/2 or so will often do wonders for pulling substance from deep shadows. For most subjects, there are better ways of dealing with the shadows. However, as one tool in the toolkit, flashing has its place.

Best,

Doremus

Hans Berkhout
20-Mar-2015, 06:03
I "encourage" some lens flare by omitting the use of lens hood/shade quite often. The result still leaves me with better separation at the lower end than I would get with a single coated older lens and I still have the advantages of modern lens design. Choice of developer helps to avoid the steel wool look that some modern lenses tend to produce.

Taija71A
20-Mar-2015, 09:28
In my "simulation" of lens flare, I applied the following curve:

Image: http://www.kenleegallery.com/images/forum/flare.jpg

Should it have been this one instead ? In other words, not just the shadows, but the whole scale simply lifted up ?

Image: http://www.kenleegallery.com/images/forum/flare2.jpg


No.
Of the two curves that you have shown Ken... The first curve is better.

But...

As you can see by the histogram for the first curve, you are still lifting not only just the shadows -- But also the rest of your Zones to a lesser but still, too significant degree.
This is not necessarily what you want to do (*If you are trying to 'simulate' Lens Flare)

Ken, can you apply a partial custom curve that only lifts the Lower Zones (perhaps not at all above Zone V)? This would even be... Much better still!

Best regards, -Tim.

Mark Sawyer
20-Mar-2015, 10:22
Just a couple of practical notes...

Pre-flashing film is actually kind of dumb. If it needs a thresh-hold fog, make a note on the holder and do it right before development. That way you can decide if and how much is needed.

Most uncoated lens flare is randomly scattered, but if you need a lens hood and you're leaving off to encourage flare, you may get defined reflections of the aperture shape.

Bill Burk
20-Mar-2015, 18:03
__

Yes!

What you have tried to 'simulate' Ken, would be exactly the same...
As adding the proper amount of 'Pre-Exposure' of your Film (*for 'Contrast Control').
--
In an 'Analog Only' workflow... This is most certainly a very, valid option.
One of the best 'treatises' on this subject... Was discussed in the Book:

"Way Beyond Monochrome" by Ralph Lambrecht.

http://www.waybeyondmonochrome.com/WBM2/TOC_files/PreExposureEd2.pdf

--
Best regards,

-Tim.
_________


Tim, You struck gold here!

Ralph Lambrecht's curves and examples prove the point.

Flashing / flare / long toe. They all sensitometrically look exactly the same to me. Doremus Scudder already explained. But I see it too.

Maybe not exactly the same, as the toe shape may differ from the flare curve. But they ARE very similar. Again if they are not exactly the same (flare versus pre-exposure) the differences are very minor and can be named. For example some trivial differences due to reciprocity / intermittence effect. Maybe a pictorially significant difference if flare includes aperture image shapes.

But by gum, I see the same thing, an upswing in the curve in the shadows... While midtones are trivially affected and highlights are barely measurably affected.

In other words, if you WANT a long-toe and the film you are shooting has a sharp toe... Pre-exposure will give it to you.

Yes. You must calculate exposure precisely... Because if you want the benefit of the curve you changed with pre-exposure... You sure do not want to put your shadows on the straight line. That would defeat the purpose!

Bill Burk
20-Mar-2015, 18:05
Drew Wiley, You were right all along!

Jac@stafford.net
20-Mar-2015, 18:10
Just a couple of practical notes...

Pre-flashing film is actually kind of dumb. If it needs a thresh-hold fog, make a note on the holder and do it right before development. That way you can decide if and how much is needed.

Indeed! There is the fact. No need to pre-flash on site.
.

IanG
21-Mar-2015, 02:01
There's an assumption in this thread that the flare from an un-coated lens (or coated) is even across the frame which is going to be rarely true in practice, often it can be quite localised. LP Clerc in his book Photography, Theory and Practice, mentions internal lens reflections causing false ghost images, and also the extremes of the central flare spot.

The false ghost images are more like adding very weak additional slightly out of focus exposures to the primary exposure so quite different to a pre or post flashing exposure.

When you look at contemporary prints made by photographers like Kertesz, Brassai etc made before WWII they have a very subtle jewel like glow from the un-coated lenses they were using, their negative were exposed comparatively more than we would now and processed to higher contrasts but the papers of the time were suited to the negatives. Mostly the printed quite small even though they often used 9x12 cameras, modern prints from the same negatives look quite different and lose the jewel like quality.

Ian

Old-N-Feeble
21-Mar-2015, 10:01
Ian, I'm not disagreeing (other than the "jewel-like" reference) but how do you know the look you appreciate is due to lack of lens coating vs. lens type? For instance, heliar vs. dagor vs. petzval vs. plasmat, etc?

IanG
21-Mar-2015, 14:42
Ian, I'm not disagreeing (other than the "jewel-like" reference) but how do you know the look you appreciate is due to lack of lens coating vs. lens type? For instance, heliar vs. dagor vs. petzval vs. plasmat, etc?

Because it's common to that era and the smaller LF formats common in Europe where it'll be most noticeable, I have seen a lot of work from that era and the Jewel like quality is because the images work best printed at quite small enlargements (usually less than 10x8 in) on the contemporary papers of the era.

It's not particularly specific to lens type but there will be small differences of course and also extremes.

Ian

Old-N-Feeble
21-Mar-2015, 14:48
I respect you, Ian, but I disagree with your premise.