Filters: Traditional or Digital?

[Viana]

It seems lots of people on CiC are struggling with the use of filters. Therefore, this thread.

Eons ago (well, it seems like eons) when I used to shoot film, I would haul around camera bags, several cameras loaded with different film speeds and have bags/pockets full of filters and other accessories. While I was young and strong, I still felt like a little pack-mule and of course, always had to think way ahead in order to shoot with the correct camera/film/filter.

With the advent of digital photography that all changed. First, the need to have different film speeds, since the digicam permits the adjustment of ISO for every picture. Therefore, few people find it necessary to carry more than 2 cameras.

Second: PS processing software has come a long, long way within just the last couple of years. These days it is possible to simulate the look of almost any filter and then some via software.

Today, I generally only use a good UV filter to protect my lenses and apply filter effects in the digital Darkroom during PS processing. Why? Because, I like complete control and I prefer not having to remove filter effects that do not work out. Case in point: The ubiquitous polarizing filters. Often the polarizers turn the sky a bizarre blue in one area, especially on wide-angle lenses, an effect that is a pain to remove via PSP! Polarizers also saturate a bit much for my taste in many cases. NDs: These sometimes produce colour casts. If the dynamic range is too much for the sensor, then I would take multiple, bracketed shots and combine these as an HDR. Even infrared effects can be done via PS processing.

Star filters are another thing with which I have a love/hate relationship. While I love the effect, these tend to soften the rest of the image. Then again, perhaps a 4, 6, 8 or other point filter might have been better? All of that can easily be done in PS processing with a filter brush.

What is your opinion? Do you use traditional or digital filters and why? Which software do you prefer for filter effects? Show us some examples of your work with either traditional or digi-filters.

[Glenn NK]

I use only two filters; UV for each and every lens (for protection), and occasionally a polarizer. As you note, the CPL is problematic with skies and WA lenses (large variation in angle between sight axis and sun). I'm not convinced (hopefully someone can convince me) that polarization can be added in PP as effectively as when shot. Because I've used polarizing sunglasses for years, I know how effective they can be in removing non-metallic reflections. Leaves, flowers, water, human skin, all have some degree of bright reflectivity which reduces saturation (my preferred time for flowers has cloudy skies).

I've never used ND filters, partly because of cost and partly because they seem to be a nuisance, but before I do, I would try the "magic cloth" trick first. It requires some considerable experimentation. With digital, experimentation isn't terribly costly, but when I became more serious about photography in the early sixties, it was quite expensive (thirteen shots of a night-lit fountain in Kodachrome II was 1/2 of a roll ).

Glenn

[Bikermike]

I always carry a polariser in my bag, but only use it rarely.

With regards to a UV on the front "to protect the lens" I'm afraid I'm from the camp which believes any extra glass will only increase the risk of glare or reflections and that the best protection is a semi-rigid lens hood.

Regards, Mike

[Dave Humphries]

I always carry a polariser in my bag, but only use it rarely.

I agree on the carrying and infrequent use of a polariser Mike.

To my mind, a CPL is no more suitable to wide angle shooting than hiking boots are when you sleep
That said, in extreme conditions, it might just be the right tool for the job.

I'm not convinced ~ that polarization can be added in PP as effectively as when shot.

I agree, when a CPL is suited to a capture though, the benefits cannot be reproduced in PP; true some aspects might be imitated, but it really isn't the same.

With regards to a UV on the front "to protect the lens" I'm afraid I'm from the camp which believes any extra glass will only increase the risk of glare or reflections and that the best protection is a semi-rigid lens hood.

I from the camp that knows the hole at the front of the lens is going to get splashed occasionally and a flat filter is so much easier to safely clean than a curved front element is - and if I scratch it, it is considerably cheaper and less hassle to replace a filter

The difference in quality is undetectable 99% of the time for what I shoot, but it might come off if I got into a night shooting situation.

So I guess it is each to his/her own.

I agree with Viana (is that your name?) about most effect filters being better applied in PP; starburst, soft focus, the colour ones for B&W conversions, etc., but also, even though I don't use 'em, I can see that if you need a 10 stop ND to smooth water in daylight, or a grad ND for taming dynamic range, these will be necessary, just as a CPL is, at the time of shooting.

Cheers,

[Donald]

... hiking boots are when you sleep

It's amazing what you find out about your colleagues here on CiC.

[Viana]

I use only two filters; UV for each and every lens (for protection), and occasionally a polarizer. As you note, the CPL is problematic with skies and WA lenses (large variation in angle between sight axis and sun). I'm not convinced (hopefully someone can convince me) that polarization can be added in PP as effectively as when shot…

I shall see what I can do. Will dig out a few recent photos to show with and without the digi-filters.

Perhaps we should have a contest where people shoot with and without the traditional filters? (My problem would be that I no longer own any that fit my current lenses.)

I would, however, be willing to process the pixs that anyone submits with a digital filter and compare those with the traditional filter captures.

To some extent, the argument claiming that digi-filters are not as good as the traditional reminds me of the argument about expensive vs. moderately priced wines. Blind tastings show that expensive does not necessarily equal quality.

I propose that we need a “blind” viewing comparing the two.

when I became more serious about photography in the early sixties, it was quite expensive (thirteen shots of a night-lit fountain in Kodachrome II was 1/2 of a roll ).

Glenn

Oh, yes! I remember what fortunes I used to spend on film and developing. So, I am thankful every day for digital media in order that I can exercise my passion for virtually no cost.

[Steaphany]

Viana,

I've never given up on filters and I still feel my collection is lacking. At present my collection consists of:

• An IR950 Infrared
• An IR850 Infrared
• An IR720 Infrared
• 25A
• G
• K2
• maxmax.com XNite CC1 Natural Light Color Correction for IR Cameras
• A Sigma CPL
• A B+W 420 2E UV Special
• A Cokin P Filter kit for custom Bokke apertures

Eons ago (well, it seems like eons) when I used to shoot film, I would haul around camera bags, several cameras loaded with different film speeds and have bags/pockets full of filters and other accessories. While I was young and strong, I still felt like a little pack-mule and of course, always had to think way ahead in order to shoot with the correct camera/film/filter.

Didn't anyone ever explain to you that was how photographers got their exercise ?

When I'm out and about with my kit, I look like an old school photographer as I keep everything protected in a pair of bright yellow Pelican hard cases, a 1600 and a 1500. Plus, if you check my signature, you'll see that I still shoot film. On that side of the lens, I know I'm lacking and need both a medium and a large format camera. See, even more to carry

What is your opinion? Do you use traditional or digital filters and why? Which software do you prefer for filter effects?

As you too have a film photography background, I'm sure you were schooled with the idea that image quality is maximized in the image capture, not the darkroom and, today, not in the computer. Obviously, post processing can yield all sorts of effects, stylized character, or image presentations, but first and fore most is ensuring that you are working with the highest quality negative or Raw file to begin with. This is why I have "speciality" filters to achieve capture effects that, while they may be approximated digitally, are best done when the shutter is open.

Show us some examples of your work with either traditional or digi-filters.

As requested, here is a Time Lapse I shot a few years back where I wanted the white clouds to be against a deep blue to black mid day sky. The effect would have been impractical post processing 1,540 individual exposures and this is, obviously, an example where a digital camera out performs film:

[Glenn NK]

Oh, yes! I remember what fortunes I used to spend on film and developing. So, I am thankful every day for digital media in order that I can exercise my passion for virtually no cost.

And I used to order 6 or 8 rolls of KII at a time to get the best price - then mailed them to a processing lab and waited a week for results. But we're much smarter now - we spend HUGE amounts on the latest sensor/AF/video/MP feature in a body. Pre-digital, I used a Canon A-1 for 26 years.

Back on topic: I do find that a CPL is very effective for macro shots of flowers - it greatly enhances the colour saturation by cutting reflections. Of course it works because the angle of view is quite narrow.

I'm surprised that no one responded to my comment about the magic cloth - perhaps it's not well known?

Glenn

[Steaphany]

Keeping the tangent alive...

And I used to order 6 or 8 rolls of KII at a time to get the best price - then mailed them to a processing lab and waited a week for results. But we're much smarter now - we spend HUGE amounts on the latest sensor/AF/video/MP feature in a body. Pre-digital, I used a Canon A-1 for 26 years.

When you clear away the marketing hype and examine the underlying technologies, digital photography and film photography both has strengths and weaknesses. There are things that can be easily done with each that is nearly impossible with the other. Besides, $ for $, or £ for £, film today is far less expensive than digital when you're after the highest possible pixel count images. Case in point, my used Sigma SA9 with two kit Zooms cost me $35US where a Sigma SD1 lists for $6,899US. The SD1's imager has a resolution equal to many fine grain modern films, so it's comparable there. But, the SD1 does not have a full frame imager, so the SA9, being just 0.5% the price can deliver the same image quality, full frame, and the film can be scanned to yield digital files of greater pixel counts. ( Sigma will have to highlight something really special for me to spend $6,899 and I'd still have to shoot over 650 24 exposure rolls before I hit that expenditure. )

Processing wise, I have a near by wet lab who also offers negative scanning ( 1 to 2 day results ), plus I still want to re-establish my wet darkroom for immediate results.

For photographers with digital cameras, apart from time, there is little post capture costs, but the new up and coming thing are super phones with supposidly quality cameras which include mandatory data service subscriptions. Users believe they have a high end camera, that they are now qualified as a photographer , and ignore the data costs for every image taken or every time that send it somewhere.

[Viana]

Viana,

I've never given up on filters and I still feel my collection is lacking. At present my collection consists of:

• An IR950 Infrared
• An IR850 Infrared
• An IR720 Infrared
• 25A
• G
• K2
• maxmax.com XNite CC1 Natural Light Color Correction for IR Cameras
• A Sigma CPL
• A B+W 420 2E UV Special
• A Cokin P Filter kit for custom Bokke apertures

Hmmm—One can do all of that and more with digi-filters Steaphany.

Put up a photo and suggest a filter or effect and I’ll show you.

The effect would have been impractical post processing 1,540 individual exposures

Actually it would have been possible to batch process those photos very quickly since the lighting was similar. I typically can sort, cull and adjust about 100 photos in an hour, that is with different lighting conditions using Lightroom. The cloud photos you put up should not have taken much time at all.

When I'm out and about with my kit, I look like an old school photographer as I keep everything protected in a pair of bright yellow Pelican hard cases, a 1600 and a 1500. Plus, if you check my signature, you'll see that I still shoot film. On that side of the lens, I know I'm lacking and need both a medium and a large format camera. See, even more to carry

, I am so glad that I gave that up a number of years ago. No more aches and pains from heavy gear.

Regarding exercise: Recently, I was on a trip with my Nikon D300 and an 18-200 lens, with lens-hood—all around my neck. This minimal gear, with a few other things, other lens, etc. in my vest pockets, allowed me to have a Nordic walking pole in each hand, walk extremely quickly and still stop every couple of minutes for a photo.

You could not pay me to go back to film. I love saving lots of money, not having to worry about schlepping or keeping track of gear, not worrying about the cost of film and developing and most especially, having complete control over the whole process.

As you too have a film photography background, I'm sure you were schooled with the idea that image quality is maximized in the image capture, not the darkroom and, today, not in the computer. Obviously, post processing can yield all sorts of effects, stylized character, or image presentations, but first and fore most is ensuring that you are working with the highest quality negative or Raw file to begin with. This is why I have "speciality" filters to achieve capture effects that, while they may be approximated digitally, are best done when the shutter is open.

True, true. Garbage in garbage out. However—it is a fact that even the super photographers in days of yore did an awful lot of tweaking in the chemical darkroom. Ansel Adams anticipated digital.

One can do even more in the digi-darkroom than with chemicals and it’s safer. A bad photo can be rescued and made acceptable while a good photo can be made great in the digi-darkroom.

I find that people who are agin’ technology often have no experience with same. I’ve heard some people say they see no reason to own a computer or access the internet.

Remember the Luddites?

[krispix]

It's amazing what you find out about your colleagues here on CiC.

Steady on now. What a man does in the privacy of his own sleeping bag etc. etc.

[krispix]

And I used to order 6 or 8 rolls of KII at a time to get the best price - then mailed them to a processing lab and waited a week for results. But we're much smarter now - we spend HUGE amounts on the latest sensor/AF/video/MP feature in a body. Pre-digital, I used a Canon A-1 for 26 years.

Back on topic: I do find that a CPL is very effective for macro shots of flowers - it greatly enhances the colour saturation by cutting reflections. Of course it works because the angle of view is quite narrow.

I'm surprised that no one responded to my comment about the magic cloth - perhaps it's not well known?

Glenn

OK Glenn. I'll bite. What's "Magic Cloth"?

[Glenn NK]

OK Glenn. I'll bite. What's "Magic Cloth"?

The best reference I have is:

http://icelandaurora.com/blog/2010/0...oth-technique/

It comes up from time to time on Naturescapes.net (of which I'm a lifetime member).

http://www.naturescapes.net/phpBB3/v...loth+#p2009535

Hope I'm not breaking any rules of this forum by posting a link to another photo forum.

I think with some practice it could be very useful as it's not limited to one area of the image. Of course it's only applicable to longer exposures (unless one has lightning quick hands).

The price is certainly attractive.

Glenn

[Steaphany]

Hmmm—One can do all of that and more with digi-filters Steaphany.

How ? Considering only the Red channel, once an image file has been created, how can you distinguish pixels bright from light whose predominant wavelength was 850nm and longer versus pixels bright from light which had a predominant wavelength less than 850nm ?

No more aches and pains from heavy gear.

I don't have aches and pains from my gear either.

... I love ... most especially, having complete control over the whole process.

Cool, found another engineer ! Are you following "Lens Design Fundamentals" by Rudolf Kingslake & R. Barry Johnson or do you have other texts for your optics design ? I've also been exploring whether I should save the bother and go with a Moravian Instruments liquid cooled G4-16000 employing a Kodak KAF16803 or just getting a Kodak KAF16803 and design from scratch. At least the G4 would save a lot of time and experimentation. As it is with my Sigma SD14, I have software that allows me to individually extract and save off the various data regions of the Foveon Raw X3F file when I need to explore byte by byte. On the wet darkroom side of things, I was looking for supplier like THE FORMULARY back in the 1970's & 80's. It's great to now be able to get the chemicals needed to mix up a developer from scratch when you need a special characteristic.

What I love about Photography is that it's not just pointing and pressing the shutter release, you can get hands on at what ever level you desire while enjoying so many fields of science and engineering all rolled into one.

I’ve heard some people say they see no reason to own a computer or access the internet.

It's always been that way, even when there was just UUCP.

Remember the Luddites?

Yes, I do.

[Viana]

How ? Considering only the Red channel, once an image file has been created, how can you distinguish pixels bright from light whose predominant wavelength was 850nm and longer versus pixels bright from light which had a predominant wavelength less than 850nm ?

In a word: Histograms. It’s not quite comparable, because the typical digital SLR cameras will generally capture light with a wavelength of anywhere from 450 to 680nm. The digital tonal range is from 0 to 255 on histograms.

The Levels tool is used to adjust brightness as is the Luminance tool while Tone Curve for lights and darks.

Take a photo without and with a filter. Post the unfiltered photo and I’ll put a digi-filter on it. Then we can compare.

Cool, found another engineer ! Are you following "Lens Design Fundamentals" by Rudolf Kingslake & R. Barry Johnson or do you have other texts for your optics design ?

I am not a lens designer and as such am not particularly interested in the detailed mathematics of lens design. However, you may want to have a look at a couple of old favorites by master Kingslake:

"Optics in Photography" or

"A History of the Photographic Lens"

[Colin Southern]

Just keep in mind that a GND filter on the camera will give you dynamic range compression at the time of capture, whereas a "digital filter" applied in post-processing won't. If one is shooting into the light then one can expect the scene to have a dynamic range that stretches the sensor to the limit ... so in this situation you're likely to have much cleaner shadows & low tones with the physical GND.

[Glenn NK]

Just keep in mind that a GND filter on the camera will give you dynamic range compression at the time of capture, whereas a "digital filter" applied in post-processing won't. If one is shooting into the light then one can expect the scene to have a dynamic range that stretches the sensor to the limit ... so in this situation you're likely to have much cleaner shadows & low tones with the physical GND.

Does this dynamic range compression occur with the magic cloth? Intuitively it would seem so as it simply masks the amount of light that strikes the sensor in one area (the brightest part of the scene).

Glenn

[Viana]

Just keep in mind that a GND filter on the camera will give you dynamic range compression at the time of capture, whereas a "digital filter" applied in post-processing won't. If one is shooting into the light then one can expect the scene to have a dynamic range that stretches the sensor to the limit ... so in this situation you're likely to have much cleaner shadows & low tones with the physical GND.

The problem with that is that one has to put on and remove the filter as one changes positions.

As I mentioned earlier, why would one not bracket and then combine the photos for the most extensive dynamic range that is possible today? That would allow the photographer much more leeway and give the ability to produce a photo as the scene actually looked or to reflect the vision of the artist.

I still think that we should test this out via actual photos.

[Colin Southern]

Does this dynamic range compression occur with the magic cloth? Intuitively it would seem so as it simply masks the amount of light that strikes the sensor in one area (the brightest part of the scene).

Glenn

Because the light reaching the sensor is reduced under whats covered by the GND filter (lets assume a typical 3-Stop attenuation) then (again, for example) a scene with an actual dynamic range of - say - 12 stops, only requires 9 stops of sensor dynamic range, thus shifting everything 3 stops further from the noise floor.

Without the GND you'd have to increase the shutterspeed by 3 stops to account for the area not covered by the GND being 3 stops brighter - thus the shadow areas at the other end of the exposure are now 3 stops closer to the noise floor (and thus are a LOT noisier).

Subsequently applying a digital GND effect in Photoshop attenuates the selected portion by 3 stops, but it can't do anything to improve the Signal-to-Noise-Ratio of the low tones, as this was locked-in at the time of capture. The net result is that shots like this can still have very clean shadow detail revealed when the captured range is further compressed into something that can be displayed on a monitor



PS: Sorry Glenn, I misread your question a bit. Short answer is, "yes".

[Colin Southern]

The problem with that is that one has to put on and remove the filter as one changes positions.

One simply needs to slide the filter up or down in the holder, but that's a trivial matter.

As I mentioned earlier, why would one not bracket and then combine the photos for the most extensive dynamic range that is possible today? That would allow the photographer much more leeway and give the ability to produce a photo as the scene actually looked or to reflect the vision of the artist.

It's a valid technique (and one that I use often) - but it doesn't work well when there is movement.

Here's a good example where the capture is straight-forward using a physical GND filter ...

I still think that we should test this out via actual photos.

I use both techniques extensively; they're simply different tools - sometimes they're interchangeable, sometimes they're not.