Menu
#1259. Red filter for b&w photography. Glass or digital?
By pascaljappy | How-To
The question came up in a recent comment exchange. I had to try it out.
The author of that comment mentioned using colour filters on his digital camera for b&w photography. Given the very different nature of a filmstock and a Bayer-filtered digital sensor, that really surprised me (although it would make perfect sense on a digital monochrome camera).
Most of my filters were sold long ago, when my own move to digital happened. But I never found the courage to pass on my cherished deep-red Linhof filter, or the Linhof’s handle or my Sekonic light meter. It’s as preserving those apendages of the main camera somehow made it more present and the turkey less cold.
During a quick walk with my daughter one cloudy January afternoon, I grabbed a few shots with and without filter to compare the results of digital filters and the ancient glass version. These are presented below. Sadly, the Linhof-assisted photographs will often be quite obvious due to the incomplete coverage of the frame (the hand-held filter was barely big enough for the Hassy lenses, and my centering now looks like drunk driving … sorry 😉 )
Note that the digitally converted image was ‘tuned’ to look as close as possible to the (glass) filtered version.
Of course, that last version also has been converted to b&w. In colour, it was largely red & black. As the image below shows.
At this point, I find the results a little underwhelming. Seeing the world through the red filter is such a vibrant experience with the naked eye, that I was hoping for more exciting b&w photographs.
Also, given it is so easy to replicate the filtered look in Lightroom with a few moves of the red, orange, green and blue sliders, why bother?
Add to that the fact that the exposure is very hard to judge through the viewfinder (see below), with the image looking overexposed in the EVF and most of the (glass) filtered photos on this page exposed at around -3EV. So why bother indeed?
I’m not sure what happened with exposure. It was never going to be easy, since only one in four pixels is sensitive to red. But, inside the EVF, the uncompensated images looked vastly overexposed. I took the one above just to check, then took all further ones at between -2.3EV and -3.7EV.
The uncompensated image above was very bright SOOC, but was easy to correct in LR. The one on the right, while better exposed actually looks a bit coarser. In the uncompensated photographs, the red channel was blown, but this doesn’t seem to harm the photo once corrected in post. It’s all very counterintuivite for me. Conversely, the properly exposed photo seems to have burried shadows that do not lift up very well in post.
My recommendation, after processing those images, would be to underexpose but less than I did. Maybe -2EV. You’d blow the reds, but not by that much, and maybe get less crunchy shadows. But is it all a bit of a pfaff for nothing? Maybe not!
Here’s the thing. While it is simpler to shoot without the filter and it is possible to replicate its tonal effects very easily through Lightroom sliders, the glass filter version still looks a little different.
Less sharp, less digital. It’s more obvious on some images than others. But there is enough of a difference to orient PP in different directions and obtain different looking results.
Whether that’s important to you is purely a subjective decision, but it’s there. Maybe it’s in my mind, but the glass-filtered images just end up looking less brittle and more film-like to me.
Here is another pair
And another
The differences are subtle, when you match the tonal curves. But they are there.
To most people, that will not be worth the effort. But if you are trying to produce a certain aesthetic, and to apply a time-honoured technique and process rather than explore monochrome conversions at random, this is certainly worth the small extra step. Just be sure to buy a screw-on filter, not one that is held in front of the lens with a shaky hand 😉
Do you see the differences on the web images? What do you think?
​Never miss a post
​Like what you are reading? Subscribe below and receive all posts in your inbox as they are published. Join the conversation with thousands of other creative photographers.
Session expired
Please log in again. The login page will open in a new tab. After logging in you can close it and return to this page.
I’m curious to see a 3rd option using a red filter on a monochrome sensor (or Pixii in monochrome mode).
Me too, Carl! I’ll try to get my hands on a Pixii camera soon. Cheers.
Pascal
I do see a difference in the B&W images, with a preference for the glass filtered images. It seems that the person that made this suggestion was on to something.
If you try this experiment again, be sure to set a custom white balance (WB) for the filter or set WB to 2500 (or lower if possible). With my full spectrum converted* camera this results in a color image that is a lot more purple-ish and a different gray scale than the red rendition. Also, set your view finder, or JPG profile, to B&W, the color rendition can mess with your mind and your exposure settings.
Now I need to go do this for a day or two. I may be able to subdue my lust for a monochrome camera. 😉
*Using the red filter on a visible light (unconverted) camera puts the transmitted light under the cut-off wavelength of the camera’s infrared filter, which is why there may be a upper limit to the effect. Though, the sensor’s blue and green channel’s do have some sensitivity as the light becomes more red.
Have more fun.
PaulB
Thanks, Paul. Yes, cooling the WB to 2500 will produce a purple image. I’ll try that in PP as well 🙂
Hi again, just tried it in a non published photo. Cooling the WB darkens the image and conversely. It also makes it more contrasty and, ti my eyes, looks nicer with greener tint. So I processed this one with 2000 Temp, -80 Tint and -40 feedback and it looks good!
Pascal
Your adjustments do provide you with a nicer rendition of the original image.
Though, set the white balance in the camera to 2000 Deg K; some cameras won’t set a WB lower than 2500 Deg K. This will adjust the sensor RGB sensitivity more in line with the red light, and should result in a deeper purple rendition than your adjusted image. This should give you better latitude for post processing.
PaulB
Hi Paul, the WB setting shouldn’t alter the sensor reading. It’s purely a rendering decision. My understanding is that a RAW file will look bluer at 2500 than another at 12000 for example, but both will look identical if pushed back to 6000 in post. So setting the WB on camera shouldn’t have an effect that can’t be undone in post, but setting it in post can indeed alter the look of the b&w conversion, as you suggested 🙂 Cheers.
Pascal
Yes, we are venturing into the Ying vs. Yang of how to get to a pleasing effect.
Enjoy the experiment, and post more results please. 🙂
PaulB
The Northern Territory’s tourist bureau has a motto – “You’ll never never know, if you never never go!”
And it resonates with our photography.
Sometimes, it costs a bit – but we learn by trying, by exploring, by doing something different, by stretching the envelope.
Ah, any comment that compares DS to NT is so very welcome 🙂 🙂
Hi Pascal
I can see the difference with the red filter. Do you normally use any filters on your Hasselblad lens maybe the extra piece of glass is softening the image. Just a thought
Hi Dave, thanks. No, I never use filters. This was the first time. So, I think you are right, the filter is softening the image a little bit, and the underexposure is probably creating a bit of “grain” as well. The second part, grain, is very camera-dependent. My camera sensor is a slightler older generation, so it gets noisy quite rapidly, but the noise is very grain like. Other cameras will show less noise, but will be more “digital looking”. So it’s hard to guarantee the same effect will be created. Cheers
Hi Pascal,
I have always found that for ‘traditional’ style monochrome street photography the red filter option produces the best results as it emphasises the human skin ( faces ) and helps draw attention to the human subject matter. Other than that either LR or Nik Silver Effex Pro allow one more leeway to focus the attention via multiple colour luminance and contrast adjustments. Nik has the added advantage of allowing the selection of a digital colour filter as well as multiple colour adjustments in monochrome.
I guess with an actual lens filter there is always the hassle of keeping it clean and avoiding lens flare – unless of course that is the intentional look one is going for.
Regards
Ian
Red filter – using film, this can be an excellent idea. Not all the time – but when it IS the “go”, the results are stunning.
That said – and I realise you are experimenting here – I much prefer the colour shots! Oops. Pardon my bad manners!
But of course it is only by experimenting that we “learn”. When someone else tells us, or we go to a workshop, or we read it in a book or a magazine, we are being “taught”. And that is NOT the same process.
Teaching can fast track the absorption of knowledge – but I’d always put my money on the “learner”. The “learner” will be the one that knows all the settings, and how they affect the creation of images.
Exposure compensation – a useful tool, grossly under-used by many. There are limits to what an exposure meter in the camera can tell you. And next day, when you do your post-processing, it will be too late to go back tot he same spot and take the same photo – the light will never be the same, even if it’s similar.
And your Sekonic. Having choked a few times on the limitations of built-in meters, I bought myself a Sekonic. The years passed. And when I went to use it, it just wouldn’t work. So I’ve replaced it with a new one. For the D850, at least (although I do use it with some of the others. And a smaller, pocket version for use with the D500 and the Zfc. Just to check and make sure. Auto metering buggers things up, way too often.
I’m not surprised you prefer the colour shots 😉 Colour is your domain!
Frankly, the goal of the experiment was above all to determine what a red filter does to a Bayer filter. I had my doubts it would be of any use, but it does produce results, and results that are different to what is achieved in PP. So it has a role to play to those who like that look, I guess.
These days, I don’t bother with the Sekonic so much. It’s easy enough to dial in -1EV to avoid blowing the highlights. But still, it is a close friend that cannot be sold on 😉