ND Filters, a quick test and surprising results

[Cage]

I'm on a very long and steep learning curve with ND filters. After today I'm a little confused as to where my use may be.

We had some rain last night and I thought Somersby Falls may be worth a look, but the heavy rain we had obviously didn't fall in that area.

It did however give me an opportunity to do some testing, and as I said in the intro, the results surprised me and maybe also confused me.

The shots below were all with the D800 and a Nikon AF-S 28mm f1.8G. They were not all taken from the exact same spot but fairly close proximity. I've cropped them all to a similar aspect, tried to bring the exposure to a comparable level and they have had minimal PP.

This shot was with no filter, ISO200, f13 at 3sec.

No Filter.jpg

This one was with an el cheapo 3 stop filter, ISO100, f13 at 3sec

Cheapie 3 stop.jpg

The third was with a B+W F-PRO ND 3.0 (8 stop). I had to run the Purple Hue slider off the page to get rid of the purple cast in one spot, and it does not seem to pick up the greens either. I'm thinking it may not be a genuine filter with that performance.

B+W 3.0.jpg

I've added this one to show what the B+W looked like SOOC.

B+W SOOC.jpg

To my eyes, the first shot, no filter, is the pick of the bunch, and seems to indicate that you can get the milky water effect without a ND filter. The second shot lacks the clarity of the first, and the third shot with the B+W filter had me shaking my head. The purple hue seemed to get worse the longer the exposure.

I'd be interested to hear the thoughts of those experienced in the use of ND filters.

[Grant S]

Are you increasing your exposure times with the addition of the filters? Looking at your second shot I'd say its slightly underexposed and the 8 stop shot is certainly underexposed. Some filters do give a colour cast though so not seeing the raw images its a bit hard to judge if you'd done any PP on them.

[wmvaux]

Do you have your view finder covered?

[hakka]

^what wmvaux said, that last shot looks like it has a light leak.

[Steve Axford]

I've got a B+W ND and it has always worked well (though I don't often use it). The fact that the colour cast is only half the frame makes me think that vmv and hakka are right.

[Cage]

Thanks for the input.

Being my first outing with ND filters, I really didn't know what I was doing. Now that it is pointed out, it does look like the purple area could very well be a light leak.

The D800 has a little shutter to cover the viewfinder and I thought I was using it, but maybe I missed a couple of shots. I'll do some more testing to see if I can duplicate the results. Hope it was my stuff-up as I've seen some interesting outcomes while playing with the B+W 10 stopper, like moving objects disappearing.

I guess the main point I wanted to make was that I was under the misapprehension that you needed a ND filter to get the milky water effect, and as the first shot shows, that is not the case.

Thanks again for the replies. That's what I love about this forum, it doesn't matter how dumb the question is, someone will always take the time to help out.

[Cage]

Just did some testing in the yard and that purple haze in the B+W shot was indeed caused by me not closing the eyepiece shutter.

Some googling revealed that the B+W 10 stopper does have a purple tinge with long exposures, but nothing that can't be fixed in PP.

Cheers

Kev

[Gonk0]

You have a light leak in the last photo as I have had a similar thing happen to me before I started to use black tape over my view finder to make sure no light gets in at all.

The biggest thing I see with your photos Kev is the range of ISO used, did you not use manual settings at all, as I use 100 ISO for all my long exposure shots, but in
saying that I'm a beginner, so I'm still learning myself.

As far as colour casts go I have the cheap cokin P setup and if I stack 2 filters I get a bad PURPLE colour cast, but if I use my screw in ND filter and even if i stack them, I don't get any cast at all.

Also how did you come to stop down by 3 stops with the cheap filter and get the same length exposure with a lower ISO as the actual stop down would have been closer to 4 with the lower ISO, were you in bulb mode and counting?

[Bennymiata]

I use an ND filter every now and again, and I was worried about colour casts, so I did a lot of research and spent probably too much money on buying a Singh-Ray variable ND filter.
I can't see any colour casts from it, right up till you have to stop increasing the effect when the black sections start coming in.
I've used it in very low light too (before sunrise) with it set to its darkest setting to get the longest exposure of the surf, and the colours were also excellent.
I guess with everything, filters included, you get what you pay for.
I thought it was good value at the time, as most of my lenses are 77mm, but now I've got a Canon 16-35 F2.8 and it has an 82mm filter thread.

However, I would certainly recommend the Sing-Ray filters.

[Cage]

Hi Gonko,

I probably shouldn't have listed any camera settings. As I said above I adjusted the exposure in PP to try to bring all shots to a comparable level. They were all underexposed, which I don't mind as I'm shooting 36MP images in RAW format.

I wouldn't read anything into the shot with the cheap filter ( 6 filters and 6 adapter rings for $17.00) as I reckon it's more like 1.5 stops than 3 stops, and to me the shot looks hazy.

re the last shot, I have a shutter over the viewfinder eyepiece which I obviously didn't use.

And yes, except for the first shot, all were in manual mode. The trip was very much a learning experience, and from that point of view, a most worthwhile exercise.

Thanks for your comments.

Cheers

[Bennymiata]

Hi Cage, I'm not knocking what you did at all, but, one thing you'll learn is that regardless of your camera and your software, if a shot is not correctly exposed when taken, every adjustment you do to it in Lightroom, Photoshop et al., will add noise and degrade the image - even in RAW.
Small adjustments to exposure are fine, but if you start trying to adjust it more than around 2 stops, you will degrade the image.
Learn to read the histogram your camera provides as this is the best clue to correct exposure.

[Cage]

Thanks for dropping in Benny.

Mate, I don't take any comments as knocking, rather I'm thankful for people taking the time to offer advice.

As I said above, this was my first use of ND filters and I learnt heaps, both from my experience, and from comments like yours. I'm excited about the potential of this genre and I will continue to explore it's potential.

Cheers

[Fruengalli]

Just did some testing in the yard and that purple haze in the B+W shot was indeed caused by me not closing the eyepiece shutter.

Some googling revealed that the B+W 10 stopper does have a purple tinge with long exposures, but nothing that can't be fixed in PP.

Got one of these & agree...nothing that can't be fixed. Started looking at a custom white balance for when I've got the filter on. Seems to work just needs tuning.

Cheers

Kev

[bobt]

I'm interested in this thread because I have just ordered a graduated ND filter. Should get it soon. It's only a cheap one because I doubt I'll use it much - but it interested me due to the fact that it rotates like a circular polariser. This means that you can "dial in" the amount of darkness it produces. Sounds cool, so I'll see how it performs. In the meantime, I've been interested in your experiments - so thanks for that!

[arthurking83]

Hi Gonko,

...... As I said above I adjusted the exposure in PP to try to bring all shots to a comparable level. They were all underexposed, which I don't mind as I'm shooting 36MP images in RAW format.

......

You can't directly compare the images then.

You may find that the quality of the images may vary to what you currently see in them.
processing them can introduce quality degradations.

All the exposures should be captured to the same exposure level .. or as close as can be.
So if you had a 3 stop ND attached then of course shutter speed should be the variable you slow it down by 3 stops compared to the non filtered exposure.

As you've noticed yourself filters don't always work the way they're rated too. So it's handy to know if you 8 stop ND is actually 8 stops worth or is it really only 5 or something.

I'm interested in this thread because I have just ordered a graduated ND filter. Should get it soon. It's only a cheap one because I doubt I'll use it much - but it interested me due to the fact that it rotates like a circular polariser. This means that you can "dial in" the amount of darkness it produces. Sounds cool, so I'll see how it performs. In the meantime, I've been interested in your experiments - so thanks for that!

Sounds more like a variable ND, not graduated ND.
Grads are the ones with half dark, half clear graduations to balance uneven exposure scenarios.
The NDs that darken as you rotate the filter's ring are called 'variable'.

[Cage]

@arthurking.... Thanks for the input Arthur.

Yes, my testing methodology was a little short of being scientific. I did however confirm that in certain circumstances you don't need a ND filter to slow down the flow of water, such as in a shady gorge, which to me was a plus, as it means I can often get the shot I want without sticking anything else in front of my lens.

I also learnt what the eyepiece shutter lever is for. Hopefully I wont forget that lesson.

For a first try with ND filters, thanks to the help on this forum, I'm pretty happy with the results. I don't mind the posted shot without a filter, other than the blue tinge in the water which I'm sure I can lose, and I can see the potential with the B+W 10 stopper, when used correctly.

Cheers for all the replies.

[arthurking83]

.....

I also learnt what the eyepiece shutter lever is for. Hopefully I wont forget that lesson.

......

I guess posting off topic replies has it's merits(sometimes).

Well then since we're taking the OT route .. one more reason to know what the viewfinder shutter does, is that it allows you to unscrew the viewfinder glass piece and either: clean it out(it gets dirty and dusty frequently) and also to replace it with another type(if you desire).

You can't unscrew the protective eye piece without first closing the shutter lever. You then leave it closed(to protect the vf glass) as you do your thing with the protective piece.

Then you replace the vf cover glass thingy .. and NEVER over tighten it!! If you over tighten it it's nigh on impossible to unscrew it next time. Just tighten it to where it stops, and by opening the shutter it locks the cover piece from undoing itself.

Hope that helps.

One thing to note about sticking things in front of your lens.
Slowing down the light through to the sensor can be a tricky affair in some ways.

Of course shooting in lower light levels is one way you can do it, but sometimes this isn't practical, for many reasons(safety, time, etc)
if you just stop down more and more, you then run the risk of diffraction causing loss of detail in the image.
And ISO is limited in how low it can go on your camera(ISO 25 or even ISO 6 would be nice to have as options)

So ND filters can be a benefit if there is a desired effect you want, when other options to achieve those effects are limiting.
Basically it's good to have access to various tools which you use when required.


I personally only have one ND filter, and it's a simple 3 stop Hoya (screw on type).
I only wanted this one for the sole purpose that with my fastest lens(50/1.2) it can cause some limitations in trying to achieve a desired look(to the photos).
Because I always have a couple of GNDs to play with, I just use them in certain configs to create 2 or 3 stop NDs if even needed(which I think I've only ever tried once).


Strangely the only time I found I needed a ND filter was once when I wanted a very shallow DOF look on a white subject in broad daylight.
I needed at least 1/16000s shutter, and most likely 1/32000s(which don't commonly exist!), so I had to stop down to f/1.4 and overexpose by a bit over 1 stop(and recover).
Even tho it wasn't what I wanted, the image still looked basically as I expected it too(not quite good! ) .. but that wasn't the point.

[bobt]

Sounds more like a variable ND, not graduated ND.
Grads are the ones with half dark, half clear graduations to balance uneven exposure scenarios.
The NDs that darken as you rotate the filter's ring are called 'variable'.

You are indeed right .... not only have I not yet used it .... I can't even identify it properly !