ND Filters & White Balance

Snapper73

Snapper73

Junior Member
Messages
1,496
Name
Scott
Edit My Images
Yes
Started a new thread from this one here http://www.talkphotography.co.uk/forums/showthread.php?t=145007&page=3 regarding using ND filters and custom white balance when shooting landscapes. so far this info from gary.

The Filters

OK, the long and short of it, Alexisonfire helped drive home something I should have probably known, but didn't. I was with him and extremely surprised at his "In Camera" shots, they do resemble very accurately his finished articles. I felt in order to get something similar, would require PP trickery and what not...

However, he simply explained - if there are even slight colours in the sky, using an ND filter, and slowing that exposure right down, would intensify said colours. Sounds simple, but it was something I was not aware of.

At one point we did a test. I had 6 stops of grad on the sky, he had 5 stops of grad on the sky, and a 3 stop ND. The vast difference in colours was absolutely incredible.

To cut a long story short, it was not long before I had set my custom white balance (another absolutely mind blowingly simple task, with earth shattering results!), and the 10 stop! I was getting photos that I could not have dreamed of within minutes.

Don't ask me about the science, I haven't a clue how it works. Woodsy tried to explain it to me, something to do with emulsion on a wall, one coat = no ND's, several coats = ND's (I think!) . Yes, he lost me

Gary.
Click to expand...

If anyone has any info on how they shoot landscape with ND filters & manual white balance settings then post in here. :)
 
Just to add to my points above, they are specific to sunrise / sunset and ambient sun related light. Not sure if using such techniques during the day would be beneficial or not. Keen to find out!

With regards to White Balance...

The white balance I would have an Auto mostly, OR I would choose sunlight / cloudy etc, depending on conditions. However, when the conditions allowed for it, for example - strong ambient light from the sun at 1am, maybe 2am, due to the "unnatural" feel of the light, I thought it OK to tweak it all the way from a nice purple to a deep red. The reality is, the light falls along those lines anyway - you have to see a "sunlit" beach at 2am to believe it.


Gary.
 
I'm also interested in this. I mostly shoot landscapes with my Cokin ND and ND8 with the wide angle holder but I recently got a 52mm screw-in 10 stop filter but haven't given it a run for it's money yet. As for white balance I shoot it in AWB and fiddle with it later in RAW if it doesn't look right but there's got to be a more accurate way of doing it!
 
Bendihossan said:
I'm also interested in this. I mostly shoot landscapes with my Cokin ND and ND8 with the wide angle holder but I recently got a 52mm screw-in 10 stop filter but haven't given it a run for it's money yet. As for white balance I shoot it in AWB and fiddle with it later in RAW if it doesn't look right but there's got to be a more accurate way of doing it!
Click to expand...

I like the "Get it right in camera" approach more than anything, and having now learned how to use in field custom white balance, I feel you can get 95% towards your finished article. A few levels tweak and some shadow boosts, job done :)

Gary.
 
Snapper73 said:
If anyone has any info on how they shoot landscape with ND filters & manual white balance settings then post in here. :)
Click to expand...

I'd love to know this as well because as far as I'm aware an ND filter will have absolutely no affect on white balance as the light source (and temperature) is the same whether the filter is on or off. The filter just changes the amount of light reaching the sensor over a given time.

Would be interested in reading the science behind the claim that the filter changes the white balance.

George.
 
BadgerLover said:
I'd love to know this as well because as far as I'm aware an ND filter will have absolutely no affect on white balance as the light source (and temperature) is the same whether the filter is on or off. The filter just changes the amount of light reaching the sensor over a given time.

Would be interested in reading the science behind the claim that the filter changes the white balance.

George.
Click to expand...

Not sure it changes the white balance as such. It without a doubt, intensifies the appearance of the light in camera. For example, a 3 stop difference was like comparing a pale peachy pink to a deep dark red. It was insane :) The white balance tweaks are a separate area, you need to use the camera settings to make them before your shot, OR you do it with the raw file later.

Gary.
 
If an ND filter has a colour case (some do) this would be corrected by using a custom white balance with the filter on the camera at the time the reference shot is taken. You would have to be careful with ND grads as only part of the picture will be corrected and may well give a complementary (as in opposite, not good) colour cast to the unfiltered part of the image!
 
Just to add it really depends on what filters you are using as well, I found the Cokin ones gave a horrible colour cast and this is why I went for the Lee system with the 100mm Hitechs. You have to remember different filters will give different results in camera.
 
Snapper73 said:
Just to add it really depends on what filters you are using as well, I found the Cokin ones gave a horrible colour cast and this is why I went for the Lee system with the 100mm Hitechs. You have to remember different filters will give different results in camera.
Click to expand...

Fair point. I think Alexisonfire used the Cokin system, obviously white balance adjustments again could undo the colour cast and / or PP? Woodsy used to use Cokin and also got around the colour cast with no great difficulty I think. Still, better to get it right in cam.

Gary.
 
As far as I know, setting a custom white balance requires you to fill the frame with a neutral colour or card.

So can I assume that you did just the same thing with the filters on.
 
Assuming the ND Grad has no colour cast, then all it's doing is reducing the light levels to reduce contrast and bring both the highlights and shadows into a range the camera's sensor can accurately record, this then gives clouds etc. more definition and detail - a bit like the difference you see trying to peek at the sun when in & out of clouds compared to looking at it with good sunglasses on

WB changes quite quickly at sunrise/sunset, so I tend to keep using the Expo every 30 mins or so, but only as a start - where I may have to adjust in ACR later

Getting WOW colours at capture is simply a case of fiddling with your custom WB settings, but raw capture allows you to 'correct' them later

I don't like coloured filters as they affect the capture, and I can always add colour later if needed - but removing it can be a PITA

:)

DD
 
DiddyDave said:
Getting WOW colours at capture is simply a case of fiddling with your custom WB settings, but raw capture allows you to 'correct' them later
Click to expand...


Just to clarify though, adding a NON Graduated ND DOES indeed create massively more rich colours around sunrise, sunset, and the ambient light around these times. So much so that I was shocked. 3 stops made such a difference it was insane, a pale pink became a deep red. Both were correctly exposed, yet one, was so much fuller and richer than the other. Maybe it's best not to think of the filter doing the work, and instead concentrate on what happens with this light when you manage a long exposure?

Gary.
 
DiddyDave said:
I don't like coloured filters as they affect the capture, and I can always add colour later if needed - but removing it can be a PITA
Click to expand...

OK, but bear in mind that the most detail exists to the right of the histogram. Intensifying red light from the middle of the histogram later yields less detail than doing it on the sensor.
 
EdinburghGary said:
Fair point. I think Alexisonfire used the Cokin system, obviously white balance adjustments again could undo the colour cast and / or PP? Woodsy used to use Cokin and also got around the colour cast with no great difficulty I think. Still, better to get it right in cam.

Gary.
Click to expand...

In which case I'd be very interested as I've recently bought these cokin filters and have been horrified at the magenta cast.

Cokin and Warehouse express told me it's beause of my digital camera (50D) and they are designed for film. WE haven't yet posted my review expressing my displeasure at the purple nastiness.
 
Byker28i said:
In which case I'd be very interested as I've recently bought these cokin filters and have been horrified at the magenta cast.

Cokin and Warehouse express told me it's beause of my digital camera (50D) and they are designed for film. WE haven't yet posted my review expressing my displeasure at the purple nastiness.
Click to expand...

Have you tried whie balance compensation yet? In CS Raw for example, could you not use the TINT slider to move from the centre point (purple on left, green on right), and simply drag upwards slightly to the green side?

Once you have mastered the fix in the white balance raw, you can then attempt to create a custom white balance in camera for use when filters are attached?

My B&W 10 stop has a real nasty cast too, I just use raw to sort.

Gary.
 
EdinburghGary said:
Just to clarify though, adding a NON Graduated ND DOES indeed create massively more rich colours around sunrise, sunset, and the ambient light around these times. So much so that I was shocked. 3 stops made such a difference it was insane, a pale pink became a deep red. Both were correctly exposed, yet one, was so much fuller and richer than the other. Maybe it's best not to think of the filter doing the work, and instead concentrate on what happens with this light when you manage a long exposure?

Gary.
Click to expand...


Ok then - as that's just increasing the length of exposure, then that has to be a factor in the increased colours, not too sure why though :thinking:

DD
 
EdinburghGary said:
Have you tried whie balance compensation yet? In CS Raw for example, could you not use the TINT slider to move from the centre point (purple on left, green on right), and simply drag upwards slightly to the green side?

Once you have mastered the fix in the white balance raw, you can then attempt to create a custom white balance in camera for use when filters are attached?

My B&W 10 stop has a real nasty cast too, I just use raw to sort.

Gary.
Click to expand...

I tried but it really didn't work, hence the idea about using a custom white balance appealed. I'd resigned myself to throwing away the filters and buying alternatives.
 
A few comments on the comments above.

ND filters often produce a colour cast because they are not neutral. This is not because they are designed for film, it's because it is hard to produce truly neutral ND, particularly a strong one. Most have a red/pink/purple/orange cast of varying degrees. Cokin get a lot of stick for this, maybe because they are not very good, maybe because different batches of filters vary, and maybe because they are the most popular in use.

An image taken with an ND grad with a colour cast is very hard to correct, as only half the image is wrong. Sometimes it matters, sometimes it doesn't.

Colour temperature and white balance is not an exact science. Sometimes it only looks right when it's wrong, eg a sunset wouldn't look right if all the red/orange was taken out. Also a pic taken under candle light looks more natural it is much warmer than technically accurate.

When you change white balance, what you are doing is changing the relative exposure of red, blue and green channels. Or more correctly, increasing/decreasing the ISO rating of red and blue (not green). I believe that this correction is applied to the Raw files as well as JPEG, but this may be specific to different cameras. (Anybody done any tests on that?) But if this applies, and if your white balance is miles out and you attempt to correct it in Raw (as is commonly done) then while you can probably get the colour back, you also run the risk of generating a lot of noise in the channel that needs boosting the most, as it might be a couple or more stops adrift of optimum.

For example, an indoor shot taken under tungsten light but with the white balance accidentally set for flash, the blue channel will be substantially underexposed. You will see it in the colour histogram (if your camera has it) even though the normal 'white' histogram looks fine. In other words, if your white balance is not at least close at the time of taking, correcting it in post will increase noise, even from Raw.
 
HoppyUK said:
For example, an indoor shot taken under tungsten light but with the white balance accidentally set for flash, the blue channel will be substantially underexposed. You will see it in the colour histogram (if your camera has it) even though the normal 'white' histogram looks fine. In other words, if your white balance is not at least close at the time of taking, correcting it in post will increase noise, even from Raw.
Click to expand...

No, that's not true.

The histogram is generated from the in-camera JPEG, not the RAW file. The white balance is applied to the in-camera JPEG and is not applied to the RAW file.
 
Cokin get a bad name because they deserve it.

This is with a ND 8 - I got up at 5am on holiday for this.
113823525.jpg


The same shot with a polariser and warming filter
113823528.jpg



I've got some much worse examples shot mid afternoon sky, rather than sunrise.
 
Blapto said:
No, that's not true.
Click to expand...


Edit: Forget this post, and my previous one about white balance. Blapto is right :) The white balance is applied to the in-camera JPEG and is not applied to the RAW file. See footnote below.


Is it not? Have you tried it? Sorry to go on, but since we are talking about some extremes of exposure here I think it's important to be right ;) and I've not actually tested it myself.

I know that the histogram is generated from a JPEG (even when you're not shooting JPEGs) but I believe that the only time that the Raw file is adjusted is because of exposure, ie shutter speed, f/number, ISO and white balance (which is an aspect of ISO). And white balance does, technically, come under the heading of exposure adjustment between different channels. It is not the same as Picture Styles in this respect, which apply only to JPEGs.

What I'm saying is that if you need to make big adjustments to Raw files in post processing to correct white balance, you will get more noise than if you have got the white balance at least approximately right in shooting. But I have never done this to prove it, only heard it from others, but it makes sense to me.

Are you saying that adjusting white balance at the time of shooting makes zero difference to the Raw file? Have you tested this? Sorry to go on about it, don't mean to labour the point, but if the Raw is not affected and white balance is just a bit of cosmetic post processing, there are some important considerations for exposure, ie, it would be fundamentally better to correct colour with filters (as you do with film) than with digital post processing.

Footnote edit: I was lead to believe that adjusting the white balance adltered the colours in the Raw file. It can only do this, I believe, by altering the exposure. Because the camera cannot increase the amount of light, in order to give more exposure to blue for example (in a shot under tungsten light perhaps) it would have to cut red, and then increase the overall exposure to compensate.

However, having done a few tests shooting a white wall with white balance at the two extremes (2500K and 10000K) the exposure remained identical, even though the colour histograms are dramatically changed (being processed JPEGs). But maybe there is still a learning point here, in that if you want optimum results while adjusting white balance, doing it with filters in front of the lens and increasing exposure accordingly will give lower noise. I think that's right now :D
 
HoppyUK said:
Is it not? Have you tried it? Sorry to go on, but since we are talking about some extremes of exposure here I think it's important to be right ;) and I've not actually tested it myself.

I know that the histogram is generated from a JPEG (even when you're not shooting JPEGs) but I believe that the only time that the Raw file is adjusted is because of exposure, ie shutter speed, f/number, ISO and White Balance (which is an aspect of ISO). And white balance does, technically, come under the heading of exposure adjustment between different channels. It is not the same as Picture Styles in this respect, which apply only to JPEGs.

What I'm saying is that if you need to make big adjustments to Raw files in post processing to correct white balance, you will get more noise than if you have got the white balance at least approximately right in shooting. But I have never done this to prove it, only heard it from others, but it makes sense to me.

Are you saying that adjusting white balance at the time of shooting makes zero difference to the Raw file? Have you tested this? Sorry to go on about it, don't mean to labour the point, but if the Raw is not affected and white balance is just a bit of cosmetic post processing, there are some important considerations for exposure, ie, it would be fundamentally better to correct colour with filters (as you do with film) than with digital post processing.
Click to expand...

I could write a long post debunking this, or I could just ask you to stop stating things as fact without testing them for yourself. Pretending that you know what you're talking about just misleads people.

If it helps, think about it this way:
A RAW file is not an image. It's a dump of what the sensor and assorted electronics collected.

Changing the white balance of a file does not affect the RAW in any way, other than tagging it with the white balance you've selected for your processing software to use as a starting point.
 
HoppyUK said:
White Balance (which is an aspect of ISO). And white balance does, technically, come under the heading of exposure adjustment between different channels.
Click to expand...

Are you under the impression that to adjust white balance the colour output of the sensor changes? That reads like you think that if you set a more blue balance then the sensitivity of the blue pixels is increased independently of the red or green pixels.
 
HoppyUK said:
Is it not? Have you tried it? Sorry to go on, but since we are talking about some extremes of exposure here I think it's important to be right ;) and I've not actually tested it myself.

I know that the histogram is generated from a JPEG (even when you're not shooting JPEGs) but I believe that the only time that the Raw file is adjusted is because of exposure, ie shutter speed, f/number, ISO and White Balance (which is an aspect of ISO). And white balance does, technically, come under the heading of exposure adjustment between different channels. It is not the same as Picture Styles in this respect, which apply only to JPEGs.

What I'm saying is that if you need to make big adjustments to Raw files in post processing to correct white balance, you will get more noise than if you have got the white balance at least approximately right in shooting. But I have never done this to prove it, only heard it from others, but it makes sense to me.

Are you saying that adjusting white balance at the time of shooting makes zero difference to the Raw file? Have you tested this? Sorry to go on about it, don't mean to labour the point, but if the Raw is not affected and white balance is just a bit of cosmetic post processing, there are some important considerations for exposure, ie, it would be fundamentally better to correct colour with filters (as you do with film) than with digital post processing.
Click to expand...

Raw data is precisely that, all recording info is tagged to the file header and in the case of canon this also includes the sensitivity of the chip, the white balance and the picture style info but the data is raw. It is the software you use to convert it that reads some/all of the header and translates it into wb, picture style etc.

For info on how a raw file is compiled see here - http://www.sno.phy.queensu.ca/~phil/exiftool/canon_raw.html

White balance and picture style has zero effect on raw recorded information.
 
This is from the Singh Ray website:

"I'm getting a color cast with my Neutral Density filter. What's the problem?

A: When using a Neutral Density filter (graduated or solid) you are generally causing your camera to make a longer exposure than normal, sometimes much longer. The software in some digital cameras uses response curves that adjust the RGB sensors sensitivity based on the exposure time, which may introduce more red into the image. Sort of a digital form of reciprocity failure. In other instances, blocking visible light and a longer exposure allows infrared light to "build up" on the red sensors leading to a red or purple-ish cast. If your filter looks neutral but you see the cast in your images, try taking some test shots using the filter with a different brand of camera, or with a film camera. If you still see the cast, and you're within the warranty period, contact Singh-Ray to have your filter evaluated and possibly repaired/replaced. "
 
CScottMcQueen said:
Raw data is precisely that, all recording info is tagged to the file header and in the case of canon this also includes the sensitivity of the chip, the white balance and the picture style info but the data is raw. It is the software you use to convert it that reads some/all of the header and translates it into wb, picture style etc.

For info on how a raw file is compiled see here - http://www.sno.phy.queensu.ca/~phil/exiftool/canon_raw.html

White balance and picture style has zero effect on raw recorded information.
Click to expand...

Ah that explains it very well. So there's extra fields/flags that the software (in my case Lightroom) uses to initially interpret the image. That would explain the blue raw pictures when I forget to reset the WB.
 
In truth, I don't know if I can put into words why it is that colours appear more saturated when using solid ND filters; I know why in my head, but that can be a strange place at times!... All I can say for sure is that upon testing it in the field, at sunset, the use of the 3 stop Lee ND filter made a huge difference to the image, and the colours were considerably more vibrant.
 
Woodsy said:
In truth, I don't know if I can put into words why it is that colours appear more saturated when using solid ND filters; I know why in my head, but that can be a strange place at times!... All I can say for sure is that upon testing it in the field, at sunset, the use of the 3 stop Lee ND filter made a huge difference to the image, and the colours were considerably more vibrant.
Click to expand...

Damn straight!!! Don't forget the emulsion! :nono: :D
 
Blapto said:
I could write a long post debunking this, or I could just ask you to stop stating things as fact without testing them for yourself. Pretending that you know what you're talking about just misleads people.

If it helps, think about it this way:
A RAW file is not an image. It's a dump of what the sensor and assorted electronics collected.

Changing the white balance of a file does not affect the RAW in any way, other than tagging it with the white balance you've selected for your processing software to use as a starting point.
Click to expand...

Apologies Blapto ;) Crossed posts. I have edited my previous post, correcting it as you describe.

I have now tested the reaonsing as I understand it, but the info I had came from a good source.

Didn't mean to start hares running. Happy to ammend ;)

Cheers,

Richard.
 
Gandhi said:
This is from the Singh Ray website:

"I'm getting a color cast with my Neutral Density filter. What's the problem?

A: When using a Neutral Density filter (graduated or solid) you are generally causing your camera to make a longer exposure than normal, sometimes much longer. The software in some digital cameras uses response curves that adjust the RGB sensors sensitivity based on the exposure time, which may introduce more red into the image. Sort of a digital form of reciprocity failure. In other instances, blocking visible light and a longer exposure allows infrared light to "build up" on the red sensors leading to a red or purple-ish cast. If your filter looks neutral but you see the cast in your images, try taking some test shots using the filter with a different brand of camera, or with a film camera. If you still see the cast, and you're within the warranty period, contact Singh-Ray to have your filter evaluated and possibly repaired/replaced. "
Click to expand...

Sounds a plausible explanation (from a filter manufacturer ;) ).

But reciprocity failure is a film thing, and doesn't apply to digital - indeed the lack of it is a major benefit for very long exposures such as astronomical.

And if the filter was fooling the camera into making white balance adjustments in AWB mode, then it would only ever apply to JPEGs and the Raws would be clean.
 
they did only say;

Sort of a digital form of reciprocity failure
Click to expand...

Re-read what they have to say. Its not about the white balance or fooling the camera at all. Its about how the camera deals with the varying sensitivity of the different colour pixels on the sensor. The RAW data isnt truly raw, its the cameras interpretation of how to amplify the seperate colour channels off of the sensor in order to produce a balanced image.
 
You must log in or register to reply here.
Back
Top