RAW and white balance + red filter

Currently, I use a Canon Powershot S1 IS. I shoot jpeg, use a red filter (or rather color correcting filter) and custom white balance using a white slate.

I’ve now bought a Canon G9 and plan to shoot in RAW format. I don’t have a lot of experience editing RAW files, but I always read RAW format allows you to do a lot of post-processing without quality loss. Does this mean I no longer have to custom white balance underwater (would save a lot of hassle imo), or is there still some advantage to setting custom WB underwater?

Furthermore, how easy is it to correct colors in RAW without quality loss? Would my red filter also become obsolete when shooting RAW? I’ve found that the red filter makes focussing more difficult, especially in low light, since it takes some ligt away. But i has generated some nices results for me when shooting jpeg with the S1 IS. Is there an advantage in using a red filter over correcting colors in post-processing?

OK, let's take this one at a time...

Currently, I use a Canon Powershot S1 IS. I shoot jpeg, use a red filter (or rather color correcting filter) and custom white balance using a white slate.

If you are doing your custom WB and then putting the filter on, the filter will have an effect. [in fact, I do this when I'm shooting JPEG on dry land with a Tiffen 812 filter.]

On the other hand, if you are putting the filter on and then white balancing, you're not accomplishinganything, as you are "correcting the corrections."

Now, let's move on to RAW...

I’ve now bought a Canon G9 and plan to shoot in RAW format. I don’t have a lot of experience editing RAW files, but I always read RAW format allows you to do a lot of post-processing without quality loss. Does this mean I no longer have to custom white balance underwater (would save a lot of hassle imo), or is there still some advantage to setting custom WB underwater?

In general, setting a custom white balance doesn't accomplish much, since you are recording everything the CCD actually sees -- The "raw" data. All it (setting a custom WB) does is write a tag into the EXIF metadata with the color settings, which Adobe Camera Raw (ACR) reads.

You'll notice in the pop-up above the color temp & tint sliders choices for "Auto" and "As shot" -- If you select As shot, ACR will read the file metadata and use it to set the sliders. If you select Auto, ACR will calculate the white point and tint, ignoring what the camera's on-board processor chose. Depending on the algorithms the camera manufacturer chooses, these (Auto and As shot) results can diverge widely.

Keep in mind, though, that in ACR, these are just suggested white point & tint settings: In fact, if you move either of those sliders, the pop-up will change to Custom.

(As a side note, often I'll look at both Auto and As shot results in the preview, noting the temp & tint values; then I take the average of the two and use this as my starting point; adjusting from there)

The problem with underwater is that everything goes to hell in a handbasket: If you use Auto WB in your camera (and As shot in ACR), it will try to "tune out" the blue cast; while if you use Auto in ACR it will ignore what your camera thought was correct and instead ACR will try to "tune out" the blue cast.

So, as Yogi Berra famously said, If you come to a fork in the road, take it! Since the actual data recorded on the CCD does not change one iota:

A] You can shoot Raw, forget about white point & color balance while you are underwater (i.e. leave the camera on auto WB) and worry about color correction on your monitor based on what you remember & what pleases your eyes;

B] Go ahead and record the Custom WB tag while underwater, use the As shot setting in ACR, and then fine tune.

If you're on dry land, setting a custom WB is quick & easy... But when shooting RAW, doesn't affect the end results all that much.

Underwater, setting a custom WB is a whole different story: Your decision rests on your diving skills and your Photoshop color correction skills.

Keep in mind that the actual image data (as opposed to the metadata tags) in the RAW file will be the same!

Also, think of RAW files as actually "digital negatives:" Just like a piece of film, you can put it in the enlarger over and over again, with different CC & exposure settings each time.

Furthermore, you can open these same RAW files a couple years later and get better results, because both the software will have improved... And so will your digital photo editing skills.

Furthermore, how easy is it to correct colors in RAW without quality loss? Would my red filter also become obsolete when shooting RAW?

Pretty much so...

I’ve found that the red filter makes focussing more difficult, especially in low light, since it takes some ligt away.

Yes, the filter factor is probably about 2 stops: This will force the camera's exposure computer to crank up the ISO (sensitivity) of the sensor, which, for the smaller sensors used in "prosumer" cameras, will result in more noise.

But i has generated some nices results for me when shooting jpeg with the S1 IS. Is there an advantage in using a red filter over correcting colors in post-processing?

In your case, leave the filter off... You don't have a Canon 5D, after all!

A] You can shoot Raw, forget about white point & color balance while you are underwater (i.e. leave the camera on auto WB) and worry about color correction on your monitor based on what you remember & what pleases your eyes;

 

B] Go ahead and record the Custom WB tag while underwater, use the As shot setting in ACR, and then fine tune.

Ok, so basically, the 'advantage' of Custom WB underwater is that you start off with a WB that is a little bit closer to where you want to be than if you Auto WB. But in the end you'll still want to fine tune the WB to what you think looks right. I figured as much, but I wasn't sure how much work it would be to get the white balance 'right' without a starting point from Custom WB. But I guess 'right' is just what you think looks nice, and from what you say it sounds pretty straight forward. Excellent explanation btw, thanks.

Underwater, setting a custom WB is a whole different story: Your decision rests on your diving skills and your Photoshop color correction skills.

It's not so much a problem with my diving skills, as with my discipline. I tend to forget to Custom WB when I descend or ascend, or the light changes. Then when I want to take a shot I remember and first have to Custom WB and tada... photo opportunity gone. So I think I'll forget about Custom WB from now on.

My photoshop skills are not great, but they are growing. That's why I ran into boundaries with the S1 IS. 3.2 megapixels jpeg really isn't much to work with.

In your case, leave the filter off... You don't have a Canon 5D, after all!

Ok, I guess I'll sell the filter along with the old camera. And no it's not a dSLR, but it'll do for now. I spend that money on dive trips for now...

It is not essential to use custom white balance when shooting raw. Using optimized white balance will result in the most accurate starting point in your raw converter, it will make the image review as good as it can be, and it will result in the most accurate RGB histograms if your camera offers them. Some very experienced shooters prefer to combine custom white balance with raw for these reasons.

It is not the case that you can do post-processing without data loss in raw. All post-processing involves data loss. With raw you simply start out with a lot more data.

Yes, adding a filter will reduce AF performance because light is lost. Yes, there are specific advantages to using color filters rather than adjusting in post. Wetpixel has had many threads on this in the past.

Now to correct some of the incorrect information Dan has provided:

The purpose of using a color filter underwater is to correct color imbalances in the light, not to impose a color cast on the final image. If you'd ever taken an underwater shot, Dan, you'd understand this. Custom white balance is ALWAYS taken just prior to a shot underwater. It is ALWAYS taken through the filter.

Not all people use ACR, Dan.

White balance settings CAN effect imager data IF auto exposure is used.

You cannot remember what an image looked like underwater when doing white balance in post. That is the most unreliable method that there is. Not only does your brain adapt to the hideous color environment underwater, it never sees the image the camera does at the moment the strobe fires. The computer is far better than you are at white balance. Use your objective tools.

Underwater custom white balance is easy and doesn't rely on diving skills or photoshop expertise. It does require a white reference and frequent fussing with the camera.

Dan, claiming that post-processing in raw doesn't result in data loss is wrong.

Dan, you have no idea what filter is being used so you can't possibly know that the filter factor is two stops. In fact, you don't even know what the filter is for. Furthermore, saying that the camera will adjust ISO to compensate is totally wrong. Adding a filter will not result in more noise unless the user is forced to bump ISO to compensate. Most cameras will never adjust ISO on their own.

Dan, I suggest you learn more about underwater shooting before advising others on how to shoot. The decision to use a filter underwater depends on a number of things, but least of all the type camera you use. A Canon 5D has nothing to do with it. Frankly, a 5D can MOST afford to do with a filter.

jkolgen, lots of experienced underwater photographers choose not employ filters and there's nothing wrong with that decision. Some types of shooting can be enhanced by filters, so there's nothing wrong with using them either. They're simply another tool in your bag. The argument for filters has nothing to do with shooting raw or using custom white balance. If all you have is JPG, though, filters and custom white balance are more compelling. Finally, I suspect you have more experience shooting underwater than Dan does, so keep that in mind when you decide whether you should abandon the filters you have based on his advice.

Craig,

What you do not know is how many UW photos from other photographers I have color corrected and printed in a commercial photo lab -- Both in Photoshop as well as in the enlarger.

And No, as long as the proper conversion is done from the 12 bit CCD-RAW file into 16 bit/channel color space, the raw pixel data is, in general, not lost. Besides, one can start all over again from the initial file; and, if need be, composite two separate conversions of the same RAW file to better extract shadow details & highlight details.

It's late at night; and I'll respond fully to your "concerns" tomorrow.

In the interim, as a Photoshop user since 1993 (v2.51) on a Mac Quadra 950, I stand fully by what I wrote above.

It is not essential to use custom white balance when shooting raw. Using optimized white balance will result in the most accurate starting point in your raw converter, it will make the image review as good as it can be, and it will result in the most accurate RGB histograms if your camera offers them. Some very experienced shooters prefer to combine custom white balance with raw for these reasons.

 

It is not the case that you can do post-processing without data loss in raw. All post-processing involves data loss. With raw you simply start out with a lot more data.

 

Yes, adding a filter will reduce AF performance because light is lost. Yes, there are specific advantages to using color filters rather than adjusting in post. Wetpixel has had many threads on this in the past.

 

Now to correct some of the incorrect information Dan has provided:

On the other hand, if you are putting the filter on and then white balancing, you're not accomplishinganything, as you are "correcting the corrections."

This is *ABSOLUTELY* wrong.

Dan, when is the last time you, personally, tested the effects of a filter underwater? Please do not give advice when you are not qualified to do so.

There is a ton of information about filtered shooting here in the forums. Do a search.

Typically, people use red/orange filters (depending on water color) to block out blue and green light, which allows for more red light to hit the sensor. Normally, underwater strobe-less shots have little or no red information, but if you choose to custom white balance without the filter, you may not have enough red light to have a clean image. You may even not have enough for a "proper" white balance. We use the filter to boost that red channel up so when you white balance, your red channel isn't full of noise.

I hate that we have to continuously correct Dan Schwartz for the misinformation he is inserting into these forums. On land, Dan may have useful information. But for underwater shots, my advise would be to ignore most of what he is writing.

We may have to take further action to ensure the integrity of our forums, if this continues. 100% of the staff is very annoyed with that is going on. We can't have misinformation presented in such a cocksure method. The next time Dan posts advise that is wrong, I am going to personally suspend him for a week. You have pushed me over the edge.

Yes even I have to butt in say something about that one.

Underwater it can get to a point where the camera will not manually WB to what you want so a red filter will help it just like what ERIC has stated. The same happens with a Video Camera.

Continuing, for a few minutes...

 

The purpose of using a color filter underwater is to correct color imbalances in the light, not to impose a color cast on the final image.

One and the same, since CCD & CMOS imagers do not suffer from color crossover. You DO know what crossover is, don't you?

If you'd ever taken an underwater shot, Dan, you'd understand this. Custom white balance is ALWAYS taken just prior to a shot underwater. It is ALWAYS taken through the filter.

See above re my experience; and add in ColorGetter drum scanning software, too.

Setting a custom WB through a filter when shooting RAW accomplishes nothing, because any filtration is compensated out. Period.

The pixel capture is linear. The filter is linear. Period.

JPEG (JFIF, EXIF) is, of course, a different story, as the transformation into sRGB or AdobeRGB color spaces is non-linear, as is the chroma subsampling.

When shooting with in-camera conversion to JPEG, setting the white balance through the filter is critical.

Not all people use ACR, Dan.

Adobe Camera Raw (ACR) is pretty much the standard, because it can be used to open and convert RAW files from most every camera and digiback, since they all pretty much use the same Bayer GRGB pattern; Fuji has supplied algorithms for their SuperCCD octagonal pattern; and ACR is also extensible, hence new versions come out quickly to support new cameras.

White balance settings CAN effect imager data IF auto exposure is used.

No, Not for RAW: Exposure is calculated based on luminance to an 18% average for the scene (& then manually offset by a certain EV, if chosen).

You cannot remember what an image looked like underwater when doing white balance in post. That is the most unreliable method that there is. Not only does your brain adapt to the hideous color environment underwater, it never sees the image the camera does at the moment the strobe fires. The computer is far better than you are at white balance. Use your objective tools.

That is true.

Underwater custom white balance is easy and doesn't rely on diving skills or photoshop expertise. It does require a white reference and frequent fussing with the camera.

JKolgen, the original poster, wrote "would save a lot of hassle imo".

And then he wrote in his followup, "It's not so much a problem with my diving skills, as with my discipline. I tend to forget to Custom WB when I descend or ascend, or the light changes. Then when I want to take a shot I remember and first have to Custom WB and tada... photo opportunity gone."

Dan, claiming that post-processing in raw doesn't result in data loss is wrong.

Again, false. Only if you clip at the low &/or high end of the pixel value will data be lost; or if you apply a transformation that improperly "bunches up" the data. Working in 16 bit/color/pixel precision -- Up from 12 bit/color/pixel precision from the CCD as written into the RAW file.

Dan, you have no idea what filter is being used so you can't possibly know that the filter factor is two stops.

Most color correcting filters tend to tend to have a filter factor in the 1.5 to 2 range.

In fact, you don't even know what the filter is for. Furthermore, saying that the camera will adjust ISO to compensate is totally wrong. Adding a filter will not result in more noise unless the user is forced to bump ISO to compensate. Most cameras will never adjust ISO on their own.

That is totally false for point & shoot and many prosumer cameras, as automatically cranking up the CCD amplifiers is used in low light situations to keep shutter speeds reasonable. This "feature" is not found, AFAIK, on dSLR cameras, though.

Dan, I suggest you learn more about underwater shooting before advising others on how to shoot. The decision to use a filter underwater depends on a number of things, but least of all the type camera you use. A Canon 5D has nothing to do with it. Frankly, a 5D can MOST afford to do with a filter.

I used the Canon 5D as an example of a camera that CAN afford to have the ISO increased, due to its low noise figure at high sensitivities.

Point & shoot and prosumer cameras -- Especially ones with lots of pixels -- on the other hand, generally fall apart at higher sensitivities, due to the much smaller photosite size.

Looking up the data on DPReview & Canon websites, we find that Mr. Kolgen's PowerShot S1 IS has 3.3 million pixels on a 1/2.7" CCD; while his PowerShot G9 has 12.1 million pixels on a 1/1.7" CCD.

Both cameras have an Auto ISO mode; and the G9 also has a "High Auto ISO" mode. I don't know if Auto ISO is the camera default setting, though.

jkolgen, lots of experienced underwater photographers choose not employ filters and there's nothing wrong with that decision. Some types of shooting can be enhanced by filters, so there's nothing wrong with using them either. They're simply another tool in your bag. The argument for filters has nothing to do with shooting raw or using custom white balance. If all you have is JPG, though, filters and custom white balance are more compelling. Finally, I suspect you have more experience shooting underwater than Dan does, so keep that in mind when you decide whether you should abandon the filters you have based on his advice.

Using a custom WB, and then mounting a filter is helpful when the camera is set to in-camera JPEG processing.

You, yourself said, and I quote:

It's not so much a problem with my diving skills, as with my discipline. I tend to forget to Custom WB when I descend or ascend, or the light changes. Then when I want to take a shot I remember and first have to Custom WB and tada... photo opportunity gone.

In other words, his image data loss is 100%.

Digital video cameras are non-linear due to the MPEG compression: I haven't seen a DV camera that writes in CCD-RAW format.

Digital still cameras produce a non-linear file when in-camera processing hands you a JPEG file: White balancing and filters are very important.

Film produces a highly non-linear file, due not only to the gamma, but also the compression on the toe & shoulder.

Digital still cameras produce linear file data (i.e. it has a linear transfer function) when writing CCD-RAW (and the image data is not clipped up in the highlights).

An optical filter has a linear transfer function, because it is a passive device.

A linear (filter) transfer function applied to a device with a linear transfer function (CCD) results in a linear output. Period.

Craig, go back to electrical engineering school and study signals & systems.

Eric: Working with CCD-RAW data requires a completely different mindset to what you are used to, either from the in-camera processing for photography, or from the gamma curves applied via scanner software; and through the gamma settings on your video adapter & monitor.

I'll try to guide you through this maze: It gave fits to the graphics & prepress industry 10-15 years ago as they moved away from halftone screens & camera-back color seps to the digital workflow; and it's giving fits to photographers today.

I'm trying to give "arm waving" or "cookbook" explanations of the results of complex engineering math, such as Fourier transforms, signal convolution through linear & nonlinear systems, and, to a degree (no pun intended), blackbody radiation.

---------------------

And Yes, I have color corrected underwater images in commercial prepress and photo lab shops, both in Photoshop and in the enlarger... Just not my own underwater images.

Yes even I have to butt in say something about that one.

 

Underwater it can get to a point where the camera will not manually WB to what you want so a red filter will help it just like what ERIC has stated. The same happens with a Video Camera.

100% of the staff is very annoyed with that is going on.

...not only the staff... :glare:

Followup:

If accurate color rendition is needed, shoot a (laminated) Gretag chart in the given lighting conditions.

Once back on your computer, start the color correction process by snapping the neutral midtones to the 18% grey patch; then mouse over the color patches, read the various color values, and adjust the color balance accordingly. You can jot down the values you get; or you can simply record this as a Photoshop Action to play back on the batch of "real" photos shot in those given lighting conditions.

So Dan,

Lets say the camera's CMOS has a 12 million dots (pixels) on it. (example only figure)

You take the picture and each one of those dots pick up the light and saves all that information as a RAW.

That is each DOT on the CMOS turns the light it picked up into a electrical signal and into a RAW format.

Now each individual dot on the CMOS has picked up a tiny bit of different light than the dot beside it has right.

So you get it into your RAW editor on your computer.... PC of course

So you are saying that when you adjust the RAW image you have full & independent control of every dots individual light variation in case you have to change one of the DOTS because the light on that DOT when taken was wrong.

So Dan,

Lets say the camera's CMOS has a 12 million dots (pixels) on it. (example only figure)

Let's back up a step, since you're a video guy; and video production evolved differently than still photography production, due to the nature of the signal in the analog days.

(Incidentally, I was a co-op electrical engineer in (1979 & `80) at RCA Camden (New Jersey), where color TV, genloc, the VCR, and the optical video disc were invented, among other things.)

Anyway, TV cameras used something generically called a vidicon (actually an image orthicon) which produced a an analog signal, synchronyzed to the horizontal & vertical sweep. Without going too far afield, the "workflow" consisted of taking this analog waveform, processing it "on the fly" for color balance & mixing with other camera signals; and then either going straight to air or to videotape.

The key point to take away from all of this is that frame-by-frame editing, and editing a portion of the individual frame was, before the advent of digitizing the signal, all but impossible. Even something as basic as a split screen entailed quite a bit of circuit wizardry.

Fortunately, we've progressed quite a bit; but even though video has gone all digital, the workflow (video editing tools, such as Premiere) has (have) evolved where one is still dealing with a stream of frames, with the emphasis on "stream." (Though, of course, each frame is now addressable, and editable in Photoshop.)

Digital still photography evolved differently, however, due to what happens to the data once it leaves the CCD: Here, instead of the 330,000 Pix Elements ("Pixels") in each NTSC video frame being streamed, we have in a monochrome camera, say, 12 million, monochrome picture elements or photosites, each of which are frozen in time from the exposure, converted from the analog charge level to a digital value from 0 to 4095 (12 bit precision), and then sent on the way.

Color is simply a variation where, using our 12 million photosite example above, we can take three successive photos while rotating a color wheel with red, green and blue filters in front of it for a complete color picture. (In fact, early digital cameras used by NASA for Mars landers did just this.)

Or, we can install red, green, and blue colored micro filters in front of each of the 12 million photosites in our example, in what is called a Bayer pattern, also known as GRGB.

So, our 12 million photosites are now grouped into 3 million GRGB color picture elements (pixels). Each of these 3 million pixels can have 4096 red, 4096 green, and 4096 blue levels.

We now have a choice where we want to send this data:

Helpfully, digital camera manufacturers install additional processor power to take this raw feed, apply color transformations to balance out the white point, and convert it into an 8 bit per color (24 bit per pixel) JPEG file that matches the response of the eye, where the red, green and blue values each take on a value from 0 to 255; and then this data is compressed (lossy compression) and written to the flash memory.

Then, you bring this JPEG file into Photoshop, and you can edit this 3 million pixel file right down to the individual pixel.

But! "Shooting JPEG" poses two problems:

First, the conversion from RAW to JPEG in the camera is irreversible: If the camera is not set properly, the shot can be ruined, with little hope of recovery;

Second, for underwater images, the "white point" (actually the entire color balance, adjusted on the yellow-blue and green-magenta axes) is completely different than what the in-camera processor is expecting. It can try to adjust; but unless manual intervention takes place by adjusting the "white balance" before exposure, one runs the risk of the finished JPEG file being beyond repair.

But wait a minute! Why not peel the 3 million pixels of raw data off the CCD and write it directly to the flash memory... And bypass the in-camera processing directly? Now, we have 3 million red, 3 million green, and 3 million blue chunks of data, each of which have a value from 0 to 4095. Each and every one of these 9 million chunks of data is individually accessable in Photoshop. (more)

You take the picture and each one of those dots pick up the light and saves all that information as a RAW.

That is each DOT on the CMOS turns the light it picked up into a electrical signal and into a RAW format.

Now each individual dot on the CMOS has picked up a tiny bit of different light than the dot beside it has right.

That is correct: Each of the red, green, and blue "dots" has a brightness ranging from 0 to 4095 on a linear scale... And here, I stress the word linear, as it pertains directly to the subject of linear response color filters placed in front of the camera lens.

So you get it into your RAW editor on your computer.... PC of course

So you are saying that when you adjust the RAW image you have full & independent control of every dots individual light variation in case you have to change one of the DOTS because the light on that DOT when taken was wrong.

That is correct... And it's a whole lot of data to be crunched each time you make any changes. Oh, and that data has to be thrown up onto the screen, too!

Back in the early days of Photoshop (c. 1993), it was not uncommon for a basic operation like Unsharp Mask to take 10 minutes to crunch on a Mac Quadra with a 33 mHz CPU!

Today, in modern point & shoot cameras, this same operation takes just a few milliseconds in dedicated processors (like the Digic III processor in JKolgen's Canon Powershot G9.

--------------

Getting back to the subject of using a filter in front of the camera lens when shooting RAW, since the raw pixel data is in linear form, it makes little to no difference if an additional linear filter is placed in front of the camera lens to "tune" the color: The same linear filter can be applied in Photoshop!

Furthermore, this filter that "tunes" the color extracts a toll in that it compromises the exposure due to light loss: The diaphragm (iris) has to be opened up, changing the DOF, the shutter speed has to be decreased, and/or the gain of the CCD's amplifiers has to be cranked up, increasing noise... And, on Mr. Kolgen's Canon cameras, that last option is programmed in as "Auto ISO"... And it may be the (easily overrideable) camera default setting, too.

Second, for underwater images, the "white point" (actually the entire color balance, adjusted on the yellow-blue and green-magenta axes) is completely different than what the in-camera processor is expecting. It can try to adjust; but unless manual intervention takes place by adjusting the "white balance" before exposure, one runs the risk of the finished JPEG file being beyond repair.

Dan could you do me a great favour and with your vast knowledge explain how light works underwater and at different depths. As well as the effects that passing through a large mass of water has on the light.

Surely that information would be the right place to start when thinking about how RAW & White Balance + red filter could be used to enhance your underwater photography.

Craig,

 

What you do not know is how many UW photos from other photographers I have color corrected and printed in a commercial photo lab -- Both in Photoshop as well as in the enlarger.

Don't care either. It's not relevant to the issue at hand.

And No, as long as the proper conversion is done from the 12 bit CCD-RAW file into 16 bit/channel color space, the raw pixel data is, in general, not lost. Besides, one can start all over again from the initial file; and, if need be, composite two separate conversions of the same RAW file to better extract shadow details & highlight details.

Same can be said for a JPG. If you import it into photoshop and convert to 16 bits, the original JPG data is not lost. Of course, you haven't done any post-processing either. Any post-processing results in data loss regardless of the original image source. Raw images just provide the maximum data available, that's all.

In the interim, as a Photoshop user since 1993 (v2.51) on a Mac Quadra 950, I stand fully by what I wrote above.

I don't see how being an old Photoshop user, one with a limited grasp of the obvious at that, qualifies you to offer advice on underwater shooting, especially since it's not clear that you are have even shot underwater before.

It seems to me that if underwater there is a blue cast the histogram would show the blue channel more to the right and the red channel more to the left. This can be corrected with white balance in your RAW converter. But if a red filter was used to realign the 3 color channels prior to the sensor recording the image, you could expose to the right further (without clipping the blue channel) increase your dynamic range/decrease your shadow noise. It would be an interesting experiment with and without the filter to see if the light loss through the filter negates any advantage to realigning the histogram and ETTR? Not worth an argument though!

Marc

I think that people should not forget that a red filter filters out cyan light. It cannot add red. The amount of red light available is fixed at any particular depth (unless you take extra with you!) For this reason red filters can only work well where there is the red part of the spectrum penetrating, where there is plenty of contrast, and where there is enough sensitivity or a wide enough lens aperture to capture it.

(Yes, I have been underwater a bit and yes I did a three-year degree course in photography too - albeit before they invented light!)

Kurt Amsler said "If you are not shooting in RAW mode, you are missing the point of digital photography." I'm inclined to agree!

One and the same, since CCD & CMOS imagers do not suffer from color crossover. You DO know what crossover is, don't you?

They do suffer from finite dynamic range though. You need to go back to school, Dan.

Just when it seems your arguments can't get any more bizarre, they do.

Setting a custom WB through a filter when shooting RAW accomplishes nothing, because any filtration is compensated out. Period.

Custom white balance is not filtration, Dan. I explained the advantages of custom white balance with raw already. Meanwhile, filters absolutely do have an effect on raw data. It's embarrassing that anyone would even suggest that they don't.

The pixel capture is linear. The filter is linear. Period.

Another irrelevant point.

No, Not for RAW: Exposure is calculated based on luminance to an 18% average for the scene (& then manually offset by a certain EV, if chosen).

Cameras take white balance into account. The effect may be of little consequence or it may not, but underwater the green channel may be greatly surpressed once white balance is set. You can't argue that exposure won't be effected under those circumstances. Actually, knowing you, Dan, I'm not sure you won't argue it. Doesn't matter, though. There are specific advantages to custom white balance with raw shooting. For many shooters, those advantages aren't worth the trouble.

Most color correcting filters tend to tend to have a filter factor in the 1.5 to 2 range.

Wrong, Dan. Why don't you name some filters commonly used underwater and list the filter factors for those? I'll make it easy for you: I frequently use a CC30M filter underwater. Why don't you tell me why I do that and what the filter factor is for that example?

That is totally false for point & shoot and many prosumer cameras, as automatically cranking up the CCD amplifiers is used in low light situations to keep shutter speeds reasonable. This "feature" is not found, AFAIK, on dSLR cameras, though.

Wrong and wrong. Nikon DSLRs do, in fact, offer auto-ISO. Meanwhile, auto exposure functions do not include auto-ISO as part of their programs unless specifically configured (if available at all). P&S cameras, by virtue of their small sensors, have very fast lenses that still offer good depth of field. DSLRs need help with shutter speeds long before P&Ss do. Regardless, this is another one of your irrelevant arguments.

I used the Canon 5D as an example of a camera that CAN afford to have the ISO increased, due to its low noise figure at high sensitivities.

That's clear, except that you used the 5D as an example of a camera where you might consider a filter. The fact is that filters have nothing to do with the camera and everything to do with the light. The 5D, by virtue of its wide dynamic range, is an example of a camera that is MOST tolerant of wide color imbalances and therefore would be the last camera that would need filters, not among the first. Your 5D example just demonstrates your lack of fundamental understanding here.

Point & shoot and prosumer cameras -- Especially ones with lots of pixels -- on the other hand, generally fall apart at higher sensitivities, due to the much smaller photosite size.

More irrelevant points.

In other words, his image data loss is 100%.

Curious that you just argued that he was wrong on that, then used his words to somehow prove some point.

It's not me that is constantly confused between custom white balance and filters. Custom white balance, as has been said before, is not essential with RAW but offers some advantages. Each shooter should decide how he wishes to shoot in that respect. Filters are a completely different matter, and one you should try to understand, Dan. These forums offer a lot of good information on filters and there are technical articles on filters on Wetpixel as well.

Dan could you do me a great favour and with your vast knowledge explain how light works underwater and at different depths. As well as the effects that passing through a large mass of water has on the light.

Surely that information would be the right place to start when thinking about how RAW & White Balance + red filter could be used to enhance your underwater photography.

Giles,

In a perfect world, we have "Sunny 16" (EV 15) shooting conditions, and loads of noise-free CCD sensitivity on a 21 megapixel chip the size of your fingernail. Then, you can shoot RAW at an ISO (equivalent sensitivity) of 6400 with as many filters in front of your lens as you please, and as deep as you wish.

Alas, we don't live in a perfect world: Sometimes the light isn't 1000 footcandles, sometimes the Canon G9's 1/1.7" 12 megapixel CCD has noise... And sometimes the light is coming through a neutral density + high-pass (blue) filter... The water. See figure 1.1a here; or the visible range slice of the absorption coefficient chart near the bottom of this document.

Yes, using a complementary low-pass or band-pass filter will yield a flatter transfer function when the transfer functions are convolved... With a lot of attenuation that has to be compensated for.

It's that attenuation -- And the compensation on a real world thumbnail-sized CCD - I am (trying to) address here: It's not only the light loss an additional (red) filter adds into the path, but also the light loss (the "neutral density" component) that has to be accounted for... And compensated for.

In the case of a Very Quiet CCD, such as that found on the Canon 5D, then the trade-off of using a filter when shooting RAW can be helpful, as image degredation from increasing the ISO is not that great.

On the other hand, with a CCD having very tiny photosites, fewer photons are "snagged" due to the smaller cross-sectional area, hence one must reduce any losses in the light path.

It is for this reason that, given the camera's 12 million pixel 1/1.7" CCD .AND. the ability to directly record the CCD data in raw form .AND. the low light conditions underwater, that I recommend omitting the optical filter, and instead apply it in software in Photoshop.

Take away ANY of these three factors -- Small CCD, RAW file capability, low light -- and then an optical color correction filter becomes a luxury, even a necessity.

My camera came to the Maldives but my flashguns and other kit went to Thailand. No filter used. No manual white balance. Shot on a Fuji S2 Pro in RAW. Reproduced this month in Diver Mag. to illiustrate CS3 RAW converter and no work done in Photoshop save to convert to a file for repro.