Nice Tip on WB for Digitals (1 Viewer)

[ManW_TheUncool]

Since John brought up the issue of people needing to be better educated in the digital age, I just found this nice tip for getting the most IQ out of our digital cameras over on dpreview.com. It's something generally overlooked by even many(?) pros who have made the switch to digital -- and is an instance of one good example that was on my mind in that other thread.

http://forums.dpreview.com/forums/re...ssage=17142860

If you want the best IQ out of your camera, then it's something you should consider doing, instead of always resorting to digital WB. Of course, you'd want a different filter for different lighting whenever feasible. Here's a suggestion for incandescent light in that same thread:

http://forums.dpreview.com/forums/re...ssage=17152013

Happy shooting...

_Man_

 

[JohnRice]

I have wondered if there is a down side to shooting under incandescents with corrected white balance done in camera instead of using filters. I know when I do it in camera I am always disappointed with the results. Maybe this is why. Could also be another reason why digital doesn't quite measure up to the hype if the white balance actually degrades the image.. I mean, on film I can shoot tungsten film with little or no filtration. On digital, it seems I may need to use heavier filtration, and thus have filter factor hassles, to get optimal results.

BTW, I don't know where she gets the CC40B and CC60B to correct for incandescents. I don't know why it wouldn't be an 80B to get you close, but since incandescents vary in temp by quite a bit, adding an 82B often gets you closer. Of course, that is with daylight film. Digital has wildcards which may make that less than ideal. Anyway, that is a filter factor of about 0.8, which is a pain in the ass when you are dealing wiith low light levels to begin with.

Guess I need to give it a try, but using gel filters on the digital is going to be a pain in the ass.

 

[JohnRice]

Hell yeah, it's overlooked. Because everything about digital claims you no longer have to filter that way. Glad to know since, like I said, I have been wondering about just that regarding why my incandescent shots don't measure up. Just understand, a significant percentage of the higher end stuff I shoot is on tungsten film, and if what that thread says is true, it eliminates a major benefit of digital.

Don't get me wrong. I'm not exactly upset, I'm just stunned that such a significant bit of information has been so subverted.

 

[ManW_TheUncool]

Maybe someday, if we ever get to the point of interchangeable digital backs for DSLRs, we will see backs designed for incandescent/tungsten light just like tungsten film was used since it's such a common situation. As it is now, it seems that most cameras are designed w/ sensors that are at their worst w/ tungsten light (unless of course you want the very warm color cast in your images).

I suppose if it ever happens, it'd happen first w/ medium format digital backs.

_Man_

 

[JohnRice]

BTW Man, thanks for posting that link. I think that issue is like digital's dirty little secret. I doubt many people know that the white balance is done that way. I certainly have never heard anything about it. Of course, now that I've seen it, I feel kind of foolish. After thinking about it for a minute, I just thought "of course that's how it does it." I can only shake my head. In the end, the whole white balance issue is something of a myth with digital, at least right now. It amazes me that what really happens is that you are always shooting daylight (or whatever the sensor is actually calibrated for) and just color correcting after the fact. What a bit of marketing BS that is. Digital just became a little less convenient, but the results will at least imporove.

 

[ManW_TheUncool]

John,

Unfortunately, there seems to be more hype along similar lines, particularly from Canon (and maybe Nikon's starting to follow along as well w/ the release of the D200), in trying to meet demand for higher dynamic range (DR) thus far. And much like the digital WB issue, most people just buy into the claims of higher DR w/out really understanding what the compromises will be.

It seems that these makers, particularly Canon, has been pushing the limits of their (12-bit/channel) RAW formats to try to produce the amount of DR that nearly all prosumers and pros have been demanding. Iliah Borg, the father of the original poster (Julia) of that thread, had posted elsewhere (in nikoncafe.com partly because he was banned from dpreview) a long while back showing some interesting differences in how Canon and Nikon differed in the way they chose to distribute the DR of their respective flagship cameras (ie. the 1DsMk2 and D2x). While it's nice to think that one can get a good deal of DR w/ the 12-bit RAW files from Canon DSLRs, the fact remains that the extra DR (just like the lattitude offered by digital WB) doesn't come from nowhere -- it costs something (ie. fine gradations in tonal steps) to increase DR w/out also increasing bit depth in the RAW files. I believe that's part of the reason why some folks have started to see how the D2x yields better/smoother/finer images when the higher DR is not needed -- afterall, it'd be wasting tonal steps to capture essentially nothing w/ part of the DR spectrum when the extra DR is not needed.

The other thing that Nikon seems to be attempting to do in this area is to try to improve the usage of their smaller DR (and also reduce the need for digital WB to some extent) by developing something they call WB pre-conditioning into the analog stage of their technology. The idea is sorta like applying an inverse filter by pre-amping the weaker color channels as necessary instead of subtracting from the stronger color channels in order to yield more efficient use of the camera's DR. So far, it doesn't seem clear how much of this idea they actually implemented into the D2x and D200, but it's certainly something they've made mention of in their marketing materials -- and also, there seems to be provisions for the necessary changes to how WB would be described in their RAW format (and one also wonders if this is not part of the reason why they chose to obfuscate the
WB info when the D2x came out).

Anyway, aside from making the cameras more efficient/smarter at making the most of the available DR, there really would need to be increase in bit depth for real gains in DR to happen (w/out sacrificing fine tonal steps that is). Fuji has essentially tried that w/ their SuperCCD w/ the duo photosites per pixel, and the diff has been noticeable.

_Man_

 

[JohnRice]

Well, Man, this white balance issue, actually the entire issue of in-camera color correction, is not a simple matter of fudging numbers to make for impressive stats. It is a complete, bald faced lie. It is no different than saying you can shoot daylight film all the time and just have it corrected in the printing or on the computer afterward. The cold, hard fact is, it doesn't work. So, this is one of the major "benefits" of digital photography that is, to put it bluntly, a load of crap. Now, under a lot of circumstances, it yields acceptable results, but if you are shooting under incandescents, you end up with serious problems. There is not only the issue of severe noise, but almost complete loss of any colors in the blue area of thhe spectrum, crossed curves (which is probably more severe than with film due to the low latitude) and who knows what else.

Anyway, tonight was my first occasion to shoot some products under incandescents (I have some situations where that just works a lot easier, I won't go into it) so this time I just shot at daylight with an 80A and the results are infinitely better. No surprise. Don't get me wrong, digital has some very nice benefits, but instances of blantant BS marketing like this really burns me.