DV Info Net

DV Info Net (https://www.dvinfo.net/forum/)
-   Photon Management (https://www.dvinfo.net/forum/photon-management/)
-   -   making tungsten look like fluorescent, not fluorescent look like tungsten (https://www.dvinfo.net/forum/photon-management/79960-making-tungsten-look-like-fluorescent-not-fluorescent-look-like-tungsten.html)

Fischer Spooner November 18th, 2006 09:12 PM

making tungsten look like fluorescent, not fluorescent look like tungsten
 
I've been having great luck shooting stuff using available, fluorescent light. See link: http://www.youtube.com/watch?v=e0GltE0_XR4

However some of the shots have underexposed faces (the shots of the male in the clip are an example of this). The usual practice is to gel the fluorescents. Since the auto color-correction in my hc1 is doing such a good job with the fluorescent lights (the CU of the female in the clip is an example of this), I would like to do the opposite: I would like to gel my softbox so that it has the same spectral quality as an overhead, un-corrected fluorescent tube.

(That way I only have to use the softbox when the fluorescents aren't good enough on their own, which equals less work for me! yay.)

I am having a hard time inferring how plus-green and minus-green gels work, based on the explanations I have read and heard. Usually people are putting the gels on the overhead tubes, and that is what the explanations are geared towards. I am wondering about the practice as opposed to theory: it seems that a plus-green will do it, but I am not sure.

Will plus-green gels do it? If I put a plus-green on my softbox, will it more or less match the color temp of the overhead fluorescents? I am interested in an approximate match, not an exact match. I am looking for "good enough". I am just trying to get acceptable exposure on shots where the available light isn't alread providing it. I am trying to avoid messing around with lights a lot, so please don't be mean and tell me how I have to spend a ton of time, money and suffer. I know how you posters like to put people down. Thanks!

PS, I am shooting on location Monday, and am short money and time. I have to walk in with a solution, and if I don't have it, I'll just shoot the scene with more underexposed faces. No lectures please! I need help. I know that a ninja is always prepared.

Emre Safak November 18th, 2006 09:41 PM

Fluorescents with a low CRI typically have a green spike. Anything that is remotely green will be accentuated. Minus-green gels try to compensate for this. Plus-green gels try to emphasize it. I would prefer to get rid of an unnaturally pronounced green response by using a minus-green rather than the other way around.

Seth Bloombaum November 19th, 2006 12:52 AM

There are many types of fluorescents, with varying color temps, even before you get to issues of a green spike.

One thing that I've done that's frequently worked well is to gel an incandescent fixture with a full CTB, sometimes 3/4 or 1/2.

Almost all the very modern fluorescents used in new commercial building construction are pretty close to daylight, and quite a few older tubes are cool-white, somewhat close to daylight. The I'd look at the daylight white balance preset, compare it to manual balance to your gelled lights.

Another trick in dealing with almost-daylight fluorescents is to manual white balance to warmcards http://www.bhphotovideo.com/bnh/cont...ughType=search

Charles Papert November 19th, 2006 12:04 PM

Quote:

Originally Posted by Fischer Spooner
I've been having great luck shooting stuff using available, fluorescent light. See link: http://www.youtube.com/watch?v=e0GltE0_XR4

I am trying to avoid messing around with lights a lot, so please don't be mean and tell me how I have to spend a ton of time, money and suffer. I know how you posters like to put people down. Thanks!
... No lectures please! I need help. I know that a ninja is always prepared.

Fischer:

I believe that the majority of people on this board are here to help, not to put people down. That's the reason I post here and not on other DV boards. I'm sorry if you have experienced otherwise. My suggestion would be to ignore the arrogant responses and focus on the helpful ones.

In my experience (and confusingly, this is a bit contrary to Seth and Emre's) most existing fluorescents in use today are closer to tungsten than daylight, and the green spike is usually not an issue with video. Generally a 1/4 CTB on your softbox should go a long way to bringing the color temperature closer together. Then white balance to the predominant source (presumably your softbox).

How to determine this? If you have time, set up a couple of sheets of white paper (or gray cards if you have them) so that one is full lit by the overheads and the other exclusively lit by the softbox. You'll need to flag them to ensure that the light from one source is not contiminating the other, and make sure that the light levels on the cards are as close as possible. Watch your monitor, and hold up different levels of CTB in front of your softbox until the color on both cards matches. This will take a bit of time and effort. The human eye can distinguish color temperature fairly easily, especially with practice, so a much easier way to do it is to hold your hand up with one side facing the softbox and the other lit by the existing flo's in the ceiling and simply judge which is warmer/cooler, then add gels as above until both sides look the same.

The magenta/green issue is a bit different (called the "mired shift"). This is harder to judge by eye. In fact, film sees the mired shift very differently than the eye and a color temperature meter is required to determine how much plus or minus green/magenta must be used. However, video is much less sensitive to mired shift and "sees" it similarly to the eye. You can use the hand test as described above to determine if you need any correction. I have shot under many types of ceiling flos without having to resort to plus/minus green correction on my lights; once again, white balance to your softbox and what little green there may be will generally not be a problem. Older facilities may have tubes that demonstrate more green, but these are rare these days.

I just shot for a few days at a Verizon store; although the ceiling flos were nominally "cool" lights by designation, we used tungsten and tungsten-balanced Kino-Flo's with no correction, white balanced to these and had no problems.

If you walk into this job with 1/4, 1/2 and full CTB for your softbox, I bet you will be covered. Having 1/8 and 1/4 plusgreen will be an added bonus.

Emre Safak November 19th, 2006 12:42 PM

Quote:

Originally Posted by Charles Papert
In my experience (and confusingly, this is a bit contrary to Seth and Emre's) most existing fluorescents in use today are closer to tungsten than daylight, and the green spike is usually not an issue with video. Generally a 1/4 CTB on your softbox should go a long way to bringing the color temperature closer together. Then white balance to the predominant source (presumably your softbox).

This is not contrary to my experience at all. In fact, I did a shoot just yesterday which looked just fine where I mixed tungstens with 1/4 CTB and office fluorescents (i.e., not Kinos or anything special). However, if I were to go messing around with green gels, it would be to remove the green spike from the fluorescents rather than to add green to the tungstens.

Charles Papert November 19th, 2006 12:54 PM

Probably therein lies the different approach--to me it's much easier for a shoot this size to correct the handful of additional units rather than have to cut and place gel for each of the ceiling fixtures, since the result is the same photographically.

Just to avoid confusion since this was part of the original question, one would of course add minus-green (i.e. magenta gel) to the flo's to correct them, whereas plus-green (green gel) would be added to the units.

Off this topic onto a geographical one...Emre, I was just wistfully thinking about how I used to stroll down Harvard Street to get a bagel-and-schmear; good bagels are few and far between in LA, just like pizza (and for that, I miss Pino's in Cleveland Circle amongst many others...and what I wouldn't give for a bowl of soup from Pho Pasteur in Allston).

Emre Safak November 19th, 2006 01:23 PM

You like Pino's? I much prefer Presto's, just down the street. In fact, I just had lunch there.

Charles Papert November 19th, 2006 02:30 PM

Don't know it, must be after my time! Pino's is probably more of a nostalgic favorite from growing up, I guess. And seriously--nearly every corner pizzaria on the east coast beats out what we get out here.

Marcus Marchesseault November 19th, 2006 11:14 PM

My guess is that many lighting situations would be much easier if people would stand on a ladder and peak into the existing lighting at a location. Use this as a guideline:

Soft White = ~3000K down to about 2700K

Cool White = 4100K If it has some sort of economical or watt-saver sort of blurb, it may have a bit of extra green.

Natural Light or Daylight = ~5500K (from 5000K to 5600K) Some specifically advertise no extra green.

Daylight Plus = ~6500K and probably only useful as "moonlight" accent lighting or to use with pure blue gel as an effect light.

Acquarium/Grow light = 6500K up to about 10,000K which is probably useless in most situations.

From my experience, there is still a whole lot of 4100K "cool white" out there in the office world. If your softbox is not matching ~3000K (warm white), the 1/4 to 1/2 CTB is probably your first choice to bring the ~3000K up to 4100K. If you happen to know what the overheads are putting out, bring full CTB for the softbox if they are called Daylight or have some sort of print shop reference. Some printing/publishing places need daylight matched fixtures with high CRI so their colors will come out right.

I would like to reiterate what Charles said about the green spike. I have not generally been having a problem with fluorescent lights. I find that if they look good to the eye that they work on video. Using too much minusgreen (magenta) can make white people look too pink. I use the pink parts of my hand (folds/crevaces) to judge if there is too much green in a light source. If the pink diminishes significantly, some minusgreen/magenta may be needed for olive-skinned talent. Be nice to your pink-skinned talent and call their complexion "ruddy". Long-term sun exposure is probably the issue and nothing but makeup and lighting can correct the problem.

If yall mailed a decent bagel to me, it would probably arrive more flavorful than what I can find here.

I remember pizza...vaguely. The best pizza in Hawaii is DiGiorno's rising crust in the frozen food section. The Safeway branded frozen rising crust pizza is also decent. I don't bother paying $25 for a delivered cracker with cheesish white stuff and a flavorless smear of sauce.

Seth Bloombaum November 20th, 2006 12:06 AM

Quote:

Originally Posted by Charles Papert
...In my experience (and confusingly, this is a bit contrary to Seth and Emre's) most existing fluorescents in use today are closer to tungsten than daylight, and the green spike is usually not an issue with video. Generally a 1/4 CTB on your softbox should go a long way to bringing the color temperature closer together. Then white balance to the predominant source (presumably your softbox)...

As Emre said, not so contrary - there is a lot of variation. I guess I've been shooting in a lot of new construction, and the high-frequency daylight-balanced tubes are really not bad for general fill and background.

Seems like we are all agreeing on having a selection of CTB on hand...

Quote:

Originally Posted by Marcus Marchesseault
...Use this as a guideline:

Soft White = ~3000K down to about 2700K

Cool White = 4100K If it has some sort of economical or watt-saver sort of blurb, it may have a bit of extra green.

Natural Light or Daylight = ~5500K (from 5000K to 5600K) Some specifically advertise no extra green.

Daylight Plus = ~6500K and probably only useful as "moonlight" accent lighting or to use with pure blue gel as an effect light.

Acquarium/Grow light = 6500K up to about 10,000K which is probably useless in most situations.

From my experience, there is still a whole lot of 4100K "cool white" out there in the office world...

Marcus, thanks for the list of temps! One can also call ahead at many large businesses and get maintenance staff to read off what is on their spare tubes.

Charles Papert November 20th, 2006 08:26 AM

Gents, there's no stigma to having different viewpoints, approaches or findings..."contrary" is not a dirty word! Perhaps I should have used "different".

Actually, in the example of the Verizon job I mentioned, the first time I shot in one of these stores the gaffer was quite sure that the flo's would be closer to daylight whereas to my eye they looked much more like tungsten. After the job he conceded this one to me! But no harm done.

The problem with the nomenclature on the side of the tubes is that the color temps given can be well off from what is listed. The only real way to measure them is with a color temp meter, not a bad thing to have in your kit but of course a bit pricey, around $500. Formulations and model nomenclature changes so often with these tubes that it's hard to keep a list of what is what. Plus, in many environments you'll find a mixed bag of tubes as the maintainence crew grabs whichever they have more of in storage--not unusual to find a cool tube nestled into a fixture with a few warm ones or vice versa.

There are a myriad of ways to use ceiling flo's to ones advantage--judiciously unscrewing individual or banks that are directly above or near the talent may reduce unwanted sheens or eye shadows; hanging some diffusion from the fixture in a "bag" formation (allowing the diffusion to droop in the middle) will create a softer source; skirting the perimeter of the fixture with duvyetene or blackwrap will make a more defined and controllable source (helpful near walls for instance).

Fischer Spooner November 20th, 2006 03:14 PM

thanks
 
Thanks very much for the thorough discussion. Particularly helpful was this bottom line:

Quote:

Originally Posted by Charles Papert
If you walk into this job with 1/4, 1/2 and full CTB for your softbox, I bet you will be covered. Having 1/8 and 1/4 plusgreen will be an added bonus.

Here's a related question which has been bothering me for a long time, related to "green spikes" and various color temperature indexes. It seems inaccurate or *incomplete* to describe a particular light source with a single number (the color temperature), when in fact a light source seems to be better described as a *color temperature wave* with an X-Y axis. In other words, for example, a ceiling flo could have a "green spike". That means that a particular color temperature (the green part) is high in relation to other color temperatures.

It seems like the proper way to describe a light source would be to talk about which parts of a its spectrum was high, and which part was low. Sort of like EQ in sound: you can have the blue part boosted, the orange part boosted, the green part boosted, etc. Is that correct?

Emre Safak November 20th, 2006 04:07 PM

Quote:

Originally Posted by Fischer Spooner
It seems like the proper way to describe a light source would be to talk about which parts of a its spectrum was high, and which part was low. Sort of like EQ in sound: you can have the blue part boosted, the orange part boosted, the green part boosted, etc. Is that correct?

They list the spectral energy distribution curve too.

Charles Papert November 20th, 2006 04:39 PM

Mired shift is listed in CC's as in Color Compensating index, as a numerical value based on how many points of green or magenta are present.

Bob Grant November 21st, 2006 06:16 AM

Just a random thought, perhaps the best answer is to change the light source in the softbox. Fluro lights seem a little problematic is softboxes but HMI although expensive initially is an option.
Osram sell 150W HMI lamps and ballasts, anymore than 150W and you're into custom built ballasts and then the cost skyrockets. I'm in the process of bulding a 150W HMI china ball, can see no reason why the same lamp and ballast couldn't be used in a softbox.
Just one word of warning, do NOT look into a naked HMI lamp, ouch!
The readily available ones do have inbuilt UV filters but the light source is incredibly bright and concentrated.


All times are GMT -6. The time now is 08:44 AM.

DV Info Net -- Real Names, Real People, Real Info!
1998-2024 The Digital Video Information Network