Story   Photographer   Editor   Student/Intern   Assistant   Job/Item

SportsShooter.com: The Online Resource for Sports Photography

Contents:
 Front Page
 Member Index
 Latest Headlines
 Special Features
 'Fun Pix'
 Message Board
 Polls
 Educate Yourself
 Equipment Profiles
 Bookshelf
 my.SportsShooter
 Classified Ads
 Workshop
 Sponsors
 Special Offers
 Our Store
Contests:
 Monthly Clip Contest
 Student Contest
 Annual Contest
 Rules/Info
Newsletter:
 Current Issue
 Back Issues
 Subscribe
Members:
 Members Area
 "The Guide"
 Join
About Us:
 About SportsShooter
 Contact Us
 Terms & Conditions


Sign in:
Members log in here with your user name and password to access the your admin page and other special features.

Name:



Password:







||
SportsShooter.com: Member Message Board

Stadium Light Color Shifts: The Low-Down
Guy Rhodes, Photographer
East Chicago | IN | USA | Posted: 4:31 AM on 06.24.06
->> I took some time tonight to experiment and illustrate why stadium lighting fixtures (mercury vapor, fluorescent, etc.) often yield unexpected and hard to control shifts from frame to frame. See my hidden gallery here: http://www.sportsshooter.com/guyrhodes/wbtests/

A large copy of the test comparison image is at:
http://www.guyrhodes.com/photo/wb_experiment_large.jpg

Pull that image open in another window if you'd like, then read my results:

In the following comparison between white balance methods, I shot the back of white sync-slate with color chips on the sticks under a typical fluorescent light fixture. I chose to use 1/500th of a second shutter speed, the ideal "bare minimum" that I'd use for fast action sports that might be shot under discharge lighting (football, basketball, etc.).

I shot a burst of six images at eight frames per second using a Canon 1D MarkII.

As you can see in rows one and two, the color is all over the place throughout the burst with both fluorscent (row 1) and tungsten (row 2) white balances.

As someone suggested in a separate thread, I slowed down to 1/30th of a second (slower than the 60Hz cycling of the light), and performed a custom white balance off the white portion of the sync slate. Running the shutter back up to 1/500th and shooting another burst (row 3) revealed the same color shifting anomaly, because again, we're still locked into one white balance (albeit a custom one) as the color of the lights goes up and down, up and down. Custom white balancing will NOT solve color shifting under discharge lights at high shutter speeds, even if you custom white balance at a speed slower than the cycle, as illustrated here.

I had to eat my own words when I tried auto white balance (row 4) as my next test. Colors still shifted through the burst, despite my thoughts that the camera would analyze each frame as it was shot.

The only way I was able to get consistent color in a burst under the fluorescent light (row 5) was to slow down the shutter speed to equal that of the cycling, which in the United States, would mean setting the camera to 1/60th or slower. This would be fine for shooting features in an office, but would hardly be adequate for available light action sports.

This experiment serves as further fuel for my loathing of discharge lighting fixtures, and the reason why I'll drag strobes to every indoor (and sometimes outdoor) sport I shoot where I know discharge light fixtures will be present.

As a last point of reference for those still hazy on 60Hz color cycling, legally download the music video by Nas featuring Puff Daddy called "Hate Me Now." The beginning of the video features several explosions on a street captured at night with a high speed film camera. You can clearly see the sodium and fluorescent lights in the background pulsing, revealed only by the extremely slow motion film.
 This post is:  Informative (39) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Pablo Galvez, Photographer
Calgary | AB | CANADA | Posted: 1:10 AM on 06.25.06
->> Guy,

This is a great explination of how difficult it is to white-balance flickering lights. Your sample photos really illustrate the point well. Kudos!

-Pablo
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Justin Kase Conder, Photographer
Fresno | CA | USA | Posted: 1:51 AM on 06.25.06
->> Rhodes you freakin rock, very through and relatively easy to grasp.

Thanks for taking the time to share.

kase
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Robert Meyer, Student/Intern, Assistant
Lincoln | NE | United States | Posted: 2:23 AM on 06.25.06
->> Guy, this is why I love you. You always do cool stuff when you're bored.
 This post is:  Informative (0) | Funny (2) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Darren Whitley, Photographer
Maryville | MO | USA | Posted: 2:37 AM on 06.25.06
->> Well done. I appreciate the technique. I was aware of the issue, but just live with it. The images showing the pulsation are especially great information.
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Chuck Steenburgh, Photographer
Lexington | VA | USA | Posted: 6:21 AM on 06.25.06
->> Did a couple of quick Google searches, and unfortunately could not find any specs for the color temperature 'range' of any of these types of fixtures. I'm thinking it might be possible to shoot in RAW and have a range of color temperature values to apply to individual frames; though it would only narrow slightly the process of trial and error, I imagine.

Thanks for the info, Guy!
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Rick Burnham, Photographer
Enfield | CT | USA | Posted: 8:20 AM on 06.25.06
->> Guy

I second Justin and Robert and I don't even know you! I shoot at Rentschler Field in East Hartford CT (UConn Football) and at night games the white balance is all over the place. In fact I think it was Andy Mead who originally asked some questions about it after shooting soccer there.

Thanks again for the info.

Rick
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Stewart Cox, Student/Intern, Photographer
Asheboro | NC | US | Posted: 11:14 AM on 06.25.06
->> One of the features of flourscent light sources is that they generate a series of wavelengths that are often concentrated into narrow bands. As a consequence those sources do not produce the continuous spectrum of illumination that is a characteristic of incadescent sources.

Maybe I did learn something in Materials and Processes.....

-Stewart
 This post is:  Informative (0) | Funny (0) | Huh? (1) | Off Topic (0) | Inappropriate (0) |   Definitions

Seh Suan Ngoh, Photographer
Singapore | SG | Singapore | Posted: 12:00 PM on 06.25.06
->> Hmm.... how about this - that all lights, so long they're powered by single-phase power will have that problem? I'm not too sure about what triple-phase power really means, but I do believe it might have less power dips at the "lows" of the sine wave.

But hey Guy, great work! Thanks for sharing!
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Guy Rhodes, Photographer
East Chicago | IN | USA | Posted: 3:50 PM on 06.25.06
->> Seh Suan,

I was going to post a follow up about single phase vs three phase power and how it might affect this all.

In the US, most larger buildings are powered by three phase alternating current. That is, the "hot" electrical service coming in has three wires (wires which subsequently go to each transformer and sub-panel), each wire carrying 110 volts, and each 120 degrees out of phase with the others.

Most big stadium lights (and gym ceilings) are not wired to the same phase of power, but rather, groups of lights are on different phases. The next time you're at an event with a big tower / ceiling full of ugly mercury lights, ramp your shutter up to 1/5000 and shoot a burst of the lights and you'll notice from frame to frame, some lights will be brighter than the others at different times.

That is because in a three phase system, one phase is peaking at a different time than the other two. This works to our advantage in dealing with the white balance problem because, if the lights are focused properly on a field (or large in number in a gym), you'll have at least a third of the total lights at their peak output in the cycle at any given time, bettering your chances of getting correct color on your subject.

What Stewart is referring to is also correct, fluorescent and mercury sources do not have the entire spectrum present. There are big pieces missing that show up as black bands when viewed throuh a slit / diffraction grating. Sodium is the worst when viewed this way, with just a tiny line of orange and purple making it out of the light source.

This is what "CRI", or color rendering index refers to. The highest CRI, or ideal specturm is 100 - that is - every color of the rainbow is present in the light source. Typical fluorscent lights have a CRI of around 60. And we wonder why people's skin looks awful under them.

CRI info:

http://www.schorsch.com/kbase/glossary/cri.html
 This post is:  Informative (9) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Alan Stewart, Photographer
Corydon | IN | USA | Posted: 1:42 AM on 06.26.06
->> Guy, awesome post!!! It's a darn shame I can only click "informative" once...this is one that should be required reading!

I have a question that I'm not really sure how to word so it can be understood. Will shooting at 1/60 ALWAYS get the entire pulse, or could someone catch a "down" cycle and continue to hit the "down" with every frame just by dumb luck?

In your hidden gallery, did you do multiple tests of the 1/60 theory, with all of the images coming out exactly the same?
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Guy Rhodes, Photographer
East Chicago | IN | USA | Posted: 11:55 PM on 06.26.06
->> Alan,

I didn't shoot multiple trials with my 1/60th shutter speed (the last row in my burst tests). I was thinking about your question and I think you'd always end up with the same brightness at 1/60th no matter what part of the wave you catch. If you catch a bright "high", you're also catching two halves of a "down" on either side. Conversely, if you happen to shoot a "down", you've still got two halves of bright "highs" on either side. Make sense?
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Jay Sinclair, Student/Intern, Photographer
Vancouver | BC | Canada | Posted: 4:55 PM on 12.19.06
->> So my question, is the best policy to shoot AWB and RAW and fix the images in Post later?
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Guy Rhodes, Photographer
East Chicago | IN | USA | Posted: 9:31 AM on 01.05.07
->> Jay,

Posted this on another thread, but here's my verdict on shooting raw:

Shooting raw only helps with the color problem, as the output of the light fixture is also rising and falling along with the color shifting. Different circuits of light above different areas of the court might be on different phases of power, peaking and falling at different times, meaning you might have a group of players in one area of the court brighter / darker than another area at any given time.

See David Harpe's suggestions here:

http://www.sportsshooter.com/message_display.html?tid=23216

He outlines your only two viable options other than using strobes.
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Guy Rhodes, Photographer
East Chicago | IN | USA | Posted: 7:49 AM on 01.07.07
->> To further illustrate this whole flickering arena lighting phenomenon, I created an animated .gif from a motor drive burst I shot in a gym lit with mercury vapor lamps earlier tonight. You can see the image here:

http://www.guyrhodes.com/photo/flicker_lapse.gif

The burst was shot @ 8fps, ISO 1600, 1/2000th @ 2.8. I had the camera in auto white balance mode. Lining these frames up next to each other (twelve total in the animation) allows us to see the problematic pulsing these lights produce.

I'll remind everyone that this flickering, including in the gym where this was shot, is TOTALLY invisible to the naked eye.

Notice the exposure and color variations across every surface of the gym as lights flicker at different times (wired to different phases of the building's power).

This is why shooting under these lights at fast shutter speeds is hit-or-miss. Sometimes you'll catch the lights as they peak, other times, you'll catch the bottom of the wave as the brightness / color are at their worst.

If you're lucky, the electricians have spread the different phases of power across the entire group of lights, rather than wiring large neighboring areas of lights on the same phase. At E.C. Central's gym (where the time lapse was shot), we've lucked out. You'll notice lights seem to be wired to different phases every third fixture. This means that by the time the light reaches floor level, the pulsing becomes very manageable, and the light stays somewhat even.

I've been to football stadiums, however, where entire poles of lights are on the same phase (meaning they all flicker at the same time), leaving entire percentages of the field in near darkness at the "bottom" of the flicker when shot in a burst, depending on how the lights are focused.

Hopefully this is making sense to everyone, especially with the animation to bring things to light - pun intended.
 This post is:  Informative (5) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Alan Stewart, Photographer
Corydon | IN | USA | Posted: 10:35 AM on 01.08.07
->> The last image on my member page shows some really ugly color shift. The first time I shot digital at this place a few years ago, I thought something was wrong with my camera, because the four schools in my county didn't have this problem.

Not much you can really do about it other than just hope you catch the lights just right ...
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Israel Shirk, Photographer
McCall | ID | US | Posted: 6:58 PM on 01.08.07
->> I'm almost through a degree in physics, so... I'm a bit of a nerd. Here's the explanation:

The cycling of AC power through a filament will cause it to release photons of all sorts of energies, most invisible (meaning that they're usually better space heaters than light bulbs, especially with tungsten)...

Molecules of a certain material will tend to release energy when they're excited to certain energy levels; between these energy levels, or voltages, they will still shoot off photons periodically, but it's *much* less common (to relase a photon would send them into what's called a 'prohibited state')... The voltages at which they will rarely emit photons are generally referred to as band gaps. If you were to try using a DC circuit and varying the voltage, you would find that what happens is that you'll still get a little variation in the colors emitted as the material isn't ever going to be uniformly excited; however, at certain voltages you will have peaks for certain colors.

What causes the variation is that as the current goes through the AC cycle, it will go through these different colors repeatedly (as mentioned above using high-speed cameras) as the voltage rises and falls - passing through different color ranges (and through band gaps) as it goes through the cycle.
 This post is:  Informative (1) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Jefferey TenHave, Photographer
South Gillies | On | Canada | Posted: 8:00 PM on 01.08.07
->> Thanks a bunch Guy... Makes great sense, but who would of thought.....
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Steven Squires, Student/Intern
Pullman | WA | USA | Posted: 6:22 PM on 02.09.10
->> Guy,

Thanks for justifying my rage!

Steven Squires
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Mark Peters, Photographer
Highland | IL | USA | Posted: 8:37 PM on 02.09.10
->> This year's Lazarus Award goes to.......
 This post is:  Informative (0) | Funny (5) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Bryan Hulse, Photographer
Nashville | Tn | USA | Posted: 10:58 PM on 02.09.10
->> It is ALIVE!
 This post is:  Informative (0) | Funny (1) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Matthew Bush, Photographer
Hattiesburg | MS | USA | Posted: 2:10 AM on 02.10.10
->> Zombie Thread..... !!!!
 This post is:  Informative (0) | Funny (2) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Dennis Wierzbicki, Photographer
Plainfield | IL | USA | Posted: 9:33 PM on 02.10.10
->> Yeah, but this is one Zombie that deserves to live on. I had this thread bookmarked, as this topic comes up all the time and I have yet to see anything better on the subject.
 This post is:  Informative (1) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

David Chandler, Photographer
East Grand Rapids | MI | USA | Posted: 5:05 PM on 12.11.10
->> Love the term Zombie thread for this one! I'm almost strictly an available light shooter in gymnasiums. I have been wondering the reason for drastic changes in sequence images when everything is set in camera. NOW I KNOW!
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Jeroen de Jong, Photographer
Waalwijk | _ | Netherlands | Posted: 8:16 AM on 12.12.10
->> I didn't discover this thread before and now I know what the shifting in my frames is. I'm also someone who only shoots with available light (strobing is not allowed during prof-sportgames overhere)
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Patrick Murphy-Racey, Photographer
Powell | TN | USA | Posted: 9:41 AM on 12.12.10
->> The Liberty Bowl in Memphis and Vanderbilt's football stadium (both state of the art in 1959) seem to have an above average amount of variance for that nasty pulsing. I see much less of it in more modern lighting setups at more modern stadiums like UT, GA, LSU, FL, etc... I seem to find more of this bad effect on the darker/older places I shoot in the SEC. For some reason the brighter ones seem to have less of that pulsation and wild exposure & kelvin shift.

Shooting at area high schools, I see this constantly as well. I wonder if adding tons and tons of heads high above the pitch creates a higher than normal "mean" of this phenomena so that we see it less in the larger venues???? What say you all to this idea?

Last, one of the reasons I've always loved the Elinchrom and Dynalite Arena kit stuff is the 2ndary capicitor that fires an insitant after the flash which cuts the tail off the main flash-tube dump. This, combined with their higher than normal internal voltage is partly how they are able to get such short, & consistent kelvin flash durations. Tube life suffers but the photos always have a perfect, clean white light with great contrast and quality. orry to be a ramblin' guy...
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Guy Rhodes, Photographer
East Chicago | IN | USA | Posted: 5:03 AM on 12.13.10
->> Patrick,

You're correct, the more heads lighting the court / field, the less noticeable this effect is down on the playing surface. The way in which groups of fixtures are wired to each of the three electrical phases will also effect how noticeable the pulsing is. I touch on this up in my last post here from 2007 (yikes, time flies).
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

David Stembridge, Photographer
Waynesboro | Ga | USA | Posted: 6:57 PM on 11.12.12
->> Great thread Guy,

I just experienced this phenomenon with my newly acquired D1 Mk3; and coming from a T2i; it has been pretty frustrating; but I now know I'm not alone.

I do wish there were consistent standards as you've "hoped" for above!
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Darin Sicurello, Photographer, Assistant
Gilbert | AZ | USA | Posted: 7:06 PM on 11.12.12
->> I assume that in the coming years, They will be replaced by LED lights.
 This post is:  Informative (0) | Funny (0) | Huh? (0) | Off Topic (0) | Inappropriate (0) |   Definitions

Add your comments...
If you'd like to add your comments to this thread, use this form. You need to be an active (paying) member of SportsShooter.com in order to post messages to the system.

NOTE: If you would like to report a problem you've found within the SportsShooter.com website, please let us know via the 'Contact Us' form, which alerts us immediately. It is not guaranteed that a member of the staff will see your message board post.
Thread Title: Stadium Light Color Shifts: The Low-Down
Thread Started By: Guy Rhodes
Message:
Member Login:
Password:


|| Sponsor Special Deals

Redrock Mircro DSLR Cinema Bundle
Available from: Samy's Camera | Price: $2443.95
Notes: SAMY'S CARRIES THE FULL LINE OF REDROCK MICRO PRODUCTS!! The Redrock DSLR Cinema Bundle is a comprehensive package for converting video DSLRs to useful cinema production tools. This bundle is compatible with most video DSLRs including the Nikon D90 and Canon 5D Mark II.
-- More Info --



Return to -->
Message Board Main Index
Red pill or blue pill ... The Matrix of photography New post on Photosique ::..