Weird color question for 1DMKIIN in burst mode

I've been noticing this for a while, but I haven't gotten too scientific in trying to find the answer, however it's beginning to bother me.


My environment:

1DMKIIn, Canon 85mm 1.8 lens, Adobe CS2, ACR3.3, indoor basketball.

My Issue:

I'll take some burst shots with the camera set on manual (it's a gym and the lighting is pretty consistent). One shot will have a different color cast than the next (which is only about 1/8 of a second later, so you can imagine not much in the environment has changed in that time).

I'll attach two images below, they show that one has considerably more warmth in the wood floor than the other, and one appears to have a pinker tint to the walls.

These images were shot in manual mode (85mm, F2.0, 1/500sec, ISO1000) raw, with the exact same processing parameters in ACR3.3 (color temp 4000, tint -2, all other calibration exactly the same and not 'auto' ) then converted to JPEG using the same exact parameters (but the difference is apparent in ACR, so it's not an issue with the conversion, which appears to be true to the ACR data).

So I would expect the two images would look almost the same from a color/exposure perspective. What am I missing?

Byte size: ZERO | Content warning: NOT AN IMAGE

Byte size: ZERO | Content warning: NOT AN IMAGE

Have you accidently set auto bracketing ?

N

My guess is that the fluorescent lighting in the gym is messing around with your colour tone. That light will change its colour 120 times a second (at 60Hz in the US), the warmer tone being whilst the voltage drop.

I'd be inclind to believe that if your camera was set to a fixed white balance as opposed to AWB.. then this would go away,. but logically if shooting RAW and processing out put with the same WB settings,. then the cameras WB setting would not matter unless ACR is importing something from the AWB camera setting and biasing ithe input based on that?

Do you get autobracketing in manual mode?

On My MkII (not N) you do.

I used it here.... https://photography-on-the.net/forum/showthre​ad.php?t=131941 even though they were 30 second exposures.

N

Thought about that, but the exposure information is identical (from the exif) so I don't think that is it.

That's an interesting concept. I could swear I've seen this in outside shots as well, but I need to actually do a test now. The gym has these big round lights, I would not have thought them florescent, but my knowledge of lighting fixtures is nil.

I'm going to go shoot some shots outside while there is more light and see what happens.

Thanks for the info,

Bob

I'm also betting on fluorescents or other gas-discharge type lamps.

I think I tend to agree with the lighting theory. I just went out back and shot several bursts in sunlight, none of which exhibited this behavior.

Too bad; sure makes the workflow a lot messier. I take tons of photos at these games and put them on on a web page for the parents. Not that the color needs to be perfect, but it looks pretty bad when sequential shots change so much... Oh well.

Thanks Adas and Jon for the enlightenment!

Bob

The problem is most likely the light source as several folks have mentioned in the thread already. With a shutter speed of 1/500 second, the lighting has been caught in various conditions.

A lot of lighting types, especially flourescent, change the intensity and color of the light as the power line voltage swings from zero volts to a positive peak value, back through zero volts, to a negative peak value, and back to zero volts. This full cycle happens once each 1/60th second in the USA (as well as some other countries) and once each 1/50th second in a lot of other places. The only real solution for photographers is to use a shutter speed that will grab either a full cycle of the lighting changes or a half-cycle of the lighting changes. In other words, 1/60th second or 1/120th second would be the shutter speed choices in the US. It's a bummer when this sort of lighting is used in sports facilities, because the tendency is to attempt to stop action with a short shutter speed. That just makes the lighting more of a problem than it is even with the correct shutter speeds for the lighting.

Another problem with a number of lighting types is the wierd color spectrum of the light in the first place. Many light types do not have a broad color spectrum and that distorts the color seen by the camera. Our brains often correct colors that we see, but the camera can't really do that.

One solution would be a high-power flash system that would over-ride the gym lights and provide a consistent color for the camera.

I have had the same thing happen while shooting night football and basketball. You see it mainly in burst mode because you see the difference from one shot to the next. Someone had a good test that showed the results but I can not find it now. He took some pictures in burst mode and you saw the WB change from shot to shot. I compared my football pictures and it was the same thing. Maybe someone can find the test.

Had the same thing happen when I shot a hocky game. I had my camera on Manual mode but I got some big color shifts in my bursts. Ice would shift from white to Little purple.

I don't know about the lighting at football games in US (don't know what you guys call football either ) , but here in Europe the lighting has evolved radically in the last few years, there are long rows of lights all around, so there's no shaddows anymore to the players, just like a nice cloudy afternoon. I'm talking about the big mathes here. The matches where all the big guys with the black or white cannons are there. They surely wouldn't tolerate such an issue. I'm sure the US lighting is twice as bright and neutral than our( as a comparison the european "GPS" is only in the project stage). You only have to go to the right matches though.

If he had been using a custom white balance setting, would there still have been differences in the images? I'm curious if the differences are a product of auto white balance changes within the camera or the actual lighting differences being captured in action.

WT