XH A1 and automatic white balance going wonky

[John Whiteway]

Hello.

I've just looked at some footage we shot today and the white balance is jumping all over the place - from blue bias to yellow and back again. We've shot in the same place before using the automatic white balance and all has come out fine. I've just been analyzing the footage and think I may have found out why, but would like some confirmation. We've had the camera set at "TV 100" in the past. When I looked at the camera now, and then the data code on the footage, I see that the "shutter speed" had accidentally been changed to "TV 500". Correspondingly the aperture was down to f1.6. I can only assume that it's this that sent the "AWB" loopy. Hope I'm right, because if so, the problem is easily solved. Could someone give me their wisdom on the matter?

Thanks.
John

[Tripp Woelfel]

John... I can't speak to your situation from actual experience but I have noticed that changing the shutter speed can change how the color appear in my footage. I've noticed this recently on stuff I've shot using the VIVIDRGB setting which is really quite saturated so color shifts would be more noticeable.

From my own experience (not voluminous) and the opinions of others in the know I've come to regard automatic settings as bad things. This includes white balance. Being a control freak, I don't like leaving things to little electronic brains that think they know what's better for me. (Perhaps this comes from years of dealing with new Microsoft product "features".)

I'd suggest that you'd be better served by setting your white balance manually. Like manual focus and exposure settings, manual WB will let you better control what comes through the lens.

BTW: I'm curious about your choice of 1/100. Do you shoot a lot of action?

[Don Palomaki]

What type lighting?

Discharge lighting (e.g., conventional fluorescent) and high shutter speeds can result in slowly drifting white balance.

[Tom Roper]

I've had satisfactory results from the AWB especially in daylight but even indoors under fluorescent or tungsten lighting.

I think there is a setting for shockless white balance that should be engaged.

[Robert Wide]

Originally Posted by Tripp Woelfel

(Perhaps this comes from years of dealing with new Microsoft product "features".)

Maybe you should try a Mac! But I agree with you Tripp, I started out on the auto settings three months ago when I just received my A1 but already now I prefer the manual controls so much more.

Robert

[Peter Ralph]

it is a mystery why there is no WBlock like exposure lock and focus lock. By all means use the cameras processors to set up a shot - but once you have it - lock it down - exposure, focus and WB. You will regret it if you don't - guaranteed.

[Ryan Postel]

I agree with Don,

I've experienced the same thing filming in an office setting with florescents and a higher shutter speeds completely expose the inconsistent color cycle the lights produce.

Then again, I recently found out from Canon that my I have a flaw in my camera that renders my Gain useless... if only I can schedule ANOTHER two weeks to be without my camera while I send it in....

[John Whiteway]

Thanks. I think you folks may have explained my problem and given me the answer. The room we were shooting in was florescently lit. As a couple of you have said, combine that with a fast shutter speed (in our case 500th sec.) and you can get drifting. Thanks. As for the use of AWB, we've shot in this room about a dozen times. Initially we were setting the white balance manually. When we forgot one day we saw no difference so have left it set a AWB when shooting in that room again.

[Don Palomaki]

You will find the drift in WB is essentially determined by the difference between the power line frequency and the field rate. The power line averages precisely 60 Hz but at any given time may be a bit fast or a bit slow, while the camcorder is pretty much locked to 59.94 based on the internal clock crystal (for NTSC SD mode).

The color balance of the output of a discharge lamp will vary somewhat over the course of the power cycle (1/120 sec).

When the shutter speed is a small fraction of a half power cycle (1/120 sec) it sees only a fraction of the light output pulse of the lamp, and if that portion of that output sample varies over time (as it will if the field rate is not equal to the line frequency), the white balance will vary correspondingly. The work around is to use different lighting, or to use a shutter speed that sees the entire half power cycle or more.