When I shoot night scenes with a non-dark background either in the automatic mode or creative modes (Av, Tv), the foreground (often people) are often overexposed. I saw somewhere the TTL evaluation is only linked to the central focus point for some cameras. But, the info I read here and there is confusing. Can you help me on this? Thanks.

-- Dave Zhou (cosymoon@yahoo.com), May 17, 2000

Answers

The 380EX with the Rebel 2000 uses E-TTL metering. It will bias exposures to whatever AF point point is used. If you had an older camera (pre Rebel G, Elan II, 3, 1V) the 380EX would revert to plain TTL metering and that may force the bias to the central AF point. Can't remember for sure.

Without seeing your pictures I can only guess. If your active AF point is not on the foreground, they certainly will be overexposed. Or if your background is too far away to keep properly exposed without blowing out the foreground, this will also force an overexposure as the camera trys to lighten the distant background. But perhaps it is a printing error. Printing machines are dumb animals, and if they see a lot of dark space they will try to lighten the print even if your camera got it right. You could ask for reprints to bias the exposure to the foreground subjects.

-- Jim Strutz (jimstrutz@juno.com), May 19, 2000.

I don't believe that the Rebel links flash exposure to active focus points (normal/ambient exposure certainly isn't), nor that it will overexpose in a vain attempt to expose the background, what makes you believe this will happen (don't suppose it matters since the user said the background isn't dark)?

I can't comment on the exposures he's seeing without knowing more information (namely, see a scan). I would definantly have a real lab make a print and/or try a roll of slide film so you can see what exposure the camera is really using.

-- Steven Fisher (srf@srf.com), May 19, 2000.

Steven & I clearly disagree on the capabilities of the Rebel 2000's metering system, for both ambient and flash exposure. I'm convinced the Rebel 2000, like the Rebel G before it, uses E-TTL metering and will in fact bias exposures to the active focusing point. Anybody else want to try to settle this?

I suppose we could ask Canon's tech. people, but they don't always answer the questions that you ask.

-- Jim Strutz (jimstrutz@juno.com), May 19, 2000.

The Rebel G does not change ambient exposure based on focus point, as I've found through experiments (white paper on one focus point, black on the other, change focus points, no change in exposure). I haven't tried this with E-TTL, some time I'll shoot some slides & find out. I was just wondering if you had any basis for E-TTL changing exposure based on focus point.

-- Steven Fisher (srf@srf.com), May 19, 2000.

Well now, this is becoming interesting to me, although I'm not sure we're helping Dave.

I wish I still had a Rebel G to experiment with, but I've upgraded to an Elan IIe, which is *supposed* to have the same metering scheme in it, from everything I've read.

I just checked an Elan IIe, 10s and A2E, they all bias the exposure (ambient readings) to the active focusing point. Hard to believe the Rebel 2000 woud be different, but if so, that would be very interesting. I also checked with a flash meter and, sure enough, the camera biases the flash exposure to the active focusing point too. Of course this is with the Elan IIe not the Rebel 2000, but I doubt the metering scheme would be different with it.

Steve, are you sure you are not checking the exposure using either maual mode or pressing the AE Lock button? Both effect the metering pattern on the Rebel G.

-- Jim Strutz (jimstrutz@juno.com), May 20, 2000.

Yes, I did the tests correctly.

-- Steven Fisher (srf@srf.com), May 20, 2000.

Thank you all for posting the comments. Since I had this problem, I read through the Cannon Rebel 2000 manual and the 380EX manual a few times, and I am still not sure how exactly 380EX correlates exposure level with focusing point(s). I did one simple test and found out 380EX ONLY links autofocus beam to the CENTRAL focus point. I wonder what happens if the focus point is not the central point. Would this also apply to the correct exposure setting be the camera? I tried to have the photos reprocessed once. The forground is still too bright. And, this is consistent with all my night photos, as along as there is a relatively bright background.

One thing might be worth mentioning: when I take photos with people, I often focus on the subject and then shift the subject off center. Since it looks like 380EX measures exposure level through a pre-flash before the main flash, when the central focusing point is shifted, will that screw up everything?

All this confusion is due to the BAD manuals. The 380EX manual was printed in 1995 and I bought it from B&H in 1999. The Cannon rebel 2000 uses E-TTL, but apparently there seems to be differences between different cameras, and of couse that is not listed in the manuals.

-- Dave zhou (cosymoon@yahoo.com), May 20, 2000.

Don't worry, the 550EX manual doesn't give much more information than how to turn on & off it's various features. :)

You might look at the Magic Lantern book for the Rebel 2000, I haven't read it so I don't know if it gives information on how the flash metering works.

I've only had the 550EX for a few weeks and haven't had time to learn much about it.

-- Steven Fisher (srf@srf.com), May 21, 2000.

"...when I take photos with people, I often focus on the subject and then shift the subject off center. Since it looks like 380EX measures exposure level through a pre-flash before the main flash, when the central focusing point is shifted, will that screw up everything?"

If I'm right that the R2K will meter the flash bisaed to the active AF point, then yes it will mess things up, since it doesn't lock flash exposure. If you focus using the central point and then shift it off the subject the metering system will try to properly expose whatever the center AF point is now on. You can use the FEL (Flash Exposure Lock) feature for this. It will fire the preflash when you press the FEL button to take a flash meter reading & hold the measurement for up to 16 seconds, I believe, while you reframe. Either do that or use an off center focus point.

-- Jim Strutz (jimstrutz@juno.com), May 22, 2000.

FEL will not help most of the time. FEL/AEL on the Rebel results in 9.5% partial metering, so unless the 9.5% area is 18% reflective you're going to have incorrect exposure. The Rebel & 380EX combo can't do flash exposure compensation, making FEL pretty much worthless.

-- Steven Fisher (srf@srf.com), May 22, 2000.

I have suffered the same annoyance, and I believe I know the answer.

The Rebel 2000/ 380 Ex combo would meter exposure on the focusing point, true. The problem is, flash exposure is not meter (ETTL) until the shutter is released. If at the time of release, the central focusing point has been moved over a dark background (done while composing), then the camera/flash think it's exposing a dark subject, pours out too much light, and overexposes the foreground.

To solve this problem, you must use the flash exposure lock. This is the button with the asterisk (*). This button locks in the flash exposure you obtained while using your central focusing point to focus on the foreground.

Finaly, the system works, but the buttons are small, and difficult to use you need the dexterity of a juggler and timming of a gymnast.

And oh yeah, the manual sucks big time. It was probably written by an orangutan.

-- Paul Obembe (paulbee@hotmail.com), May 22, 2000.