I've been using my Rebel XTi happily for about three weeks now, and while not new to photography (been using an all manual film SLR quite a while), I'm new to digital SLRs and their dodads and gizmos. What I've been experiencing looks to be underexposure. And I'm thinking the light meter (I assume there is one like my manual SLR) seems to be uh... "wrong."

I've been reading that other PPL are reporting (or complaining) about underexposure on their XTi's but with various composed scenes where there is an abundant variation in light, and also colour balancing to worry about as well, and so I believe my XTi may also be "fooled" by various scenes. BTW, I shoot with center-avg weight, I've read on other pages where Evaluative Metering is weighed on the focus point that is selected, biasing exposure to the focused area more. And since many ppl tend to focus on brightly coloured parts of the scene as their "subject" this tends to cause a bias towards underexposure, as the camera tries to make the bright subject "in the middle" of the exposure gradient. Fine by me, I just wish Canon mentioned this somewhere in the literature.

So with all this "variations" that come into play, I decided to go about and do a standard test. I pulled out my 18% grey card, being after all, "middle grey". So using the grey card, slightly tilted to the light source, colour balanced, filling most of the frame, the spike falls within the 2nd 5th of the histogram. Definately not in the middle, as I had expected. Nor is it "just slightly left of the middle".

Can somebody explain this? Do other Canon dSLR bodies behave this way? And why do I ask if other bodies ask this way? Continue to read this epic 1st post to find out!

-------------------------------------

According to this page I read:

http://www.luminous-landscape.com/tu...stograms.shtml

The spike is supposed to be in the middle of the histogram (18% grey is indicated in the middle of the histogram). That would mean my XTi is acting kinda... slightly off.


However on further searching, I come to this page:

http://www.bythom.com/graycards.htm

Who argues that these new digital cameras are set at *gasp* 12% grey and so in order to get the "proper exposure" using an 18% grey card, you take multiple shots with exposure compensation at third-stop increments, and pick the shot that gives you the spike in the middle...

So wait a moment. Is my XTi underexposing by design, and I need to dial in a positive exposure compensation of about half a stop because of this 12% grey? My 18% grey isn't middle because it's reflective, instead of luminance. Would this be the same on other Canon dSLR bodies

BTW, after dialing in 1/2 a stop, the histogram is "slightly to the left" of the centre, while dialing in 2/3 of a stop gets me almost dead on center, with a fraction slightly to the right.

For xTi Users...have anyone discovered this???

From my 1 month+ experience of using my 400D, the photos that come out when exposure compensation is 0 do look kind of dark, compared to those that come straight out of a P&S. So most of the time the EC is set to +1/3 or +2/3. Even the shops where I tested this camera before buying had it set to +1/3 or +2/3.

That's the workaround I'm using to negate the darkness, other than using the GIMP to brighten those shots that seemed too dark.

But then, some websites are arguing that the 400D is exposing correctly, and the previous incarnation of the Rebel-the 350D, actually overexposes.

I just ignore most of what's said, and just turn the exposure compensation to suit my taste.

Every camera has different presets mah...
Some cameras might be too bright at 0 while some too dark..
As said, there's always an exposure compensation =)

i found that this usually happen if i use evaluative metering. Try using partial metering. Most of the time it turns up to be better.

Can also lock the exposure (AE lock? i think) for what you wanna expose, recompose your shot then shoot.

One of my friend bring it back to canon.. and when it come back out.. it is much much better now..

I was told it was "design" of 400D but alot ppl said their got their "fixed"...i still doubt about it.

http://www.pbase.com/jrmiami/canon_fixed_m...i_underexposure

http://www.stevesforums.com/forums/view_to...535&forum_id=37

http://www.digitalhome.ca/forum/showthread.php?p=535312

so, can we get canon to fix it for us or we just have to live with it?

try calling canon up and see how's their reply

after reading http://www.stevesforums.com/forums/view_to...535&forum_id=37

and tried it out with my sigma 28-200mm and 50mm f1.8 lens. Confirm this is wat i am getting too. kitlens is underexpose, other lens normal exposure... need to call canon liao

This post has been edited by vincent_audio: Apr 22 2007, 02:17 AM

a simple tip, the histogram may be useful for correct exposure, but never rely too much on it. you might end up reading the histogram too much.

Oh kitlens is under exposure ?? Shix no wonder most of my photos is look dark if EV 0 ...hmm i think better dun send back to canon for "tuning" since like been said others lens is ok + if we send back to canon they might just "overclock" the EV internally from the camera where no longer match what we say from LCD...

Just like u telling a TV repair man said "I wan my TV brightness more even i have tune it to max bright"...AFAIK TV all kinda of device they have inner adjustment can be done..long term might be no good for the camera sensor ler...

Just my though....

I'm kinda regret to choose Canon over Sony Alpha..

The "60 years old experience" thing is nothing but a gimmick...and because of alot of "die-hard" fans advices...

... ended up buying something with alot of problem lar...

will shout at canon tomorrow morning!
wanna him to fix all my problems~

First time heard ppl buy canon 400D regret

I love my 400D

poor Nonac Owner
good luck~
shout at Saujana!

This post has been edited by shinchan^^: Apr 23 2007, 08:49 AM