Auto/Manual Mode

[canongear]

Just wondering if someone can explain these results to me.
Why is it that I got different histogram results when I did the following.
Using a tripod and the camera's(Canon S95 ) timer, I took two pictures of the same scene in my living room.
The first picture was taken using auto mode.
Then, I took the second picture using manual mode but, used the exact same aperture, shutter speed, ISO and white balance (AWB) that the camera used in auto mode.
The same focus point was picked by the camera in both pictures and the flash was not used in either picture.
The pictures were taken within 7 seconds of each other and the room lighting was very controlled, no changes.
Altough the two pictures look the same scene wise, one is just slightly overexposed compared to the other one, or the other way around if you prefer. The histogram display shows this and it's noticeable just looking at the two pictures.
JPEG file size is slightly different as well.
2.01 vs 2.04
2.01 a bit overexposed compared to 2.04
The camera settings are at default settings so, why wouldn't the results be the same for the two pictures?

[Wilt]

Results should be the same if everything was identical in terms of settings and light levels. If light from outside was contributing to the levels inside, passing cloud could affect the amount of light in the room by 0.5EV, for example, even though your brain perceives 'same light' with only seven seconds separating the two shots.

Can you post the shots with EXIF information not stripped off the images (not 'save for web')

[canongear]

Originally Posted by Wilt

Can you post the shots with EXIF information not stripped off the images (not 'save for web')

My living room is a mess right now so, i'm not going to post the pictures.
I understand what you are getting at regarding the possible changes in light but, the blinds on the windows are closed except for one little spot and it's a overcast day out.
That's why I'm spending quality time inside taking pictures of my living room!
Maybe what I'll do is, later tonight when there is no chance of outside light maybe affecting the outcome, I'll try the same thing again and see what happens.

[tonylong]

I can speculate that the S95 would have a bit of "leeway" in the electronic shutter and such, but I wouldn't want to make an argument for such. I really don't know how those cameras are "spec'd" for such things but I wouldn't be surprised if they are a bit looser than we might wish, so you just have to work with things.

[RichSoansPhotos]

Originally Posted by canongear

Just wondering if someone can explain these results to me.
Why is it that I got different histogram results when I did the following.
Using a tripod and the camera's(Canon S95 ) timer, I took two pictures of the same scene in my living room.
The first picture was taken using auto mode.
Then, I took the second picture using manual mode but, used the exact same aperture, shutter speed, ISO and white balance (AWB) that the camera used in auto mode.
The same focus point was picked by the camera in both pictures and the flash was not used in either picture.
The pictures were taken within 7 seconds of each other and the room lighting was very controlled, no changes.
Altough the two pictures look the same scene wise, one is just slightly overexposed compared to the other one, or the other way around if you prefer. The histogram display shows this and it's noticeable just looking at the two pictures.
JPEG file size is slightly different as well.
2.01 vs 2.04
2.01 a bit overexposed compared to 2.04
The camera settings are at default settings so, why wouldn't the results be the same for the two pictures?

You have to understand the exposure triangle, google this, or read a book by Bryan Peterson, get a grasp of exposure and turn to manual exposure rather than Auto

[tzalman]

Originally Posted by tonylong

I can speculate that the S95 would have a bit of "leeway" in the electronic shutter and such, but I wouldn't want to make an argument for such. I really don't know how those cameras are "spec'd" for such things but I wouldn't be surprised if they are a bit looser than we might wish, so you just have to work with things.

I don't know the S95 but with a lot of digicams the electronic shutter can calculate and perform a more accurate exposure than the conventional numbers. Although it reports 1/10 it may actually be activating the sensor for 0.095. Shooting manual would constrain it to 0.1.

[canongear]

Originally Posted by tonylong

I can speculate that the S95 would have a bit of "leeway" in the electronic shutter and such, but I wouldn't want to make an argument for such. I really don't know how those cameras are "spec'd" for such things but I wouldn't be surprised if they are a bit looser than we might wish, so you just have to work with things.

Originally Posted by tzalman

I don't know the S95 but with a lot of digicams the electronic shutter can calculate and perform a more accurate exposure than the conventional numbers. Although it reports 1/10 it may actually be activating the sensor for 0.095. Shooting manual would constrain it to 0.1.

I think you are both on the right tract about this.
It seems that the camera itself is producing the different results and not any possible changing light conditions.
I just tried my experiment again, and got the same results as mentioned in my original post.
There was no chance of any outside light changes this time as it's night time and the window blinds are completely closed.
Took 4 series of auto/manual shots of the same scene and the first picture of each series is slightly overexposed compared to the second picture.
It's not a big difference but, it is noticeable.
Anyway, I think I know now the reasons for the results.
Thanks for the info.