Does a full frame see less light?

[JakAHearts]

Post pictures with Exif intact.

[Numenorean]

What's your shutter speed?

[Wilt]

Originally Posted by snapshot2011

tests were conducted in manual and flash bounced off ceiling.

I have made the test very fair, but the 5d just doesn't see the light like the 60d does

Just to prove that "larger format does not need 'more light' "...ISO 100, 1/160 f/7.1, manual flash power

40D

5D

If anything, these two shots would seem to indicate "smaller format does need 'more light' " is true!

[btweller]

Originally Posted by Wilt

Just to prove that "larger format does not need 'more light' "...ISO 100, 1/160 f/7.1, manual flash power

40D

5D

If anything, these two shots would seem to indicate "smaller format does need 'more light' " is true!

Did you physically move the 5D closer to get the same framing, or did you switch lenses? If you moved the camera closer, I'd argue that is the reason the 5D shot is brighter. Assuming a speedlight in the hotshoe, on manual power your flash-to-subject distance just decreased.

[tkbslc]

24mm on FF is very wide, 24mm on crop is not. So your light bounce would have to cover more area of the room to fill the entire scene with light. The flash (in ETTL) should increase output to compensate for the wider angle, but sometime you are at the limit or metering is fooled. If you used 38mm on the FF camera (24mm x 1.6 = 38mm) you should get the same light bounce and exposure.

[dmward]

Come on guys.
The physics of light and lens design haven't changed just because the light receptor is a digital sensor instead of film.
Think about this for a second; Tri-X film in a 35mm, 2 1/4 sq, and 4x5 camera, all took the same exposure. Now there were shutter and lens transmission differences that might have some minimal impact. And I could put all three sizes of film in the same developer for the same amount of time to get negatives with the same characteristics.
Same is true today with different sized sensors. There may be manufacturing tolerances that minimally affect the outcome but different cameras, with different sensor sizes, using a zoom lens on a tripod with manual metering, exposure and consistent light output will deliver essentially the same digital negative to Lightroom (or whatever digital "developer" you use). If you are doing the test with JPGs then all bets are off, because there is no way to be certain the in-camera raw processing to JPG is consistent.

Better film example would have been Ektachrome 64, which I did many times on multiple formats for clients

[Wilt]

Originally Posted by btweller

Did you physically move the 5D closer to get the same framing, or did you switch lenses? If you moved the camera closer, I'd argue that is the reason the 5D shot is brighter. Assuming a speedlight in the hotshoe, on manual power your flash-to-subject distance just decreased.

I used the Tamron 28-75mm f/2.8 zoom. Set it to 28mm for 40D shot, then zoomed it to frame the same area for the 5D shot. both cameras put on the same tripod for the shots.
Metz 54MZ flash on Manual power. No change to camera position, zoom head set to 28mm direct flash for both shots (to cover the same physical area from the same distance)

Shots captured in RAW, imported into LR. no postprocessing done to the shots.
I had to increase exposure of the 40D in postprocessing by +0.33EV to equal the brightness of the 5D shot, but I did not post that alteration.

[dmward]

Originally Posted by Wilt

I used the Tamron 28-75mm f/2.8 zoom. Set it to 28mm for 40D shot, then zoomed it to frame the same area for the 4D shot. both cameras put on the same tripod for the shots.
Metz 54MZ flash on Manual power. No change to camera position, zoom head set to 28mm direct flash for both shots (to cover the same physical area from the same distance)

Shots captured in RAW, imported into LR. no postprocessing done to the shots.
I had to increase exposure of the 40D in postprocessing by +0.33EV to equal the brightness of the 5D shot, but I did not post that alteration.

Wilt,
the 1/3 stop and white balance differences are more than likely sensor characteristics between the two cameras. Would be interesting to put a color checker in the middle of the image and see what dng profiler did to get the two to a common baseline.

[Curtis N]

There are differences between camera bodies in how E-TTL works.

The point was made that, assuming the autozoom on your 580EX II is set to compensate for the camera's sensor size, it will zoom to a shorter focal length on the 5D and the flash range will be reduced as a result. However,
1) The E-TTL flash metering system will automatically account for that, and
2) It's irrelevant in ceiling bounce configuration.

You may want to check your camera's custom functions regarding E-TTL metering (Evaluative vs. Average), but other than that, chalk it up to differences in camera firmware and adjust FEC accordingly.

[Wilt]

Originally Posted by dmward

Wilt,
the 1/3 stop and white balance differences are more than likely sensor characteristics between the two cameras. Would be interesting to put a color checker in the middle of the image and see what dng profiler did to get the two to a common baseline.

David, I agree about your comment about sensor characteristics between two models.

In response to your request for AWB using LR eyedropper on 4th grayscale patch of ColorChecker (no other post processing differences between the shots). Manual flash power for both, direct flash, fixed zoom head coverage. ISO 100, 1/160 f/9 both both shots.

40D

5D

[drvnbysound]

Curious... rather than trying to do these test scenarios with manual flash, couldn't you also eliminate the same variable by removing the flash all together? Shoot the same shot with both cameras using ambient light only? I would think this would be a better way to determine if one sensor is more sensitive than another...

[Wilt]

Originally Posted by drvnbysound

Curious... rather than trying to do these test scenarios with manual flash, couldn't you also eliminate the same variable by removing the flash all together? Shoot the same shot with both cameras using ambient light only? I would think this would be a better way to determine if one sensor is more sensitive than another...

Yes. Given the fact that the OP questioned the need for 'more light' from the flash with the larger sensor, I sought to directly address that perception with my test. But you are right...light is light!

Since the OP was using flash in ETTL mode, I wanted to remove the variable of the bodies telling the flasj to output a different amount of light, via Manual flash (known to be consistent to within 0.1EV, based upon measurements made with a flash meter). Keeping the flash in the test scenario would eliminate any opportunity for OP to voice issue with 'but I used flash, your tests did not'.

[SkipD]

Originally Posted by snapshot2011

Is this just a moving up to full frame thing or have I an issue? To counter act this, I have to increase my FEC by +1 to get near to the 60D look.

Are you using the same focal length on both format cameras or are you changing the focal length to get the same framing with both cameras?

The reason I ask this question is because if the framing is different for the two cameras (same focal length, for example), the lighting may be analyzed as different by the two cameras because there is potentially a significant difference in the reflective qualities of the scene being analyzed by each camera's metering system.

[dmward]

Originally Posted by Wilt

David, I agree about your comment about sensor characteristics between two models.

In response to your request for AWB using LR eyedropper on 4th grayscale patch of ColorChecker (no other post processing differences between the shots). Manual flash power for both, direct flash, fixed zoom head coverage. ISO 100, 1/160 f/9 both both shots.

40D

5D

Wilt,
It looks to me like they are identical. I'm looking at them on a Mac Book Air so that may add some discrepancy.

Why did you choose the 4th gray square? It's my understanding, from comments made by people in other fora that are involved in Lightroom development, that the Adobe WB tool is based on readings from the second square from the left. Although, in my experience, there is very little variance across the gray scale row.
What is interesting is the shape of a tone curve when adjusting it so that all of the squares read their defined value.

[Wilt]

Originally Posted by dmward

Wilt,
It looks to me like they are identical. I'm looking at them on a Mac Book Air so that may add some discrepancy.

Why did you choose the 4th gray square? It's my understanding, from comments made by people in other fora that are involved in Lightroom development, that the Adobe WB tool is based on readings from the second square from the left. Although, in my experience, there is very little variance across the gray scale row.
What is interesting is the shape of a tone curve when adjusting it so that all of the squares read their defined value.

Here are the two histograms. I can see a very slight difference in the photos and the histograms. Certainly not 0.33EV worth, as the prior results showed.
40D
5D

I chose the 4th square simply because it is same as 18% gray. That square I have found to match exposure and WB results obtained with Kodak, Douglas Gray Card, and PhotoVision target, so I have no reason to doubt its value or neutrality. I had not heard previously about the Adobe WB tool.
Got any reference link that I can read?