ETTL-IIb?

While playing with the 1D3 I noticed that the flash pictures taken with my 580EX looked nice and bright on the large LCD screen (set to factory default in the middle). Checking the histogram showed that it wasn't the LCD brightness but the actual exposure that made the image bright.

Now, from previous experience, I know that my copy of the 580EX habitually underexposes (more so than the 550EX, the 420EX and the Metz 58AF-1) and the images never come out looking as bright as the one from the 1D3 without some +FEC dialed in. Curious, I pulled out my 1D2 and mounted the lens and the 580EX from the 1D3 and set the settings exactly like those on the 1D3. As expected, the image was a tad underexposed. Did the same with the 30D and got similarly underexposed pictures. I took 8 sample shots with each camera just to make sure it wasn't a fluke.

HOSTED PHOTO
please log in to view hosted photos in full size.

Underexposure isn't your flash - it's the in-camera ETTL system. I can account for variances in your flash exposure to the camera. I can put my Canon camera on a tripod, fixed subject and press the shutter release 3 times and get different selected focus points for each image.

Note that flash exposure in E-TTL mode, for the Canons, is based upon the subject brightness at the focus points. If you have multiple focus points selected, the brightness at those points will be averaged. If you have one point focus, the brightness at that single point will determine exposure. I have done extensive testing with my 20D to prove it.

The best thing Canon users can do is to disable EVALUATIVE flash exposure when flash is the sole light source (i.e., flash is NOT is the fill mode). Using the AVERAGE setting for E-TTL will virtually eliminate flash underexposure with Canon flashes. Outdoors remember to reset back to EVALUATIVE. On my 20D, it's Custom Function 14.

Note that the severe flash underexposure problems was what "ETTL II" was all about - it merely added the AVERAGE setting to the firmware. I was having 40%-50% severe flash underexposure.

RVsForFun wrote in post #3499136
Underexposure isn't your flash - it's the in-camera ETTL system. I can account for variances in your flash exposure to the camera. I can put my Canon camera on a tripod, fixed subject and press the shutter release 3 times and get different selected focus points for each image.

Note that flash exposure in E-TTL mode, for the Canons, is based upon the subject brightness at the focus points. If you have multiple focus points selected, the brightness at those points will be averaged. If you have one point focus, the brightness at that single point will determine exposure. I have done extensive testing with my 20D to prove it.

The best thing Canon users can do is to disable EVALUATIVE flash exposure when flash is the sole light source (i.e., flash is NOT is the fill mode). Using the AVERAGE setting for E-TTL will virtually eliminate flash underexposure with Canon flashes. Outdoors remember to reset back to EVALUATIVE. On my 20D, it's Custom Function 14.

Note that the severe flash underexposure problems was what "ETTL II" was all about - it merely added the AVERAGE setting to the firmware. I was having 40%-50% severe flash underexposure.

I think you have your ETTL and ETTL-II confused. The 20D is an ETTL-II camera. Hence, it's flash exposure will not be determined by the AF point but, rather, by what the camera determines to be the subject in the scene.

No, I don't have E-TTL and E-TTL II confused. I use the E-TTL term for both as E-TTL isn't available anymore. And E-TTL II *does* use reflectivity at the active focus point(s) for exposure, and I can prove it. The basic difference between E-TTL and E-TTL II is the ability to turn off this programming and get an averaging result, hence the CF 14 on the 20D. Until the 580EX II, no E-TTL flash from Canon could actually determine its own light output, the camera does.

If you'd like to prove that E-TTL II bases its exposure upon the focus point reflectivity, here's how to do it. Make sure your E-TTL II camera is set to the default of "0:EVALUATIVE" setting:

1) Shoot a photo of a person a few feet away using a single, multiple focus points. This is your control photo. Make sure the subject(s) is wearing a neutral shirt - not black and not white.

2) Shoot the same subject, same distance, same flash setting (bounce or not bounce, whatever) but use a single, central focusing point. You may or may not see a difference in this exposure and the first one.

3) Shoot the same subject again, but have them hold up a white sheet of paper where the central focusing spot occurs. You'll see markedly reduced exposure, in the order of 2-3 stops.

4) Shoot again, but hold up a very dark patch of cloth where the central focus point occurs. If it's so dark you can't focus, use manual focus. You'll note gross over-exposure in this image.

5) Change the custom E-TTL function to "1:AVERAGE" and shoot again. You'll see a difference.

RVsForFun wrote in post #3508106
E-TTL II *does* use reflectivity at the active focus point(s) for exposure.

That statement runs contrary to official canon documentation, and my own experiments.

Here are two screen captures from EOS Viewer Utility showing the active focus point for each. The left side of the image is black cardboard and the right side is white paper. I took three shots each using right, left and center AF points, as well as manual focus. There was no difference detected between the various AF points.

File Name
IMG_0653.CR2
Camera Model
Canon EOS 20D
Shooting Date/Time
7/8/2007 2:56:11 PM
Shooting Mode
Manual Exposure
Tv( Shutter Speed )
1/250
Av( Aperture Value )
8.0
Metering Mode
Evaluative Metering
ISO Speed
100
Lens
70.0 - 200.0 mm
Focal Length
200.0 mm
Image Size
3504x2336
Image Quality
RAW
Flash
On
Flash Type
External E-TTL
Flash Exposure Compensation
0
Red-eye Reduction
Off
Shutter curtain sync
1st-curtain sync
White Balance Mode
Auto
AF Mode
One-Shot AF
Parameters Settings
Contrast Mid. High
Sharpness Mid. High
Color saturation Mid. High
Color tone 0
Color Space
sRGB
Noise Reduction
Off
File Size
7386 KB
Custom Function
C.Fn:01-4
C.Fn:02-0
C.Fn:03-0
C.Fn:04-1
C.Fn:05-0
C.Fn:06-0
C.Fn:07-0
C.Fn:08-1
C.Fn:09-0
C.Fn:10-0
C.Fn:11-1
C.Fn:12-0
C.Fn:13-1
C.Fn:14-0
C.Fn:15-0
C.Fn:16-1
C.Fn:17-0
C.Fn:18-0
Drive Mode
Single-frame shooting
Owner's Name
Curt & Becky Newport
Camera Body No.
1721004204

HOSTED PHOTO
please log in to view hosted photos in full size.

HOSTED PHOTO
please log in to view hosted photos in full size.

RVsForFun wrote in post #3508106
No, I don't have E-TTL and E-TTL II confused. I use the E-TTL term for both as E-TTL isn't available anymore. And E-TTL II *does* use reflectivity at the active focus point(s) for exposure, and I can prove it. The basic difference between E-TTL and E-TTL II is the ability to turn off this programming and get an averaging result, hence the CF 14 on the 20D. Until the 580EX II, no E-TTL flash from Canon could actually determine its own light output, the camera does.

If you'd like to prove that E-TTL II bases its exposure upon the focus point reflectivity, here's how to do it. Make sure your E-TTL II camera is set to the default of "0:EVALUATIVE" setting:

1) Shoot a photo of a person a few feet away using a single, multiple focus points. This is your control photo. Make sure the subject(s) is wearing a neutral shirt - not black and not white.

2) Shoot the same subject, same distance, same flash setting (bounce or not bounce, whatever) but use a single, central focusing point. You may or may not see a difference in this exposure and the first one.

3) Shoot the same subject again, but have them hold up a white sheet of paper where the central focusing spot occurs. You'll see markedly reduced exposure, in the order of 2-3 stops.

4) Shoot again, but hold up a very dark patch of cloth where the central focus point occurs. If it's so dark you can't focus, use manual focus. You'll note gross over-exposure in this image.

5) Change the custom E-TTL function to "1:AVERAGE" and shoot again. You'll see a difference.

OK, I stand corrected. You do not have ETTL and ETTL-II confused. It seems like you just don't know what the difference between the two are. I think you may benefit from the section called "How ETTLII works" in the following thread:

https://photography-on-the.net/forum/showthre​ad.php?t=46599

Re ETTL not being available anymore, it may be true that new EOS cameras won't be employing them anymore but ETTL is very much available in the older cameras that are still out there, like the 10D, the 1D, DRebel and all those EOS film cameras.

As they say, pictures are worth a thousand words. I did the "test" you suggested I do. Although it wasn't exactly the same test, the principal is the same. And as shown by my results, ETTL-II is, in fact, subject based and not AF point based. Notice that none of the pictures show exposure differences of 2 or 3 stop.

HOSTED PHOTO
please log in to view hosted photos in full size.

HOSTED PHOTO
please log in to view hosted photos in full size.