Chuck Westfall's comments on the Canon E-TTL II flash system.
Chuck Westfall of Canon was a frequent poster to the usenet, providing useful information about how things really work. Here are some of his comments on E-TTL as copied from the various forums. I put them here on this web page to preserve them in case the source, a forum post at: http://photography-on-the.net/forum/showthread.php?t=46311 expires.
Also note that Canon now has a good web page with specifics on E-TTL II here:
Here are the postings that are believed to come from Chuck Wesfall:
E-TTL II is never linked to the active focusing point. In fact, that is one of the main differences between E-TTL II and the original version of E-TTL. There's a brief description of E-TTL II with sample images on Canon Inc.'s EOS-1D Mark II web site here.
Here's what we published in the EOS-1D Mark II White Paper document:
ADVANCED E-TTL II
New algorithm gives greater flash exposure control
For improved flash control using Canon EX series Speedlites, a new algorithm has been developed which enables superior E-TTL flash accuracy and reliability. In previous cameras, evaluative flash metering was based on the assumption that an autofocus point would cover the subject. When this is not the case, inaccurate flash exposures result. The EOS-1D Mark IIâ€ôs evaluative flash metering is not dependent on the active AF point.
In the new algorithm, ambient light is measured when the shutter button is pressed. Next, a pre-flash is fired and the metering sensor takes readings at the central 17 metering zones. The ambient and pre-flash readings are compared. The metering areas having a small difference are selected as the flash exposure metering areas. (Areas with very big differences between ambient and pre-flash readings are excluded or down weighted because they are assumed to contain a highly reflective object or that the subject is not in that part of the frame. The algorithm avoids chronic underexposure problems in such situations.) These readings are weighted, averaged and compared with the ambient light reading, and the main flash output is then set and stored in memory. E-TTL II weights and averages the flash metering for the subject and all other objects at the same distance as the subject. Even if the subject's position, reflectance or size changes, the flash output will not change radically. The flash exposure will be highly accurate and stable.
Most EF lenses provide distance information, and this data is also considered in determining if there is a highly reflective object, once again lessening the chance of underexposure.
Additional info on E-TTL II:
In essence, distance information is not required for E-TTL II. But when it is available and the flash is direct, then it can be used as a reference to modify the flash exposure if necessary.
Additionally, the EOS-1D Mark II is provided with a new Custom Function (C.Fn 14-1) that allows photographers to select between evaluative and averaged flash metering in E-TTL II. Averaged flash metering may be preferable when using direct flash with lenses that don't have a built-in distance encoder, but we encourage Mark II owners to try both settings on C.Fn 14 to see which metering method they prefer.
E-TTL II is functional with all EF lenses, not just the ones with distance data. If you use a lens without distance data, the only thing you lose is the Mark II's ability to modify the flash exposure based on distance data. Everything else works the same. The Mark II *never* falls back to E-TTL.
There's never any "full-frame" flash metering in E-TTL or E-TTL II. In both cases, with EOS cameras that use a 45-point focusing system/21-zone metering sensor, all flash metering is carried out by the 17 metering segments within the Area AF ellipse shown in the viewfinder. Subject matter outside the ellipse is completely ignored in terms of flash exposure control.
C.Fn 14-0 on a Mark II allows E-TTL II flash metering to be subject-based, so it can use anywhere from 1 to 17 metering segments depending on the camera's analysis of the pre-flash information. This is not a spot meter reading, unless the camera determines that the subject is so small that it occupies only one of the 17 metering segments. Most subjects will cover a larger area than that. The big improvement here over the original E-TTL algorithm is that the size and location of the primary flash metering area can change dynamically according to the size and location of the subject. In the original E-TTL algorithm, the size, location and weighting of the primary flash metering area was linked to the active focusing point.
C.Fn 14-1 on a Mark II applies the E-TTL II flash metering algorithm equally to all 17 metering segments within the Area AF ellipse, so in comparison to 14-0, 14-1 is not subject-based.
As I've mentioned in previous posts, the setting of the focus mode switch on the lens has no bearing on E-TTL II flash metering, so C.Fn 14-0 and 14-1 are effective whether the lens is set for AF or manual focus.
There are two differences between original E-TTL and E-TTL II in this particular comparison:
1. E-TTL II will factor in distance information when it is available during direct flash operation, regardless of the C.Fn 14 or C.Fn 4 setting. Standard E-TTL does not use distance info.
2. When C.Fn 14-1 is active on the Mark II, E-TTL II flashmetering is averaged for all 17 metering segments regardless of the C.Fn 4 setting or the focus mode set on the lens. This gives the photographer more flexibility in setting up the camera according to their personal preferences. Original E-TTL can't average its flashmetering unless specific camera settings are used.
Director/Technical Marketing Dept.
Camera Division/Canon U.S.A., Inc.