E-TTL II - implications for Canon flash systems

[mark u]

There is a rather cryptic reference to the new E-TTL II system in the

releases that have prematurely sneaked out of Canon France about the

EOS 1D II. It's clear that the new system uses focal distance

information in assessing the exposure (like the Nikon system), but

that's about all. Supposedly the EOS 1D II itself remains compatible

with EX speedlites, but it's not clear if existing speedlites will

handle the new system on the new body, or just rely on E-TTL I

standards. A post on fredmiranda.com suggests the latter is the

case, and that there is also a new flashgun in the works. Since the

new system relies on focal distance information, it could well lead

to problems with any lenses that fail to report this correctly.

 

Until we get more hard information, I'd suggest that anyone with a

Type B body (with A-TTL suited to EZ flashguns) not waste money on an

EX Speedlite with the idea that this can be used with a future

digital body, but stick to the cheaper EZ guns for now. I suspect

that the new gun(s) may well turn out not to support TTL or A-TTL

modes, even though I'd expect them to support E-TTL I.

 

How the new E-TTL II system will percolate into the rest of the range

will also be a topic of interest.

[omar_torres]

The way I understood the distance information was that indeed Canon's lenses could report the information, but that it just wasn't allowed to be used in flash exposure calculations because of the patents. So I guess that the lenses will work as usual but for E-TTL II you would need a new body and flash. Unless current bodies/flashguns can be firmware updated to the new calculations.

 

This is all just speculation and guesswork of course.

[omar_torres]

Let me add something to that; Some Canon bodies have DEP mode, so indeed they can and do use the distance information because it is required to calculate the hyperfocal distance in DEP mode. If current bodies/flashguns can be 'flashed' with new firmware then tell me where to send the body :D

[bill_goldman]

This is only speculation on my part, but I believe that E-TTL II is only a refinement of the E-TTL system and not a major change. Since EF lenses, at least most of them if not all, supposedly have the capability to report the focused distance to the camera, it would appear that the addition of that data into the camera's algorythms can only improve the performance of the E-TTL system.

 

Since the camera computes the proper flash exposure, present EX flash units will not be compromised. I do not believe the distance data will be used to calculate exposure but to verify subject position or unusual subject reflectance.

 

Of course we will have to wait and see, when the details become available.

[mark u]

Bill, you may be right. As you say, until we get real details we're left with speculation/reading between the lines on the available information. I do think that Speedlites will be re-vamped. There's really nothing that's ideal for mainstream DSLR (10D, 300D) owners. And there's no doubting that flash has been a cause of much angst from those communities, so improvements are to be expected. Film users are better served.

[vesa_perala]

I'm with Bill here.

 

I think the flash doesn't calculate the exposure, it is the body which just tells the flash(es) what output power to use.

 

Vesa

[omar_torres]

Didn't know that the flashes didn't play a part on the exposure calculation. In any case, the important part is that there won't be a different batch of lenses with distance information, a la Nikon, in order for it to work. So it could be just a matter of flashing the body's firmware. Can't wait!

[eosdoc]

<blockquote><small><i>Some Canon bodies have DEP mode, so indeed they can and do use the distance information because it is required to calculate the hyperfocal distance in DEP mode</i></small></blockquote>

This statement is wrong. DEP mode does not use subject distance.

Instead, image defocus is used, according to:<BR>

<a href=

http://bobatkins.com/photography/eosfaq/eosfaq24/9miscellany.html#q27

>http://bobatkins.com/photography/eosfaq/eosfaq24/9miscellany.html#q27</a>

[mark u]

An interesting aside in Chuck Westfall's description of DEP mode is this paragraph:-

 

"The presence or absence of distance encoders in the lens is irrelevant for the purpose of EOS Depth AE calculations. Keep in mind that the first EF lenses with distance encoders were the EF 35-135 USM, 70-210 USM & 100-300 USM, introduced in 1990; OTOH, Depth AE was introduced at the beginning of the EOS series, with the EOS 650 in 1987."

 

This implies that at least some older lenses would not be able to give distance information to the system. OTOH I could imagine that flash exposure metering could be linked to the defocus amount and DOF so that it would be optimised for the "subject" but not dependent on whether the point of sharpest focus fell on a bright or dark portion of the scene.

[steve_dunn2]

<p>Many, but not all, Canon EF lenses return distance information, so a distance-based flash metering scheme can't be used with all lenses (I'm not sure about third-party EF lenses but I'd be surprised if it weren't true that some do and some don't). But this is nothing new; Minolta and Nikon both have distance-based flash metering systems and a collection of lenses including some which do not return distance information. So it's easy to see how that would work: if your lens returns distance information, you get E-TTL II; if it doesn't, you get E-TTL I.</p>

 

<p>The flash does not do metering; that's all done in the body. Any of Canon's E-TTL-capable bodies has all the hardware it needs to do E-TTL II; all they would need would be a firmware upgrade (which Canon surely will not do, with the remotely possible exception of the 1V). For that matter, even non-E-TTL bodies that have multi-zone ambient exposure metering sensors have all the hardware they need to do E-TTL II or E-TTL I metering; again, they're simply not programmed to know about E-TTL. No change is required to the flash units, because all the flash does is emit a pre-flash when told to do so, and emit the main flash at whatever power level it's told by the body.</p>

 

<p>I'd agree with the suggestion not to buy a new flash between now and PMA unless you need it in that time, but not for the reason suggested. I'd say it's because Canon may introduce a new flash unit (or they may not; I haven't seen rumours of one but I don't see all the rumours). That new unit may be the best fit for you. Or it may replace an existing unit, whose price would then drop and you could pick it up for less. I'd say the same thing about bodies and lenses, too; wait until you know for sure, not just from rumours, what new stuff will soon be on the market.</p>

 

<p>Canon has already stopped supporting A-TTL in its new flashes; the last new flash to support A-TTL would have been, um, I believe the 540EZ back in the early-to-mid 1990s. AFAIK, all Canon flashes since the introduction of E-TTL circa 1995 have been EX flashes, which support E-TTL and TTL but not A-TTL. I suspect Canon will continue to support TTL for a while yet, partly for compatibility and partly because it's low-tech, long since debugged, and is therefore very cheap to include in flash units.</p>