Canon Wireless Flash Ratios

There has been a discussion on DPReview lighting forum about the ratios on wireless flash. The thread starts at: http://forums.dpreview​.com …rum=1025&messag​e=15185401 In addition, there were some test by TomCee posted at: http://tomcee.smugmug.​com/gallery/841767

As you can see, there is some question on how Canon sets the ratios and what the resultant H:S (Highlight:Shadow) ratios are.

Can anyone...Scott, Robert, etc....shed some light (no pun) or knowledge on this.

Pls note these tests were done with a DRebel, so ETTL, not ETTL-II, was in play. Don't know if that would make a difference. I guess it would be helpful if someone had similar tests on a ETTL-II camera.

Thanks....JOHN

The Canon Ratios are (A:B):C where C is for a background light and not calculated by the camera as exposing the subject. If you point C at the subject it will be overexposed.
Here are a number of threads on ratios:
http://www.photography​-on-the.net/forum/showthre​ad.php?t=95285
http://www.photography​-on-the.net/forum/showthre​ad.php?t=92411
http://www.photography​-on-the.net …?t=96081&highli​ght=ratios
http://www.photography​-on-the.net …?t=87673&highli​ght=ratios
https://photography-on-the.net …ad.php?t=46778&​highlight=
Can you be more specific about what you would like to know about wireless ratios? ETTLII hasn't changed ratios, just how the preflash is measured. In addition, the ratios you are setting up are not Highlight/Shadow ratios as much as Key/Fill ratios as you are not setting up a ratio between flash and ambient light but between flashes.

Thanks, Scott. I have read, followed, and done some minor posting on the threads you listed above.

Chuck Gardner has raised, what I think, are good questions, the answers which would be valuable.

He states them, much better than I could, in this post: http://forums.dpreview​.com …rum=1019&messag​e=15232905

John

The ratios used on EX speedlights are different then what you measure with studio lights. In studio lighting, the Key light is measured with the fill light on. That's why a 3:1 ratio is a one stop difference. You are taking an incident flash reading at the subject. With wireless ETTL, the output of each flash is determined by the preflash reflected back to the camera from each flash. A 2:1 ratio here is a one stop difference because only one flash is measured at a time. It also isn't going to be as accurate as studio lights and a meter, but it is pretty damn good. The only way to replicate the accuracy of studio lights is to put the EX lights on Manual with Pocket Wizards and then you can set ratios with an incident flash meter. You could also use wireless Manual, but then you can't get a reading with both lights firing because the wireless controller will set off the meter prematurely. You can meter each light individually using the Pilot button.

Scott, follow on question....

As you have stated, the wireless settings (made on the on-camera unit) are (A:B):C.

I assume that after doing it's ambient and then pre-flash metering, ETTL-II will set the level of the over all flash (with each unit sending its set ratio) to properly expose what it determines is needed to properly illuminate the "subject".

Can you still use camera FEC to rasie or lower the combined light output (of all the groups, with ratios maintained)? if not, how would you adjust, using wireless not manual flash, if the "subject" were under or over exposed?

John

Excellent question and one that took an entire afternoon to work out the details of. If all the units are on different groups as you would with ratios, FEC set on the camera or the Master affects the total flash output.

If you want to use wireless ETTL and don't like ratios, you can set all the flashes to Group A and set FEC individually on each flash.

For total control, use wirelesss Manual. Hold down the Mode button of each flash for >2 secs until M flashes. Then set each one to the desired output using a flash meter. I set the on camera flash to Master with it's flash set to off so only the controlling pulses are fired from the camera position. Set the other flashes to Slave.

Allow me to narrow down on the quoted statement. What happens if I have 2 (or more) units in a group.

For example, I assume your "off" master is set to group 'A' and you have another unit set to 'A' to actually flash for, let's saw fill light from above the camera position (or where ever).

Or, using a bow/umbrella, you need more power so you set to flashes to 'B' and bounce them both into the same umbrella.

Will the second 'A' unit be increased by FEC set on the master? Will both 'B' units be equally increased?

I am sure the combinations and permutations could get rather confusing, so I will assume like units and not to expand this further.

John

Al this setting to manual, and setting the power can be done from the master flash I believe (at least with a 580EX): Leave the slave in whatever mode, se the master to 'M'. slaves will follow and their output can be set on the 580.

You can also set the power to the slaves on the master. I prefer to do it on the slave itself. It makes it easier just to walk over to the slave and adjust it with the flash meter.

When you set FEC on the Master, and the slaves are assigned A,B and C, all the groups are increased in a manner that preserves the ratios you set. If there are three flashes set to A and one to B and you set FEC to +1, the total output of the A slaves will be increased by one stop as will the single B slave. I suspect that each A is increased 1/3 as much as the single B. If all the slaves are assigned to the same group, then you can set FEC individually/independe​ntly on each flash.

Just a quick question:
If I set my 580 (master) to _not_ fire, the output of the slave (550) will still vary with the setting of a ratio. Is this related to the amount the master 'would have' flashed if set to 'on'?
About setting the slaves output from the master: It seems the only way to get a 550EX to do 1/3 stop increments... (at least I think it does, don't have a flashmeter at hand)

Thanks again Scott...

Slightly different direction, but still on wireless ETTL (not manual)...

Would I be correct in assuming that, on a 20D or similar ETTL-II camera) that the camera determines the intencity of each flash by the change in light from the ambient reading before the pre-flash and the during tyhe pre-flash....on each metering point. Effectively giving a "spot metering" of the areas illuminated by 'A'....and then does the same for 'B'.

This is the only way I can imagine that it could get fairly accurate readings of each lighting groups relative intensite to be able to set the ratios. I cannot imaging that determining the ratios based on the change in overall average metering would work, as the areas of one might be greater than another and have a significant effect. How many meter points are involved...I seem to remember the number being the central 17. (If true, does that mean one should never use CF14=1 in this set up).

John

Yes, sort of, although the more accurate answer is that the amount of output from each flash depends on the group that it is in, the number of flashes in the group, and the ratio that has been set. The master does send control signals to the other flashes but it's the camera itself that's actually controlling all the flashes and their outputs. The outputs are not controlled relative to what the master puts out or would put out in the case when it's set not to fire. Remember you can have multiple slaves in group A and the combined total of these flash output determine the output level from group A, not what is output by the master along. And the same goes for the other groups. All flash outputs are relative to the total output required for the shot, taking into consideration the set group ratios and the total number of flashes in each group.

There is no reason why CFn14=1 should not be used if there was a reason for using it in a non-wireless remote situation. Obviously, for most portrait work, CFn14=0 (Evaluative) would probably work out best since we want the emphasis to be on the subject, without background lighthing (or lack of) influencing the exposure.

However, if we were taking, say, a group shot of a lot of people, then CFn14=1 (Average) may work out a little better.

The center 17 metering points applies to the 1D (mark II) series cameras. On the 20D, all the metering points are used.