Reading mode: Light Dark

ETTL tested on 10D - 9 images - you may be surprised

Go to last post
That's my point over and over. The AF point was center in both cases. And yes, the ETTL image is underexposed... but the Auto-flash is not. The ETTL set the exposure wrong compared to the Auto flash. Setting the exposure correctly is the job of ETTL. What does that say, besides less effort but better result. Same basic setup, drastically different results. ETTL may be great for many things, but for the person that is just taking photos at a party, or having to take PhotoJournalistic style photos, often it just doesn't work very well. Why should I have to go through all kinds of extra effort to attain similar results?

Here is where ETTL shines. You're at a party, and your subject is in the center, walking up to you between people forwards and on the sides. With ETTL, you can get a good exposure of the person walking up, while Auto, will probably underexpose the subject and overlight the people around him. If your focus point is on the subject, ETTL is great.

In studio, I use manual on the camera, several individually adjusted studio flashes, and a separate flashmeter. I wouldn't use an Auto flash, nor would I use ETTL. They are both the wrong tools for the job.

Remember, this whole thread started when someone asked why they were getting better exposures with an Auto flash than with ETTL. The answer has been revelaed in various ways... but it all boils down to knowing how to use ETTL, and going through the gyrations of getting it setup right. And sometimes that's just too much work, when an Auto setting (which we lack) would do the job as well (or better depending on the situation) and easier.
JeffW said:
I'm no pro but isn't your top picture just under exposed...not
necessarily having less exposure range, in general?

How did you set up your ETTL? Do you have it tied to your AF? If
so, what AF sensor was used? As I'm sure you know and have read
from the rest of this post...this makes a huge difference...if AF
is active when you took the picture, then ETTL will bias the active
AF point...and if that is over the left side of your test pic, then
it is understandable the pic is under exposed.

I am curious how the pic would come out of you set CFn4-3 and
CFn13-4 (on a 10d...allows AF to be set independent of
exposure)...focus the shot by pushing and releasing the " "
button...then take the shot. I think the two pics would look the
same since ETTL would evaluate the entire seen as opposed to just
around the active AF sensor.

Anyways...can you post another pic with ETTL and AF not tied
together??
Click to expand...
 
mostly weddings with the same situation of solid black and solid white and have to say ETTL does a better job in these situations. Auto-flash will give you an average of the image based on 18% gray like all meters but depending on the percentage of white or black the results will differ. ETTL lets me identify the specific area I want to set my exposure for and keys on that area, so I can get more of what I want without having to offset the f-stop. Sure you may have to adjust in PS but there is less of a chance of losing detail.
This is all IMHO, for what it is worth.
Ben Strasser said:

Doug Walker said:
Your examples are not compelling. With the extreme amount of
options and flexiblility that Canon's ETTL system offers and the
benefits that result it also requires some intelligence to use. Are
you saying you lack this intelligence or are just incapable of
understanding the benefits of ETTL or are just too lazy to spent
the time necessary to take advantage of the benefits? How in the
world did you ever become a "pro"?
Seems you have an attitude of "my cup is half empty". Maybe a
paradigm shift in thinking about ETTL is in order for you.
Take an aspirin, go to bed and rethink this in the morning. You're
looking pretty silly at the moment.
--
Doug Walker
Check my profile for equipment list.
Click to expand...
Click to expand...
 
You didn't use ETTL at all, you used FEL. Different animal altogether. Different way of shooting, different metering pattern, different everything.

I have no issue with FEL, except that it inconvenient and is difficult to use with action shots or Photojournalistic photos, which is my area of work.

I think that this is a source of confusion... FEL and ETTL are completely separate and different. ETTL is problematic, FEL works great when it's possible to use it. Some people keep mixing them up.
Ben Strasser said:

Doug Walker said:
Your examples are not compelling. With the extreme amount of
options and flexiblility that Canon's ETTL system offers and the
benefits that result it also requires some intelligence to use. Are
you saying you lack this intelligence or are just incapable of
understanding the benefits of ETTL or are just too lazy to spent
the time necessary to take advantage of the benefits? How in the
world did you ever become a "pro"?
Seems you have an attitude of "my cup is half empty". Maybe a
paradigm shift in thinking about ETTL is in order for you.
Take an aspirin, go to bed and rethink this in the morning. You're
looking pretty silly at the moment.
--
Doug Walker
Check my profile for equipment list.
Click to expand...
Click to expand...
 
I've written in this forum of the many differences between ETTL and Autoflash.

It boils down to ETTL being the equivalent of spotmetering your flash exposure and Autoflash mode being the equivalent of centerweighted metering of flash exposure.

Spotmetering flash exposure requires a lot of experience to determine which are in the photograph to meter off of, and how much flash exposure compensation to add or subtract. It is based on a Color Zone System of exposure. This is the primary reason people have a lot of difficulty with ETTL. Many people expect automation in flash exposure. But this is not what ETTL offers.

The things you have to think about for ETTL are myriad including:

1. IS THE LENS ON AUTOFOCUS OR MANUAL? If manual, then the camera will use the autofocus point with the brightest object to meter. It assumes that the brightest object is your subject. It does not necessarily use your selected focus point or the center focus point.

2. WILL YOUR SUBJECT BE PRECISELY UNDER A FOCUS POINT? If not you will have to use FEL in order to meter your subject then recompose your shot.

3. WHICH FOCUS POINT HAVE YOU SELECTED TO METER THE SUBJECT? If you are using FEL, then on only the center focus point is metered. If you don't use FEL, then your selected focus point is what is metered. You have to place your metered subject under this focus point before taking the photo.

4. ARE YOU GOING TO USE FEL? You choose FEL if you can't get your subject precisely under a focus point and you have to recompose.

5. IS YOUR SUBJECT THE COLOR EQUIVALENT OF MIDDLE GRAY (18% grey)? If not, you have to add or subtract flash exposure compensation. If your subject is a color other than gray (e.g. yellow or pink), you will have to convert in your mind that color into the equivalent gray and see if it is darker or lighter than middle gray. This is similar to the Zone system of photography, but in color. This conversion is where the ETTL becomes complex for people to grasp since people don't have experience with a Color Zone system of exposure. For example, a blue sky is a very bright color, but a plant's green leaves are close to middle gray. Is your subject a person who has white or black or brown skin? You have to adjust flash exposure compensation to get the right exposure. You have to rely on a lot of experience to get this right. Or you have to make several bracketed flash exposures.

6. WILL YOUR SUBJECT BE MOVING? Metering a moving subject is difficult because the light values may change depending on the position. You may have to try your luck, because even FEL will be unreliable depending on lighting conditions.

7. WILL YOU HAVE ONLY ONE CHANCE TO TAKE THE PICTURE OR CAN YOU BRACKET FLASH EXPOSURES TO GET IT RIGHT? If you don't have the option of bracketing flash exposure or have a lot of experience with spotmetering flash exposure with ETTL and FEL, you may have problems getting reliable exposures. Examples of this include wedding photography, children at play, and other quickly moving subjects and changing environments. If you have clouds moving on the sky so that the amount of sunlight varies quickly, then you may have exposure problems.

I have four Canon 550EXs and an ST-E2. For doing on-location studio photography with umbrellas and light stands, this is a very nice, highly portable, wireless flash system. If you have time, experience, and understanding of ETTL, you can get well-exposed flash photos with this system. The problems occur when you don't have these factors. ETTL can be unreliable when you don't have time to set your flash exposure settings (FEL, flash exposure compensation) for example. For example, the time it takes to make your measurement, the distracting preflash, the use of FEL and recomposition, makes candid photography much more difficult with ETTL than Autoflash.

Auto-Flash mode is highly useful in instances where ETTL is more difficult to use. I wish that Canon had included Auto-Flash mode in addition to ETTL on it's 550EX flash. Nikon did with it's SB-28, SB-50, SB-80, and SB-800 line of flashheads. It would have made people much happier with Canon's flash system. For once-in-a lifetime photo opportunities, with generally centered subjects (though you can set exposure compensation for off centered subjects also in Auto-flash mode), such as in wedding photography, I resort to using my Nikon SB-28s in Autoflash mode on my Canon 10D to get reliable flash exposures. For casual photography, snapshots, candids, auto-flash mode is a lot easier to use than ETTL.

For wedding photography, to get quick, reliable flash exposures, I prefer using my two Nikon SB-28s on a bracket, each in Autoflash mode, one for bounce, the other for fill, to take photographs. Since Canon did not include autoflash mode in the 550EX, I have to carry both the 550EX and Nikon SB-28 with me to get the best of both worlds.

People may not realize it but Nikon's own TTL Flash system is very similar to ETTL - it fires a preflash to determine flash exposure. The difference is that Nikon also uses distance information from the lens (the "3D" part of its flash system) to determine exposure. Nikon's system combines ETTL with distance information. This gives Nikon a huge advantage because assuming your subject is middle gray, the distance to your subject automatically gives you the flash exposure value. The ETTL-like preflash part gives you information allowing you to set exposure compensation. Unfortunately for Canon, Nikon owns the patent on using distance information from the lens. Until this patent expires, Canon can't use distance information. Patents do expire - for example, Nikon could not use ultrasonic motors on its lenses until Canon's patent on them expired. Nikon could not do image stabilization until Canon's patent on it expired.
 
You still have to do exposure compensation with Auto-flash mode to get a precise exposure. Both ETTL and Autoflash mode require experience to determine how much compensation to add. Since Autoflash is similar to centerweighted metering, the precision for determining exposure compensation, I think, does not have to be as precise as with ETTL - making it easier to determine.
Ben Strasser said:
Ben Strasser said:

Doug Walker said:
Your examples are not compelling. With the extreme amount of
options and flexiblility that Canon's ETTL system offers and the
benefits that result it also requires some intelligence to use. Are
you saying you lack this intelligence or are just incapable of
understanding the benefits of ETTL or are just too lazy to spent
the time necessary to take advantage of the benefits? How in the
world did you ever become a "pro"?
Seems you have an attitude of "my cup is half empty". Maybe a
paradigm shift in thinking about ETTL is in order for you.
Take an aspirin, go to bed and rethink this in the morning. You're
looking pretty silly at the moment.
--
Doug Walker
Check my profile for equipment list.
Click to expand...
Click to expand...
Click to expand...
 
Very well written summary, balancing the advantages and disadvantages of ETTL -vs- FEL -vs- AUTO.

I find it interesting that you have resorted to using a Nikon flash in auto mode for certain situations, due to the lack of a true "Auto" mode on Canon systems. Why do you think Canon didn't include an Auto mode... it just doesn't make any sense to me.

I agree, ETTL, FEL and Auto are all appropriate in different situations. It's amazing to see how some people here are arguing that Auto is not necessary, and that ETTL and FEL + some intelligence are all you need. Well, even intelligence doesn't overcome the lack of an auto mode in some cases, since the process is unusable for some types of photography.

Thanks again for the post.

BTW... people are going to jump on you for your post... let me just mention item (1)... Canon inidicates that the metering is an average of all 35 zones. If you can point to any authoritative info that says that the brightness of a particular item influences or biases the exposure, please provide a link.

Also, be ready for a bunch of cf4-1, Cf4-3, CF13-something, etc settings thatr you should play with to overcome your 'use a Nikon on Auto' solution.

Edward
 
Thanks for your help. I have a much better understanding of how the flash metering works now.
 
Thanks for your insight. After reading this thread as a novice trying to understand how ETTL works, I agree with Unbound that people seen to have different ideas regarding your statement #1.
marianco said:
1. IS THE LENS ON AUTOFOCUS OR MANUAL? If manual, then the camera
will use the autofocus point with the brightest object to meter. It
assumes that the brightest object is your subject.
Click to expand...
Disregarding whether ETTL or Autoflash is better in some undefinable absolute terms, what happens when the lens is set on manual seems to be the primary point of contention.
--
Regards,
Carl
 
The only thing that I've gotten out of this thread is that a new 550 (560?) with an auto mode, as the Nikons have, would make everybody happy. Would settle every debate about this.
Scott
 
Thanks. Very helpful.

I am confused though with respect to 5. I thought the primary purpose of ETTL is to properly expose the subject whether it is 18% gray or not. I understand that other parts of your picture may be under or over exposed if the subject is not 18% grey. However, if you add or subtract exposure for non 18% grey subjects you may get a better overall exposure, but won't the subject then be under or over exposed? Is not the decision one of whether to expose the subject correctly or compensate to get better overall exposure and dynamic range and then use PS on the subject. In some cases "perfect" exposure of the subject may be desired and you may not care about the rest of the picture in which case shouldn't ETTL do the job (assuming the subject is under the selected focus point and you do not recompose) without compensation even if the subject is not 18% grey?

Frank B

marianco wrote:
nd you have to recompose.
marianco said:
5. IS YOUR SUBJECT THE COLOR EQUIVALENT OF MIDDLE GRAY (18% grey)?
If not, you have to add or subtract flash exposure compensation.
If your subject is a color other than gray (e.g. yellow or pink),
you will have to convert in your mind that color into the
equivalent gray and see if it is darker or lighter than middle
gray. This is similar to the Zone system of photography, but in
color. This conversion is where the ETTL becomes complex for people
to grasp since people don't have experience with a Color Zone
system of exposure. For example, a blue sky is a very bright
color, but a plant's green leaves are close to middle gray. Is
your subject a person who has white or black or brown skin? You
have to adjust flash exposure compensation to get the right
exposure. You have to rely on a lot of experience to get this
right. Or you have to make several bracketed flash exposures.
Click to expand...
 
marianco said:
I've written in this forum of the many differences between ETTL and
Autoflash.

It boils down to ETTL being the equivalent of spotmetering
Click to expand...
snip...
marianco said:
Spotmetering flash exposure requires a lot of experience
Click to expand...
snip...

You've hit the nail on the head...

E-TTL does require the ability to spot meter. That means learning to judge reflectance and the ability to pick an 18% point to meter, or the ability to dial in EC.

I learned spot metering, EC, and FEL with a T90 and 300TL...so the move to E-TTL on the 10D was no big deal. Getting 35 properly exposed low iso slides from a 36 exposure roll out of my T90 was a BIG deal...with or without flash.

I just pulled out my old Canon 199A flash, shot a few test images, and it works great, too, as an auto flash...and it's a cheap alternative to the SB-28.

So...those of you who don't like E-TTL, there is a cheap easy low-tech fix.

Go to eBay and buy up all the old Canon 199A flashes.
--
Rob Wierman
http://www.pbase.com/weirdrob/root
 
If you were to take a third shot in this test using a Metz thyristor flash, would you expect the results to look more like the first or second shot?

Frank
Banana Chips™ said:
Banana Chips™ said:
I was shooting in AF and simulating MF by using CF4-1 with the *
button released.

Frank
Click to expand...
yep. i gotcha.

i did a similar test. i did this in a dark room with white wall,
attached the cap to the wall, M exposure...1/200s @ f8 using 50mm,
10D, 550EX

the first one with the center AF point enabled:



and the 2nd one with no AF point enabled (MF):



this is very much consistent with Frank's test results.

later dudes.
Click to expand...
 
Straight from the horse's mouth: This is Canon's statement about Manual Focus and E-TTL Flash (from http://www.cps.canon-europe.com/articles/article.jsp?articleId=6202&pageId=7 )

Manual focusing with E-TTL flash

Evaluative metering is calculated for the subject area covered by the active focusing point. But what if you focus manually, and there is no active point? Or what if you alter the composition of the image after the shutter release button is partially depressed (locking the focus)?

The camera knows if you have switched to manual focus. But it can also tell if you move the camera after the focus has been locked by partially pressing the shutter button - the meter readings taken as the button is being pressed, and just before the button is fully depressed will be different.

In these cases, the camera cancels the focus point linkage to the flash exposure control. It then moves the metering zone to the area which received the most light during the preflash on the assumption that this is where the main subject will be.

---------------

Another place people have problems understanding flash photography is that flash photography is essentially taking a DOUBLE EXPOSURE:

1. THE AMBIENT LIGHT EXPOSURE: The Canon 10D uses its 35 light metering zones to determine the light falling on the subject and its environment. It then sets the camera's exposure based on the metered information (aperture and/or shutter speed - unless you are using manual exposure).

2. THE FLASH EXPOSURE: The Canon 10D uses only one of the 7 light metering zones under the focus points to determine exposure. It uses only the center light metering zone if you use FEL. This determines the duration of the light from the flash when you press the shutter button. It is called "Fill" when the flash duration is reduced (flash exposure compensation) by the photographer from the original duration set by the camera.
unbound said:
Very well written summary, balancing the advantages and
disadvantages of ETTL -vs- FEL -vs- AUTO.

I find it interesting that you have resorted to using a Nikon flash
in auto mode for certain situations, due to the lack of a true
"Auto" mode on Canon systems. Why do you think Canon didn't include
an Auto mode... it just doesn't make any sense to me.

I agree, ETTL, FEL and Auto are all appropriate in different
situations. It's amazing to see how some people here are arguing
that Auto is not necessary, and that ETTL and FEL + some
intelligence are all you need. Well, even intelligence doesn't
overcome the lack of an auto mode in some cases, since the process
is unusable for some types of photography.

Thanks again for the post.

BTW... people are going to jump on you for your post... let me just
mention item (1)... Canon inidicates that the metering is an
average of all 35 zones. If you can point to any authoritative info
that says that the brightness of a particular item influences or
biases the exposure, please provide a link.

Also, be ready for a bunch of cf4-1, Cf4-3, CF13-something, etc
settings thatr you should play with to overcome your 'use a Nikon
on Auto' solution.

Edward
Click to expand...
 
I use Nikon SB-28s for Autoflash mode because:

1. They are safe to use with the Canon flash system (they have a trigger voltage less than or equal to 6 volts )

2. I had them from my Nikon system already. I didn't have to purchase new flash heads.
3. They are well-built - better than Sunpak, and other less expensive models.
4. They have surprisingly good power for their small size.

5. You can attach a power pack to the flash for extended use and to reduce flash recycle time.
6. I can use them on Nikon cameras.

7. Autoflash mode on Nikon Flashheads is really good. Many Nikon fans prefer it to the 3D TTL flash mode that Nikon touts.
marianco said:
Manual focusing with E-TTL flash
Evaluative metering is calculated for the subject area covered by
the active focusing point. But what if you focus manually, and
there is no active point? Or what if you alter the composition of
the image after the shutter release button is partially depressed
(locking the focus)?
The camera knows if you have switched to manual focus. But it can
also tell if you move the camera after the focus has been locked by
partially pressing the shutter button - the meter readings taken as
the button is being pressed, and just before the button is fully
depressed will be different.
In these cases, the camera cancels the focus point linkage to the
flash exposure control. It then moves the metering zone to the area
which received the most light during the preflash on the assumption
that this is where the main subject will be.

---------------

Another place people have problems understanding flash photography
is that flash photography is essentially taking a DOUBLE EXPOSURE:
1. THE AMBIENT LIGHT EXPOSURE: The Canon 10D uses its 35 light
metering zones to determine the light falling on the subject and
its environment. It then sets the camera's exposure based on the
metered information (aperture and/or shutter speed - unless you are
using manual exposure).
2. THE FLASH EXPOSURE: The Canon 10D uses only one of the 7 light
metering zones under the focus points to determine exposure. It
uses only the center light metering zone if you use FEL. This
determines the duration of the light from the flash when you press
the shutter button. It is called "Fill" when the flash duration is
reduced (flash exposure compensation) by the photographer from the
original duration set by the camera.
unbound said:
Very well written summary, balancing the advantages and
disadvantages of ETTL -vs- FEL -vs- AUTO.

I find it interesting that you have resorted to using a Nikon flash
in auto mode for certain situations, due to the lack of a true
"Auto" mode on Canon systems. Why do you think Canon didn't include
an Auto mode... it just doesn't make any sense to me.

I agree, ETTL, FEL and Auto are all appropriate in different
situations. It's amazing to see how some people here are arguing
that Auto is not necessary, and that ETTL and FEL + some
intelligence are all you need. Well, even intelligence doesn't
overcome the lack of an auto mode in some cases, since the process
is unusable for some types of photography.

Thanks again for the post.

BTW... people are going to jump on you for your post... let me just
mention item (1)... Canon inidicates that the metering is an
average of all 35 zones. If you can point to any authoritative info
that says that the brightness of a particular item influences or
biases the exposure, please provide a link.

Also, be ready for a bunch of cf4-1, Cf4-3, CF13-something, etc
settings thatr you should play with to overcome your 'use a Nikon
on Auto' solution.

Edward
Click to expand...
Click to expand...
 
From Canon itself: http://www.cps.canon-europe.com/articles/article.jsp?articleId=6202&pageId=7

Manual focusing with E-TTL flash

Evaluative metering is calculated for the subject area covered by the active focusing point. But what if you focus manually, and there is no active point? Or what if you alter the composition of the image after the shutter release button is partially depressed (locking the focus)?

The camera knows if you have switched to manual focus. But it can also tell if you move the camera after the focus has been locked by partially pressing the shutter button - the meter readings taken as the button is being pressed, and just before the button is fully depressed will be different.

In these cases, the camera cancels the focus point linkage to the flash exposure control. It then moves the metering zone to the area which received the most light during the preflash on the assumption that this is where the main subject will be.
CBenson said:
marianco said:
1. IS THE LENS ON AUTOFOCUS OR MANUAL? If manual, then the camera
will use the autofocus point with the brightest object to meter. It
assumes that the brightest object is your subject.
Click to expand...
Disregarding whether ETTL or Autoflash is better in some
undefinable absolute terms, what happens when the lens is set on
manual seems to be the primary point of contention.
--
Regards,
Carl
Click to expand...
 
Thanks for this reminder. I had read this before. But then someone pointing to the following confused me.

http://www.robgalbraith.com/ubbthreads/showflat.php?Cat=&Board=UBB8&Number=138289&Forum=f8&Words=35%20zones%20receive%20equal%20weighting&Searchpage=0&Limit=250&Main=137317&Search=true&where=bodysub&Name=210&daterange=0&newerval=&newertype=&olderval=1&oldertype=m&bodyprev=#Post138289

The link states:

When the lens is set to manual focus, all 35 zones receive equal weighting for flash metering in E-TTL.

Am I misunderstanding or are these two sources contradictory.

--
Regards,
Carl
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top