Showing results for 
Show  only  | Search instead for 
Did you mean: 

70D bounce flash


Recently I upgraded my 650D to the 70D and I noticed that the pictures are a lot darker (underexposed) when bouncing the flash. I have tested this with several other cameras, the 600D, 650D and the 700D using the 430EXII and 580EX all with the 18-135 STM. All but the 70D resulted in acceptable exposure.

Is there a setting I am missing? I used both the [A+] and P setting.


@Tim wrote:

Hello socal13,

There have been several answers, tests, suggestions and the like represented in this thread.  Even you yourself seem to have found a work around that maybe works for your needs.  That being said, this is not a known issue within Canon regarding the EOS 70D. 

If you feel as though your camera would benefit from the expertise of a service technician, it's urged that you set up a service evaluation with Canon.  That can be done through our web site, linked here.  Often times issues can only be truly understood under factory conditions, as there we could ascertain the true cause of this issue, if there is a malfunction with your particular camera.  

Obviously it depends on your definition for "known issue within Canon" - I have an open support ticket with Canon Canada and I was informed that my ticket was discussed with Canon USA. In my ticket I did refer to this thread, which is on Canon USA web site. I spend several hours just talking to Canon support. I spend hours writing about it on this forum. You, TIM, seem to be working for Canon USA yourself.


What should we do in order to make this issue "known within Canon"?


I cannot send you the 70D body since I already returned both of them to the store. May be someone else will. I have personally reproduced the issue with 4 different 70D bodies in Toronto - 2 that I bought and returned to FutureShop, now I am just testing them in the store. Out of all experiments that I did only one, at the Exposure Expo organized by Henry's, I got something that looked like a good exposure, using the camera body and lens that the Canon personal provided (it worked with all flashes, including mine, but did not work with my lens). I have checked and the firmware of the camera is the same for all cameras that I tested. But it was in the Henry's store a couple of days ago that the store guy did the testing and he was able to reproduce my findings.


I don't believe you need to see the actual body, since the issue happens with any 70D body that I can get my hands on. And since people as seening the same  behaviour with 5D Mark III it seems clear that this is the new normal, it is not a defect.


By the way, my Nikon buddies, who were laughing at me for all the troubles I went through the last months with this issue, confirmed (I saw the pictures) that Bounce flash still works as a main/key light in D600. That is the solution for me.




Hello Tim,


Like the others who replied on your post, I don't understand why this is not a known issue.


I really enjoyed my previous 600D en 650D, thinking that the 70D would be even better with the dual pixel technology. As some things did improve, the first pictures I took indoors I was thinking "I must be doing something wrong". But when I used my brothers 700D taking the same pictures in the same condition I almost returned the camera. After reading more about this issue, I knew this must be a software bug. And as I did not have my 650D anymore I did not returned the camera.


As I take most of my pictures indoors, I never had to think about flash compensation or FEL with my 600D/650D. Now I have to compensate up to 3 stops and the worst part is a lot more pictures need post correcting hurting the quality.


And again, I don't understand why this is not a known issue as this is really easy to reproduce comparing it with a 700D and the 18-135 STM lens.Until this issue is resolved, I cannot recommend the 70D to anyone.


I just purchased a 700D (T5i) and the bounch flash works fine, including with wireless TTL triggering.

It is true that the camera feels a lot cheaper than 70D, but I paid less than half of the price of 70D after all.


In all fairness 700D has a bit more noise at high ISO (6400), but I will not need to go there at all - the TTL flash gives me more than enough light at ISO 400. Also 70D AUTO ISO goes to 1600 once you move the flash head from the direct position, while 700D stays at 400 and still has enough light for proper exposure.


Until Canon fixes the TTL in 70D I would recommend 700D for anyone who does not need the extra 10 focus points. With the rest of the money you can buy another lens or just go on vacation.'s pretty sad when a camera half the price works better than the higher end camera.


I don't know why its so freaking hard to release a new firmware update...I mean I fully understand that they need to do a lot of testing on any new firmware updates...but come on.  It'll soon be a year since release of the camera.  Just fix a few of the bugs and push out a firmware already.


The 650D/700D is the better buy. Although it has less features then the 70D, it is has them better implemented


Some of the features of the 70D are not implemented quite user friendly. Like HDR and multi exposure, why are the not in the Scene [SCN) menu? Easer to access and no need to digging into the menu.


There is also another bug I haven't seen anyone mentioned before. With the 18-135 STM, when I zoom to 135 and focus into a object close by, then try to focus something in the distance, it refuses to focus. I have to zoom out first, focus, zoom back to 135 and refocus before it works. I haven't seen this on the 650D. Hope this gets fixed too in a firmware release.



I would like to add that the focus problem I mentioned does not occur in Live view.

Tim:  Can you confirm whether Canon is currently working on a firmware update to address this problem with the 70D?  Do they need more data from users related to this issue?  The problem is actually pretty easy to document, even without a laboratory.  First you reset the 70D and a Speedlite 430EXII to factory settings.  Then in a uniformly dark room, set the camera to full auto or P and the 430EXII to ETTL and take a picture while bouncing the flash off the ceiling.  The result is a significantly underexposed picture.  Then keep all the settings the same and take another shot after depressing the FEL button.  The result is a correctly exposed picture.  The evidence suggests that if you don't select FEL, the camera does not calculate the correct power level for the primary flash (you can actually tell with your peripheral vision that it's a weak flash--much weaker than the flash after selecting FEL).  I've run the same experiment using my XSi body with the same lens and flash and you don't need to select FEL to get a correct exposure.  I'm new to the process, so I'd appreciate any advice you can give for getting this issue resolved.  Since many other 70D users have reported the same problem, it seems to be a 70D issue rather than a malfunction in any individual camera.  It's fortunate that there's a workaround, but Canon should fix the problem so we don't have to rely on a workaround.  Thanks! 

I've gone round and round with Canon on this problem.  It was escalated to their engineering group.  I sent many examples of the problem with simple instructions on how to duplicate.  They did duplicate it, and responded, basically, that this is the way the camera operates.  They never really addressed the actual issue of underexposure without FEL.


Their response: "We were able to reproduce a similar result with a little testing.  The cause of the difference you are seeing is that the metering used with FEL is essentially spot metering.  If you are zoomed into an area with a lot of light colors, the image may be underexposed because the light colors are "tricking" the metering system.  Using FEL essentially allows you to pinpoint where the flash exposure metering will happen, and if you're using FEL over a mid tone region such as a person's face, it will produce a correct result, even if everything around them is bright white.  I would recommend using FEL if at all possible in tricky shooting situations. Please let us know if we can be of any further assistance with your EOS 70D. Thank you for choosing Canon."


This response doesn't make any sense to me, in that the samples I sent were of a blank sheet of light brown cardboard.  There were no "light colors" to :"fool" the metering system.  On a uniform piece of cardboard, spot metering should NOT produce a different result than evaluative metering.  They did not answer why the standard Evaluative metering was not being done correctly and was always underexposed.   I, too, believe that the 70D flash system does not expose correctly when using bounce flash or any flash modifier.  My 40D did not have this issue.  They have also suggested that because I wear glasses, light it entering the viewfinder and messing up the exposure.  This is not the problem as I have totally blocked the viewfinder, also testing in LV mode, and the same problem exists.


On a brighter side (pun intended), I did find that this problem with the 70D can be mitigated somewhat by using "Average" rather than "Evaluative" for the ETTL flash metering (in Flash control menu, pg 195 of 580exii manual).  I'm pretty much able to use a flash modifier or bounce flash with my 580exii, when using ETTL Average, without having to do an FEL first.  This is only a bandaid, and the downside to this is that if you may occassionally overexpose your subject if the rest of the scene is very dark, as the flash tries to make the total image average out.

I, too, wish that Canon would correct their problem with the 70D grossly underexposing with modified flash.

Thanks for the info--and for trying to fix the problem with Canon.  It's unfortunate that they're not willing to fix the underlying issue.  I agree that they're obfuscating the issue when they say it has something to do with FEL's use of spot metering.  As you point out, you get the same (incorrect) result even in a room of uniform darkness.  So it's not that the meter is being fooled.  Evidently the algorithm the camera uses to calculate the primary flash power in FEL mode is different from what it uses when you're not using FEL. 


Thanks for the tip about using Average vs Evaluative mode.  I'll give that a shot.  I'll also chime in with Canon.  If they get enough feedback from unhappy customers, they'll be more inclinded to provide a fix.


Other than this issue, the 70D's a wonderful camera! 


Thanks for the tip about using Average vs Evaluative mode.  I'll give that a shot.  I'll also chime in with Canon.  If they get enough feedback from unhappy customers, they'll be more inclinded to provide a fix.


Other than this issue, the 70D's a wonderful camera! 

I like the 70D as well, but it has two significant drawbacks, and they interfere with each other.  First is this flash underexposure issue when using a modifier/bounce, that can be mitigated somewhat by using FEL.  The second is erratic focus with the viewfinder on 2.8 or faster lenses (many of us have experienced this issue, MANY forum threads about it) that can also be mitigated, by using LiveView.  Unfortunately, you can't use FEL when in LiveView.  So... if you want consisten focus and correct flash exposure, these two workarounds cannot be paired together.  You get one or the other.


A fix for this ETTL Evaluative flash issue would make this camera reliable, especially if photographing an evening event.  As is, it is not.  It requires lots of trial and error to get a correct flash exposure, hardly suitable for a fast moving event.


DOES ANYONE FROM CANON EVEN READ THEIR OWN FORUMS?  Your engineering group has already recreated the flash problem yet has not provided a viable explanation or a fix.  How about some answers?