Flash diffuser and flash power

modchild

Suspended / Banned
Messages
3,914
Name
Stuart or just Stoo!
Edit My Images
Yes
I've got a party coming up and there's not enough room to set up my studio lights etc so I'm going to have to use a basic Stofen type diffuser with a Canon 580EX II and a 5D3.

Now this is the first time, and probably only time, I'll be using these diffusers and I've got a White, Yellow and Blue set. What I really need to know is if I use it bounced or direct at the subject and if I'll need to increase some flash EV to compensate for the light loss. I've got some decent fast lenses (f1.4 prime and f2.8 zoom) and I'll be shooting in AV to control DOF for portraits or group shots. I have got time for some practice but any help on where to start would be helpful.

The last time I used the flash was at a gig where the lighting was almost non-existant so I used direct flash but ended up with highlights etc from the sweating musicians and would like to avoid that again. Some of the gig shots have been used by some of the groups for promo work so they must of liked them but I'd like to get better.

Any constructive help would be very gratefully received.
 
Thanks for your suggestion but I tried in the living room at home in TTL and it seemed to come out too dark whether I bounced it or had it direct. I wont be able to bounce it at the party as the room is too high and not white which is why I got the diffusers. I'll have another play around this afternoon to see if I can nail it.
 
Thanks for your suggestion but I tried in the living room at home in TTL and it seemed to come out too dark whether I bounced it or had it direct. I wont be able to bounce it at the party as the room is too high and not white which is why I got the diffusers. I'll have another play around this afternoon to see if I can nail it.

Then you wasted your money. Stofen-type diffusers need a white ceiling and light toned surroundings to work. With out that, all they do is waste light by blasting it everywhere with no hope of any bouncing back to soften the shadows. And since they don't make the light source itself any larger, there's no softening effect from the gun. All they do, with the gun pointing upwards, is raise the light source a little and that can be slightly helpful at close range.

TTL should take care of the exposure, though fitting a Stofen substantially reduces brightness. Be sure to have the head tilted, which you want to do anyway, but it also disables the camera-to-subject distance information that usually takes priority with direct flash. In TTL, the pre-flash gets skewed when using a Stofen, because the camera doesn't know it's there.
 
Thanks for that, as I said I've not tried it before so maybe it's not the best idea. In that case would I be better off just getting a mini softbox that attaches to the flashgun and work that manually and practice till I get the exposure right. I had a bit of a play around with the diffusers today and got some decent shots of the kids but it was in much brighter ambient light so I'll have another play around later.
 
... I had a bit of a play around with the diffusers today and got some decent shots of the kids but it was in much brighter ambient light so I'll have another play around later.

As Richard said, the key issue with the diffusers is the surroundings. Stick a fong dong on a flash in a very small space and you get a lovely soft light. Try it in a large room away from the walls and you get something close to a bare flash.
 
Thanks for that, as I said I've not tried it before so maybe it's not the best idea. In that case would I be better off just getting a mini softbox that attaches to the flashgun and work that manually and practice till I get the exposure right. I had a bit of a play around with the diffusers today and got some decent shots of the kids but it was in much brighter ambient light so I'll have another play around later.

Since there is no bounce surface to work with and make the light softer (by making the light source area, eg ceiling, much bigger) then you have to provide your own.

Bigger is better, and softer, but with on-camera accessories you quickly hit the physical limit of what's practical. Lumiquest makes some good on-camera accessories, but there are plenty of others.

What kind of party is it? If it's not a dark disco type affair, the usual way of softening the light is to reduce the flash power and drop the shutter speed (raise ISO/low f/number) to pull up the ambient as a bit of fill-in. Then gel the flash to balance colour. You're looking for the best compromise in difficult conditions.
 
Thanks for the help guys, it is a dark disco type party so I'll have a look through the Lumiquest link and see if I can find something there. It's not super critical as it's for a friends birthday, but I would like to do the best I can for her.
 
Thanks for your suggestion but I tried in the living room at home in TTL and it seemed to come out too dark whether I bounced it or had it direct. I wont be able to bounce it at the party as the room is too high and not white which is why I got the diffusers. I'll have another play around this afternoon to see if I can nail it.

Also a thought if they came out dark do you have Flash Exposure compensation (FEC) activated?

I'd expect TTL on a 5D mkIII to be pretty good :bonk:
 
Also a thought if they came out dark do you have Flash Exposure compensation (FEC) activated?

That was the point, whether there's a rule of thumb that you use x amount of FEC for a certain type of diffuser or softbox etc. Having only used bounced, direct flash or off camera speedlites before, or my studio lights, I just wondered if there was a standard sort of setup for diffusers. I think I'll just use a speedlite softbox, TTL and FEC and check the histagram regularly. I'm going to take my 70D and Nissin Di866 MkIII as well, just to give it another test.


Thanks again to all who contributed, specially the links for Lumiquest and the gels, will have a good look through them. :thumbs:
 
That was the point, whether there's a rule of thumb that you use x amount of FEC for a certain type of diffuser or softbox etc. Having only used bounced, direct flash or off camera speedlites before, or my studio lights, I just wondered if there was a standard sort of setup for diffusers. I think I'll just use a speedlite softbox, TTL and FEC and check the histagram regularly. I'm going to take my 70D and Nissin Di866 MkIII as well, just to give it another test.


Thanks again to all who contributed, specially the links for Lumiquest and the gels, will have a good look through them. :thumbs:

Using auto-TTL, you shouldn't need any flash compensation - in theory. The pre-flash is fired at a known output, the light coming back is measured regardless of the situation, and the main flash exposure is calculated from that.

As mentioned above, there is one situation that can fox it, and that's when the gun is fired direct with the head in the straight ahead position. Here the camera knows that subject distance is a very accurate guide to exposure and it then takes lens focusing distance as an overriding consideration, rather than what comes back from the pre-flash. If you have a Stofen fitted that the camera doesn't know about, that will upset it.

Having said that, there is no official info on how Canon's flash algorithms work, and there is sufficient evidence to suggest that it changes with different/new cameras. And it's clouded by the fact that flash exposure calculation is far more complicated than normal ambient metering, and additionally it often has to take ambient exposure in to account too. Lots of variables and room for error.

What I do is make sure that distance data can't mess things up by tilting the head. This disables it (the flash range info on the back of Canon guns will switch off) and the camera is forced to use pre-flash feedback only. I also use centre-weighted metering for flash, that seems to be less volatile than evaluative. And chimp a lot :D
 
Just been trying out different methods and attachments and have come to the conclusion that it's going to be a mini softbox and check the EV with a bit of chimping to check the histogram a fair bit. I wont know the actual room conditions till I get there so I've practiced in the house in a good few different lighting setups.

Thanks Richard for your post above, it looks like tilting the head even slightly makes a big difference in keeping the exposure accurate by using preflash. I've also come to the same conclusion with centre-weighted metering, it seems to give a better light to the main subject. The ladies are all in 'ball gowns' so it looks like there'll be a lot of central subjects to get all the dress in too.
 
Just been trying out different methods and attachments and have come to the conclusion that it's going to be a mini softbox and check the EV with a bit of chimping to check the histogram a fair bit. I wont know the actual room conditions till I get there so I've practiced in the house in a good few different lighting setups.

Thanks Richard for your post above, it looks like tilting the head even slightly makes a big difference in keeping the exposure accurate by using preflash. I've also come to the same conclusion with centre-weighted metering, it seems to give a better light to the main subject. The ladies are all in 'ball gowns' so it looks like there'll be a lot of central subjects to get all the dress in too.

Tilting or swivelling disables the distance info, and on my Canon 580 guns, you only have to press the head release button to disable it, even without moving the head at all.
 
Back
Top