|
i thought rim lights come from the back?
Posted 54 months ago.
(permalink)
|
|
angle of incidence is one thing to always consider, but if you're using a hot shoe/canon ex/sb what-have-you, as i imagine you are being a strobist reader, it's probably to do with the zoom. adjust and take tests set to different lengths... 24mm-105mm if it's a speedlite. i usually keep mine set between 50mm and 70mm unless i'm trying to light up an entire room.
Posted 54 months ago.
(permalink)
|
|
I'd say they're A) too far forward, B) too low and C) on too high of a setting and D) too close to his face. I'd first get the camera and flashes set to manual, then work on balancing them.
Posted 54 months ago.
(permalink)
|
|
I think this one is called a "kicker" and I think I read somewhere it is usually around half the brightness of the main light.
Edit: Found it in LSM.
Originally posted 54 months ago.
(permalink)
~ Aidan Dunbar edited this topic 54 months ago.
|
|
how and where are you metering them from?
Posted 54 months ago.
(permalink)
|
|
How did you meter main light and the kicker? Where did you point the meter?
I guess the specular reflection from skin is much stronger at an angle close to the surface.
Posted 54 months ago.
(permalink)
|
|
To me, the kicker seems to be more powerful than the main light.
Increase distance to kicker light or turn down power.
If you increase distance by a factor of 1.4, then the power is halved.
greetings,
Posted 54 months ago.
(permalink)
|
|
... and did you have the dome down when you metered the kicker?
I don't have a clue about using a light meter, but as I ordered one myself I don't let a lack of knowledge get into the way of spreading some confusion ;-) :
(1) I would assume that you set the light meter to incident metering. Then you put the dome down and hold the light meter to camera left of the subject facing the kicker light and measure, say, f/5.6.
(2) I also believe that the hair/rim/kicker should be at the same level as the key light.
(3) Hence I would hold the light meter in front of the subject facing the key light, which seems to be on camera right high, and adjust the power of the light until it will also meter at f/5.6.
(4) Then I would put the dome up again and set the light meter to reflective metering. I would put the light meter in front of the lens and face the subject. The metered aperture should then be set on your camera.
(5) With the dome up and incident metering mode again it might also make sense to put the light meter directly in front of the subject facing the camera and meter there. This should yield the same result as (4), right?
(6) Maybe it makes more sense to exchange key and kicker light in my rant, but that doesn't make a difference for our example.
Here is a link to a very nice introduction to using a light meter. I don't remember from whom I got it, but here on the forums, probably from you? super.nova.org/DPR/Ratios/
Btw. which light meter do you use?
Posted 54 months ago.
(permalink)
|
|
|
from the look of your images, I doubt they would both meter the same.
are using the flashes on manual?
for a quick check of the amount of light hitting your subject, meter each light source separately at the subject's head's location, with the meter facing each light source.
Originally posted 54 months ago.
(permalink)
targophoto.com edited this topic 54 months ago.
|
|
My question for Till is, your key light is diffused, is your your "rim", "kicker" light diffused?
Looks like your rim light is direct with out any diffusion, this will cause it to look more blown out.
I would try to soften the rim light some, I like it one stop over my key light but thats a matter of taste.
Posted 54 months ago.
(permalink)
|
|
Till, yes, it is due to the angle of incidence coupled with the hard quality of light. It is specular reflection.
I'll bet, by the way you phrased your question, that you already know of several ways to mitigate, alter or eliminate this effect.
Posted 54 months ago.
(permalink)
|
|
you can test it by taking a shot from the position of that flash... if it is still blowing out you know it's because the flash is set too high/too close etc... if it gives you the right exposure from there it's the specular
Posted 54 months ago.
(permalink)
|
|
|
Guys, please excuse my dumb question, but how can this be a specular highlight? Looking at the neck the light is coming from almost 90° (camera left), probably 100-120°, right? My limited understanding would be that the camera to subject angle needs to be about the same as the light to subject angle to produce a specular reflection.
And on top of that we are talking about light skin. Wouldn't this reduce the potential of a specular highlight by a great deal?
I think there is a lot left to learn for me?! (I guessed that much anyway ;-)
Posted 54 months ago.
(permalink)
|
|
|
it all depends on the skin composition and don't forget it's a curved surface...
take a photo of a metal ball or some shiny jewellry... the "family of angles" (had to get that in there somewhere lol) increases because of the curvature
the light isn't coming from 90 degrees though otherwise it would like the whole of the shoulder and and the whole side of the face including the nose :)
Originally posted 54 months ago.
(permalink)
GarethDix edited this topic 54 months ago.
|
|
Yes it is specular. Due to hard light/small source and skin oils and surface texture.
Posted 54 months ago.
(permalink)
|
|
Mmm interesting:
I sense a lot of interesting conceptual stuff being touched upon here:
Three areas that this makes me want to comment on:
(1) ETTL control
(2) Spectral reflections
and (3) limitations on the flashes ability to throttle back its power
so in turn:
(1) ETTL: From the way you refer to the "power" setting on the speedlite: are you using canon ETTL here?
I take your point (and like your style) that you've checked what you've got with a light meter; to get some kind of objective calibration to the set up; but presumably there is also a certain issue here of how the camera is deciding to tell the flash what to do:
This will presumably partly depend on what group the flash is put into (assuming you are working with Canon ETTL here): If the rim light source is in group "C" then the poor camera control will get mighty confused anyway given group C is now putting illumination on the foreground; so notions of +/- 1EV become rather less indicative of what's going on; the camera can't decide what's gong on in the foreground/ background: leading to somewhat random results irrespective of your setting of EV compensation for the group…. The rim/ kicker light must be set in groups A or B to work using ETTL.
(2) Spectral reflections:
All that computerised electronic control logic aside: the issues of spectral reflections always appearing the same brightness irrespective of the distance of the light source to the illuminated object is worth considering here as a previous poster touches on. (Light science and Magic 3rd edition, discusses this on p38; as I was coincidentally reading last night) and simply changing in size as you move the light source nearer/ further away doesn’t control the reflections apparent brightness. Paradoxically, the inverse square law is satisfied by the size of the reflection changing; so the point brightness can stay the same for all points within the reflection, and the inverse power with distance fall off being satisfied as the area of the reflection changes: but it looking like the brightness of the reflection doesn't change. This is a very interesting point; it had never occurred to me this could well occur rather than only with simple polished metal surfaces, to a lesser degree with dark objects (e.g. afro caribean skin perhaps) but also with caucasian skin (especially if damp of course it would be more likely). As opposed to Despite knowing that all reflected light contains "spectral" as well as diffused reflections, I hadn’t ever thought to consider you could have a geometrical set up; which with a surface one thinks of as producing “only diffuse” reflections, could nonetheless in this special circumstance be producing mostly spectral reflections:despite that normally we think of the ratio of the two depending on the nature of the surface: and for light coloured surfaces, diffuse reflections dominate... but clearly not totally...
Presumably if the light source were softened with diffusion (of any kind; umbrella; soft box; bounced off wall etc), then what happens is the ratio of spectral to diffused reflections changes; and hence in a somewhat round about way; the brightness in this situation becomes more controllable.... as once diffused much more of the light will be hitting the surface outside the family of angles; which given this surface is not dark; or mirrored; or metallic; will actually be bounced back as diffuse reflections: and hence exhibit inverse square fall off characteristics.
Is it just that the way the rim light is hitting the neck; you're bang "in the zone" sorry.. lol; family of angles; to maximise the spectral reflection content; and the narrow nature of the neck means you don't have any surface that would only be giving off diffuse reflections anyway; hence the neck is behaving like a "direct reflection only" type surface...?
I'm not too sure what point I'm making here; but would be interested if anyone has observations on whether I'm barking up completely the wrong tree or not still trying to wrap my head round this stuff.
(3) Lastly: could it be that the ETTL of your camera is trying to dial down the power output of the flash; but simply has got to the bottom of the control range of the flash; so can’t dial it down any further? I have ringing in my ears here; DH comments about using laser printer paper in front of his strobes when doing macro work; for its ability to absorb light; and hence act as “neutral density” to drop the power output of the flash down; beyond where the manual power setting allows one to go, as otherwise they over expose in an uncontrollable way. Could this be a problem on this kind of picture? Esp. if an older speedlite; which had fewer power settings ? (something tells me NO! Don’t be silly Jonathan: clearly speedlites must be designed without such a silly limitation when working in this extremely common set up). But perhaps if you had the camera set up with the lens wide open (for super limited depth of field) and ISO sensitivity up really high? (though doesn’t look like it in this picture) Does anyone know; can you hit this situation without realising it in real world portrait sessions?
Sorry to ramble, but thoughts anyone?
BTW: Happy new Year
Jonathan
Posted 54 months ago.
(permalink)
|
|
Hi guys, thanks for your answers!! I love this forum... I would have replied earlier but friends came over and now it's quite late.
What have I learned:
1) It's called "kicker", not rim light (will look that up in LSM tomorrow)
2) Kickers are usually set to one stop below mainlight - that's what I would have done seeing the above results
3) Some of you agree that this is connected to direct reflection. I'll have a look through my filter sample pack tomorrow and see if I can find a polarizing sheet to play with (thanks StroboSapien, you got me thinking there...)
As for your questions concerning the method of measurement: Take a look at this nicely illustrated article pointed out by Mariano above (BTW you were right Mariano, I posted this link here a few days ago). In short: Two measurements with globe down facing the lights and after that a measurement with globe up pointing to the camera. All measurements from head-position with a Sekonic L-358.
Setup
Manual mode for flashes and camera - no TTL of any sort
Mainlight: Bare SB-26 from above camera right @70mm
Kicker: Bare snooted SB-24 from behind model camera left @85mm
I did not note the power settings (they were in the 1/8 to 1/32 area) this time because the f-stop really says it all - but when I think back half a year I wouldn't have understood that and may have profited from that info... so next time I'll do better. Same for light positions...
Originally posted 54 months ago.
(permalink)
Till_Santiago edited this topic 54 months ago.
|
|
Kicker/rim/white line are all acceptable terms. Or would you like to go over snoot/cardboard box? Flag/gobo?
LSM is not the Websters dictionary of photography.
Posted 54 months ago.
(permalink)
|
|
This is a great question. I am going to do a full post about it.
Posted 54 months ago.
(permalink)
|
|
|
Great!! Looking forward to that one!
Posted 54 months ago.
(permalink)
|
|
|
Thanks for featuring this on the strobist blog!! I'll continue with different sizes of rim lights and let you know the results!
You're right, I mixed up those two last readings, they have to be reversed. Two lights only btw, the background light is spill from the main light.
I'm leaving for a short holiday in 5 minutes, people getting nervous around here ;-) otherwise I'd certainly stick around for comments!
Posted 54 months ago.
(permalink)
|
|
It's specular and is dependent mostly on the oil content of the skin.
Seriously, take the same shot, then get your wife of girlfriend to put some powder make up or cover up on the right side of your face and take the exact same shot again. The difference will be obvious.
Kind regards
DWBell
Posted 54 months ago.
(permalink)
|
|
|
The blog post is here.
Posted 54 months ago.
(permalink)
|
|
|
Till, please do as dwbell advised. I'd love to see you wear makeup ;-)
Posted 54 months ago.
(permalink)
|
|
And a kilt!
Posted 54 months ago.
(permalink)
|
|
Dean Collins solved this one for me years ago. The specular value of the light does not change when you move the light further from or closer to the subject. Remember, you are looking a reflection of the light source. If you lower the power of the light and move it closer, the specular stays the same but the incident value, sometimes called the diffused highlight will increase and move closer to the specular. I know this is counterintuitive but it works. You can get the same results by making the light larger with a light box. I like a strip light for this.
Posted 54 months ago.
(permalink)
|
|
liefr1943 is correct. I spent some time working with Dean and moving that rim light back would do nothing to reduce that blowout. A modifier such as an umbrella or bouncing it from a large surface would soften it up. You would have to reduce it by quite a bit or move it more to the side to eliminate the harshness.
Posted 54 months ago.
(permalink)
|
|
|
That's the secret world of the MUA, together with a good tog they make the subject look like million dollars.
Of course if you don't want to dress up like a girl, you could scale the whole set up down and use an 8 ball, then spray the 8 ball with matt spray - same effect, same result.
Kind regards
DWBell
PS - I'm holding out for Till in make up non the less! ;-)
Posted 54 months ago.
(permalink)
|
|
|
Yeah, well... makeup... we could try something like that at our Frankfurt Meetup. Preferably not on my face? And what's an 8 ball?
In the meantime I continued the experiments with a softbox as kicker. See for yourselves:
Posted 54 months ago.
(permalink)
|
|
I haven't seen it addressed anywhere above or on the blog, but I think how you meter is important. The key and fill lights are the only lights that should be metered with the meter facing the lens. For the kicker/rim lights I always point the meter directly at the light for a reading.
Clearly the light level of the kicker is higher than they key or it wouldn't be blowing out. If your meter is facing your camera but the light is striking the meter from the side, only a small portion of the dome is receiving light and will therefore give you an inaccurate reading. When metering your key or fill lights, the meter is facing the lens because the lights are essentially facing the meter. You are getting a reading of the direct light. To be able to compare this and create a ratio between key and kicker, the kicker must be metered the same way, with a direct light reading.
If you aren't using a meter, just shoot each light individually and see how and where they register on your histogram.
If you are using ETTL, there is no way to honestly know precisely what your ratios are because although you can adjust the "ratio" between the lights the lights are dealing with different physical variables, being the distance from the subject and the use of modifiers on one light and not on the other. If you're using your umbrella or softbox on your key light and using a bare flash on your second group, it will meter for you key light and then add whatever amount of light you've requested for your 2nd group to your kicker. The light loss accomodated by ETTL to the key because of the umbrella is also applied to the bare kicker, making it that much brighter than the key and thereby creating a ratio different than what the flashes are telling you.
Originally posted 54 months ago.
(permalink)
Steve.Korn edited this topic 54 months ago.
|
|
|
We discussed measurement technique:
As for your questions concerning the method of measurement: Take a look at this nicely illustrated article. In short: Two measurements with globe down facing the lights (for incident light difference) and after that a measurement with globe up pointing to the camera (for exposure). All measurements from head-position with a Sekonic L-358.
This is neither about measuring technique nor about camera or flash automatics. It's direct reflection effects which can be controlled by different means, one of those being the size of the lightsource.
Posted 54 months ago.
(permalink)
|
|
|
I guess I'm not clear as to exactly where you are positionaing your meter when you are measuring your kicker. As the article you posted says:
'Readings for accent lights should be taken where the light hits the subject with the dome down. For a hairlight that means the top of the head, not down in front of the face."
You said " All measurements from head-position with a Sekonic L-358."
I'm not sure what you mean by this. Did you point the meter at each light or are all of the lights being meterd from the front of your face? I've no question regarding the specular reflection, I'm just interested in understanding how you arrived at your exposure readings.
Originally posted 54 months ago.
(permalink)
Steve.Korn edited this topic 54 months ago.
|
|
|
I've been away a few days...
For measuring the kicker I held the meter approximately at the side of the head facing the kicker with dome down and pointed it towards the kicker.
I don't see how measurements with globe down facing the lights (this is meter-direction information) plus at head position (this is meter position information) can be ambiguous. Even if the meter position is something like 5-10 cm off it will change the reading only about a tenth of a stop or so (I took more than one reading the second time around, so I can judge how sensitive the readings are to changes of position).
When metering the kicker the meter did not receive direct light from the other flash and indirect light reflected off walls and/or ceiling must have been many stops darker.
Hope this clears things up :-)
Posted 53 months ago.
(permalink)
|
Would you like to comment?
Sign up for a free account, or sign in (if you're already a member).
|
Strobist.com /
Discuss
Feed – Subscribe to Strobist.com discussion threads
