MayaTlab Offline Upload & Sell: Off
|
p.1 #10 · p.1 #10 · Deep "parabolic" umbrella vs regular umbrella use cases? | |
Lots of things to cover here. Let's start with the general bits :
a) Regarding the term "parabolic" : it has a precise definition that entails a precise result, and as Mark_L wrote, most modifiers sold as "parabolic" simply aren't shaped at all like a paraboloid and aren't even trying. Parabolic doesn't mean that the modifier is deep, it means that it conforms to a specific mathematical definition of the arc shape - wikipedia's article on the matter is explicit enough in that regard.
As a result, NONE of the deep umbrellas on sale today are parabolic modifiers, because their arc shape is nowhere near the definition of a parabola. In fact, a shallower Paul Buff PLM is closer to that definition.
A perfectly parabolic reflector with a perfect mirror finish has a focal point, and if a light source is at this focal point, it will reflect the light rays from the light source in a perfectly parallel beam, like this :
Courtesy of Parabolix.
An example of a modifier that tried to get as close as possible to that definition is the Broncolor Satellite, and it's very shallow :
As a result, a shallower Paul Buff PLM can send light at a rather tight beam angle. Below, a Chinese copy of the extreme silver PLM vs a 20° grid :
b) White materials will scatter the light in all directions, regardless of the angle of the light ray that hits them. A perfect mirror will bounce the light at the opposite angle at which the light ray hit it, a bit like a ping pong ball (without spin ). Silver materials are, in a way, imperfect mirrors. They vary in terms of reflectiveness. Some of them are quite reflective and won't scatter the light much, others are designed to scatter it a bit more. To sum it up in a over-simplified way :
The result of this is that :
- there isn’t such a thing as a white parabolic modifier. Since the main property of a parabolic modifier is its ability to re-direct light rays in a specific direction, a material that scatters light rays in all directions negates the interest of the paraboloid shape. Some would argue that Paul Buff’s soft silver PLMs aren’t parabolic reflectors either as the material scatters light too much, but I have a less finite opinion.
- Moving the light source inside a white reflector will not change much the beam angle of that modifier : it’s always going to scatter light in all directions. Differences will be marginal.
- On the contrary, silver reflectors are more or less heavily relying on where the light source is positioned relative to them. Moving the light source relative to the umbrella, either laterally or closer/further out, can very significantly change the beam angle, or the effective size of the modifier (from the subject’s point of view). The more scattering the silver material is, the less sensitive it is to light source position.
Paul Buff used to sell the exact same umbrella shape (the PLM) in white, and two silver materials, with different scattering properties, soft silver, and extreme silver, to benefit from the different characteristics of each material.
Broncolor’s smaller 88 and 133 paras use a slightly more scattering material than their bigger ones, because it improves the quality of the illumination from the subject’s point of view (see “d”) and most people don’t really like the look of a really hard parabolic modifier anyway.
c) "Spill" can have various origins, but one of the most common causes with umbrellas is bare flash tube spill, i.e. when the flash tube isn't covered on the sides by a spill kill / umbrella reflector. In the following photo :
- the blue arrow corresponds to the area that's illuminated by the light reflected by the umbrella,
- the red arrow to the area that's illuminated by the strobe's bare flash tube if it hasn't been properly killed by a spill kill reflector or by other means. Very often people say that umbrellas, even silver ones, have massive spill, while they don't bother to effectively kill the spill from the bare flash tube,
- and the green arrow to the area that's illuminated by the light reflected from the subject (if your subject is a white wall, it will matter).
That's been taken with a Profoto B2 without a spill kill modifier, and with the Cactus umbrella slightly forward from the head (hence why you see a streak of light coming from the bare tube). If I had used a flash with an external flash tube and without a spill kill modifier, bare tube illumination would be massive and cover a much wider angle.
Bare flash tube spill can be useful, for example if you want to benefit from the flash tube’s exposition to use a bounce reflector in addition to the umbrella. But it’s a good thing to eliminate it properly, at least when comparing spill from different modifiers .
d) effective modifying size and appearance from the subject’s POV (i.e., what it will look like on your subject in terms of specularity, shadows, etc.) can be different from the physical size or appearance the modifier.
With silver reflectors, since they bounce and re-direct light, only the areas of the umbrella that effectively re-direct light towards your subject will directly contribute to the light quality falling on your subject. From the latter’s point of view, these areas will be illuminated. The areas of the umbrella that re-direct light somewhere else, or don’t receive light in the first place, will look dark from its point of view. With deep umbrellas, what you get usually is this (compared here a Profoto deep with a shallower PLM-like silver umbrella) :
The result is that for most intents and purposes, deep silver umbrellas are much smaller effective modifiers then shallower pseudo-parabolic umbrellas like PLMs.
You can also notice that both umbrellas are illuminated in a bicycle wheel pattern. That’s because of two things :
- the silver material used in both of these umbrellas is quite reflective, and
- because they are 16 sided fabric modifiers, they are imperfectly shaped, and only the middle area of the fabric between the ribs is effectively in the right orientation to be able to re-direct light rays towards the subject.
The result is that catchlights might exhibit this bicycle wheel pattern, and you may see multiple, stepped shadows appear.
Paul Buff’s soft silver umbrellas use a silver material that’s just scattering enough that this bicycle wheel pattern doesn’t appear. It will look like this (extreme on the left, soft on the right) :
But since the material scatters light a little bit more, it isn’t quite as directional as its extreme silver cousin (soft on the left, extreme on the right) :
That being said, you can still see in the above shot that the soft silver can have a hot spot that more or less corresponds to the area lit by the more extreme version, with the head in the same position. So some of its pseudo parabolic DNA is still there.
All of these comparisons were made with the light modifier on axis with the camera’s lens (i.e., the subject), but you may very well want to feather the light (albeit with highly directional silver umbrellas, there isn’t much feathering room to be honest). when you feather the light, you will see the effective modifying area of the umbrella change, move to the side, etc…
Since white reflectors scatter light in all directions, their fabric doesn’t have to be in a precise orientation to re-direct light towards your subject. As a result they don’t suffer from this bicycle wheel pattern look. But how your strobe is positioned, how its flash tube is designed and which reflector is used can still matter in terms of illumination evenness.
e) Regarding diffusion : I’m sorry, but adding diffusion over a parabolic, pseudo parabolic, or deep silver reflector will instantly negate its directional aspect and turn it into a sort of softbox . Here is, on the left, an extreme-like silver umbrella, without diffusion, and on the right the same umbrella, but with Paul Buff’s front diffuser :
Some diffusion materials might be weak enough that they will scatter the light only a tiny little bit, but I don't think that they're sold as front covers for umbrellas.
And now your questions more specifically :
The following presumes that bare flash tube spill has been effectively eliminated.
A deep white umbrella won’t send light at a significantly narrower beam angle than a shallow one. It might be a tiny, tiny bit narrower, if you’re illuminating only the centre of the umbrella (by shoving the light deep inside - with the same reflector you used to kill bare flash tube spill when the light was further out -, or using a narrower reflector for example), since the un-illuminated sides are “flagging” the light a little bit. As a consequence though, you’ve just slightly reduced the effective size of your modifier from the subject’s point of view, since only the centre of the umbrella is illuminated. Shadows will be a tiny bit harder. Overall we’re taking minute differences here. Personally I wouldn’t buy a deep white umbrella if I wanted a narrower beam angle.
An example of a deep white umbrella with a Profoto B2 head :
A deep silver umbrella won’t send light at a narrower beam angle than a pseudo-parabolic one, such as Paul Buff’s PLM.
That being said, a lot of variables come into play here : if you manage to hit only the centre of the deep silver umbrella with your light source, it becomes, in a way, a smaller shallow pseudo parabolic silver umbrella, since the central portion of deep silver umbrellas can be seen as being reasonably parabolic. But you have to manage to light the entire central portion and only the central portion from a specific point within the umbrella, since, as written before, silver materials bounce light rays at the opposite angle. You can’t just randomly shove the light inside, or use any reflector, like you would with a white umbrella. The problem with this approach is that you’re basically carrying useless extra fabric and bits. A 160cm deep silver umbrella, in that scenario, will look, from the subject’s point of view, basically like a 110cm shallow pseudo parabolic umbrella, and shadows won’t be softer.
Putting diffusion over a silver reflector, including parabolic ones, turns them into softboxes.
Putting diffusion over a white umbrella... is a waste of power IMO .
As far as I’m concerned :
- I prefer to use deep white umbrellas over shallower white ones, but that’s mostly because I use Profoto heads with an internal flash tube design, that sends light in a quite directional way. Using a deep white umbrella allows me to light the inside of the umbrella more evenly when positioning the head at the umbrella’s rim than a shallower one, which would require the head to be much further back to be evenly illuminated, and, as a result, require the use of a spill kill reflector. That’s because the shape of a deep white umbrella means that the distance between the head at the rim and either the sides or the centre of the umbrella is more constant. With a shallower umbrella, the head is much closer to the centre of the umbrella than to its sides, which is why you get a hotspot inside the umbrella.
- I have yet to find a good reason for deep silver umbrellas to exist, at least in their current form. So I don’t use them. Instead I mostly use PCB’s soft silver umbrellas, and if I need an even tighter light beam, I also have more directional shallow silver umbrellas. If you carry two 54” PCB PLM, one extreme silver and the other one soft silver, you can do basically everything a 160cm deep silver umbrella does and more, without the drawbacks, and without a significant difference in weight. A soft silver PLM produces a light that isn’t far from a typical gridded softbox (there are differences, but they’re small), but it’s way more efficient, so for battery-operated situations it’s very useful to me. I rarely use the more directional silver pseudo-parabolic umbrellas, as I don’t like the stepped shadows and the bicycle wheel pattern, but that’s a personal preference here.
Edited on Sep 20, 2017 at 07:28 AM · View previous versions
|