I have a confession to make. I realize what I am about to say will shock and offend many of you. I expect some friendships will be irreparably damaged; I understand I may be forever unwelcome at future industry functions. Marian Sandberg will soon disable my account to prevent this blaspheme from marring an otherwise fine publication. It's quite possible I will lose my job. Going to LDI would be unthinkable, since the name calling and assaults on my character should I attend would be overwhelming. However, what I have to say I've wanted to say for awhile. It needs to be said, and though the backlash may force me and my family into hiding I cannot demur any longer. For God, King, and country I and I alone must speak out and accept the ugly consequences of my actions. Should your constitution be weak, skip this entire article. Are you ready? Okay. Here it goes: I think LEDs suck.
They suck for a lot of reasons, for example, color. At the Lighting Master Classes held here in NYC, Holder did a shoot out with various LED lekos. Sometimes, the colors between units matched. Most times it didn't, sometimes by a wide margin and other times only subtly. Why the discrepancies? Physics holds the answer.
Subtractive color mixing works by blocking (or subtracting) all the frequencies in the visible spectrum except the color the designer picked, i.e., the color of the gel. If the source emits nothing in the green part of the spectrum, putting a green gel in front of that source will yield zero light. You can't subtract what isn't there. Unlike their incandescent counterparts, LEDs do not emit a full spectrum. Where as an incandescent source emits all colors in a gentle, rolling curve resembling the profile of an Appalachian mountain, white LEDs emit very different wave form. After all, white light from a LED is created by a blue LED shining through a yellow phosphor layer. This creates issues with subtractive color mixing, as the designer might be trying to get a certain color the source does not emit. While I applaud Lee's innovation for creating a cool-LED gel series (and it is pretty awesome), do we really have to have an entirely separate gel library for colors we already own?
Single, white LED fixtures only make up a fraction of the market. Most fixtures use additive color mixing. Enter the red, blue, green LED fixture. Ta-da! So now we have three chips which emit very specific frequencies of red, blue, and green. What could be greater? Well, it turns out the whites look “funky.” Intuitively we sense something's amiss with the quality of it. No problem! Enter the red, blue, green, amber LED fixture. Ta-DA! Instead of three spikes we have four. Better? Well, we can create more colors but the whites still look off. No problem! Enter the red, blue, green, amber, and white LED fixture. TA-DA! Problem solved! Thanks, everybody, that's a wrap!
Except, no, it isn't. Huge chunks of the visible spectrum are still M.I.A. Are we going to keep adding LEDs to fill in the entire visible spectrum? Turns out, we already invented that type of source and called it a tungsten-halogen lamp. Furthermore, how many more chips can we practically cram onto the existing housings? Instead of more red, blue, and green LEDs in a given instrument, there are fewer red, blue, green, amber, and white LEDs in the same size fixture. Full red or blue ain‘t so bright, and so what if I can mix a ridiculous color like Lavender Sunrise Surprise using varying intensities of every LED in the unit. Why do I need to additively mix what Rosco, Lee, and GAM have already created? Yes, I understand manufactures are creating bigger circuit boards and housings to hold more LEDs which gooses the output. So now we've got heavier (because more LEDs mean a larger heat sink) fixtures with edges that still shift towards totally different colors than the beam center. Bring out the heavy diffusion! Huh ... where did my light output go? Don't even get me started on short waves. I did a party once in an all white venue where everyone had a blue outline. I remember the resulting migraine to this day.
Another issue with color is standardizing it across multiple units which span multiple generations. I remember a gig years ago where we up-lit the perimeter of a white tent with LED strip lights. To save time, we addressed every unit the same. We lost flexibility but were hoping to gain time. Sadly, it back fired. Every single unit emitted a slight variation of the same color -- some where slightly warmer and some where slightly cooler. In the all white tent, the differences were painfully obvious. The programmer, who had more important things to do, had to assign each unit an individual address, re-patch, and color balance every single unit one by one. Some he couldn't see because of the console's location and relied on an electrician over radio to “see” for him. The shop guys were surprised by my ire over the whole thing. Never occurred to them that mixing two different generations of the same light would cause a problem. Honestly, it never occurred to me either. Nowadays everybody knows.
The units were supposed to be identical, but that isn't completely possible. The way LEDs are produced and categorized, called binning, means there will never be an absolute perfect match. To do so would be cost prohibitive, and those few, perfectly matching LEDs generally go to airports who pay top dollar for them. We get a lesser grade LED. Most of the time no one will notice a difference, until the day when everyone notices a difference. You'll just have to figure it out on-site should that happen, and as we've discussed in the past I really, really hate that.
LEDs suck for other reasons. Every time a LED unit slowly dims out, a designer dies a little on the inside. Okay, okay, I exaggerate. Everything dims pretty well for a time until near the end of the curve. Suddenly, your beautiful salmon color flashes to Bodily Fluid Green and begins to step, step, step, step, step, and ... wait for it ... off. It's painful to watch. Gear manufactures have created amazing software algorithms (and other trickery) to correct the problem, even adding faux lamp inertia, and most units have it fairly well licked.
How much R&D went into that? Very smart and crafty software engineers aren't cheap, and their expense must surely be layered into the purchase price of every unit. So all of us spent exactly how much money to meticulously and artificially add what incandescent lamps do automatically? Couldn't we have spent that money making a more efficient incandescent bulb?
Furthermore, different iterations of tungsten-halogen lamps don't require firmware updates to work cohesively together. They just need regular, standard electricity -- no voltage transforming or hardware patches needed! In my experience most of the New York theaters barely keep their lekos in good physical working order. Some still have asbestos pig tails and lens tubes with caked dust dating from the Carter administration. I highly doubt any real effort is going to go into updating software and firmware on a regular basis. If the rinky-dink theaters cannot afford, manage upkeep, or mix and match any of these fancy LED fixtures (with either each other or other sources), then how big can the market really be?
Many say, “Just wait, LEDs will get better,” but I have my doubts. They may always suck. Years ago I worked a large party on Long Island. The Lighting Designer wanted to use LEDs to wash the tent, which would have saved considerable power and effort. The florist (and also the client) immediately shot it down. I asked him about it later, and his answer surprised me. He said, “Humans have been eating around fire for tens of thousands of years. Restaurants put candles at every table for a reason. [Incandescent] lamps are essentially fire. LEDs make people uncomfortable because they aren't fire and they aren't light we evolved to see in.” He makes a lot of pertinent points.
LEDs create light via a wafer where the energy state of an electron changes, which causes a photon to be emitted. That's not fire or anything resembling fire. Also, new research suggests the human eye evolved to see wavelengths around the red end of the spectrum in order to quickly ascertain the health or emotional state of others. Mark Changizi talks about this idea in his book, The Vision Revolution: How the Latest Research Overturns Everything We Thought We Knew About Human Vision. In short we continually read each others faces for subtle changes in skin pigment which give us clues about how someone is feeling. For the skin to reflect those all important wavelengths, they must first be present in the light that hits our skin -- as with sunlight and firelight. With LEDs those wavelengths may not be there, blinding us in a sense we unconsciously rely on, causing the discomfort many experience in an all LED environment.
TV gets away with using LEDs on people, and in some cases creating an all LED environment. None of my complaints really apply in that realm. Still ... TV shows can get into real trouble when a designer tries to mix cool LEDs and warm, traditional lamps. Film and concerts are their own markets, with different exposure to LED technology. In my limited experience with film, they still mainly use the same fixtures types used on Citizen Kane. However, for events and performance, all LED really, really, really sucks. I realize similar arguments were made when we transitioned from gas light to electric light. The outcry then probably echoes my outcry here. I realize the inherent unfairness of comparing an established and honed technology -- tungsten-halogen lamps -- to a new and still improving technology -- LEDs.
I must admit in some ways LEDs do work. LED walls in all their glorious flavors are amazing. LEDs buried in light boxes, behind a quarter of an inch or more of milk-plexi, designed into a set, look beautiful. Washing a wall of a building, some LED products work really well. Unfortunately, LEDs are expanding well beyond those limited applications.
We do need greener alternatives to traditional lamps, which gobble electricity and inadvertently produce enormous amounts of waste in the form of heat. Everyone knows the savings, but only a few environments realize those savings -- specifically TV studios, permanent installations, and shows powered off generators. Until the market can fully bear “green tech,” which in my experience it largely cannot, LEDs are the answer to a question few can afford to ask. The contortionist act they require us to perform make no sense for large swaths of the market. Maybe, then, LEDs would suck a whole lot less if we kept them where they belong: the periphery.
Lance Darcy is a Lighting Director and Director of Photography for The Lighting Design Group, based in New York City.