May2012
www.tvbeurope.com
“The science hasn’t changed, the software has. It is much more user friendly”
TVBEurope 19
News & Analysis Low Energy Lighting
TV standards: Getting the light right
A new world standard — a Television Lighting Consistency Index that is relevant to TV and film — could replace the Colour Rendition Index by the end of the year. David Fox reports
A LOT of new, low energy, low-cost lighting produces inaccurate colours that can’t even be corrected by colour grading. But, it is difficult to assess each light without testing — and there isn’t even a standard test. Until now. As TVBEurope reported in the September 2011 issue ‘Creating a Television Lighting Consistency Index’, independent consultant and colour scientist Alan Roberts has created a measuring system and measuring tools that will make it easy to assign a number between 0-100 to any light, to show how suitable it is for television or film production. He has now joined an EBU technical group (FTV-LED) to take this work forward, which the EBU hopes to publish as a standard during the summer. Roberts wants the metric finished before IBC, as there is a CIE (International Commission on Illumination) meeting directly after IBC where he hopes the CIE will adopt it as an international standard.
The EBU team has agreed to
compare any input light source with the EBU’s own correlated colour temperature using the ColorChecker Macbeth chart, “which isn’t ideal, but it is so widely available.” Two areas of the parameters
aren’t yet fixed: The camera responsivity (the response of the red, green and blue channels to different wavelengths), which is important because it affects the look of the colours; and the colour difference metric, the mathematical way of describing the visual aspect of the colour difference.
A Standard camera Per Bøhler, the NRK’s principal engineer, is currently doing the camera responsiveness measurements. “This will produce spectral data that will become the EBU Standard Camera — a mathematical embodiment of a hypothetical average camera,” in the same way that the CIE has a Standard Observer (an average of 17 different people taken in 1927), explained Roberts.
deliver a paper on this at IBC. Roberts is doing the work on how to measure colour differences. The EBU group has agreed
the matrix gamma curve (24), which he says doesn’t make any difference to the observed results, and the display primaries (ITU 709 for HD). The CIE has six different metrics for colour difference created over the years, so Roberts is testing to find which of these delivers the results that most closely match the human eye. That metric will produce a value, showing the difference in subjective lightness, chroma and hue in whatever light is being examined. The software will then provide a single number, from 0-100, so that video or film lighting users know how suitable the light is for their production. A value of 90+ would mean that no colour correction would be necessary for that light; between 75-90 would mean some grading would probably be needed (although you might not get every colour correct); from 50-75 would indicate that grading would be
According to Stoke’s Law, a phosphor can’t emit energy at the same wavelength that it receives it at, so “it is not possible to make a white LED that has an even spectrum”
In the US, the Academy is
“very interested” in adopting it, as this is an area it has been interested in for several years. He then hopes to convince lighting manufacturers to put the Q number on their lights, so that anyone can see what they are suitable for. Since September, a lot of the
parameters in the testing have been fixed, and although “the science hasn’t changed, the software has. It is much more user friendly,” said Roberts.
“This will allow you to model things without measuring any more cameras. It also gives the camera manufacturer something to shoot for” when developing new cameras. They can’t take any
single existing camera as their standard, as Sony, Panasonic, Ikegami and others all have their own look, “so there is no generic description of a camera.” Bøhler, who is chairing the
FTV-LED group, will probably
required if used for broadcast work, but the pictures probably won’t look right due to pushing the cyan and getting noise in those areas; while below 50 would “be OK for weddings and news”, but it would be difficult to match the pictures, even with grading. However, as Roberts says, “any light is better than no light. Even a low pressure sodium street light shouldn’t score negative.” The output from the software will not only give a formal Q
Colour check: Alan Roberts hopes to make it simple to check how suitable a light is for broadcast
number, but also information that a colourist could use to correct for that light either in lightness, chroma or hue. However, as the various grading software use different numbers, all Roberts can do is be consistent with the report, which will simply give a plus or minus (between -5 and +5) or a 0, which indicates that nothing needs to be done. Much of the industry is
moving towards LED lights, but in his testing so far Roberts says that LEDs are generally the most difficult of commonly used lights to correct for, as they produce a large blue spike. According to Stoke’s Law, a phosphor can’t emit energy at the same wavelength that it receives it at, so “it is not possible to make a white LED that has an even spectrum.”
The only way to make an LED light with good colour response is to do what Gekko Technology does, which is to use differently coloured LEDs and mix them together. “It’s as simple as that — or rather, as expensive as that.”
In his testing the Gekko lights have been the best he has seen, although he hasn’t yet tested the Arri LEDs, which he expects to be at least as good. He believes the minimum number of differently coloured LEDs needed for a broad- spectrum light is five (white, red, green, blue and cyan — required to fill the gap left by the Stoke’s shift). Gekko uses six and Arri 11 or 12. It would also be possible to use an ultraviolet LED (with a UV filter) to move the Stoke’s gap to the edge of blue.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68