Is CRI really dead?

Marty Welk · January 9th, 2010, 03:06 AM

indeed there is a lot of "white" LED stuff that "can" light for video, depending on how important All the colors and all aspects of the colors are.
There are white leds with higher CRI things, if you want to do the "chart" thing :-) Nitchita is one i am thinking of (5mm).
but yes if were to get real, they tossed in some 1W china blue leds into a box tossed in a amber flip filter on the front, and are selling them for a tidy profit, and they arent very professional. but high powered led is not cheap, even UGLY high powered led is not cheap, cooling and drivers and all.

are the CRI Claims even REAL at all? Hype, do manufatures Hype and BS people to sell them things. naw never happens. Ex: 100,000hr phosphor based 5mm leds dont last 9000hr to 50%, First complete fraud lie. Overdriven or overheated in some consumer items they are sold in they are at 50% in less than 300hr. (not the same for non-phosphor based, or "high-powered")

All the high powered leds still suffer from reduced spectrum, even if they can light a CRI chart, most of them you can see it with the lowest form of color chart the camera being aimed at. Rebel LED with CRI? only one i know that could do even a good CRI is the RGB array, and it isnt $6?

They have RGB leds, and RGB combined lighting for stage and screen and all, its very expencive. but they arent efficent. if i put arrays of rebel RGB on my On-Cam light it will take more than the 20-25W they give me on the cam (for example)

i would like to know exactally what happens to the picture if i output 3 SPIKES of red green and blue , into 3 seperate filtered photon collection units 3CCD 3CMOS kinda thing? even if its not efficent or cheap.

Tama Berkeljon · January 9th, 2010, 10:20 PM

Chris,From a manufacturers perspective, When it comes to CRI and Leds, I believe that a certain portion may be as you say "spin" from the LED manufacturers, but certainly there is also some truth to the theory that standards change and become outdated, and need revision. Untill I did research myself into CRI, I did not realise that this was NOT a be all and end all standard, Rather something constructed to serve a purpose as best it could at the time.

As far as fair pricing goes, there are many ways of looking at the factors that make a price "fair" including- build quality and reliability, efficiency of the LED's used, Quality of the Led's used and whether they've been implemented to get the maximum specified life or are being overdriven for short term gains (output), and what support and backup you get if there is a problem with your light, spectral quality, lensing if any, and light properties such as homogenity and even colour distribution.

That said there is a simple way of looking at apparent value, but it only takes ONE of these factors into account.
That is to look at $ per watt, How much do you spend per watt of light you get.

For Example,

Litepanels using their 5mm leds - approx $ 1800 for 45watts (1800/40) = $40 per Watt.

Cool Lights USA 5mm Leds - Approx $529 for 48 watts (529/48) = $11 per Watt

KometLed 12 - using 12 x 18 watt Leds, approx $6400 for 216 watts (not including PSU consumption) (6400/216) = $ 29 per Watt

Outsight Creamsource using 84 high power Luxeon Leds, Approx $7000 for 400 Watts ($7000/400) = $17.5 per Watt

(all pricess approximated in USD.)

This is a very rough comparison, as it obviously does not take into any of the abovementioned factors, nor does it have any way of measuring the EFFICIENCY of the LEDS Used, which is a Major factor relating to consumption of power vs output.
(also,all of these kits have very different accesories, just for further confusion.. )

Sure, Its true that you can buy a Rebel LED for as little as $6 or $8, but this takes into account nothing for the drive circuitry needed to regulate current, the PCB needed for mounting the LEDs, the Heat management system, lensing, DC supply, mounting and rigging, connectors, cabling, or design.

Unfortuntely I think that the lack of standardisation will continue in this area for quite some time, but all we can do as a manufacturer is try to supply the highest quality information about our products so that people can make the best choices for themselves and their needs at that point in time.

Marty, Yes this is a terible thing that some manufacturers are doing, but unfortunately it is partially due to their own ignorance of the properties of LED and drive currents/ heatsinking as well as using the wrong leds for the job also.
Im not condoning it. Im just acknowledging that it is a sad situation. Many manufacturers have jumped onto the LED bandwagon without understanding the technology, and unfortunately this hurts the consumer as much as it does the LED industry.

As with any new technology, especially at the beginning, there will be large extremes between the ends of the market, some manufacturers get the price/quality point right, and others do not.

Good luck to you all in your search for the right LED's for your situation ! there are many good ones out there for different applications/ expectations.

T

Chris Ficek · January 9th, 2010, 11:48 PM

Isn't the rating of wattage/$ somewhat of a less than ideal standard by which to metric the performance of an LED lighting device. Many manufacturers also list light performance using a lumen reference--completly not appropriate for evaluating lighting device performance. The only way to really evaluate the light output (not quality) is via a Lux or Foot Candle reading at a given distance from the fixture. What matters in terms of brightness is the amount of light falling on our subject. Of course the evenness and falloff characteristics are another matter, some lights are way bright in their central sweet spot but fall off so quickly that they're only good for the "deer in the headlights" look:)

So really what we are saying over a few of these posts is that a method of metricing LED lighting devices light output for video purposes is a somewhat ellusive science. To do the job right we have to consider all of the following--and in some cases if one component is poor it impacts the other considerations so greatly that the entire fixture should be written off:

1. Colour rendering quality, now is that as seen by the eye or the sensor?
2. Light intensity, and the homogenous nature of its pattern
3. Colour temperature, for consistancy and matching with other light sources
4. Full spectrum output, free of peaks and valleys

What else did I miss?

Chris Ficek · January 9th, 2010, 11:51 PM

And another thing, why hasn't anybody mentioned the fact that LED lights currently seem to be a disaster at single shadow rendering. It's probably not possible to deal with this as long as the fixtures keep using hundreds of LEDs.

Or is single shadow rendering just a bunch of bull put out by the hard light manufacturers to put people of the LED revolution? Comments?

Tama Berkeljon · January 10th, 2010, 12:51 AM

Quote:

Originally Posted by Chris Ficek

Isn't the rating of wattage/$ somewhat of a less than ideal standard by which to metric the performance of an LED lighting device. Many manufacturers also list light performance using a lumen reference--completly not appropriate for evaluating lighting device performance. The only way to really evaluate the light output (not quality) is via a Lux or Foot Candle reading at a given distance from the fixture. What matters in terms of brightness is the amount of light falling on our subject. Of course the evenness and falloff characteristics are another matter, some lights are way bright in their central sweet spot but fall off so quickly that they're only good for the "deer in the headlights" look:)

So really what we are saying over a few of these posts is that a method of metricing LED lighting devices light output for video purposes is a somewhat ellusive science. To do the job right we have to consider all of the following--and in some cases if one component is poor it impacts the other considerations so greatly that the entire fixture should be written off:

1. Colour rendering quality, now is that as seen by the eye or the sensor?
2. Light intensity, and the homogenous nature of its pattern
3. Colour temperature, for consistancy and matching with other light sources
4. Full spectrum output, free of peaks and valleys

What else did I miss?

Too right.. We've just built a tool to start showing the test values of some of these metrics- both of our own lights, and those made by traditional lamp manufacturers plus other LED Manufacturers.. It really is a work in progress, but with only three lights up, you can already see where the idea is heading.. Im thinking of something that very simply allows you to profile or compare several lights to each other quickly.

We've started by only showing power output (lux/ft-candles/f-stop) but im keen to add other metrics also, to help give a better overall picture of what a light might do..
(one of the next simple things to add is watts consumed / efficiency)
If anyone wants to have a play and give some feedback that would be very helpful !

the "Compare" tool is at Outsight
Its very newly up, if you break it, let me know how and we will get it fixed !

Chris, I really like your list of consideration metrics.. I wonder if somehow we can all come up with a way of scoring each, then translating that into an overall score for a light, an overall metric. (obviously you would want to show it's makeup also, how you arrive at that number when presenting it for it to be REALLY useful)

I have added a couple of metrics that I think may be helpful - and rearranged a little

1. Colour rendering quality, now is that as seen by the eye or the sensor?
2. Light intensity, and the homogenous nature of its pattern
3. Colour temperature, for consistancy and matching with other light sources
4. Full spectrum output, free of peaks and valleys
5. Colour homogenity with other units from the same product line (Taken from X individual pairings)
6. Lumens per watt total efficiency (how much light do you actually recieve at the target for a cost of x watts)
7. Homogenity of the beam (how even the distribution of light is)

So if you added all of these 6 up, say like this :

1. gets 20% (colour rendering, to a chip, film, or eye. )
2. gets 20% (output power is important, right ? )
3. gets 10% (CCT should be within range of tungsten or daylight.)
4. gets 10% ( my weighting would suggest color rendering more important than spectrum coverage)
5. gets 10% (colour homogenity is also a slightly less important metric but still something to be aware of..)
6. gets 20% (efficiency )
7. gets 10% (homogenity)

Thats a very rough hack at the idea...

My question now is : How far can we push this ? Are there enough people interested in the truth about their LED's to get in to scorings of this sort ? Seems like it might be worth the work for us all. :)

Chris Ficek · January 10th, 2010, 11:32 AM

Now we're cookin'
What does everybody else think is important for LED to rule the roost of location and studio lighting.

Tama Berkeljon · January 12th, 2010, 12:30 AM

I doubt my mention of doing some work could have scared everyone away from the thread,
I thought we were all chomping at the bit to solve some stuff, sorry if I seem overly enthusiastic !

Any further thoughts anyone ?? :)

T

Bob Grant · January 12th, 2010, 02:38 PM

I think the majority of DVInfo members are simply after "a light".
Once you get into the more indepth considerations you're outside the orbit of the one man operation because of cost and portability. This is what's making LED lighting popular, one can buy an affordable, convenient light. The quality of the light it provides might not be perfect but it's better than no light. Once you get beyond the $500/light pricepoint the landscape changes and the size of the market decreases.
For studio lighting the majority have already moved away from tungsten, at least down here. This is a conservative market, lights last a long time and the current fluro and HMIs work well. For LEDs to gain market share they need to offer something dramatically better or a new feature. So far I see the ability to change color without gells as being the strongest selling point.
Sorry if this is way off the original topic, more addressing the issue of why this isn't raising a huge amount of interest.

Chris Ficek · January 13th, 2010, 08:38 PM

Tama, Oh yeah I'm in, just been shooting 24/7 for the past week. The show wraps tomorrow night so I'll be returning to the forums over the weekend. Been pondering a game plan, let's see if anyone cares. I agree that many just want a fairly priced light that works but if you look at the hundereds of posters over at the candlepower forums CRI is a hot topic. Probably not an area of interest for many but some inquiring minds surely need to know.

Cheers 'til then

In the mean time someone care to enlighten me on the difference between a spectrophotograph and a spectroradiograph?

January 9th, 2010, 03:06 AM	#31
Marty Welk Major Player Join Date: Sep 2008 Location: East Bay Cali Posts: 563	indeed there is a lot of "white" LED stuff that "can" light for video, depending on how important All the colors and all aspects of the colors are. There are white leds with higher CRI things, if you want to do the "chart" thing :-) Nitchita is one i am thinking of (5mm). but yes if were to get real, they tossed in some 1W china blue leds into a box tossed in a amber flip filter on the front, and are selling them for a tidy profit, and they arent very professional. but high powered led is not cheap, even UGLY high powered led is not cheap, cooling and drivers and all. are the CRI Claims even REAL at all? Hype, do manufatures Hype and BS people to sell them things. naw never happens. Ex: 100,000hr phosphor based 5mm leds dont last 9000hr to 50%, First complete fraud lie. Overdriven or overheated in some consumer items they are sold in they are at 50% in less than 300hr. (not the same for non-phosphor based, or "high-powered") All the high powered leds still suffer from reduced spectrum, even if they can light a CRI chart, most of them you can see it with the lowest form of color chart the camera being aimed at. Rebel LED with CRI? only one i know that could do even a good CRI is the RGB array, and it isnt $6? They have RGB leds, and RGB combined lighting for stage and screen and all, its very expencive. but they arent efficent. if i put arrays of rebel RGB on my On-Cam light it will take more than the 20-25W they give me on the cam (for example) i would like to know exactally what happens to the picture if i output 3 SPIKES of red green and blue , into 3 seperate filtered photon collection units 3CCD 3CMOS kinda thing? even if its not efficent or cheap. __________________ ----------------sig----------------- Re-learning everything all over again, one more time.

January 9th, 2010, 10:20 PM	#32
Tama Berkeljon New Boot Join Date: Nov 2009 Location: Sydney, NSW Posts: 11	Chris,From a manufacturers perspective, When it comes to CRI and Leds, I believe that a certain portion may be as you say "spin" from the LED manufacturers, but certainly there is also some truth to the theory that standards change and become outdated, and need revision. Untill I did research myself into CRI, I did not realise that this was NOT a be all and end all standard, Rather something constructed to serve a purpose as best it could at the time. As far as fair pricing goes, there are many ways of looking at the factors that make a price "fair" including- build quality and reliability, efficiency of the LED's used, Quality of the Led's used and whether they've been implemented to get the maximum specified life or are being overdriven for short term gains (output), and what support and backup you get if there is a problem with your light, spectral quality, lensing if any, and light properties such as homogenity and even colour distribution. That said there is a simple way of looking at apparent value, but it only takes ONE of these factors into account. That is to look at $ per watt, How much do you spend per watt of light you get. For Example, Litepanels using their 5mm leds - approx $ 1800 for 45watts (1800/40) = $40 per Watt. Cool Lights USA 5mm Leds - Approx $529 for 48 watts (529/48) = $11 per Watt KometLed 12 - using 12 x 18 watt Leds, approx $6400 for 216 watts (not including PSU consumption) (6400/216) = $ 29 per Watt Outsight Creamsource using 84 high power Luxeon Leds, Approx $7000 for 400 Watts ($7000/400) = $17.5 per Watt (all pricess approximated in USD.) This is a very rough comparison, as it obviously does not take into any of the abovementioned factors, nor does it have any way of measuring the EFFICIENCY of the LEDS Used, which is a Major factor relating to consumption of power vs output. (also,all of these kits have very different accesories, just for further confusion.. ) Sure, Its true that you can buy a Rebel LED for as little as $6 or $8, but this takes into account nothing for the drive circuitry needed to regulate current, the PCB needed for mounting the LEDs, the Heat management system, lensing, DC supply, mounting and rigging, connectors, cabling, or design. Unfortuntely I think that the lack of standardisation will continue in this area for quite some time, but all we can do as a manufacturer is try to supply the highest quality information about our products so that people can make the best choices for themselves and their needs at that point in time. Marty, Yes this is a terible thing that some manufacturers are doing, but unfortunately it is partially due to their own ignorance of the properties of LED and drive currents/ heatsinking as well as using the wrong leds for the job also. Im not condoning it. Im just acknowledging that it is a sad situation. Many manufacturers have jumped onto the LED bandwagon without understanding the technology, and unfortunately this hurts the consumer as much as it does the LED industry. As with any new technology, especially at the beginning, there will be large extremes between the ends of the market, some manufacturers get the price/quality point right, and others do not. Good luck to you all in your search for the right LED's for your situation ! there are many good ones out there for different applications/ expectations. T __________________ Tama Berkeljon Inventor/Designer - Creamsource www.outsight.com.au Built in Australia ! My blog http://tama.typepad.com/blog/

January 12th, 2010, 12:30 AM	#37
Tama Berkeljon New Boot Join Date: Nov 2009 Location: Sydney, NSW Posts: 11	I doubt my mention of doing some work could have scared everyone away from the thread, I thought we were all chomping at the bit to solve some stuff, sorry if I seem overly enthusiastic ! Any further thoughts anyone ?? :) T __________________ Tama Berkeljon Inventor/Designer - Creamsource www.outsight.com.au Built in Australia ! My blog http://tama.typepad.com/blog/

January 9th, 2010, 11:48 PM	#33
Chris Ficek Regular Crew Join Date: Feb 2007 Location: Ontario Canada Posts: 113	Isn't the rating of wattage/$ somewhat of a less than ideal standard by which to metric the performance of an LED lighting device. Many manufacturers also list light performance using a lumen reference--completly not appropriate for evaluating lighting device performance. The only way to really evaluate the light output (not quality) is via a Lux or Foot Candle reading at a given distance from the fixture. What matters in terms of brightness is the amount of light falling on our subject. Of course the evenness and falloff characteristics are another matter, some lights are way bright in their central sweet spot but fall off so quickly that they're only good for the "deer in the headlights" look:) So really what we are saying over a few of these posts is that a method of metricing LED lighting devices light output for video purposes is a somewhat ellusive science. To do the job right we have to consider all of the following--and in some cases if one component is poor it impacts the other considerations so greatly that the entire fixture should be written off: 1. Colour rendering quality, now is that as seen by the eye or the sensor? 2. Light intensity, and the homogenous nature of its pattern 3. Colour temperature, for consistancy and matching with other light sources 4. Full spectrum output, free of peaks and valleys What else did I miss?

January 9th, 2010, 11:51 PM	#34
Chris Ficek Regular Crew Join Date: Feb 2007 Location: Ontario Canada Posts: 113	And another thing, why hasn't anybody mentioned the fact that LED lights currently seem to be a disaster at single shadow rendering. It's probably not possible to deal with this as long as the fixtures keep using hundreds of LEDs. Or is single shadow rendering just a bunch of bull put out by the hard light manufacturers to put people of the LED revolution? Comments?

January 10th, 2010, 11:32 AM	#36
Chris Ficek Regular Crew Join Date: Feb 2007 Location: Ontario Canada Posts: 113	Now we're cookin' What does everybody else think is important for LED to rule the roost of location and studio lighting.

January 12th, 2010, 02:38 PM	#38
Bob Grant Trustee Join Date: Nov 2005 Location: Sydney Australia Posts: 1,570	I think the majority of DVInfo members are simply after "a light". Once you get into the more indepth considerations you're outside the orbit of the one man operation because of cost and portability. This is what's making LED lighting popular, one can buy an affordable, convenient light. The quality of the light it provides might not be perfect but it's better than no light. Once you get beyond the $500/light pricepoint the landscape changes and the size of the market decreases. For studio lighting the majority have already moved away from tungsten, at least down here. This is a conservative market, lights last a long time and the current fluro and HMIs work well. For LEDs to gain market share they need to offer something dramatically better or a new feature. So far I see the ability to change color without gells as being the strongest selling point. Sorry if this is way off the original topic, more addressing the issue of why this isn't raising a huge amount of interest.

January 13th, 2010, 08:38 PM	#39
Chris Ficek Regular Crew Join Date: Feb 2007 Location: Ontario Canada Posts: 113	Tama, Oh yeah I'm in, just been shooting 24/7 for the past week. The show wraps tomorrow night so I'll be returning to the forums over the weekend. Been pondering a game plan, let's see if anyone cares. I agree that many just want a fairly priced light that works but if you look at the hundereds of posters over at the candlepower forums CRI is a hot topic. Probably not an area of interest for many but some inquiring minds surely need to know. Cheers 'til then In the mean time someone care to enlighten me on the difference between a spectrophotograph and a spectroradiograph?