Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: WayneLarmon on February 03, 2014, 06:45:12 am
-
I want to measure the characteristics of light sources. Spectrum plots, CRI, color temperature, etc. Like this
(http://farm4.staticflickr.com/3695/11880380353_91aa8f397b_o.png)
(The above screen shot came from http://photo.net/casual-conversations-forum/00cHcE?start=50 (http://photo.net/casual-conversations-forum/00cHcE?start=50) and scroll down to the post by Lad Lueck , Jan 10, 2014; 10:06 p.m.)
I'm trying to figure out better sources of continuous light. I've already made a DIY spectroscope
http://wlarmon.smugmug.com/Other/DIY-Spectroscope/32099985_nMBQnS#!i=2788846720&k=BLnwZzj (http://wlarmon.smugmug.com/Other/DIY-Spectroscope/32099985_nMBQnS#!i=2788846720&k=BLnwZzj) which works for determining spikiness, but doesn't tell me the shape of the spectrum, the color temperature, or the CRI.
I'm assuming that I need some kind of spectrophotometer with matching software. The post I linked to mentions a "Gretag Macbeth i1 Spectrophotometer." But Gretag Macbeth doesn't exist anymore. XRite spectrophotometers seem to be oriented towards profiling monitors and printers. I don't need that (I already have a NEC PA241W, Spectraview II, and an i1DP for my monitor and I don't have a printer.) I just want to measure light spectrums.
Lab grade spectrophotometers seem to cost "send for a quote", which probably means that I can't afford one. Budget is an issue. What is the cheapest way to do what I want to do?
TIA.
-
I'm assuming that I need some kind of spectrophotometer with matching software. The post I linked to mentions a "Gretag Macbeth i1 Spectrophotometer." But Gretag Macbeth doesn't exist anymore. XRite spectrophotometers seem to be oriented towards profiling monitors and printers.
Same company now, the older and newer X-Rite i1Pro can do this, best mated with BableColor.
http://www.babelcolor.com/#CTandA
-
So if I bought new, I'd buy the cheapest XRite package that has an i1Pro spectrophotometer, which I think is the
X-Rite i1Basic Pro 2 Color Profiling Software Bundle $1,149
http://www.bhphotovideo.com/c/product/854476-REG/X_Rite_EO2BAS_11_Basic_Pro_2.html (http://www.bhphotovideo.com/c/product/854476-REG/X_Rite_EO2BAS_11_Basic_Pro_2.html)
+ the BabelColor software?
Or haunt eBay looking for a used i1Pro (1 or 2) Spectrophotometer? ( + the BabelColor software.)
And this wouldn't do me any other good, because I already have the i1DP for my PA241W and I don't have (or intend to get) a printer. Correct?
Um, do you know of anybody that is measuring LED lamps? The only reason I want this is because nobody else is showing spectral plots and CRI ratings of LEDs. We all pretty much know what CFLs are good for by now but LEDs are a wild card. (i.e., some of them now have smooth spectrums, just like tungsten. But maybe not shaped quite right. Which may be fixable with a Rosco gel. Maybe.)
Thanks.
-
And this wouldn't do me any other good, because I already have the i1DP for my PA241W and I don't have (or intend to get) a printer. Correct?
Don't understand the question. The i1Display-Pro is a colorimeter so no, you can't use it for the work the OP is asking about.
Um, do you know of anybody that is measuring LED lamps? The only reason I want this is because nobody else is showing spectral plots and CRI ratings of LEDs. We all pretty much know what CFLs are good for by now but LEDs are a wild card. (i.e., some of them now have smooth spectrums, just like tungsten. But maybe not shaped quite right. Which may be fixable with a Rosco gel. Maybe.)
I have measured and built plots of LED's using BableColor and the i1Pro-2 Spectrophotometer. BableColor will produce all kinds of reports to the qualities of the light measured.
-
And this wouldn't do me any other good, because I already have the i1DP for my PA241W and I don't have (or intend to get) a printer. Correct?
Don't understand the question. The i1Display-Pro is a colorimeter so no, you can't use it for the work the OP is asking about.
I meant that if I bought the i1Basic Pro 2 Color Profiling Software Bundle (in order to get the i1Pro 2 Spectrophotometer to measure lights) that this Spectrophotometer would only be useful to me for measuring lights. Because I already have the i1 Display Pro calorimeter to measure my PA241W (with Spectraview II). I wish XRite wouldn't name everything i1[whatever] because it causes confusion real quick when discussing several different XRite products.
I have measured and built plots of LED's using BableColor and the i1Pro-2 Spectrophotometer.
Have you published the results? And have you found any LEDs that are useful? (Enough so that SoLux bulbs aren't needed anymore?)
In another thread (Medium Format DSLR as film scanner) you linked to a thread on photo.net about DSLR camera scanning.
http://photo.net/digital-darkroom-forum/00cKbq?start=0 (http://photo.net/digital-darkroom-forum/00cKbq?start=0)
I participated in that thread. If you read that thread you will know what I am concerned about. I'm trying to improve the SOTA for camera scanning. Because quality film scanners have gone away and aren't coming back. But garden variety DSLRs are already great imaging engines. I know that great results are possible with camera scanning (which I demonstrated in that thread), but the techniques need perfecting.
-
I only just recently measured the LED's in my office ceiling for a tutorial on DNG camera profiles. Absolutely doable in BableColor software.
-
the older and newer X-Rite i1Pro can do this, best mated with BableColor.
The ColorMunki Photo with Argyll can do this also and is a much less expensive package.
-
The OP can download the software shown in the first post for free. Even better? Of course, way more useful stuff in BableColor but if all he wants to do is measure like that screen capture, register i1Pro on X-rite site and download, it's free.
-
But I thought he didn't have a spectro yet? The i1 Pro is much more expensive than the ColorMunki.
-
The ColorMunki Photo with Argyll can do this also and is a much less expensive package.
You mean this?
X-Rite ColorMunki Photo Color Management Solution $449.00
http://www.bhphotovideo.com/c/product/550833-REG/X_Rite_CMUNPH_ColorMunki_Photo_Color_Management.html (http://www.bhphotovideo.com/c/product/550833-REG/X_Rite_CMUNPH_ColorMunki_Photo_Color_Management.html)
It is a spectrometer but the Argyll docs says that ColorMunki Design or Photo is "UV cut only." What is the significance of that?
I've used Argyll to convert ICC profiles to 3-D VRML files (using iccgamut and viewgam) but I didn't know it could be used to make spectral plots, do a CRI rating and give the color temperature of an illuminant. Before I buy a ColorMunki Photo, could you give me some hints on how to do this with Argyll?
http://argyllcms.com/doc/ArgyllDoc.html (http://argyllcms.com/doc/ArgyllDoc.html)
-
Better deal (better instrument):
http://www.ebay.com/itm/X-rite-i1-eye-one-Pro-Color-Management-Tool-/251440747621?pt=LH_DefaultDomain_0&hash=item3a8b095465
A few of them on ebay.
-
I've used Argyll to convert ICC profiles to 3-D VRML files (using iccgamut and viewgam) but I didn't know it could be used to make spectral plots, do a CRI rating and give the color temperature of an illuminant. Before I buy a ColorMunki Photo, could you give me some hints on how to do this with Argyll?
http://argyllcms.com/doc/ArgyllDoc.html (http://argyllcms.com/doc/ArgyllDoc.html)
Wayne, I've used argyllCMS for making spectral distribution plots. You want to use spotread — it's part of the argyll package: http://argyllcms.com/doc/spotread.html
-
Better deal (better instrument):...
Ah, thanks. I was searching eBay for "spectrophotometer" and only getting lab type instruments. That price is more reasonable for experimenting.
Wayne, I've used argyllCMS for making spectral distribution plots. You want to use spotread...
Thanks.
-
I only just recently measured the LED's in my office ceiling for a tutorial on DNG camera profiles.
The reason I am curious about LEDS is
(http://wlarmon.smugmug.com/Other/Spectrum-3/i-zzvmMFk/0/S/60D_11272-S.jpg)
Standard soft white CFL
(http://wlarmon.smugmug.com/Other/Spectrum-3/i-Bpr6FV6/0/S/60D_11258-S.jpg)
Expensive CFL with a CRI of 93
(http://wlarmon.smugmug.com/Other/Spectrum-3/i-SbFTg5z/0/S/60D_11277-S.jpg)
Regular old tungsten bulb (which we can't buy anymore)
(http://wlarmon.smugmug.com/Other/Spectrum-3/i-F28BPSb/0/S/60D_11446-S.jpg)
Cree daylight LED. CRI 80
It is this last that has me interested. Note the smoothness of the spectrum. This is only CRI 80. They have a new "TW" version that they rate at CRI 93 and refer to as "full spectrum" But it isn't available in my local Home Depot yet. (It is only in California Home Depots at present.)
SoLux bulbs are neat and all, but they are very hot. Trying to put four or so SoLux bulbs in a softbox would be, um, challenging. (I can swing four 100 watt (equiv.) CFLs in a softbox. But CFLs, well...
B&H (et al.) have inexpensive small LED panels that might do for camera scanning, if I put one close to the film. But LED lights (under $1,000) have no CRI rating (nor spectral plots) I'd like to experiment with those LED panels, but I'd really like to know what I am getting. Hence, the interest in being able to do my own spectral plots.
-
Spotread produces this:
Result is XYZ: 48.533259 51.313619 46.795263, D50 Lab: 76.868594 -2.560387 -5.444825
Ambient = 51.3 Lux, CCT = 5564K (Delta E 6.324134)
Suggested EV @ ISO100 for 51.3 Lux incident light = 4.4
Closest Planckian temperature = 5386K (Delta E 5.505304)
Closest Daylight temperature = 5499K (Delta E 2.036443)
Color Rendering Index (Ra) = 94.1
And the attached plot of the spectrum (which I can't seem to preview so I don't know if it will show up or not...)
UV cut means it won't measure into the UV spectrum (partly that is the LED and partly it is the receive path). The plot shows what it measures.
Anyway, just giving you some options. I spoke up since it seemed you were heading down an unnecessarily expensive path.
-
Spotread produces this:...
Yeah, this looks like all the info I need.
And the attached plot of the spectrum...
Is this a plot of a CFL? Spiky but still has a CRI of 94.1. Which makes me wonder a lot about the meaningfulness of CRI ratings. (The spectrograph of the CFL I displayed a few posts ago is very gappy, but is still rated at CRI 93.)
-
Yes, it is a CFL with daylight mixed in.
FWIW, here is a plot of a Cree 100 Lumen LED (flashlight) I have sitting on my desk (measured CRI 72).
I don't think CRI is very useful. Spectral plots tell the story.
-
I don't think CRI is very useful. Spectral plots tell the story.
It isn't useful and you're correct, the plots are.
-
Here's what I get from the office LEDs. Nothing special, about 2 years old and purchased at either Home Depot or Lowes, can't remember.
(http://www.digitaldog.net/files/BableColor_LED_office.jpg)
-
If anyone's interested, here's a set of plots of a Philips Hue variable color LED bulb set to various various color temperatures.
-
http://www.olino.org/advice/us/overview
Repeat the jobs they have done so thoroughly?
BTW: probably expensive but a LED with 93 CRI by Philips for medical tasks:
http://www.burtonmedical.com/products/examination-lights/nova-exam-led/
--
Met vriendelijke groet, Ernst
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
January 2014, 600+ inkjet media white spectral plots.
-
http://www.olino.org/advice/us/overview (http://www.olino.org/advice/us/overview)
Repeat the jobs they have done so thoroughly?
It looks like they have only tested consumer bulbs and not on LEDs that are designed for photographers (http://www.bhphotovideo.com/c/search?Ns=p_PRICE_2|0&ci=12248&setNs=p_PRICE_2|0&N=3988592074&srtclk=sort), which remain an unknown wilderness. (Until somebody buys one and tests and reports on it.) And I don't see many (any?) bulbs that are for sale in stores in the US.
But thanks for the link. I did not know about this site. There is lots of information there. It is nice that they also include regular tungsten bulbs for reference.
-
It looks like they have only tested consumer bulbs and not on LEDs that are designed for photographers (http://www.bhphotovideo.com/c/search?Ns=p_PRICE_2|0&ci=12248&setNs=p_PRICE_2|0&N=3988592074&srtclk=sort)
Where do they specify the output? Are they just super bright but using similar technology to tested consumer blubs?
-
The typical 80 CRI (useful value or not) seem to be a glued fact to almost any LED I've measured.
The spectral curve that Andrew shares is exacly the same I get for both cheapos as well as "better brand" LED's. At the moment I'm on my way to check a Ferei divinglamp (to be combined with a GoPro cam) which holds 3 LEDS (the extremely powerful Cree XLamp XM-L2 LED). The diving lamp has three settings and the most powerful setting deliver about 2900 lumen! Un piccolo bastardo di luce! Madonna! You can get a nice suntan even underwater.
Italian or Finish site, make you choice ;)
http://www.fratellicordi.it/1/w163_7401102.html
http://www.valostore.fi/tuote/sukelluslamppu_ferei_w163/
(Some people build divinglamps based on Soluxbulbes, but 50 W on a 12 V battery…not surprised that the LED is a real gift nowdays)
Ferei's own spectral information for the LED mentioned, show almost exacly the same spectralcurve that Andrew posted.
http://www.cree.com/led-components-and-modules/products/xlamp/discrete-directional/xlamp-xml2
It has been a lot of talk about that little LED, which can be bought for 8 USD each for making your own stuff if you know how to deal with it:
http://www.ledgroupbuy.com/cree-xm-l2-white/
http://www.ledgroupbuy.com/exotic-leds/
Some people use this LED and combine it with other singlecolors LED's to maximize spectral enviroment for reef /aquariums. These guys measure spectras like maniacs to tune light in for corals and all the rest for an optimium environment. You see, geeky crazy people everywhere. Who cares about a shitty aquarium? I don't get it. But I don't mind the info though!
http://reefbuilders.com/2012/12/11/xml2-crees-highest-performing-led/
http://playsofrays.blogspot.in/2012/12/leds-made-easy.html
http://reefbuilders.com/2014/01/30/giesemann-futura-s/
I doubt that spectral distribution would be different with any of the Photo panels. The choice of CCT seem to be the only option. Anyone measured any of the photopanels in regards of spectra? Another question – or more of a wish – …. if it's possible to convert a Solux to a LED….. shoulden't be impossible, no?
Off topic: does anyone know what part of the spectral distribution that gets lost under water and which is enhanced? I'm talking Maldive's water now. Perhaps a dip in the cyan part of the spectracurve (of a LED lamp) would be a good thing in that water nightdiving? I'm not sure how far from the lightsource the "filtering" from the water sets in on the spectral distribution of any light. Would be nice to know. Anyone at the Maldives here at this board? (Ah… so you've got better things to do there – I don't blame you)
Mi
-
...Anyone measured any of the photopanels in regards of spectra?...
Well, I think that is the question of the thread...
Spectra of conventional LED's are well known and show the characteristic we published earlier.
Personally, I doubt the "photography LED panels" are much better. I say that because none seem to specify the output spectra. I can measure the spectra and would be happy to do that if someone local (Vancouver, BC) had one to measure...but I don't expect to buy one and measure it any time soon (given my low expectations and their high prices).
The TW Cree bulb mentioned earlier should be an improvement. But based on the patent application they are very focused on raising the CRI number only (not producing a good spectra!). The patent is focused on a single very narrow width notch filter (10-20nm only) optimally placed at a wavelength to generate maximum boost in the CRI number. Of course they have weasel words to generalize the patent but it is clear that isn't their focus. So the TW bulb isn't likely good in absolute terms although probably better in relative terms. That's just IMO, if I can get one to measure I'll do it. It isn't available so far in the local HD stores.
-
It's a shame that CRI is used as it is, mostly for marketing to the unsuspecting buyer. It's much like lumen for selling projectors, megapixels for selling cameras, huge contrast ratio's for selling displays or the billon's of colors it is said to produce. All marketing, much not useful.
-
It's a shame that CRI is used as it is, mostly for marketing to the unsuspecting buyer. It's much like lumen for selling projectors, megapixels for selling cameras, huge contrast ratio's for selling displays or the billon's of colors it is said to produce. All marketing, much not useful.
The olino.org site gives at least a good explanation of the CRI flaws and adds enough other information to judge the spectral distribution of a high CRI lamp.
The development that might give us full spectrum LEDs is the one done by Rosenthal at Vanderbilt Uni.
http://news.vanderbilt.edu/2012/05/quantum-dots/
Philips meanwhile makes hybrid LED Fluorescent tubes that could become high CRI lamps in time. Expertise enough there.
--
Met vriendelijke groet, Ernst
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
January 2014, 600+ inkjet media white spectral plots.
-
The olino.org site gives at least a good explanation of the CRI flaws and adds enough other information to judge the spectral distribution of a high CRI lamp.
They reference a CIE Technical Committee studying this question: "The result of the three investigations show that there is a low to no correlation at all between the computed CRI and the visually determined differences in color. The TC (CIE Technical Committee) comes to the following remark: the CIE CRI is in general not applicable for a ranking on color rendition when white led sources are part of the observed illuminants."
The CIE Technical Committee apparently recommended developing a new color rendering index standard.
-
...the CIE CRI is in general not applicable for a ranking on color rendition when white led sources are part of the observed illuminants.
Ya think? In case the quoted statement is a bit opaque, here are two graphs from the referenced page:
(http://www.olino.org/wp-content/uploads/2009/09/400x211xnpl_lev23_lmp91_powerspectrum_at_1m_distance-400x211.png.pagespeed.ic.cSMv6AU7Ru.png)
Spectrum of light with a color temperature of 3084 K and CRI = 94
(http://www.olino.org/wp-content/uploads/2009/09/400x211xnpl_lev27_lmp105_powerspectrum_at_1m_distance-400x211.png.pagespeed.ic.6ccOKpH1Xh.png)
Spectrum of light with a color temperature of 3076 K and CRI = 83
http://www.olino.org/us/articles/2009/11/30/a-close-look-at-the-color-rendering-index-cri-or-ra (http://www.olino.org/us/articles/2009/11/30/a-close-look-at-the-color-rendering-index-cri-or-ra)
Meet the new boss,
Same as the old boss.
Won't get....
Ah, you know.
-
The test panel results quoted in the link had a link to the PDF of the test conditions The test was done with equal 200 Lux illumination on the objects for every Kelvin grade choice of lamps. The PDF did not refer to the Kruithof curve of preferred Kelvin grades for different light levels and there was no compensation for that phenomenon in the test . The low rating of the tungsten and halogen lamps shown there is more a (negative) confirmation of Kruithof's discovery 80 years ago than bringing CRI to its knees. Not that the CRI index is good but this test was worse.
http://www.olino.org/us/articles/2009/11/30/a-close-look-at-the-color-rendering-index-cri-or-ra
--
Met vriendelijke groet, Ernst
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
January 2014, 600+ inkjet media white spectral plots.
-
Um, do you know of anybody that is measuring LED lamps? The only reason I want this is because nobody else is showing spectral plots and CRI ratings of LEDs. We all pretty much know what CFLs are good for by now but LEDs are a wild card. (i.e., some of them now have smooth spectrums, just like tungsten. But maybe not shaped quite right. Which may be fixable with a Rosco gel. Maybe.)
Thanks.
I don't think anyone is interested and CRI isn't the most reliable way to measure light quality. The usual consumer doesn't care about light quality. The best LEDS are from Sharp (Japan), they are called Mega Zenigata. Have fun.