Approximating color temperature with smart phones and camera

46 posts / 0 new
Last post
djozz
djozz's picture
Offline
Last seen: 4 hours 58 min ago
Joined: 09/07/2012 - 17:04
Posts: 17620
Location: Amsterdam

Good stuff contactcr! I like that a few more people on BLF have the opportunity to measure the colour information of our flashlights.

Bob_McBob
Online
Last seen: 5 min 1 sec ago
Joined: 08/14/2016 - 04:53
Posts: 589
Location: Canada
contactcr wrote:
* It said (Caution) in the output on some lights/modes but still gave me somewhat reasonable output. Should I be concerned?

From the spotread documentation. I think this basically means any really rosy or green LED will always have the “caution” warning.

Quote:
If the notation (Caution) is displayed after the CRI or TLCI, then this means that the the spectrum white point is beyond the standard tolerance distance from the black body or Daylight illuminant locus. This may make the CRI or TLCI measurement less accurate, and indicates that the illuminant may be a lower quality source of light.
bmengineer
bmengineer's picture
Offline
Last seen: 2 months 1 week ago
Joined: 01/26/2018 - 11:38
Posts: 1030
Location: Ontario, Canada

I downloaded the "White Balance Color Temp Meter" to my Pixel 3 and tried it out. Readings were taken on lower modes, in a dark room, with a sheet of printer paper as the backdrop.

  Rovyvon Aurora A8 Olight S1R Nitecore TUP, LH351D mod FW3A (3D) Nitecore P18, XHP35 Hi swap Armytek Elf C2 (Warm)
Expected 5000 6000 5000 5000 4500 4000
Test 1 4600 5320 4660 4540 3810 3620
Test2 4510 5380 4440 4440 4300 3750
Average 4555 5350 4550 4490 4055 3685
Difference -445 -650 -450 -510 -445 -315

I have no source for the Elf C2 CCT and guessed at the expected value. If I shift all the values up by 500, using the average of two readings:

  Rovyvon Aurora A8 Olight S1R Nitecore TUP, LH351D mod FW3A (3D) Nitecore P18, XHP35 Hi swap Armytek Elf C2 (Warm)
Expected 5000 6000 5000 5000 4500 4000
Corrected average 5055 5850 5050 4990 4555 4185

These now fall within about 150K of the values I was expecting. Taking more samples would likely improve the consistency as well. Very promising!

Find all my reviews of flashlights and more gear at www.bmengineer.com

contactcr
Offline
Last seen: 12 hours 36 min ago
Joined: 05/19/2017 - 18:52
Posts: 3590
Location: US

I dont have any of these lights but the real measured CCT is never even numbers.

For example the Wuben with 5000K 90CRI LH351D: CCT = 4816K

So your numbers could already be closer than you think.

bmengineer
bmengineer's picture
Offline
Last seen: 2 months 1 week ago
Joined: 01/26/2018 - 11:38
Posts: 1030
Location: Ontario, Canada

contactcr wrote:
I dont have any of these lights but the real measured CCT is never even numbers.

For example the Wuben with 5000K 90CRI LH351D: CCT = 4816K

So your numbers could already be closer than you think.

Yes, of course. I know, for example, those 4500K XHP35 Hi LEDs tend to measure cooler than advertised. I mostly want to be able to loosely quantify CCT for my reviews, and for that I think a 250K or 500K range gives people enough of an idea.

Find all my reviews of flashlights and more gear at www.bmengineer.com

maukka
maukka's picture
Offline
Last seen: 2 weeks 4 days ago
Joined: 12/31/2015 - 04:15
Posts: 2175
Location: Finland
bmengineer wrote:
Yes, of course. I know, for example, those 4500K XHP35 Hi LEDs tend to measure cooler than advertised.

Most flashlights have a warmer hotspot and cooler spill. Averaged and integrated over the beam the results are usually spot on at binned CCT.

Kame Sennin
Kame Sennin's picture
Offline
Last seen: 11 months 7 hours ago
Joined: 09/24/2018 - 14:38
Posts: 347
Location: France

I might get a good price on a used xrite I1 Pro (looks like the one on the firsts pictures from OP as far as i can see)

Is that one still recommanded ?

maukka
maukka's picture
Offline
Last seen: 2 weeks 4 days ago
Joined: 12/31/2015 - 04:15
Posts: 2175
Location: Finland

Yes. It’s still good for LED measurements although the i1Studio/Colormunki Photo may be cheaper and practically just as good for what we do. I like the Babelcolor CT&A, which doesn’t work with the i1Studio.

Kame Sennin
Kame Sennin's picture
Offline
Last seen: 11 months 7 hours ago
Joined: 09/24/2018 - 14:38
Posts: 347
Location: France

That one was cheaper than I1studio /Colormunki Photo so i am testing it now Smile

 

I see in the OP that you were using spotread with -a (ambient), was that with the I1pro ?

I am asking because despite having an ambient measurement cap with my I1pro, Argyllcms returns a "Requested ambient light capability, and instrument doesn't support it." error when i try to use the -a instead of -e

An other question : Do you use the ambient measurement cap (ref 609192) ? If i don't use it the sensor is easily saturated but when i try to use it the CCT seems to be off

 

maukka
maukka's picture
Offline
Last seen: 2 weeks 4 days ago
Joined: 12/31/2015 - 04:15
Posts: 2175
Location: Finland

If you run spotread.exe -v, what does it say?

Kame Sennin
Kame Sennin's picture
Offline
Last seen: 11 months 7 hours ago
Joined: 09/24/2018 - 14:38
Posts: 347
Location: France

The label says x-rite, eye-one Pro, rev D

 

maukka
maukka's picture
Offline
Last seen: 2 weeks 4 days ago
Joined: 12/31/2015 - 04:15
Posts: 2175
Location: Finland

That is strange. Your meter is only a year older than mine, but mine has firmware 506 and reports Measure Ambient ?: Yes.

Check the meter’s functionality with x-rite’s i1diagnostics: https://www.xrite.com/service-support/downloads/i/i1diagnostics_v4_1_2

This will install the x-rite drivers and you have to go back to argyll drivers manually afterwards from device manager.

If this doesn’t help, contact ArgyllCMS’s creator at graeme@argyllcms.com He’s very helpful.

Kame Sennin
Kame Sennin's picture
Offline
Last seen: 11 months 7 hours ago
Joined: 09/24/2018 - 14:38
Posts: 347
Location: France

 

Thanks for your help!

I did run the diagnostics pgm and everything seems to be fine except that the "ambient" doesn't appear in what i think is functions list :

[ Emission  Reflectance  Reflectance Scan ] but no ambient !

I will try to ask Greame if he knows something about special versions (according to Argyll documentation the B, C & D rev are supposed to have the ambient measurement included) but i fear that i'll have to take the I1pro back to the seller

 

Wild guess : From what i understand about ambient measurement mode it is supposed to use a calibration table matched to the cap that is furnished with the i1pro. Maybe the calibration table in my I1 is damaged, causing it to disable that function ?

Does the number on your cap match the sn from your I1 ?

maukka
maukka's picture
Offline
Last seen: 2 weeks 4 days ago
Joined: 12/31/2015 - 04:15
Posts: 2175
Location: Finland

Ah, yours is the rarer no (UV) filter version of the i1Pro. That’s probably why it doesn’t work.

Mine reads: Measurement Modes: [Reflectance Spot, Reflectance Scan, Emission, Ambient Light Spot, Ambient Light Scan]

Kame Sennin wrote:

Does the number on your cap match the sn from your I1 ?

Yes. And also the white tile.

Kame Sennin
Kame Sennin's picture
Offline
Last seen: 11 months 7 hours ago
Joined: 09/24/2018 - 14:38
Posts: 347
Location: France

Ok must be that !

What got me wondering was that the ambient measurement cap was included, but obviously the sn does not match the I1pro one and according to the seller he previoulsy had an other I1 that was discarded and the cap might be from this one.

Well ... failed attempt but at least now i know what to look for and the seller (nice and helpfull guy) agrees to take it back so it's ok.

Thank you for helping sorting this out

Pages