This is a list of notable changes on the fluxometer/rainbow website.
June 2024: Manchester model
- “As bright as daylight” now uses a model derived from the TM Brown et al. Manchester Recommendations paper. This model is slightly more sensitive than our previous one (based on Zeitzer 2000).
- Added some new filters, mostly eyewear
March 2021: CS 2.0
- Added new mode for the Rea 2021 “CS 2.0” model. Don’t rely on this yet, it is not verified.
December 2020: comparison to a blackbody (melanopic)
- A minor feature appears under “CCT” now: we compare your illuminant to a blackbody of the same CCT, and say whether it makes less (or more) melanopic content than a planckian radiator would.
October 2020: upload your own files
-
We’ve created a new format for writing an SPD in a URL. URLs can only be about 2Kb, so just writing them in the usual way doesn’t fit, and that’s why we made this new format. Read about it here: spdurl.
-
You can also drag and drop your own CSV onto the webpage directly, and then share the spdurl. Here’s a sample CSV. You can drag it to our main page here.
-
You can view any spdurl-formatted spectrum using our “data” URL: https://fluxometer.com/rainbow/#!id=data/[spdurl]
.
-
Also, the Windows f.luxometer software has been fully updated to support CIE S 026, and it has a utility to upgrade existing CSVs.
July 2020
CIE-compatible interpolation
-
We upgraded all upsampling operations (e.g., 10nm to 1nm) to CIE-compliant Sprague interpolation, per Westland 2012.
-
Results may shift, but likely by a tiny amount.
May 27 2020
Changing units: CIE S 026
-
The new default metric is now CIE S 026 Melanopic EDI. This is a major change, as it makes the numbers 10% smaller.
- CIE S 026 gives us two ways to talk about the melanopic response:
- using “effective watts” (the portion of a light’s power that is absorbed)
- by considering how much Daylight (D65 lux) would be needed to elicit the same response
-
We are using method 2, “effective daylight illuminance” (EDI for short) metric as our main result, because we think it translates best to everyday use. It is, after all, simply daylight lux, a measurement most people know.
-
The Lucas 2014 paper used Illuminant E (5455K) rather than D65 as its reference, which is the main thing that changed.
-
We added an option to show the LRC CS metric for any light source.
-
Our desktop software has not yet incorporated these new metrics, but we will soon.
- All this means: where we used to say “melanopic lux” we now say “melanopic EDI (lx)”, and the values are 10% smaller. Some are also using “mEDI” when space is limited, but it means the same.
About S 026
-
We expect the first metric (“effective watts”) will be used frequently in research applications, and you can find this information in our Circadian/Color tab.
-
Also, you can choose the “Lucas 2014” metric using “Melanopic (Lucas)” in “Action spectrum”, and we’ll continue to show these values, for use with older published literature.
-
The cone responses are also updated in CIE S 026, using Stockman and Sharpe cone fundamentals, rather than the approximations used in Lucas 2014. We have indicated these as EDI also, so you will see “S-cone EDI (lx)”.
-
We have scaled M/P to correspond to D65 effective illuminance vs. measured illuminance (so D65=1.0 M/P).
On the way to consensus standards
-
Because the new CIE standard follows international rules about units, we should avoid saying “melanopic lux” (because lux is a unit that cannot be modified). Instead we use “effective illuminance” which means “this has the same absorption rate as daylight would at this lux value”.
-
There are other at least four different proposals about what M/P should represent. More changes are likely as standards are written.
-
It has been suggested that we can use the shorthand “EDI” for “effective daylight illuminance” and we think this is the most likely abbreviation to catch on, because it is possible to put this many letters on a handheld meter.
2018-2019
- Added total quantal units
- Added CIE S/026 (preliminary, not main metric yet)
- Added PPF (power units)
- Added VIIRS (satellite sensitivity)
- Added Bierman eyelid (via Manuel Spitschan code)
Nov 2016
- Added Quantal melanopic units
- Added preliminary bilirubin action spectrum
May 16 2016
- Added Macular pigment filter (based on Snodderly) for MPOD 0.5
Aug 25 2015
- Added S/P ratio and Scotopic lumens
Aug 3 2015
- Added NOAA/NGDC data (linked below), ReTimer, BluTech and TechShield
July 17 2015
- Added SI units per CIE TN 003:2015 (also linked from references)
July 3 2015
- Added photosynthesis via DIN 5031-10
July 1 2015
- Added tabs and “compare” feature
- Renormalized Brainard so illumE matches Lucas
June 15 2015
- Fixed bug that prevented Brainard action spectrum from working
- Added Kindle Paperwhite (full and half dimming), Candle, and Reduce Whitepoint tests
- Represented M/P as ratio rather than percentage (more like S:P)
- Added some references
- Text edits
June 7 2015 Initial release.