Deuteranopia: strong color tint towards pink in neutral shades of gray

[suit4]

After testing SD and comparing it to the output of the Photoshop color blindness simulation, we think that neutral grays should look the same in all simulations. The Deuteranopia setting seems to have a strong tint to pink in the neutral grays. See attached image bottom right.

[michelf]

This indeed appears suspicious. The filter algorithm Sim Daltonism uses has been developed a long time ago by the HCIRN (which no longer exists). Others tools based on the same algorithm will exhibit the same pinkish tint for light grays, for instance Coblis. I haven't investigated too much to find a solution, or how correct this is or not.

I could switch to another algorithm. I've been told once the algorithm I use is better than most though, despite its flaws. I certainly welcome opinions (especially scientific opinions) on this topic. If someone wants to share tweaks to the code I'll certainly look at the result too.

[habere-et-dispertire]

I have seen the same issue. Firefox is using a new algorithm for colour vision deficiency simulations. The matrix transformation work it is based on appears to be freely available and physiologically based:

https://www.inf.ufrgs.br/~oliveira/pubs_files/CVD_Simulation/CVD_Simulation.html
https://bugzilla.mozilla.org/show_bug.cgi?id=1655053

Perhaps this can provide some new direction.

[parkersweb]

I'm having a similar problem - originally wondered if this was related to the new MacBook Pro screen, or the fact that I'm using OSX Monterrey. Interestingly though using a screenshot of the same page and passing it through jsColorBlindSimulator http://mapeper.github.io/jsColorblindSimulator/ doesn't seem to suffer the same problem?

Original screenshot:

Output of jsColorBlind Simulator using protanopia setting:

(oddly the 'non-commercial only' algorithim is slightly more pink - but not as pink as the one below:

Screenshot of Sim Daltonism output:

Screenshot of Firefox emulation: