I was reading about Kruithof's graph, dated 1947, on the matching of illumination level and color temperature for pleasant indoor lighting, and found it is considered obsolete / incorrect by many. This open-access paper concludes illuminance is way more crucial than color temperature.
ABSTRACT
Kruithof’s graph identifies combinations of illuminance and correlated color temperature (CCT) alleged to yield pleasing visual conditions for interior lighting. Though in research terms the support provided by Kruithof is insufficient, it is widely cited as a design rule and has been the focus of many experimental studies despite evidence against Kruithof since at least 1990. The current article examines the trends displayed in those studies considered to provide credible evidence: these do not support Kruithof. For pleasant conditions, these data suggest only avoiding low illuminances and do not favor any CCT.
This is quite in line with my recent observations, that the same fluorescent tubes hurt if run at low power / low illuminance, and do not cause major eyestrain / fatigue at high power in brighter rooms. It sounds a bit simplistic though, as my symptoms are pronounced and their onset immediate. I surfed thru the article and could not clearly understand what the metrics employed for judging pleasantness are.
Just shooting in the dark. I am pretty sure you have already explored the illuminance variable, @KM?