3 ms·
Have you read the research? It doesn't really support the claims people make about it. A good starter thread (albeit it's hard to follow all the replies in the
by Jach 6y ago
Have you read the research? It doesn't really support the claims people make about it. A good starter thread (albeit it's hard to follow all the replies in the nested conversations): https://twitter.com/danluu/status/1115707741102727168 https://twitter.com/danluu/status/1115707741102727168
- LocalPCGuy 6y agoI've read quite a bit of it. It really depends on the claim. Speed is generally faster with longer lines, although it does seem like there is an upper limit. But comprehension and subjective judgement (yes, that last is hard to quantify) seem to be best between 40-60 characters (actually less than the commonly recommended 65 as seen in the original post here). Also, it's generally not line-length, but characters per line, that matters, and keeping the visual angle from changing as the user reads. So it depends on what exactly is being measured, which is why I said the overall UX and not something more specific. This is a bit older, well prior to the days of 4K displays, but lots of interesting stuff nonetheless: http://images3.wikia.nocookie.net/__cb20060729105544/psychology/images/e/eb/Dyson,_M_C_(2004).pdf http://images3.wikia.nocookie.net/__cb20060729105544/psychol...