4 ms·

Roughly 1/3 of those would be annoying and upsetting, due to wasting the users' time when interacting with software. Otoh, some of these are nice examples of ju

• andybak 12y ago
  Ideally animations that take an amount of time will be used to mask essential delays rather than introduce additional ones. I don't know how much this is actually true in a lot of real world implementations of 'material design' style UI but I hope it to be so. Casually speaking - material apps don't feel any slower to me. And it's important to remember that 'perceived speed' is the only relevant metric when discussing usability.
  • tenfingers 12y ago
    Most animations nowdays don't mask any actual delay, but actually introduce new ones. Often, they cannot be skipped. Even more often, they're not properly chained, in a way that breaks the interaction if you're doing something else while the animation is still completing its cycle. If the animation is is actually masking a delay, you can bet the animation is not interrupted and/or sped-up when the data is ready. There are very few cases were animations provide visual cues which are actually an improvement. Very, very few. Touch interaction benefits for more visual feedback (such as sliding), but note that sliding in my eyes is not an animation as should be directly linked with the position of the finger. In the last years I've been aggressively disabling all kind of animations in software which serve no actual purpose. Sadly, this kind of stuff often cannot be disabled.
• dandelany 12y ago
  That's the point of experiments - implementing them to find out which are useful and which are annoying.