3 ms·

The 100 Millisecond Rule

5 points by mgav 6y ago

"The 100 millisecond rule: every interaction should be faster than 100ms. Why? Because 100ms is the threshold where interactions feel instantaneous. The team at Superhuman obsesses about speed and uses this rule to inform every single product decision..." source: https://www.typeform.com/blog/inspiration/superhuman-product-market-fit/

1cvmask 6y ago
That is part of the new norm of UX. When it is “slow” use an activity wheel.
- rolph 6y ago
  perhaps an on demand heads up, such as click on the spinner and get a fly out stating whats happen[ing][ed] eg files located...3 files parsed...data moving to local storage...cleaning up..
patatino 6y ago
I think context matters a lot, if I know a lot of stuff is happening I don't care about waiting a couple of seconds, but when something should be instant it gets annoying waiting for even one second.
eesmith 6y ago
Other than that throw-away comment, nothing in the article was about achieving sub-100m UX performance or measuring the effectiveness of decreasing latency. Quoting https://danluu.com/input-lag/ https://danluu.com/input-lag/ , "The most commonly cited document on response time is the Nielsen group article on response times, which claims that latencies below 100ms feel equivalent and perceived as instantaneous." [Typos fixed.] That article is at https://www.nngroup.com/articles/response-times-3-important-limits/ https://www.nngroup.com/articles/response-times-3-important-... . Note that the danluu essay continues with "One easy way to see that this is false is .." and that section of the essay starts with "For very simple tasks, people can perceive latencies down to 2 ms or less", with a link to the research. FWIW, the original literature reference is likely: Miller, R. B. Response Time in Man-Computer Conversational Transactions. In Proceedings of the December 9-11, 1968, fall joint computer conference, part I; ACM: San Francisco, California, 1968; pp 267–277. Under "Topic 1. Response to control activation", Miller comments cases where response should be in "No more than 0.1 second", and "no more than 0.1 to 0.2 seconds".
flurly 6y ago
Everything is a tradeoff. For instance if you think about optimistic mutations - where when you post on FB, it instantly reacts and moves on, the tradeoff is speed for a potentially more confusing user flow in the future (oh whoops our backend was down, and that post you made 5 minutes ago, yeah that's not around anymore). Blanket philosophies like 10,000 hours or 100 millisecond, boil all of the nuance out of these tradeoffs. I'm a bit torn as I do see the benefit of simplicity, but am also a big proponent of thinking from first principles.
psyklic 6y ago
I am suspicious of this 100ms rule. TVs operate at ~30 frames per second (33ms/frame), and humans can see obvious differences (e.g. when scrolling fast on smartphones or playing a FPS game) even at 60 fps (17ms/frame). The Remarkable 2 digital writing tablet has a 21ms writing delay that is still noticeable (but only slightly).

5 points by mgav 6y ago

1cvmask 6y ago

That is part of the new norm of UX. When it is “slow” use an activity wheel.

rolph 6y ago
perhaps an on demand heads up, such as click on the spinner and get a fly out stating whats happen[ing][ed] eg files located...3 files parsed...data moving to local storage...cleaning up..

patatino 6y ago

I think context matters a lot, if I know a lot of stuff is happening I don't care about waiting a couple of seconds, but when something should be instant it gets annoying waiting for even one second.

eesmith 6y ago

Other than that throw-away comment, nothing in the article was about achieving sub-100m UX performance or measuring the effectiveness of decreasing latency. Quoting https://danluu.com/input-lag/ https://danluu.com/input-lag/ , "The most commonly cited document on response time is the Nielsen group article on response times, which claims that latencies below 100ms feel equivalent and perceived as instantaneous." [Typos fixed.] That article is at https://www.nngroup.com/articles/response-times-3-important-limits/ https://www.nngroup.com/articles/response-times-3-important-... . Note that the danluu essay continues with "One easy way to see that this is false is .." and that section of the essay starts with "For very simple tasks, people can perceive latencies down to 2 ms or less", with a link to the research. FWIW, the original literature reference is likely: Miller, R. B. Response Time in Man-Computer Conversational Transactions. In Proceedings of the December 9-11, 1968, fall joint computer conference, part I; ACM: San Francisco, California, 1968; pp 267–277. Under "Topic 1. Response to control activation", Miller comments cases where response should be in "No more than 0.1 second", and "no more than 0.1 to 0.2 seconds".

flurly 6y ago

Everything is a tradeoff. For instance if you think about optimistic mutations - where when you post on FB, it instantly reacts and moves on, the tradeoff is speed for a potentially more confusing user flow in the future (oh whoops our backend was down, and that post you made 5 minutes ago, yeah that's not around anymore). Blanket philosophies like 10,000 hours or 100 millisecond, boil all of the nuance out of these tradeoffs. I'm a bit torn as I do see the benefit of simplicity, but am also a big proponent of thinking from first principles.

psyklic 6y ago

I am suspicious of this 100ms rule. TVs operate at ~30 frames per second (33ms/frame), and humans can see obvious differences (e.g. when scrolling fast on smartphones or playing a FPS game) even at 60 fps (17ms/frame). The Remarkable 2 digital writing tablet has a 21ms writing delay that is still noticeable (but only slightly).