Re: [w3c/gamepad] Should fire events instead of using passive model (#4)

xingri left a comment (w3c/gamepad#4)

Based on our preliminary evaluation, we did not observe any functional regressions from moving to an event-based approach, other than the loss of touch-related events that has already been discussed.
We also compared the current polling-based approach with the event-based approach (without any coalescing mechanism) on Chrome running on Windows. Our tester did not notice any meaningful difference in input responsiveness or performance between the two approaches.
This leads us to suspect that there may already be synchronization, locking, or other buffering mechanisms within the browser implementation that mitigate potential event-rate concerns. However, this is only an observation from our testing, and we would be interested in hearing whether browser implementers have additional insights into how input events are handled internally.
From our perspective, the preliminary results appear encouraging, although broader testing across platforms and implementations would be valuable.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3c/gamepad/issues/4#issuecomment-4930223034
You are receiving this because you are subscribed to this thread.

Message ID: <w3c/gamepad/issues/4/4930223034@github.com>