You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There's a slight difference in velocity reported by iOS's pan gesture recognizer's velocityInView vs our own VelocityTracker. Flutter's calculation is ~20% faster.
On Android, the velocity tracker is distinct from the gesture callbacks, but our flings are a bit slower.
Probably better to have a platform agnostic velocity and correct the difference in the friction simulations.
As of last time I checked in spring, it is still and issue. This and #13171 are the 2 remaining scrolling related issues I know of.
I left them deprioritized and not slated for Q3-Q4 since I don't think they're noticeable enough. Let me know if that should change or if anyone is blocked on it.
It's still and issue.
I feel ListView has too much friction when deceleration on iOS.
But https://github.com/xster/flutter-test/tree/master/t95_ios_settings_page performs well.
Is this my illusion?
I went deep into the source code, and they all used the same ScrollPhysic.
Really don't know why.
Flutter does not align with iOS when holding down the drag
They scroll stop at different times. Flutter stops earlier
As a iOS developer.
Based on my development experience, we hardly need a Cupertino UI.
But something is important:
Standard dimension. eg: NavigationBar with 44.0. Although I can use AppBar+PreferredSize to implement it.
The scroll experience.
The transaction. Especially pan to swipe back. I have tried it in t95_ios_settings_page, It's also a little weird.
Few users like a Cupertino UI. Because design should be diverse.
But most users hard to tolerate app with inconsistent interactions.
So I still prefer to use MaterialApp and I'm not going to use CupertinoApp.
I guess there will be a lot of iOS developers like me.
So I suggest that Flutter should pay more attention to interactions instead of CupertinoUI.
Users building custom designs is indeed our vision though we've recently focused on building an extensive set of Cupertino widgets based on our quarterly survey results. Note also that the Material/Cupertino interest ratio is also skewed between China and the rest of the market.
I do agree this is important however. We just haven't fine tuned it because no one noticed so far :) (as mentioned above). It shouldn't be difficult to do and it's in our plans for Q2. You can also send us a PR based on that comparison test.
This thread has been automatically locked since there has not been any recent activity after it was closed. If you are still experiencing a similar issue, please open a new bug, including the output of flutter doctor -v and a minimal reproduction of the issue.
Activity
Hixie commentedon Aug 29, 2017
I've tentatively put this at post-1.0 but we can bring this back if people notice.
xster commentedon Aug 29, 2017
Agreed. I think it's lower prio than the last 4 we dealt with so far.
eseidelGoogle commentedon Apr 6, 2018
This came up in a review from an internal team today. I presume this shows up in:
https://github.com/flutter/scroll_overlay
eseidelGoogle commentedon Aug 31, 2018
@xster do we know if this is still an issue? Has anyone played with scroll_overlay recently to tune our physics further?
xster commentedon Aug 31, 2018
As of last time I checked in spring, it is still and issue. This and #13171 are the 2 remaining scrolling related issues I know of.
I left them deprioritized and not slated for Q3-Q4 since I don't think they're noticeable enough. Let me know if that should change or if anyone is blocked on it.
molon commentedon Apr 10, 2019
It's still and issue.
I feel
ListViewhas too much friction when deceleration on iOS.But
https://github.com/xster/flutter-test/tree/master/t95_ios_settings_pageperforms well.Is this my illusion?
I went deep into the source code, and they all used the same
ScrollPhysic.Really don't know why.
eseidelGoogle commentedon Apr 10, 2019
https://github.com/flutter/platform_tests contains a project one can use to test ListView against an iOS scroll physics if you'd like to confirm.
molon commentedon Apr 11, 2019
Very useful tests.
Three problems can be clearly seen from the test:
As a iOS developer.
Based on my development experience, we hardly need a
CupertinoUI.But something is important:
AppBar+PreferredSizeto implement it.t95_ios_settings_page, It's also a little weird.Few users like a
CupertinoUI. Because design should be diverse.But most users hard to tolerate app with inconsistent interactions.
So I still prefer to use
MaterialAppand I'm not going to useCupertinoApp.I guess there will be a lot of iOS developers like me.
So I suggest that Flutter should pay more attention to interactions instead of
CupertinoUI.xster commentedon Apr 11, 2019
Users building custom designs is indeed our vision though we've recently focused on building an extensive set of Cupertino widgets based on our quarterly survey results. Note also that the Material/Cupertino interest ratio is also skewed between China and the rest of the market.
I do agree this is important however. We just haven't fine tuned it because no one noticed so far :) (as mentioned above). It shouldn't be difficult to do and it's in our plans for Q2. You can also send us a PR based on that comparison test.
1 remaining item
LongCatIsLooong commentedon Jun 26, 2020
Looked into this (prompted by #59623).
UIPanGestureRecognizeruses a far smaller sample size to calculate the velocity.Reduce damping in iOS scrollbehavior for list with differently sized …
github-actions commentedon Aug 17, 2021
This thread has been automatically locked since there has not been any recent activity after it was closed. If you are still experiencing a similar issue, please open a new bug, including the output of
flutter doctor -vand a minimal reproduction of the issue.