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
The .NET Core app is using about 2-3x more CPU time on startup and from my measurements it is taking about 400-500ms more to show the window. The render time seems to have been fixed but startup/construction time still seems to have regressed significantly.
With an empty WPF project on .NET Core and .NET Framework the results are similar.
The text was updated successfully, but these errors were encountered:
jtg15
changed the title
Startup Performance Worse than .Net Frameworrk
Startup Performance Worse than .Net Framework
Mar 8, 2020
jtg15
changed the title
Startup Performance Worse than .Net Framework
Startup Performance Worse than .Net Framework 4.8
Mar 8, 2020
This is a function of the difference between ngen and ReadyToRun/crossgen. We already have a tracking bug in WPF (#1748) for this and a companion bug here: dotnet/runtime#13339.
I'm going to close this as a duplicate, please add your comments and upvotes to the dotnet/runtime bug.
.NET Core 3.1.101
Windows version 1909
Does the bug reproduce also in WPF for .NET Framework 4.8?: No
I noticed that .NET Core WPF feels noticeably more sluggish than when using .NET Framework WPF especially with startup times.
I searched around for some other issues related to performance and I found this #1549. Using the same repo here: https://github.com/rladuca/NetCorePerformanceTest, I profiled it and I screenshotted the results.
The .NET Core app is using about 2-3x more CPU time on startup and from my measurements it is taking about 400-500ms more to show the window. The render time seems to have been fixed but startup/construction time still seems to have regressed significantly.
With an empty WPF project on .NET Core and .NET Framework the results are similar.
The text was updated successfully, but these errors were encountered: