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
To elaborate: I'm mostly curious what it means in the context of PhpStorm. TTR being 20 minutes to do what exactly? Before indexing is done? To run static analysis on a project?
Even then, 20 minutes is a huge amount of time that doesn't seem realistic
That one, from a first time opening a the project. More specifically at which point code completion is fully functional (the project being used for testing is Laravel based), PhpStorm will start up faster then that, but it has various stages where different parts of code completion gradually comes online.
Even then, 20 minutes is a huge amount of time that doesn't seem realistic
It's CPU time, not wall time. I found it a more fair yard stick since else it would be a reflection of my hardware more then anything else. Using CPU time means it's proportionally transferable to other systems. It also reflect the "jet engine" effect, 22m spread over 16 cores will probably serve you 2 minutes of being very very hot.
One point where PhpStorm is being unfairly measured here is that I couldn't easily separate the language engine from the rest of the application, but I am working on that since I got some helpful feedback from JetBrain regarding that, but the difference isn't more then 12% from what I saw.
Activity
brendt commentedon Jul 2, 2026
To elaborate: I'm mostly curious what it means in the context of PhpStorm. TTR being 20 minutes to do what exactly? Before indexing is done? To run static analysis on a project?
Even then, 20 minutes is a huge amount of time that doesn't seem realistic
AJenbo commentedon Jul 2, 2026
That one, from a first time opening a the project. More specifically at which point code completion is fully functional (the project being used for testing is Laravel based), PhpStorm will start up faster then that, but it has various stages where different parts of code completion gradually comes online.
It's CPU time, not wall time. I found it a more fair yard stick since else it would be a reflection of my hardware more then anything else. Using CPU time means it's proportionally transferable to other systems. It also reflect the "jet engine" effect, 22m spread over 16 cores will probably serve you 2 minutes of being very very hot.
One point where PhpStorm is being unfairly measured here is that I couldn't easily separate the language engine from the rest of the application, but I am working on that since I got some helpful feedback from JetBrain regarding that, but the difference isn't more then 12% from what I saw.