I was wondering if anyone toying with .NET 2.0 and recompiled their applications with it is noticing slowdowns. It seems like something's wrong with some BCL functions. While I was writing some optimized byte array converters I noticed that some .NET 2.0 base classes were slower.
Some raw output, while fiddling with some stuff:
ToInt32 Tests
============================
Filling data...
Testing BitConverter...
BitConverter took 1515,625 ms
Testing Superluminal...
Superluminal took 562,5 ms
Data check... OK!
ToInt64 Tests
============================
Testing BitConverter...
BitConverter took 1468,75 ms
Testing Superluminal...
Superluminal took 984,375 ms
Data check... OK!
ToInt32 Tests
============================
Filling data...
Testing BitConverter...
BitConverter took 3218,75 ms
Testing Superluminal...
Superluminal took 656,25 ms
Data check...
OK!
ToInt64 Tests
============================
Testing BitConverter...
BitConverter took 3156,25 ms
Testing Superluminal...
Superluminal took 984,375 ms
Data check... OK!
Notice how the BitConverter class, which is internal to the CLR (means no managed code, but natively built into the CLR), takes almost double the time. Same applies my own ToInt32 function, which takes a tad longer under .NET 2.0. So is anyone witnessing similar stuff?
Question
Glowstick
I was wondering if anyone toying with .NET 2.0 and recompiled their applications with it is noticing slowdowns. It seems like something's wrong with some BCL functions. While I was writing some optimized byte array converters I noticed that some .NET 2.0 base classes were slower.
Some raw output, while fiddling with some stuff:
Notice how the BitConverter class, which is internal to the CLR (means no managed code, but natively built into the CLR), takes almost double the time. Same applies my own ToInt32 function, which takes a tad longer under .NET 2.0. So is anyone witnessing similar stuff?
Meh.
Link to comment
Share on other sites
5 answers to this question
Recommended Posts