Hard Light Productions Forums
Modding, Mission Design, and Coding => FS2 Open Coding - The Source Code Project (SCP) => Test Builds => Topic started by: portej05 on September 08, 2009, 05:21:07 am
-
This test is now closed. Thanks everyone!
In order to remain necessarily cryptic about what's changed with this one (it's a massively subjective test, hence why I'm putting it out to all of you), I'm not going to document what has changed.
Folks who were in IRC when this was discussed should not post here, or reveal what the test is about.
Just spend 5 mins playing the game on some of the heavier missions and let me know what you think.
Here they are:
Executable One
Executable Two
All will be revealed!
EDIT: Read above.
-
_Is_ one of the exes a control or are both different somehow from trunk?
-
I think at this point I should announce that this test is now closed :D
Thanks everyone who provided feedback.
It was testing the difference between a build with a global exception handler and a build without the global exception handler (_v1 was the trunk build, _v2 was my build).
I wasn't sure that it would make much difference, however The_Es first comment was 'v2 seems slightly faster', so I pursued a little further :D (don't ya love subjective tests - in all fairness to The_E, maybe some caching had kicked in somewhere :P )
For those who want technical information on how __try/__catch works in VS: Kevin Frei: How much does Exception Handling cost, really? (http://www.nwcpp.org/Downloads/2006/ehc.ppt)
EDIT: Should have mentioned that in the instance of a C++ exception throwing function (like 'new') things could have been a little slower on the prologue and epilogue of functions, but I'm still unsure as to how much