The dedicated server doesn't time scale the same way the in-game server does. I believe it also depends on what scaling is set in the track .GDB.
OK, I had not looked at this part of the .gdb when choosing tracks to test with. Thanks for the info Dave, I will investigate further.
Not a criticism as I can only imagine the amount of work put in to creating changeable weather, but imho it has to be more reliable if we are going to use it.
I totally agree. The weather system is totally missing the point of its original intended purpose at the moment.
Don't worry, it will not be used in championship races until/if it is sorted properly. I can handle criticism as long as its constructive or highly humorous.
I like the fact that the changeable weather is unreliable. We don't know what the real weather will bring although, we have a good idea within 30 minutes. In the real world, guys don't dive to the pit lane for wets till it is raining.
I used to enjoy the races when we used Dave's system but, some numpty started a conversation about how it could be improved:
https://www.simracing.org.uk/smf/index.php?topic=7297.0In the end it was decided that it was better to have a bit more predictability in the weather that would be used so that people did not have to practise for all sorts of different weather possibilities. Also the races could be too much of a lottery some times for peoples liking.
After running the actual weather file used in last nights race on my own server, I discovered that I had the same looking sky and no overcast clouds. What happened was the weather system decided it will not rain but, we would have overcast conditions. The conditions value that it gave for overcast was 60 but, it should have been 62.5 - 65. At 60 we still get blue ish sky. It totally changes to overcast at 62. I am sure that some of the add on tracks are a bit tighter with the conditions values.
Conclusion; conditions values still need a bit of tweaking for add on tracks.
Also, the forecast was miles off. If we had overcast conditions then it should have reported light rain at the most, not normal rain which is what it did report.
Conclusion; Level of potential error needs to be reduced but, there may be more to this issue than that. Looking into it now.