I think he meant the way, the traffic lights switch to green, when you manually set a starting-point. Unlike the usual way, there is a 3 second countdown that makes the starting-procedure very predictable ...
So maybe give it a randomizer and keep the countdown "silent" ?!
Well, this isn't actually a bug, but a further request:
Is it possible to remove those tyre-stacks (and actually the cones too) underneath the highway in SO?
As you can see on the screenie, the area is perfectly drivable, so that won't be a limitation ...
That's the way, I do it now ... but actually this will produce an error, whenever a track (and thus it's length) is changed, as happened to BL1 not too long ago
Will have a look at it (had downloaded it even earlier but forgotten about it somehow :shrug.
Nevertheless, I'd appreciate an official list, as you are missing at least one of those in your file (see your thread) and god knows how many we haven't sumbled upon ...
Oh, just one more thing:
How about giving the textfile with all those options (basically those on post 1), if no or maybe "help" action is requested? that one maybe even without pubstat-key?
And I'd appreciate a list of all the error-messages (maybe as action=errorcodes ?)
Seems to me, IE with just above 2000 characters is the limiting factor (or was, not sure, how it is now with the new versions) as servers seem to be able to cope with those lengths: http://www.boutell.com/newfaq/misc/urllength.html
I actually like flame's middle one on the left!
It gets the modular approach, is "simple" though appealing.
And the "i" gives a real good logo/icon without loosing anything
Just got another idea:
How about a history of laps driven by a racer (giving a list containing server, track, car, laptime and timestamp for within some time from the query - say a day or 12 hours or so)
Just if it is possible that is ...
Oh, btw:
For some days each morning around 6am (CET) I'll get this from pubstat:
Well, what I would really like is:
- have action=fuel combined with action=pb, thus having both datasets with a single query
- have action=pb altered in a way, that outdated PBs are returned with their actual times instead of giving all zeros (perhaps have an "outdated flag" somewhere) as it can be seen on lfsworld
A real kickass-feature (though I doubt it would be implemented, but one should be allowed to dream) would be:
give pubstat an array of racers and fetch, whether they have done any laps (maybe even on a particular set of car/track combos) within a set threshold,
thus enabling one to update his/her database only, when it is neccessary.
(This would for example avoid me to fetch the hostlist every 2 minutes just to see, if any of the racers might need an update.)
Or (even more dreaming): Have only PB-data (PB and laps) of a racer gained up to a specific date (Though I'm quite sure, this won't be possible as it requires to store a racers pb-history (which would actually be veeeeeeeery cool))
... well ideally I'd take a combination of all of that ... as to be realistic, I think the first two could be worth a try anyway
Yeah ... I aggree, that that one might need a bit of imagination ... In retrospective I can't seem to understand, why I believed the inner section of those tetrahedron to actually be a prism ...
Well ... though I'm still quite fascinated by that polyhedron-approach, I did a quick shot at a similar thing to result in a prism ... and here it comes ... (the grey thingy in the middle obviously representing the prism core )
EDIT: Just put both together for a better comparison
Well, without all that lightning stuff and text, one could always shrink the "basic shape" down to a convenient size as 16 or 32 pixels ... something like those:
(as before mere "quickshots" just to get an idea of what i meant )
Hm ... somehow I was thinking of the prism actually being created by modules, plugins etc.
something like this ... only done by someone capable of doing it
Well ... the prism-thing there is actually an anti-prism due to being made out of tetrahedrons but well ... it's just an example ...
and obviously I wasn't capable of doing that rainbow-stuff in a manner that it "expands" ... you'll have to imagine that yourself ...
This morning there seemed to be some access-issues with lfsworld, as a pubstat-script of mine threw the following messages: (times are CET)
07:50: "no route to host"
07:55: "connection timed out"
08:00: "network is unreachable"
11:10: "no route to host"
I didn't check at 08:00, but on the event at 11:10, I checked lfsworld myself via webbrowser (and got a timeout), so I am quite sure, it was not a problem of my server ...
on any other attempt, everything was fine ...
So nothing severe, just wanted to make sure, you are aware of this
EDIT:
Well, actually it might be more severe than I thought, as the masterserver was down too a few minutes ago ...
Well, you could retrieve the wr alongside with the racername (which is provided) and afterwards do a seperate query on all that racer's hotlaps (?action=hl), where you pick out the one for your desired track/car, as there is a timestamp there.
Of course his produces an additional query for each wr (or at least each wr that is done by one, who didn't do another one), so it would be considerable keeping track of wrs and only fetching the timestamp, when one is changed ...
Not sure, if this the right area, but I'll give it a try
Playing a little with SMX-files I noticed something that i find quite strange:
Apart the fact, that checkpoint objects (start/finish or splits) are in most cases "areas" instead of mere "lines", which I might even understand, some of them appear to be "doubled". That means: For a given checkpoint you encounter on track there might be more than one checkpoint object in the SMX.
As an example I have attached an image of Blackwood where the objects lebelled "0" and "1" are the first split of BL1 & BL2 ("original direction"), "2" and "3" are split 2 of BL1 an "4" is start/finish for both.
(Don't be bothered with the red arrow. that indicates driving direction an the blue "strip" is the road of BL1)
Now the questions are:
1: Why are there multiple objects for a given checkpoint? I could even understand that, if it was for different tracks, but since even for split 2 of BL1 there are two it doesn't make sense. Or is that "normal" and "reversed"? But that would not explain, why there are checkpoints with only one object (not here apparentely, but for other tracks there are), nor would it explain, why for example start/finish of SO3 consists of 4(!) objects (or s/f of KY1 and KY2 being 3 plus one inside pitlane ...)
2: Is there any way to determine, which is the correct one to use for a given track and/or direction (especially if there are multiple objects for a singel checkpoint)? This information does not seem to be included either SMX or PTH-files ... Any other info is appreciated ...