Page 1 of 1

Cars running consecutive races w/ Perfect-n

Posted: Wed Mar 23, 2011 11:20 pm
by Husker
We have a 6 lane track and use Perfect-n scheduling. We typically have 6-12 cars in each rank. The time between heats is rather long because we have to wait for cars to be returned to the staging table to be sorted. Obviously the time would be greatly reduced if the next heat cars were ready right after the current heat starts.

I considered reducing the number of lanes to get away from this problem. 4 lanes gets a lot of love around here so I worked though the number of cars to see where there isn't cars racing consecutive heats. At 16 cars, every other heat has a repeat and above 17 (from what I worked through) there aren't repeats.

So 4 lanes doesn't do much for us. I don't see any advantage in our case to switch to 4 lane racing over 6 lane. Am I missing something?

Also, does anyone know how many cars are needed for 3, 5, and 6 lane racing to stop having cars consecutive heats?

Re: Cars running consecutive races w/ Perfect-n

Posted: Wed Mar 23, 2011 11:32 pm
by Stan Pope
PPN has the option to sort the heats so as to minimize racing in successive heats. The down side is that the nice "sequential nature" of cars in lanes goes away.

Each combination of lane- and car-counts has its own characteristics.

If you are racing by time and do not want good opponent balance, you can opt for a different scheme. I personally like good opponent balance even with timed racing because it helps folks justify some of the otherwise incongruent results from timed racing.

Re: Cars running consecutive races w/ Perfect-n

Posted: Sat Mar 26, 2011 8:03 pm
by Husker
I will say that the sorting method used in GPDM works very well (I'm guessing that's your algorithm). I've looked at s few schedules for 6 lanes. Many of them have patterns like consecutive races always in certain lanes. When you know the pattern, it makes sorting for the next race a bit easier.

And I have to say I'm impressed by your 12 minute response time. At 1:30 AM!!¡!

Re: Cars running consecutive races w/ Perfect-n

Posted: Sat Mar 26, 2011 10:15 pm
by Stan Pope
Husker wrote:I will say that the sorting method used in GPDM works very well (I'm guessing that's your algorithm). I've looked at s few schedules for 6 lanes. Many of them have patterns like consecutive races always in certain lanes. When you know the pattern, it makes sorting for the next race a bit easier.

And I have to say I'm impressed by your 12 minute response time. At 1:30 AM!!¡!
That sorting algorithm is mostly the concept and effort of Cory Young and was integrated into both the Web Page version and DLL version. As for response time ... well, happily, I was here when the question arrived!

Re: Cars running consecutive races w/ Perfect-n

Posted: Mon Mar 28, 2011 9:52 am
by Rukkian
One option in GPRM that really helped us this year was the master scheduling. I don't think we had a single car that ran in 2 consecutive races all day, and was able to run all of the races on average in less than 40 seconds per heat. The dad that ran the cars back to the start got a very good workout!