Uh, how could it go by "desination" sector number?  There's only one destination in one path search.
Since I'm obviously still confused, I'll use an example.  The following two paths exist:
1 > 2 > 10 > 3247 > 3819
1 > 2 > 9 > 18328 > 3819
Which will it report?  The second has a lower sector number at the point of variation, but the first has a lower sector number at the point just before finding the target sector.  How does it build its open list?  If it's a queue, 9's adjacents will be inserted into the queue before 10's and 18328's adjacents will be checked before 3247's.  That's what I meant by "prefers lowest".  Works the sectors in numeric order.
However if the open list is implemented as a heap, it'd report the first, because 3247 and 18328 are both in the heap at the point it starts checking through the 4th-deep nodes and it'd check 3247 first and find THAT connection before 18328's.  It sounds like you're saying this is how it works instead of my original inference?

It's going from adjs out from the start, to the destination in a bfs. So if 9 is an earlier door than 10, the 9 would come first. Usually that's the case, but it isn't always the case... especially w/ edited warps.

In this case it would stack all of the adjs to 1, then all of the adjs to 2. Work thru them till it finds 9 and 10. It would cache all of the adjs to 9, and all of the adjs to 10. Work thru them till it reaches 18328 and 3247 respectively. Cache their adjs and as it caches the adjs to 18328, boom it as the path. Assuming that 9 comes before 10 as a door to 2. If 10 comes before 9, then it would be the other way around.

What I meant by destination was destinations along the path, sorry if I munged that.

Okay so it's just implemented as a queue, and all of 9's adj. would be tested before all of 10's adj.
It works the way I thought based on just lower sector # but you do have a point about edited warps.  They always display in numeric order, and by default ARE in numeric order in the files, but edited warps may allow "first" to not be "lowest".  I'll have to test that when I get home from work to make sure tedit doesn't sort before storing.  It'd be tons faster than sorting before each display, so I'm kinda surprised it doesn't, but I've never seen *displayed* adjacents in any order but lowest to highest, even after editing warps.

Cool, I finally had the chance to test a new bubble-finding approach, and the results are promising. It can perform a single-door search in a 5K universe in under 10 seconds. What's unique about it though, is that it will find bubbles of any size at that rate.

I *think* it will work on a multi-door search at a much higher rate than conventional approaches that incur exponential rate growth. Untested as of yet though.

I'll be holding the scripting session on Sunday, Jul 1, at 9-11 PM EDT. I'll start another thread soon as well, stating the same, along with brief details about Skypecasts.

Stay posted.

+EP+

sounds good

EP,

Did you ever set up a database of your past Skype tutorials?

Nah, and IIRC there was only 1. I think the material from that class is still around... I just don't remember the link.

I think this class was cancelled due to a lack of interest.

Truth is, I haven't thought of that search technique since the class was cancelled, and may have forgotten it. Typically, there are only one or two people interested in such things at a given time

+EP+

Im always interested, though it may seem that im not since I dont speak up too much.   Still, Im always reading what you have to say...and I always have something new to learn.

btw.. I hear and see that swath has a better bubble technique than twx. I often find that twx has missed/skipped over a larger bubble that swath may find.

I dont want to give in to swath just to find the largest bubble in the game.   Any truth to this? Or is it just my eyes...

Hi guys, I thought I would chime in.
I may be the only one, but in a game, I find a big bubble, I avoid it like the plague. It just puts a big x on you.

Go for dead ends. You will always do better.

Witness the perpetual Game...
Imagine building +200 SBB ports, upgrade them, create +200 planets with production to match port-production and then NEGO'n them (sometimes) twice a day...  no need to hide with that kind of setup
 

Who builds that much? Get in and take over early. That is a good idea when you are planing a long haul. But in unlims where it last 24-48 hours. Watse of time. Build up 2 or 3 different dead ends so if 1 gets found you still have 1. Then stay mobile. Staying in 1 place is asking for trouble. But in longer games I see it.

If the game last longer than 2 days, I wont play.

Plus being retired makes it hard lol

WS, He's talking about really long truce games and the ol' turtle strategy.

sing you not log on to icq anymore or are you invisible?? have a couple questions

I play the same in a long one too. What ever the max planet per sector is, I keep it at that and scatterd. Saty mobile wins. You can always cash and buy figs.

But anywho