Author |
Message |
Singularity
Veteran Op
Joined: Thu Jun 02, 2005 2:00 am Posts: 5558 Location: USA
|
Ok. So then... do a basic bubble search (breadth search) and find gates after the last pass? Then once gates are found plot to... someplace close? Someplace you think is inside the bubble?
_________________ May the unholy fires of corbomite ignite deep within the depths of your soul...
1. TWGS server @ twgs.navhaz.com 2. The NavHaz Junction - Tradewars 2002 Scripts, Resources and Downloads 3. Open IRC chat @ irc.freenode.net:6667 #twchan 4. Parrothead wrote: Jesus wouldn't Subspace Crawl.
*** SG memorial donations via paypal to: dpocky68@booinc.com
|
Mon May 15, 2006 10:50 pm |
|
|
PHX
Lieutenant
Joined: Sat Mar 03, 2001 3:00 am Posts: 592 Location: USA
|
Today in about 30 minutes time i wrote a ztm that plots from sector to 1, catches first warp, voids it, plots again. When it can no longer get a path if clears the avoids and moves on to next sector, and skipping any sectors already known to have 6 warps out. This is very slow but it seems to me that it should be 100% accurate. And I would think accuracy before speed?
It bursts the avoids and bursts the avoids clearing, but only sends 1 path at a time, thatd be what makes it slow.
I was just coming here to ask a question and found this thread in progress.
Anyway, my question was gonna be. Does TWX in fact catch this data automatically? It seems that question is answered already here. Anyone have input or advice?
|
Tue May 16, 2006 1:21 am |
|
|
Singularity
Veteran Op
Joined: Thu Jun 02, 2005 2:00 am Posts: 5558 Location: USA
|
Yes TWX catches the data provided you allow it to. No it's not 100% accurate. What happens when setting 1 avoid causes more than 1 adj to fail? Like in the case of a bubble where the first sector in (door) is more than a 2-way but setting that avoid prevents you from plotting out along most paths.
That approach is also slower than layering passes on top of each other. If you go thru all of the sectors that have 0 warps out you will plot a lot of courses which in turn reduces the number of courses you have to plot on subsequent plots. Covering areas all over the universe means you'll eliminate the easy warps (higher traffic areas) first allowing you to reduce your overall plot count later on. This is why I use a 6 pass approach. Pass 1 plots to and from 0 warp sectors, pass 2 plots to and from 1 warp sectors, pass 3 to and from 2 warp sectors, and so on. It's very fast (6 passes in under 6 hours) and accurate down to a tiny number of sectors (see above). What I'm working on now is adding that little extra bit to find those particular sectors.
If you plot from 1 to 6, while there's a course, it also means you have to set a trigger for each and every plot you make to find out whether or not to continue on. Using a layered approach you can also cache several plots (I do 10 at a time) on the server in a burst, eliminating ping from the equation (which even on fast servers can add quite a bit of time to the process).
About the backplot thing, tested it and have decent numbers. Despite the extra 20000 plots you'd think you have to do, in reality it only added about 1000 plots to my first pass. The reason being that it found additional warps early on, meaning later in the pass it didn't have to do those plots. It also added afew thousand warps to the DB during the first pass, which makes later passes more efficient. Overall (the entire process) it added about 700 plots and 6 minutes, but also added a few new warps and made the first few passes much more efficient (good for a fast ztm algo). I'll continue testing this to see if that holds over different maps.
_________________ May the unholy fires of corbomite ignite deep within the depths of your soul...
1. TWGS server @ twgs.navhaz.com 2. The NavHaz Junction - Tradewars 2002 Scripts, Resources and Downloads 3. Open IRC chat @ irc.freenode.net:6667 #twchan 4. Parrothead wrote: Jesus wouldn't Subspace Crawl.
*** SG memorial donations via paypal to: dpocky68@booinc.com
|
Tue May 16, 2006 2:15 am |
|
|
PHX
Lieutenant
Joined: Sat Mar 03, 2001 3:00 am Posts: 592 Location: USA
|
hmmm...good point...why didnt i think of that
HEY i just got another idea that will make this method work.....ill report back with my findings.
|
Tue May 16, 2006 5:43 am |
|
|
PHX
Lieutenant
Joined: Sat Mar 03, 2001 3:00 am Posts: 592 Location: USA
|
Someone mentioned twx will catch the data if its not busy doing something else. So if I am grabbing one word off the line of warps, is that sufficient to cause it to miss the path data.
|
Tue May 16, 2006 5:45 am |
|
|
Promethius
Ambassador
Joined: Mon Feb 09, 2004 3:00 am Posts: 3141 Location: Kansas
|
quote:Originally posted by PHX
Someone mentioned twx will catch the data if its not busy doing something else. So if I am grabbing one word off the line of warps, is that sufficient to cause it to miss the path data.
I have nothing more to offer than a guess on that, but I have not seen it miss data. I have seen it run faster than the data was being displayed. The ZTM data window finished with file updates written while data was still being displayed in the SWATH window. I am guessing that SWATH is buffering the data as the TWX database contained all warps even when forcing a disconnect while SWATH was scrolling the ZTM. Maybe something else was going on, I don't know.
_________________
/ Promethius / Enigma / Wolfen /
"A man who has no skills can be taught, a man who has no honor has nothing."
|
Tue May 16, 2006 7:28 am |
|
|
Singularity
Veteran Op
Joined: Thu Jun 02, 2005 2:00 am Posts: 5558 Location: USA
|
I cache plots and just waitOn one of the cached plots. 2 reasons: 1, it guarantees it will get parsed. 2, it allows me to kill-script if I need to stop a gridder.
_________________ May the unholy fires of corbomite ignite deep within the depths of your soul...
1. TWGS server @ twgs.navhaz.com 2. The NavHaz Junction - Tradewars 2002 Scripts, Resources and Downloads 3. Open IRC chat @ irc.freenode.net:6667 #twchan 4. Parrothead wrote: Jesus wouldn't Subspace Crawl.
*** SG memorial donations via paypal to: dpocky68@booinc.com
|
Tue May 16, 2006 10:57 am |
|
|
ElderProphet
Commander
Joined: Tue Oct 07, 2003 2:00 am Posts: 1131 Location: Augusta, GA
|
If you recieve the plot from the server, and set about some task that is going to take a full second of processing to complete, you aren't going to reliably capture any data that arrived in the interim, while that processing took place.
So if you que 10 plots against the server, don't start creating 20k arrays or reading 20k line files in the background. Keep your processing tasks quick, and it won't be a problem. The sort of stuff you are talking about doing PHX is insignificant time-wise, and will not cause any problems.
_________________ Claim to Fame: only guy to ever crack the TW haggle algorithm, and fig/shield/hold price formula, twice.
|
Tue May 16, 2006 1:54 pm |
|
|
PHX
Lieutenant
Joined: Sat Mar 03, 2001 3:00 am Posts: 592 Location: USA
|
Sounds good. well my idea was, before avoiding the first warp on each plot to get a new path, i plot from that first warp back to source, which most of the time is 1 hop. Does this sound like it would create a 100% perfect ztm? I estimated this ztm to take about 16 hours in a 20k game....OUCH. Oh well I was just goofing around with the idea, there are plenty of ztm scripts already available.
|
Tue May 16, 2006 2:03 pm |
|
|
Promethius
Ambassador
Joined: Mon Feb 09, 2004 3:00 am Posts: 3141 Location: Kansas
|
quote:Originally posted by PHX
Sounds good. well my idea was, before avoiding the first warp on each plot to get a new path, i plot from that first warp back to source, which most of the time is 1 hop. Does this sound like it would create a 100% perfect ztm? I estimated this ztm to take about 16 hours in a 20k game....OUCH. Oh well I was just goofing around with the idea, there are plenty of ztm scripts already available.
lol - PHX, the reason to write a ZTM is because you CAN and it is one of those puzzles to figure out. Actually the reason I started was because I wanted to capture data after the ZTM and the scripts I had would not do that. I wanted to start the script and come back to have the 7-warps in, dead end sectors and one or two other things that aren't in the scripts I release as public.
EP, I try not do do anything other than plot and get the sectors for a plot when the ZTM is running so hopefully I am not missing data. Am I correct when I see TWX finish the file writes ahead of SWATH displaying the plots when I say the data has already been caught? That seems to be the case, but I know all too well that what appears to be happening is sometimes not accurate.
_________________
/ Promethius / Enigma / Wolfen /
"A man who has no skills can be taught, a man who has no honor has nothing."
|
Tue May 16, 2006 5:28 pm |
|
|
ElderProphet
Commander
Joined: Tue Oct 07, 2003 2:00 am Posts: 1131 Location: Augusta, GA
|
I'm sure that the amount of processing that you are doing is fine. TWX can do a ton of processing in a half-second, so you'd have to be really trying to over do it.
_________________ Claim to Fame: only guy to ever crack the TW haggle algorithm, and fig/shield/hold price formula, twice.
|
Tue May 16, 2006 6:31 pm |
|
|
PHX
Lieutenant
Joined: Sat Mar 03, 2001 3:00 am Posts: 592 Location: USA
|
quote:Originally posted by Promethius
quote:Originally posted by PHX
Sounds good. well my idea was, before avoiding the first warp on each plot to get a new path, i plot from that first warp back to source, which most of the time is 1 hop. Does this sound like it would create a 100% perfect ztm? I estimated this ztm to take about 16 hours in a 20k game....OUCH. Oh well I was just goofing around with the idea, there are plenty of ztm scripts already available.
lol - PHX, the reason to write a ZTM is because you CAN and it is one of those puzzles to figure out. Actually the reason I started was because I wanted to capture data after the ZTM and the scripts I had would not do that. I wanted to start the script and come back to have the 7-warps in, dead end sectors and one or two other things that aren't in the scripts I release as public.
I have a script out that will get some of the stuff you are talking about by checking the twx dbase. Sector List Generator is available at the Grimy Trader. It could stand to have alot more options added to it, someday ill get around to it.
|
Wed May 17, 2006 5:24 am |
|
|
|
Who is online |
Users browsing this forum: No registered users and 19 guests |
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot post attachments in this forum
|
|