| www.ClassicTW.com https://mail.black-squirrel.com/ |
|
| ProZTM_20 https://mail.black-squirrel.com/viewtopic.php?f=15&t=20455 |
Page 1 of 1 |
| Author: | Promethius [ Thu Mar 27, 2008 12:55 pm ] |
| Post subject: | ProZTM_20 |
I am rewriting the ProZTM script that is currently at version 1.4A (ProZTM14A) and this is a complete rewrite of two of the pass methods. I have also changed the way I detect certain items, and cleaned up the menu. I am needing to know if anyone uses the traffic analysis report that is in v1.4. I am thinking of removing it for 20. Some of the other data creation options will remain such as 7-warps in and dead ends. The exporting of a SWATH warp spec is being eliminated unless I hear that someone uses it. In testing of ProZTM2.0 vs ProZTM1.4, I found the following results against a 20k sector, 52,620 verified warp universe: --------------------- I am running 2.0 again to verify the time and accuracy. I do have a method to get 99.084% accuracy in 3 hours and 30 minutes (approximate) in a 20k bang, but that leaves too many warps not found. I may update ZTM_Speed with that method. |
|
| Author: | Thrawn [ Fri Mar 28, 2008 9:04 am ] |
| Post subject: | Re: ProZTM_20 |
Excellent news Promethius. I find this script very useful. Both Daala and I use the traffic analysis but we can work without it if you remove it. The same with the exporting of a SWATH warp spec. I use ProZTM1.4 in my script tutorials and can use it for further needs of the removed items. I don't expect you to keep it for just us. 
|
|
| Author: | Parrothead [ Fri Mar 28, 2008 10:24 am ] |
| Post subject: | Re: ProZTM_20 |
ill take 99% in 3 hours...heh...Usually only a partial is needed in a unlim anyway. Except for Zarkhan and Lonestar who hang on to the truce game BS. |
|
| Author: | LoneStar [ Fri Mar 28, 2008 2:02 pm ] |
| Post subject: | Re: ProZTM_20 |
Parrothead wrote: ill take 99% in 3 hours...heh...Usually only a partial is needed in a unlim anyway. Except for Zarkhan and Lonestar who hang on to the truce game BS. What's that I smell... Smack? Sniff, Sniff... Nope. Just Parrot Feces. |
|
| Author: | Parrothead [ Fri Mar 28, 2008 4:40 pm ] |
| Post subject: | Re: ProZTM_20 |
Not Smack...but a full ztm in an unlim is usually not required. Ergo the faster less accurate one has advantages. |
|
| Author: | Promethius [ Fri Mar 28, 2008 5:37 pm ] |
| Post subject: | Re: ProZTM_20 |
Parrothead wrote: Not Smack...but a full ztm in an unlim is usually not required. Ergo the faster less accurate one has advantages. One of the things that is changing in the new version is the addition of displaying in near real time the number of warps that have been found. ProZTM_14A showed the number of warps found only when changing to a new pass or completion. The near real time on the number of warps found will let the user decide if 45k or 50k warps is sufficient and shut the script down manually or they can let it run until it finishes. Yes, the near real time display has a time penalty involved - maybe adding 2-3 minutes overall. |
|
| Author: | LoneStar [ Fri Mar 28, 2008 7:16 pm ] |
| Post subject: | Re: ProZTM_20 |
Parrothead wrote: Not Smack...but a full ztm in an unlim is usually not required. Ergo the faster less accurate one has advantages. Yep. which is why I have a script that only maps an existing grid.. helps to have those adj's for times when people think '..no way can anyone get the drop on me'  Btw, I studied the output of 1.4A vs a 100% explore/map. found that there's problems mainly with One-ways. wish I had Promp's source code.. but my guess is that during the oneways check newer oneways are created in the process of checking for oneways.. So. Prom. Make a list of checked oneways.. then at the very end check for newer oneways and against the already made list. Based on my 'looksie' and not having the source, I'd say this will make your maps 100% accurate. |
|
| Author: | Singularity [ Sat Mar 29, 2008 6:32 am ] |
| Post subject: | Re: ProZTM_20 |
Quote: One of the things that is changing in the new version is the addition of displaying in near real time the number of warps that have been found. ProZTM_14A showed the number of warps found only when changing to a new pass or completion. The near real time on the number of warps found will let the user decide if 45k or 50k warps is sufficient and shut the script down manually or they can let it run until it finishes. Yes, the near real time display has a time penalty involved - maybe adding 2-3 minutes overall. I do that, it takes a while to do the warp summation... Code: :count_warps setVar $warpcount 0 setVar $sect_idx 0 :top_of_count_loop add $sect_idx 1 if ($sect_idx > SECTORS) return end add $warpcount SECTOR.WARPCOUNT[$sect_idx] goto :top_of_count_loop Little snip of code I use. Not much to it, but you need to find a good time to do it. If you take too much CPU writing to the window it'll effect TWXproxy's runtime performance, so I only do it every 1k sectors or so. Quote: So. Prom. Make a list of checked oneways.. then at the very end check for newer oneways and against the already made list. Based on my 'looksie' and not having the source, I'd say this will make your maps 100% accurate. A 6 pass ZTM is very accurate with small bubbles, but large intra-linked bubbles are much trickier... so getting 100% is a lot harder than it sounds. But the extra 1 or 2 warps isn't usually important. If you use the plot&avoid method new one ways will come along naturally and you don't to expand them. All you need to do then is reverse-check them to make sure they're actually 1-way. |
|
| Page 1 of 1 | All times are UTC - 5 hours |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|