ProZTMSpeed v2.0 has been released on http://www.grimytrader.com The script is not as accurate as ProZTM2.0, but:

---------------------------------------
20k Sector, 100% warp count = 52620

ProZTMSpeed v2.0
  Pass  Warps  Elapsed Time       Accuracy   Missed  
   2    52315  4 hrs 02 minutes   (99.420%)    305 
   1    52059  2 hrs 55 minutes   (98.934%)    561 
              

ProZTM v2.0
  Pass  Warps  Elapsed Time       Accuracy   Missed  
   3    52603  6 hrs  05 minutes  (99.967%)     17
   2    52167  4 hrs  11 minutes  (99.139%)    453 
   1    51227  2 hrs  05 Minutes   (97.353%)  1393


ProZTM v1.4A
  Pass  Warps  Elapsed Time       Accuracy   Missed  
   3    52555  5 hrs 32 minutes  (99.876%)      65
   2    51423  4 hrs 58 minutes  (97.725%)    1197
   1    43813  1 hr  02 minutes  (83.263%)    8807

I will be changing part of ProZTM v2.0 to include a method I came up with for ProZTMSpeed that "should" speed it up also. I am not sure how it will affect overall accuracy and if it degrades ProZTM v2.0 then I will not use it.

               / Promethius / Enigma / Wolfen /

"A man who has no skills can be taught, a man who has no honor has nothing."

Have you done any comparisons of time and accuracy vs other people's scripts?
Yours sounds good to me, but I'd be curious to know the numbers from some of the other popular ztm scripts. Just a curiosity thing.

If the data above are accurate, the "3 pass" that took 6 hours is accurate w/ both Ck's 6 pass and my own in both time and warp count.

The odd thing I found is that using the method for ProZTMSpeed 2.0 in ProZTM 2.0 resulted in lost accuracy even though more sectors were found in the first two passes. It does not make sense to me, but I will not be adding the code from the "speed" version.

ProZTMSpeed 2.0 does not use any avoids and that is where it gets it main speed.

               / Promethius / Enigma / Wolfen /

"A man who has no skills can be taught, a man who has no honor has nothing."

Well there ya go, unless you're specifically targeting low traffic sectors you'll miss them w/o avoids.

I can see why a ZTM method that doesn't use avoids would result in faster pass completion (no setting/clearing delay) and higher warps found (all plots succeed) in the first 2 passes. Ultimately though, when you only have a few warps missing, the most sure (though not completely sure) way of finding them is by voiding known warps and plotting.

Those charts are very nice. On the surface, it would seem like a good idea to use a no-avoid method for the first 2 passes, then an avoid method for subsequent passes. The flaw in that idea though is that you would really have to rerun pass 2 with avoids for 1-warp sectors, probably negating any time improvements.

And I think there is still room for improvement in our current ZTM methods.

Both Sing and EP hit everything on what I see with the ZTMs. The "Speed" version was written for just that, speed in getting as many warps possible in a short period of time while sacrificing accuracy. The ProZTM2.0 version was written to get accuracy while sacrificing speed. I really don't know of a way to get speed and accuracy (at least to the point I want). Accuracy requires the use of avoids and that slows everything down.

Sing's thoughts on targeting low traffic sectors is dead on with what I am doing in the "Speed" version. I also throw a bit of a twist into the 2nd pass for this (Sing, hope the ICQ got through).

EP's thoughts about the use of avoids to get sectors mirrors what I have found also. You get a large percentage of sectors, but not all of them. If someone wants to watch grass grow, do a single pass using avoids without any other passes before that. Slow and not 100% accurate.

From what I've seen of my ZTMS: ProZTM14A is a middle of the road ZTM - neither the fastest or most accurate possible (ProZTMSpeed2.0 is faster, but not as accurate, and ProZTM2.0 is more accurate, but not as fast for completion).

               / Promethius / Enigma / Wolfen /

"A man who has no skills can be taught, a man who has no honor has nothing."

I really haven't tested against the other ZTMs that are out. The main reason is the amount of time, and I think you would need to test against a 5k bang in addition to a 20k bang. It is possible that the method I use might be less accurate in a 5k universe than a 20k or the opposite could be true (doubt that tho). The main reason I test against 20k is that is the most common game size I play.

I would love to see someone test ZTMSupG, CK's, and SWATH against ProZTM14A and ProZTM20 and give the results in a chart format similar to what I've used for my ZTM comparison. When I first started working with ZTMs, I was told that SupG's ZTM (which is based on The Reverand's ZTM I think) was the best out. I haven't tested against it and really should. To do the test, a person would need to have new databases created for each script and run as the same player in a game. Figure for a 20k that would be 6 hrs approximate times the number of scripts.

Regardless of the ZTM a person uses, they need to do a CIM for found warps if the ZTM does not do that. The CIM for warps provides free accuracy and speed (if the script is written to take advantage of explored sectors). My testing is a player in a sector without moving so the CIM is taken out of the test.

               / Promethius / Enigma / Wolfen /

"A man who has no skills can be taught, a man who has no honor has nothing."

For what it's worth.. I did compare you ZTM agaisnt a 100% explore and found only 1-ways to be 'missed' for example:


....so. my suggestion is do a recheck of 1-ways to get any *newer* 1-ways generated.

hehe reading about ztm algorithms makes me want to play tw again. pioneering new ztm algorithms was fun. good times.

No joke. Man I miss pioneering cool new techniques and tricks, was the best part of the game.

Thank you for the info on that, and I will see what I can do on adding that check. It also gives me another idea to try.

               / Promethius / Enigma / Wolfen /

"A man who has no skills can be taught, a man who has no honor has nothing."

Rev, Sing... I know what you mean. I've never been able to read or comment on ZTM techniques without pondering new methods for the next few days. Rev, your method was revolutionary, and quite brilliant. Everything since has merely been tweaks on your pass technique. Still, there may be another revolutionary idea out there to be uncovered.

Prom, keep the interest alive man

+EP+

I have to agree with you on Rev's method since I patterned off of SupG's which came from his. A ZTM is something that appears so simple and is really frustrating to get right - kind of like Trade Wars as a whole.

The issue of balance on time required vs accuracy is also a concern.

               / Promethius / Enigma / Wolfen /

"A man who has no skills can be taught, a man who has no honor has nothing."

yeah my method is very orderly and rigid - it is very right-brained. but i agree with you - i often wonder if there isn't another angle that slices through all that and simplifies the problem - something more left-brained.