Hmmmm the expansion check code in 0.9.2 must be causing the loop somehow. I will revisit tonight and I will put up 0.9.0 in the meantime.

Version 0.9.3 released

Okay there was one more very minor thing I had changed in 0.9.1 and that was a firebase parm. It may have been causing the locking up.

Now the only difference between 0.9.0 and this release is the expansion base check and Im certain there is no way that this check is causing the lock.

Please let me know asap if you have the lock and make sure you use the updated SCD file.

I had a crash-lock without your mod as mentioned earlier. It's rare, but maybe there's a similar issue in Sorian?

it might well be someone reported the exact same kind of crash as the one you describe in sorian feedback thread a few pages ago. as i said here before the lock up is
probably a loop thing
it is
probably coming from a mod
and it
might be duncanes 0.91

as long as you have several packs installed there is not much chance knowing from where it comes with 100% accuracy. it is very unlikely it comes from stock lua code though but there were typos in stock lua too it was not perfect at FA release.

I just want to rule out the other changes I made which could be causing the lock to happen more often. I would really like to see the sim speed situation improved so I want to see if reducing the expansions helps.

9.3 Looks ok to me. Had several games now and no lock ups, plus no slow downs.

Obviously AI doesn't expand as much as before, so abit less challanging.

Good stuff. I would like to increase the expansion slightly but no idea how. Somewhere between vanilla and what you did in 9.3.

Thanks Jace11.

I would be keen for others to try an let me know.

Im pretty sure 0.9.3 doesnt cause lockups.

Can you also post your CPU with your feedback....

hehe nice release duncane.

I tested rush and adaptatie on ambush pass.

rush needs to rush harder on such a map and ensure it has enough factories if its expansion policy does not go as planned.

adaptative is a mean badass.....

both seemed to have used their commander as a forward defense , shielding the base and fending off enemy paltoons with great effectiveness. I dont know if this is a side effect or if you added that behaviour but it is really working wonder against a rushing opponent. adaptative bet to tech fast even on a small map does seem to pay off once a t2 pd +t1radar is up, any rushing opponent is doomed to try to catch up on tech most of the time it is too late and the counter rush is deadly. a very good antirush archetype on a 10x10 map. teching turtling and counterrushing all in a matter of 15 minutes.

This seems like a very powerful AI. I've tested it against the newest Sorian build, and it's won every time. Keep at it! (You too Sorian )

@pyro786: Im surprised it won against Sorians. I find its usually the other way round. What maps/AIs did you try it on? Are you certain you wernt selecting AIx for the standard AIs. Anyway I appreciate the positive feedback.

@Ajunta: I have tweaked the com behaviors by using a couple of the improvements Sorian made. I think the key one is that it ignores the enemys com when deciding if to attack. In the stock AI's this meant the com hardly ever left the base as it counted the enemy com as a big attack force to avoid. I can't take credit as I got it off Sorian, plus he has many more com behaviour additions.

I can confirm this. I let the AIs play 3 games on Field of Isis. Using Dunane 0.9.3 and Sorian 9.6c. First one was Sorian vs Duncane Adaptive. Second one was Sorian Rush vs Duncane Adaptive. Third one was Sorian rush vs Stock Adaptive. Sorian was Seraph and Adaptive was UEF for all three games. First two games were won by Duncane Adaptive. Sorian Rush fought a lot better then Sorian normal on that map but still lost to Duncane Adaptive. Last one was won by Sorian as expected.

One interesting thing I notice with that last game (running without Duncane fix) was how choppy and laggy (with sounds hickups and intermitent freezes) it was compare to the first two games (where duncane fix was running). What ever you are doing Duncane, it seems to be working . Can't thank you and Sorian enough for all the hard work you two are putting into this!

Hmmm ... that is surprising that it beats Sorians. Anything in particular that seem to make it win? Also can you try it with Sorians as UEF and the stock as Seraphim?

As for the performance gain... an early fix I did reduced the bunching of units at base and the recent expansion base numbers reduction has helped too I think.

Thanks heaps for the positive feedback. Always gives me a warm feeling when someone says something nice

In general, it just seems to get map control better than Sorian. Usually Sorian goes a mix of ground/air and Adaptive beats it on the ground, but I've seen your adaptive taking the air a few times and beating Sorian around quite a bit. I keep factions set on random to keep it as fair as possible, and your adaptive continually beats Sorian.

Hmmm.. surprising (but cool of course). You are selecting sorian rush if its 10km or less yeah?

My PC was a beast 18 months ago, still pretty good now...

AMD Athlon 64x2 4800
can't remeber my MOBO, its an ASUS Crossfire deluxe.
2GB CORSAIR DDR 800 RAM
150GB RAPTOR X
2 x ATI HD2900XTs in crossfire.
WINDOWS XP
etc

Anyway, I was getting slowdowns, with vanilla etc late game like others have reported, not seen any real slow-downs recently with 9.3

Thanks Jace11....

Once we are sure about the expansion bases, I will increase it a bit. At the moment I want to keep it as low as possible to confirm it helps the sim issue.

yes i can only confirm what people say about 0.93 adaptative.

It is a bloody badass. first time i saw it play against Ajunta Exp009 i ended up thinking , wth !!! is it cheating or what ?

map control is okay and at the same time teching is very fast. it is all relying on the commander blowing up everything in the middle of the battlefield it is deadly against a rush type since most attack platoons dont make it to 0.93 adaptative main base. it is a considerable mass sink to rush archetypes like Sorianrush096 or Ajuntaexp009 and it cost almost nothing to Adaptative 0.93 who can tech up unscathed and struggle for map control on the side of the map.

occasionnally Ajuntaexp009 manages to slip a platoon through blow all pgens and take control of the game but time is slipping fast too and if adaptative gets unscathed to 10 minutes mark it wins systematically.

i had the same duel game restarting 4 times in a row yesterday night between adptative 0.93 and ajuntaExp009 on ambush pass and the outcome of the duel is absolutely random sometimes exp wins around minute 15 sometimes exp cant trouble adaptative and is doomed to lose before the 20 mark.

the decisive factor seems to be the number of wasted engineers (stuck, poped, or idled) one of the AI has to bear. I think it is time we put our efforts into these engineer buold functions with sorian, i think it is the main cause of poor effectiveness of any of our archetypes now. Commander sometimes stalls early game, there are literally gangs of dummy engineers waiting to be slaughtered here and there by a well adjusted artillery shell, and countless engineers preferring to go straight for a mex into an anemy stronghold or military platoon instead of a cornered lonely mex because the former mex is 2.5 gu closer than the latter...

i definetevely have to test the 4 non cheating Ai mods to have an idea of how more enjoyable they are to play from a human point of view than the version of the last month. i expect them to be very competitive right from the strat of the game i ll try to give feedback about that as soon as possible.

Most of the time the way I test the AI is to have AI fix and Sorian AI fight it out. What I have been doing lately is setting up a lan game and set the speed to adjustable with cheating turned on. Then once the game starts I use the setarmyfocus (I believe that is what it is called). Then I crack up the game speed, plus I can watch the economy. I noticed a few engs sitting around doing nothing. They would just sit there. So, I click on one of them and tell them to move or reclaim a dead unit near by. It will do the action that I told it and then it is off doing something else. This basically gets it unstuck. I am not sure if that helps figure out what is going on or not. I have see engs get stuck with the Sorian AI as well but not as many.

Matt

Is there a way to tell that the mod is installed properly and doing its thing? As I understand it, it replaces the normal AIs so you dont have any new options to select in the AI menu?

yes endrek mod and duncane mod overwrite everything they modify but are silent mods. and so they cant coexist.

there is nothing to tell stock duncane or endrek from one another save the way they play.

these two mods are still active if sorian pack is hooked to them. sorian pack adds 2 ai to those you have wether they are stock of duncane/endrek modified versions.

never thought about doing that indeed, i ll keep that result in mind it is very intresting.

Sorry duncane to parasite your thread.

Hmmm very interesting. Thanks for confiming the engineers are getting stuck. It seems quite random. Sometimes none get stuck other times several do. Often all in the same spot which is odd.

Im trying to resolve this next. Thanks for tip about the getting going again. So if you give them an order they get unstuck...

I reran the game on Field of Isis with Sorian Rush (9.6c) as UEF and Duncane Adapative (9.3) as Seraph and Duncane still won pretty easely. It keep overwhelming sorian with land units and gained land control pretty quickly.

I also tried the new Sorian 9.6d. Same map, with Sorian Rush as Seraphim and Duncane Adaptive as UEF. The new Sorian finaly Won! It was a pretty interesting game. Sorian started up strong and gained control of 2/3 of the map until mid game where adaptive got back the upper hand with waves of land units. Eventually, Sorian got back at him with 2 platoons of T3. His early economic push made Sorian finish his experiemental first which won it the game.

For good measure I redid the last test by reversing the roles. Sorian UEF got beaten pretty badly by waves and waves of land units from Dec. Adaptive Seraph. One funny thing happened: Adaptive was about to blow up sorian commander's but it reached his 500 unit limit and blew most of his attack force! That happened four times! The AI should prolly stop making T1 units after the firts blow up or at least limit itself to artillery. Eventualy Adpative got his experimental out and finsihed up the new Sorian.

Finaly I redid that last one, keeping the AIs with the same faction, but reversing their position on the map and the outcome was the same: Adaptive won over Sorian with rush of land units.

The one thing with this map that may hurt Sorian is the lack of Firebase positions. I havent seen Sorian nor Adaptive build any of those. Even it's expension bases were not that well defended. One good firebase would take care of all the waves of T1 units from either side.

Yeah I agree the AI doesnt build enough defenses at its expansion points. I will look at adding more.

I dont think of this as firebases as such. I think of firebases as artillery or TML points positioned in range on the enemy bases. The AI does do these but not often which is good as its a risky tactic.

You're right, I meant defense bases.