My test map

I want to use the concept in my map, but I stumbled into a problem. Of course, I tried asking the creator himself, but he doesen't seem to be active, so I turn to the community for assistance.

My problem is that it doesen't always deal damage (about 50% of the cases), where as the original concept map by SelfPossesed forks perfectly. In my test map, you deal 20 damage to Jim by swarming anywhere. It seems that, in a nutshell, I copied everything he does, but something is wrong.

I hope I'm not the only one that was playing around with this wonderful concept. Any help would be greatly appreciated. Thanks.

Original topic.

I think selfpossesed had a timer.
1/8 of a game second wasn't long enough...
If I remember right he used 2 or 3 trigger cycles.

...testiing...

uhm, I have no idea...
sry, maybe in a few hours when I'm sober again...

Post has been edited 1 time(s), last time on Dec 19 2008, 1:01 am by Ahli.

You are not alone. I am planning to use it in a map.

Like ahli said, did you try lengthening the period of time the scarab is allowed to stay there?

Yes, I used 2 trigger cycles. The result is the same, in a nutshell: each trigger cycle the scarab is under the target it has the same chance of dealing the damage. Besides, the original concept map held the scarab for only 1 trigger loop under the target anyway.

I went through your triggers. Move the constant AI Script after the trigger to move the scarabs and it should work beautifully.

Many tests have shown that it works perfectly if the unit that is getting the damage is owned by a computer for the duration of the trigger cycle. I will, of course, follow your suggestion to check that possibility also.

ADD: Nope, running the AI command after the command to move the scarab does not help at all.

Post has been edited 1 time(s), last time on Dec 19 2008, 5:33 pm by JaFF.

I guess it was just coincidence then, that when I tested it normally I got a 7/15 % success rate versus when I moved the Constant trigger under the Deal damage trigger for Player 3 I got a success rate of 13/15 %.

I compared the triggers of both maps and they both work the same way. I really can't find any reason one should work and not the other.

You made a relatively small amount of tests for those result to be accurate (I'd take atleast 100 as a more or less accurate result). I had cases when I got 10 successes in a row with the flawed system, so those figures mean almost nothing.

I've made enough tests to conclude that if the victim unit is owned by a computer for the time the damage is made, the system is flawless (WRT to 1 unit, of course).

Original Topic.

True. The first time I tested the other way, I had a 100% success rate over maybe 16 tries, so I stopped. Then I decided I should at least get some actual percentages. I don't have the kind of patience to get 100 test trials per method on a map that isn't mine, unfortunately.

However, you can make it automatic: add 1 gas when you move a scarab and add 1 mineral when you detect the motion of the target unit (if you detect motion, move the target unit back to its starting place). Repeat with intervals of about half a second (enough time to detect motion if it occurs). That way you'll get 100 experiments in just under a minute and the minerals/gas ratio will be your success rate.

Or just give a unit 1000 life and run the trigger 1000 times, then afk.

Yeah, that would be a bit simpler than what I said.

Make the Defiler invincible. The Scarab is targeting your Defiler instead of the Marine.

EDIT: I just tried with invinc, it isn't working properly even with it. Gimme a sec, testing a bit more.

EDIT 2: I removed the "remove" trigger and the "constant move" trigger. Apparently, the Scarab sometimes targets the Lurker Egg instead of the Marine. The best shot is then to see if there's an alternative unit to use as a target.

EDIT 3: Just in case, turn on vision so that Player 3 can always see Player 1. I also moved the Reavers to a further location. Even still, the Scarab sometimes targets P2's units instead. I need to find the lowest priority ground unit there is.

EDIT 4: Gah, I'm not sure what ground unit would be preferable. I tried a couple buildings, I tried spider mines, I even tried critters. But even still, the Scarab sometimes prefers to attack P2. Why a P2 unit sometimes has higher priority is really confusing. If I had to give you a solution, it would be to use a computer player as a dummy target and use burrowed units moved under the damage location. Even if I found a unit that would allow your test map to work, there's the possibility of a nearby unit taking precedence over the intended target during actual gameplay. Having the exact same type of burrowed unit as a target should cause the Scarabs to attack the nearest target, like in my test map. It's the best thing I can come up with at the moment.

Post has been edited 4 time(s), last time on Dec 20 2008, 10:48 am by SelfPossessed.

Your attack priority idea is the first thing that makes sense to me. How about this: In my current version of the test map, I constantly craete reavers, let them shoot one target, move their scarab to a 1x1 box, remove the reaver and repeat. That way I have 9-12 scarabs at all times in that box. What if I craete a new target for each of the scarabs as well?

EDIT: Actually, try the new target thing on the damage trigger. That or set P2 to ally then. It might cause the Scarabs to choose a new target.

---

New targets for each Scarab shouldn't do anything. Why is probably due to how Random Suicide Mission works (I'm no expert on it though).

I have vision turned on, so it isn't due to not being able to see the target.
I even tried switching P1 to P4, which didn't work, so it isn't due to player number.
I moved the Scarab area to the corner of the map, so it isn't distance either.
I also tried changing up the unit to be attacked by Reavers to the same Marine, and still they liked to target P2's stuff.

Again, your best hope is to use a target. As in, use a burrowed Devouring One or something for P2 as the Reaver target AND create one for P2 underneath the Marine to get hit.

Just tried that, didn't help. I'll try moving the exact same target underneath the marine.

THE MAP

Yup, moving the target of the particular scarab under Jimmy works. Of course, this is a primitive version with sloppy triggers, but don't worry, I'll start working on a scarab holding system that will allow instant access to much more scarabs and their specific targets.

I just tried moving a Devouring One under Jimmy (NOT the one the Reavers are attacking) and it works fine. It takes less locations and triggers to do it that way.

EDIT: Attached a modified version of your first test map using a burrowed Devouring One. Notice that this version stacks the Scarab so it isn't offset when moved on top of the Marine. You need several trigger cycles to stack Scarabs, so you cannot rely on creating a Reaver to shoot each time like in your second version.

Attachments:

Post has been edited 3 time(s), last time on Dec 20 2008, 8:28 pm by SelfPossessed.

Odd, I tried creating a zergling under the marine and it didn't help. Maybe it has something to do with a zergling being less of a priority than a devouring one.

Lings have the highest priority for all ground units