Poll

What changes would you like to see for the F2 rules ?

No "Exploits", meaning no use of dirty tricks labeled as exploits. (A good example would be instant killing through use of uncapped sysvars).
7 (46.7%)
No use of .tieloc or .tieval.
1 (6.7%)
Only negative .tieloc values allowed. (Would remove the rule against tie feeding.)
3 (20%)
No use of .tieloc or .tieval if your bot creates ties at any point for any purpose. (This would allow the current bots to keep their tie defences)
4 (26.7%)

Total Members Voted: 7

Author Topic: Pole for the new F2 rules.  (Read 3393 times)

Offline Moonfisher

  • Bot Overlord
  • ****
  • Posts: 592
    • View Profile
Pole for the new F2 rules.
« on: November 03, 2008, 01:18:08 PM »
Now the ups and downs as I see them :

1. It's a grey area... when is somthing an exploit. I would say a good rule of thumb is that if you're using uncapped sysvars to kill a bot faster than you should be able to, then it's an exploit.
But there's probably other situations, personaly I think raping alge is an exploit (But sexrepro is only allowed in F1, so it doesn't matter).

2. This would mean a lot of maintenance for all the old F2 bots since most have tie defences.

3. This would also cause maintenance of old bots, and it would break the rule against tie feeding in F2. I think this is too drastic a change considering that all bots in F2 are buildt with the knowledge that tie feeding was illegal in F2.

4. About the same rule as #2, but only if your bot creates ties (For instance if it's a multibot). So basicaly if you at any point store a value in .tie then you're not allowed to use .tieloc and .tienum in any way. This would allow the current bots to keep their tie defences, and allow MB's in F2, but still prevent the use of dirty tricks through ties.



Personaly I'm favoring rule 1 and 4... I think that would be enough to keep F2 more interesting and it wouldn't require any changes or rerunning of the league. (Except I think the 2 top F2 bots would need to be refactored, but I believe that's a good thing)

Offline Peter

  • Bot God
  • *****
  • Posts: 1177
    • View Profile
Pole for the new F2 rules.
« Reply #1 on: November 03, 2008, 02:26:55 PM »
I agree completely with you. Altrough the first one is a little tough to control. As yo said it is a gray area.
Any uncapped sysvars should be handycapped as soon as possible anyway.

The second point, I think most bots should be able to have some tie-defence when running in F1-anyway. But the main point is ,who would want to do the maintaince.

Four, I agree with that. It will effectively shut down instatie-attacks and it will accept multibots.

I think that is nice for me  , a first prototye tie-moving MB of mine defeats MB-league with ease. And it could do fine in F2.
Oh my god, who the hell cares.

Offline Commander Keen

  • Bot Builder
  • **
  • Posts: 91
    • View Profile
Pole for the new F2 rules.
« Reply #2 on: November 03, 2008, 07:11:30 PM »
I agree with 1/4. They make the most sense and are by far the easiest to implement.

Offline Peter

  • Bot God
  • *****
  • Posts: 1177
    • View Profile
Pole for the new F2 rules.
« Reply #3 on: November 04, 2008, 05:26:41 AM »
I agree with you all, but this poll just misses the option 5. F2-league is just fine as it is right now.
Oh my god, who the hell cares.

Offline Peter

  • Bot God
  • *****
  • Posts: 1177
    • View Profile
Pole for the new F2 rules.
« Reply #4 on: November 06, 2008, 01:28:44 PM »
1.
About point4. No tieloc and tieval. I would like to use them in a MB. It could be that only tin/tout isn't enough for communication between ties, may I use the area of 50-150(if those are free) for tieval storing. Or any other free area for the couse, including racial memory.

2.
Second, in some cases even direct storing in active sysvars could be easier and more effiecient. Like one part of a MB tells directly at the other part it should move upwards.


Ofcource both options are all regulated with some tin/tout conspec recocnition. Should a bot that uses tieval that way be allowed in F2. I think it may be allowed.


I would like to change the rule towards, no tieloc/tieval stores in other bots then your own specie when controlling ties.

What do you all think?
Oh my god, who the hell cares.

Offline Moonfisher

  • Bot Overlord
  • ****
  • Posts: 592
    • View Profile
Pole for the new F2 rules.
« Reply #5 on: November 06, 2008, 05:52:52 PM »
There's 10 .tin/.tout sysvars.... I don't see that nr 11 would make a decisive difference... if you're short have your bot aim at whatever it's tied to and it'll be able to read .in/.out aswell bringing you up to 20... .tieloc and .tieval CAN be used for comunication... but will almost always be used as a weapon....

But sure if you're only doing it to your own species... only problem is it's a lot harder to check for, and you never know if you're tied to your own species... Seasnake used to tie to spinner because it had the cameleon behavior... then you can argue that you wouldn't try to harm your own, but still... storing anything in a sysvar could pose a problem for oponents, and when you have something in .tieloc/.tieval then you need to make sure you set them to 0 every time you used them or tieing to an oponent by accident (wich will happen at times) will cause an action on it.
I don't see the harm in allowing it, but it makes the rules more complex and leaves loopholes, and if you use the & operator you can just stack bits and transfer a ton of actions anyway... unless you need to transfer values in the area of 30000 every time on every in/out leaving no space for extra data....

The other way all you need to do to check a bot is search for .tie and .tieloc/.tieval, if both are there it's illegal for F2, if only one is it's cleared...

But if you wan't to store a value in .up for instance, then the value will be rather low and you can just use bit shifting to send more data along. You'll get a lot more data transfered by shifting bits or aiming at the target and using .in/.out than by allowing .tieloc/.tieval for comunication.