MartinBraendli2 wrote:

CodeLegend wrote:

jokebookservice1 wrote:

I think every run should be taken 3 times, not just the final

Definitely. (in certain circumstances, RunRunProtect has beaten STINGER!)

Yes, I agree. You don't even have to take the average of the ranks, the sum of ranks (the lower the better) does the same trick with less math.

Also, I'm happy to hear that STINGER can be beaten. RunRunProtect is IMO currently the best public bot.

@MATU
Why do you store the protections in a list? I think the look up is quite expensive. Why don't you use arrays? This would also make it easier to analyse, what lines/variables of a bot are protected.

Pseudocode:

protect(var){
   if (var is a number){
      protection[var]++
   } else {
      protect[getIndex(var)]++
   }
}

CodeLegend wrote:

@MATU

I tried to test my bot against all of the ones competing so far, and it glitched out on me… I downloaded the page and put it on google drive:
https://drive.google.com/open?id=0B-LI96qMKG3bdUVzVlVOeWRFVmM

EDIT: I added them in one at a time, and it crashed when I added InfiniteLoop.

jokebookservice1 wrote:

I don't like the idea of a knockout as mentioned: Some bots require other specific bots to work optimally

MartinBraendli2 wrote:

jokebookservice1 wrote:

I don't like the idea of a knockout as mentioned: Some bots require other specific bots to work optimally

So you want a hack and smash with 30+ bots at once?
If there are a lot of bots around, there will be a lot of weak bots around. The winner then wont be, who performs best against good bots, but who performs best against victim bots. This is contrary to all forms of competition I know. The chess world champion is not determined, by who can beat a 4 year old boy in the least amount of rounds, but by who can beat other world class chess players (hope you get my analogy). To avoid such a thing, you have to eliminate weak bots.

MartinBraendli2 wrote:

jokebookservice1 wrote:

I don't like the idea of a knockout as mentioned: Some bots require other specific bots to work optimally

So you want a hack and smash with 30+ bots at once?
If there are a lot of bots around, there will be a lot of weak bots around. The winner then wont be, who performs best against good bots, but who performs best against victim bots. This is contrary to all forms of competition I know. The chess world champion is not determined, by who can beat a 4 year old boy in the least amount of rounds, but by who can beat other world class chess players (hope you get my analogy). To avoid such a thing, you have to eliminate weak bots.

jokebookservice1 wrote:

MartinBraendli2 wrote:

jokebookservice1 wrote:

I don't like the idea of a knockout as mentioned: Some bots require other specific bots to work optimally

So you want a hack and smash with 30+ bots at once?
If there are a lot of bots around, there will be a lot of weak bots around. The winner then wont be, who performs best against good bots, but who performs best against victim bots. This is contrary to all forms of competition I know. The chess world champion is not determined, by who can beat a 4 year old boy in the least amount of rounds, but by who can beat other world class chess players (hope you get my analogy). To avoid such a thing, you have to eliminate weak bots.

Can we have an elimination system, but at the beginning all bots play against each other

gtoal wrote:

MartinBraendli2 wrote:

My propositon:
- All bots are randomly distributed on 4 groups (A,B,C,D), so that no 2 bots of the same creator are in the same group.

Isn't this verging into Kirkman's School Girl Problem territory?

G

IcyCoder wrote:

IcyCoder wrote:

IcyCoder wrote:

Submission: TTAIB (the totally awesome icy bot)

SET DIRECTION RANDOM
MOVE
IF *@RANDOM%2 EQUALS 5 5 3
SET *@5 @4
TAG
SET EXECUTION_POINTER 1

Version 2:

SET DIRECTION RANDOM
MOVE
IF *@RANDOM%2 EQUALS 5 5 3
SET *@4 @5
TAG v2
SET EXECUTION_POINTER 1

Version 3:

PROTECT EXECUTION_POINTER
SET DIRECTION RANDOM
MOVE
IF *@RANDOM%2 EQUALS 6 6 4
SET *@4 @6
TAG IcyBot
SET EXECUTION_POINTER 1
SET DIRECTION RANDOM
MOVE
IF *@RANDOM%2 EQUALS 6 12 10
SET *@4 @12
TAG IcyBot
SET EXECUTION_POINTER 8
SET DIRECTION RANDOM
MOVE
IF *@RANDOM%2 EQUALS 6 18 16
SET *@4 @18
TAG IcyBot
SET EXECUTION_POINTER 14
SET DIRECTION RANDOM
MOVE
IF *@RANDOM%2 EQUALS 6 24 22
SET *@4 @24
TAG IcyBot
SET EXECUTION_POINTER 20
SET DIRECTION RANDOM
MOVE
IF *@RANDOM%2 EQUALS 6 30 28
SET *@4 @30
TAG IcyBot
SET EXECUTION_POINTER 26
SET EXECUTION_POINTER RANDOM

Credit:
@iamunknown2 for help
@gtoal for his edit of mine
@jokebookservice for a line of his bot

My bot beat everyone but SPRAYnPRAY and CopyBot (CopyBot won) but v3 will be better

I found out that Revision 6 was problematic, so I reverted back to revision 5 and guess what:

iamunknown2 wrote:

Better watch out, Stinger…

iamunknown2 wrote:

Better watch out, Stinger…

jokebookservice1 wrote:

iamunknown2 wrote:

-snip-

have you played against runrunrun?

CodeLegend wrote:

@MATU

I tried to test my bot against all of the ones competing so far, and it glitched out on me… I downloaded the page and put it on google drive:
https://drive.google.com/open?id=0B-LI96qMKG3bdUVzVlVOeWRFVmM

EDIT: I added them in one at a time, and it crashed when I added InfiniteLoop.

TheMonsterOfTheDeep wrote:

Submission: Snipar
-snip-
I'm not sure if this submission is valid, because it sort of really builds off of RunRunRun and RunRunProtect (i.e. protecting all its code ). I can mark it as non-competitive if you guys want.

As it were, this isn't even the real bot - I lost the real bot because I only had its code stored in the controller and I deleted it. Remember kids to never assume your work is saved!