--- Log opened Tue Feb 17 00:00:23 2004
00:12:48 #corewars: < grabek> exit
00:12:51 -!- grabek [~grabek@nat-b.acn.pl] has quit [leaving]
00:37:53 !irc.koth.org *** Looking up your hostname...
00:37:53 !irc.koth.org *** Checking Ident
00:37:53 !irc.koth.org *** No Ident response
00:37:53 !irc.koth.org *** Found your hostname
00:37:54 !irc.koth.org *** Your host is irc.koth.org[216.231.108.170/6667], running version 2.8/hybrid-6.3.1
00:37:54 -!- Mode change [+i] for user jk_log
00:37:54 -!- jk_log [~jakub@134.gymko.ba.gtsi.sk] has joined #corewars
00:42:59 -!- grabek [~grabek@nat-b.acn.pl] has joined #corewars
00:58:28 #corewars: < brx> done
01:00:32 #corewars: < grabek> g'night
01:00:36 #corewars:  * grabek waves
01:00:37 #corewars: < brx> night!
01:00:41 #corewars: < grabek> by the way, brx
01:00:49 #corewars: < brx> yes?
01:00:50 #corewars: < grabek> post either to r.g.cw
01:00:57 #corewars: < grabek> or url it
01:01:06 #corewars: < brx> oh i am reading the newsgroup
01:02:24 #corewars:  * grabek waves once again
01:02:28 #corewars:  * brx waves back
01:40:34 -!- brx [~brx@pD9EAAAD3.dip.t-dialin.net] has quit [leaving]
02:06:54 -!- Mr_Alert [1001@cpe-66-27-200-2.socal.rr.com] has quit [Remote closed the connection]
02:11:02 -!- Mr_Alert [~joe@cpe-66-27-200-2.socal.rr.com] has joined #corewars
02:21:24 -!- Mr_Alert [~joe@cpe-66-27-200-2.socal.rr.com] has quit [Read error: 54 (Connection reset by peer)]
02:22:31 -!- Mr_Alert [~joe@cpe-66-27-200-2.socal.rr.com] has joined #corewars
02:55:49 -!- phunk [~daquik@s86.dial2.sne.nac.net] has joined #corewars
07:03:59 -!- Mr_Alert [~joe@cpe-66-27-200-2.socal.rr.com] has quit []
07:06:34 -!- Mizcu [Mizcu@dsl-hkigw4m4a.dial.inet.fi] has joined #corewars
08:54:38 -!- _phunk [~daquik@s104.dial1.sne.nac.net] has joined #corewars
08:55:32 -!- phunk [~daquik@s86.dial2.sne.nac.net] has quit [Ping timeout: 180 seconds]
08:55:41 -!- _phunk is now known as phunk
08:56:53 -!- willvarfa [~Will@arken-16-57-105.ip-pluggen.com] has joined #corewars
09:10:00 #corewars: < phunk> Hello Will
09:12:40 #corewars: < willvarfa> hello phunk
09:12:45 #corewars: < willvarfa> 5mins
09:20:38 -!- will_varf [~Will@arken-16-57-54.ip-pluggen.com] has joined #corewars
09:22:57 -!- willvarfa [~Will@arken-16-57-105.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
09:30:20 #corewars: < will_varf> back
09:30:27 #corewars: < will_varf> so, phunk, whats up?
09:39:47 -!- _phunk [~daquik@s135.dial1.sne.nac.net] has joined #corewars
09:41:16 -!- phunk [~daquik@s104.dial1.sne.nac.net] has quit [Ping timeout: 180 seconds]
09:51:47 -!- willvarfa [~Will@arken-16-57-107.ip-pluggen.com] has joined #corewars
09:53:58 -!- will_varf [~Will@arken-16-57-54.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
09:57:38 -!- _phunk is now known as phunk
10:03:27 -!- phunk [~daquik@s135.dial1.sne.nac.net] has quit [Ping timeout: 180 seconds]
10:23:12 -!- will_varf [~Will@arken-16-57-37.ip-pluggen.com] has joined #corewars
10:25:46 -!- willvarfa [~Will@arken-16-57-107.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
10:54:13 -!- willvarfa [~Will@arken-16-57-69.ip-pluggen.com] has joined #corewars
10:56:42 -!- will_varf [~Will@arken-16-57-37.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
11:30:15 -!- will_varf [~Will@arken-16-57-76.ip-pluggen.com] has joined #corewars
11:33:09 -!- willvarfa [~Will@arken-16-57-69.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
12:01:21 -!- willvarfa [~Will@arken-16-57-120.ip-pluggen.com] has joined #corewars
12:04:18 -!- will_varf [~Will@arken-16-57-76.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
12:14:05 -!- willvarfa [~Will@arken-16-57-120.ip-pluggen.com] has quit [Lämnar]
12:18:08 -!- willvarfa [~Will@arken-16-57-120.ip-pluggen.com] has joined #corewars
12:35:18 -!- willvarfa [~Will@arken-16-57-120.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
12:42:55 -!- willvarfa [~Will@arken-16-57-34.ip-pluggen.com] has joined #corewars
12:55:16 -!- joonas [jpihlaja@kruuna.helsinki.fi] has joined #corewars
12:56:05 #corewars: < Mizcu> hi joonas
12:56:16 #corewars: < joonas> hi miz
12:56:23 #corewars: < joonas> on holiday?
12:56:31 #corewars: < Mizcu> yes
12:56:54 #corewars: < joonas> nice.  done any skiing yet?
12:57:08 #corewars: < Mizcu> i dont plan to do any
13:01:00 #corewars: < willvarfa> hello folks
13:02:02 #corewars: < joonas> hi will
13:03:36 -!- will_varf [~Will@arken-16-57-117.ip-pluggen.com] has joined #corewars
13:06:19 -!- willvarfa [~Will@arken-16-57-34.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
13:22:06 -!- will_varf [~Will@arken-16-57-117.ip-pluggen.com] has quit [Lämnar]
13:25:33 -!- willvarfa [~Will@arken-16-57-113.ip-pluggen.com] has joined #corewars
13:25:34 #corewars: < willvarfa> back
13:53:45 -!- will_varf [~Will@arken-16-57-116.ip-pluggen.com] has joined #corewars
13:56:29 -!- willvarfa [~Will@arken-16-57-113.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
14:25:12 -!- willvarfa [~Will@arken-16-57-18.ip-pluggen.com] has joined #corewars
14:27:47 -!- will_varf [~Will@arken-16-57-116.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
14:56:14 -!- will_varf [~Will@arken-16-57-41.ip-pluggen.com] has joined #corewars
14:58:33 -!- willvarfa [~Will@arken-16-57-18.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
14:58:56 -!- Roy [~roy_van_r@guest-dhcp-216.si.hhs.nl] has joined #corewars
14:59:02 #corewars: < Roy> Hi
14:59:11 #corewars: < joonas> hi roy
14:59:52 #corewars:  * Roy just send his Frenzy entries
15:00:02 #corewars: < Roy> Are you going to compete?
15:00:10 #corewars: < joonas> roy: what kind of read/write limits would you put in a r/w limited hill?
15:00:14 #corewars: < joonas> no
15:00:33 #corewars: < joonas> the read limit doesn't have to be the same as the write limit.
15:01:04 #corewars: < Roy> Uhm...read: 1000 write: 200
15:01:12 #corewars: < Roy> Would be fun I guess
15:01:28 #corewars: < Roy> What where you thinking? (and others?)
15:03:17 #corewars: < joonas> along the same lines. read limit >> write limit.
15:03:40 #corewars: < joonas> although perhaps in a bigger core?
15:03:57 #corewars: < Roy> Yes, then get the problem: You know where they are, but how do you get there :)
15:04:02 #corewars: < Mizcu> It would be interesting to see what kind of information-relaying solution could be done in such enviroment.
15:04:37 #corewars: < Roy> Bigger core -> Lot more brainless clears/papers I guess...?
15:04:58 #corewars: < Mizcu> for example: large(no big or x) coresize, first one to find "target" and to mark it succesfully wins.
15:05:09 #corewars: < joonas> that's why I thought that the read limit could be CORESIZE
15:05:36 #corewars: < Roy> Yes, but then you only have a write-limit hill :) would also be fun though
15:05:58 #corewars: < Roy> In that case you encourige scanners though
15:06:08 #corewars: < joonas> yes.
15:06:24 #corewars: < Roy> But when you have readlimit<CORESIZE they have to copy themselfs too, maybe more fun
15:06:35 #corewars: < Roy> Maybe readlimit 2667?
15:06:37 #corewars: < Mizcu> but hoppers are deadly to scanners and hop's distance can be << write
15:07:02 #corewars: < joonas> they have to copy themselves anyway to be able to stun/kill the enemy.
15:07:30 #corewars: < joonas> well, copy something, anyway.
15:07:34 #corewars: < Roy> Well yes, but then you still have one main program
15:07:52 #corewars: < Roy> with read limits you encourage new kinds of warriors
15:08:14 #corewars: < joonas> oh, and in a write limited hill you could have truly multiprocess scanners (many instances of a scanner.)
15:08:48 #corewars: < Roy> Yes, but you don't need multiprocess yet without read limit
15:09:04 #corewars: < joonas> hm.. my only worry is that a severly read limited hill would degenerate into paper likes
15:09:12 #corewars: < Roy> Then you could have one main scanners that sends helpers that kill
15:10:07 #corewars: < Roy> Thats why maybe 2000 would be nice (4 instances scan everything)
15:10:32 #corewars: < Roy> But the problem with the write limit is getting the scannerinstance to the 2k place away :)
15:10:50 #corewars: < joonas> heh... you need a multistage bootloader. :-D
15:11:28 #corewars: < Roy> Hmm, how about 2000/200 that could be fun
15:12:17 #corewars: < joonas> sure you wouldn't like a bigger hill?
15:12:45 #corewars: < Roy> Well, maybe a bigger hill like 16000 or something, and warriorsize 200..? not too big
15:12:51 #corewars: < joonas> stones: paper, paper: scanner
15:13:58 #corewars: < Roy> We should have a website with a poll :)
15:14:00 #corewars: < joonas> I mean by that that there could be a kind of reversal of the redcode trichotomy with suitable r/w parameters.
15:14:51 #corewars: < joonas> good idea.
15:15:07 #corewars: < joonas> maybe a poll on the newsgroup
15:15:08 #corewars: < Roy> Heh, maybe it would be fun to have a shrinking core contest once, you start with read/write==CORESIZE but every 2 cycles read/write -- :)
15:15:20 #corewars: < joonas> LOL
15:15:47 #corewars: < joonas> you've a sick mind roy :)
15:16:02 #corewars: < Roy> But it would be "fun" (or torture)
15:16:07 #corewars: < joonas> claustrophobic warriors :)
15:16:21 #corewars: < joonas> oh no, the sky is falling!
15:16:41 #corewars: < Roy> Hmm, shriking isn't good, then qscans -> stone -> papers would rule...
15:17:09 #corewars: < joonas> unless you make the core big enough that qscans aren't an option.
15:17:17 #corewars: < Roy> Maybe an expending core? that would encourage smart warriors, first read/write==WARRIORLENGTH and then it gets bigger :)
15:17:39 #corewars: < joonas> yeah, make paper really easy to spot too.
15:17:50 #corewars: < Roy> You start without knowing anything about the core, and nowhere to go, but you get more and more space to expend
15:18:04 #corewars: < will_varf> but how to know the current size of core?
15:18:16 #corewars: < will_varf> writing your own counters is messy
15:18:26 #corewars: < Roy> It just wraps?
15:18:30 #corewars: < will_varf> how about a certain pspace location containing the counter?
15:18:30 #corewars: < Roy> Counters if your own problem :)
15:18:31 #corewars: < Roy> if=is
15:18:41 #corewars: < Roy> You could have that yes..
15:19:19 #corewars: < joonas> hm... one problem with the shrinking/expanding idea: the limits would have to increas/decrease by jumps
15:19:54 #corewars: < Roy> What do you mean?
15:20:10 #corewars: < joonas> well the read/write limits have to be divisors of coresize.
15:20:31 #corewars: < joonas> otherwise you get really unintuitive semantics for reads/writes outside the limits.
15:20:39 #corewars: < Roy> Ah...
15:20:58 #corewars:  * Roy didn't yet look beyond the idea :)
15:21:46 #corewars: < Roy> I didn't even know if you could make a normal warrior with only 1,0,-1, but when I proposed the tournament I saw papers/scanners/oneshots etc....
15:22:26 #corewars: < joonas> that was a nasty idea.  sure you don't have a bit of a sadistic streak? ;)
15:22:57 #corewars: < Roy> You should ask my girlfriend ;)
15:23:15 #corewars: < joonas> lol... perhaps I don't want to know after all.
15:24:20 #corewars:  * joonas goes off to post a poll on r.g.c.
15:25:07 #corewars: < will_varf> back in 10min
15:25:11 -!- will_varf [~Will@arken-16-57-41.ip-pluggen.com] has quit [Lämnar]
15:27:29 -!- joonas [jpihlaja@kruuna.helsinki.fi] has quit [Ping timeout: 180 seconds]
15:28:47 -!- joonas [jpihlaja@kruuna.helsinki.fi] has joined #corewars
15:28:59 #corewars: < Roy> Ok, you already know my aswer :)
15:31:59 -!- joonas [jpihlaja@kruuna.helsinki.fi] has quit [Ping timeout: 180 seconds]
15:34:42 -!- Roy [~roy_van_r@guest-dhcp-216.si.hhs.nl] has quit []
15:41:30 -!- willvarfa [~Will@arken-16-57-110.ip-pluggen.com] has joined #corewars
15:56:40 -!- will_varf [~Will@arken-16-57-2.ip-pluggen.com] has joined #corewars
15:59:35 -!- willvarfa [~Will@arken-16-57-110.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
16:11:52 -!- CoreChild [~d432ba35@brparcunix.cidadeinternet.com.br] has joined #corewars
16:11:55 #corewars: < Mizcu> CC
16:16:44 #corewars: < CoreChild> Hi Mizcu.
16:17:24 #corewars: < CoreChild> Just a quick visit.  I will be back later (about 6 p.m. GMT)
16:19:30 #corewars:  * CoreChild still agrees read limit > write limit.
16:27:42 -!- willvarfa [~Will@arken-16-57-31.ip-pluggen.com] has joined #corewars
16:30:03 -!- will_varf [~Will@arken-16-57-2.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
16:38:42 -!- CoreChild [~d432ba35@brparcunix.cidadeinternet.com.br] has quit [WebChat -> www.vIRCio.org  (Ping timeout)]
16:57:32 -!- willvarfa [~Will@arken-16-57-31.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
17:11:21 -!- willvarfa [~Will@arken-16-57-112.ip-pluggen.com] has joined #corewars
17:26:32 -!- will_varf [~Will@arken-16-57-55.ip-pluggen.com] has joined #corewars
17:29:31 -!- willvarfa [~Will@arken-16-57-112.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
18:02:32 -!- willvarfa [~Will@arken-16-57-13.ip-pluggen.com] has joined #corewars
18:05:08 -!- will_varf [~Will@arken-16-57-55.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
18:33:36 -!- will_varf [~Will@arken-16-57-45.ip-pluggen.com] has joined #corewars
18:36:07 -!- willvarfa [~Will@arken-16-57-13.ip-pluggen.com] has quit [Ping timeout: 180 seconds]
18:51:30 -!- CoreChild [~John@217.158.137.194] has joined #corewars
18:51:34 #corewars: < CoreChild> Hi All
18:52:43 #corewars: < Mizcu> cc
18:53:06 #corewars: < Mizcu> not using the same webchat, i see
18:53:18 #corewars: < CoreChild> Hi Mizcu.
18:53:50 #corewars: < CoreChild> No, I'm in the netcafe. I have installed Xircon (a freeware Windows IRC prog) on a couple of the machines
18:54:07 #corewars: < CoreChild> (It refuses to install on one of them for some reason)
18:54:56 -!- will_varf [~Will@arken-16-57-45.ip-pluggen.com] has quit [Lämnar]
19:00:10 #corewars: < Mizcu> Shall we continue discussion about the clp-hybrid 
19:02:01 #corewars: < CoreChild> Ah yes.
19:02:11 #corewars: < CoreChild> I didn't have chance to look into it.
19:02:20 -!- phunk [~daquik@s132.dial2.sne.nac.net] has joined #corewars
19:02:30 #corewars: < CoreChild> I have only had 1 hour of redcoding time, and I used that for my RF16 entry no. 2
19:02:33 #corewars: < CoreChild> Hi Phunk
19:03:27 #corewars: < CoreChild> 57 characters, 126 Wilfiz
19:03:53 #corewars: < CoreChild> This screen has a severely annoying flicker
19:06:34 #corewars: < Mizcu> a)  Is one dodge enough after fence has been damaged?   b) Should there be dodge while clearing?  c) Should there be creation of additional fences after dodgind?
19:07:39 -!- joonas [jpihlaja@kruuna.helsinki.fi] has joined #corewars
19:07:54 #corewars: < Mizcu> hi joonas
19:07:59 #corewars: < joonas> hi mizcu
19:08:06 #corewars: < phunk> Hi, Corechild
19:08:35 #corewars: < joonas> hi phunk
19:09:45 #corewars: < CoreChild> Hi Joonas
19:09:48 #corewars: < joonas> for some reason mailing the COREWAR-L list is way slower than just posting to the newsgroup.
19:10:03 #corewars: < CoreChild> Possibly one jump over the fence would be enough
19:10:18 #corewars:  * CoreChild doesn't know why, but had noticed
19:10:57 #corewars: < phunk> Hello, joonas
19:11:05 #corewars: < joonas> ah, the one shot that clps?
19:11:22 #corewars: < CoreChild> I found out how Michal makes his CLP 8 lines.  He relies on empty core.
19:11:33 #corewars: < joonas> sounds dodgy
19:12:07 #corewars: < joonas> one pass of the opponent and bye bye clp :-/
19:12:09 #corewars: < CoreChild> Yes. Oneshot+CLP = CLP-Shot?
19:12:19 #corewars: < Mizcu> What would be the optimal jump distance..  4000 might/would be bad in case of a bishot
19:12:50 #corewars: < CoreChild> Just past the fence I think :-)
19:13:11 #corewars: < joonas> unless it sees a djn stream.
19:13:31 #corewars: < CoreChild> For example, decoy at -100, fence at -80.  If something hits the fence jump to -160
19:13:39 #corewars: < Mizcu> but then is needed a detection on which fence is jumped over
19:14:28 #corewars: < CoreChild> Opponent oneshot becomes clear one finding the decoy, then overwrites fence.  We dodge the clear and have at least CORESIZE cycles to dispose of him
19:14:52 #corewars: < CoreChild> Just have one fence.  Most clears move forward.
19:15:08 #corewars: < joonas> hm. would the oneshot keep on clearing at the spot it was before dodging?
19:15:25 #corewars: < CoreChild> DJN streams move backwards though... but are they much of a worry?
19:15:35 #corewars: < joonas> i.e. should it re-adjust the wipe ptr when dodging?
19:16:14 #corewars: < CoreChild> Well, it could just do the CLP bit in the scan loop
19:16:33 #corewars: < joonas> ah, of course. otherwise most of the CORESIZE cycles will be wasted rewiping.
19:16:36 #corewars: < CoreChild> Can CLP be combined with bomb-dodger?
19:17:10 #corewars: < joonas> heh.. that would be nice: an intelligent one shot that instea of ignoring bombs, squats a clp/shot on it. :-)
19:17:22 #corewars: < Mizcu> dodgind clear or a dodging scanner?
19:17:24 #corewars:  * CoreChild introduces another idea before we have discussed the first one properly!
19:19:16 #corewars: < joonas> it should probably only bomb dodge once.
19:19:26 #corewars: < joonas> and then go into clp dodging mode.
19:19:46 #corewars: < CoreChild> Is Lukasz's page up?
19:20:19 #corewars: < CoreChild> Ah yes, perfect against the stone -> clear
19:20:32 #corewars: < joonas> so you have bomb identifying code in the oneshot's scan loop: if it finds a bomb -> boot the clp/clear on it, if not -> just start the clp/clear.
19:20:34 #corewars: < CoreChild> Dodge the bomb, then wait for the clear and dodge that too :-)
19:21:18 #corewars: < joonas> heh... Digitalis, I have your number! :-)
19:21:19 #corewars: < CoreChild> Well, my initial idea was:
19:21:29 #corewars: < CoreChild> loop add   inc,   ptr
19:21:46 #corewars: < CoreChild> ptr  seq   a,     b
19:21:57 #corewars: < CoreChild>      jmp   oneshotfoundsomething
19:22:12 #corewars: < CoreChild> jmz.f loop,  fence
19:22:30 #corewars: < CoreChild> ;if reach here, clp found something
19:23:26 #corewars: < joonas> ah. hm. yes. so ideally you want the code at "oneshotfoundsomething" to try to identify stray bombs/decrements.
19:24:38 #corewars: < joonas> the clp bit would have to be an spl -> dat clearer, no?
19:24:41 #corewars: < CoreChild> Yes, that would be a good addition.
19:25:11 #corewars:  * joonas worries about imps
19:25:20 #corewars: < CoreChild> Yes.  We could have a spl -> dat clear whatever happens.
19:25:32 #corewars: < CoreChild> If the oneshot finds something we clear there
19:25:50 #corewars: < CoreChild> If we find a bomb, we move the clear onto it and jump to it.
19:26:09 #corewars: < CoreChild> If the fence is overwritten, we put the clear the other side of it and jump to it
19:26:28 #corewars: < joonas> right. now how do we deal with imps?
19:27:06 #corewars: < joonas> the dat clear could be a <2667, <5335 wipe.
19:27:28 #corewars: < joonas> only a kludge though.
19:27:47 #corewars: < joonas> see, Roy's got a nasty habit of stuffing 7 point imps everywhere .-/
19:29:02 #corewars: < CoreChild> Use a spiral clear :-)
19:29:09 #corewars: < joonas> doh
19:29:13 #corewars: < joonas> of course.
19:29:25 #corewars: < CoreChild> Have you ever tried oneshot -> spiral clear by the way?
19:29:33 #corewars: < joonas> not that I recall.
19:29:53 #corewars: < joonas> I think I would have entered into the KOFACOTO tournament one if I had.
19:30:08 #corewars: < joonas> the one where michal's tiny spiral wipe slaughtered RotJ.
19:30:18 #corewars:  * CoreChild wonders if it works.  Probably not well, since no-one has bothered so far (that I noticed)
19:30:33 #corewars: < joonas> see above. :-)
19:31:07 #corewars: < CoreChild> I wonder if '88 spiral clear would be any good.
19:32:10 #corewars: < joonas> hm.. I think so.  even an add/sub spiral clear works if core is otherwise datted.
19:32:36 #corewars: < CoreChild> Zooom-trick now has a lexicon entry
19:33:12 #corewars: < CoreChild> We were also discussing a oneshot which changed it's step each round (via p-space)
19:33:22 #corewars: < CoreChild> It would change step regardless of win / loss
19:33:32 #corewars: < joonas> why?
19:34:01 #corewars: < Mizcu> capability to beat multiple different types of opponents
19:34:02 #corewars: < CoreChild> The purpose being to use it in benchmarks, so you would be optimizing against a generalized oneshot
19:34:38 #corewars: < joonas> ah. hey, that's a good idea.
19:34:42 #corewars: < CoreChild> Rather than against a specific oneshot.  Scores against oneshots are highly variable
19:35:09 #corewars: < CoreChild> And, try to ensure it doesn't get brainwashed!  Overwrite the STP
19:35:13 #corewars: < joonas> for papery things, sure.
19:36:00 #corewars: < CoreChild> The steps could be in a table.  (maybe some other constants too, such as boot distance)
19:36:15 #corewars: < CoreChild> Is there anything else worth generalizing?
19:36:45 #corewars: < joonas> hm. sounds simpler o just have lots of different oneshots in your benchmark.
19:37:15 #corewars: < joonas> a few different steps in one warrior would be good, but many steps and boot distances too?
19:37:24 #corewars: < CoreChild> Then oneshots would make up a higher percentage of your benchmark.
19:37:45 #corewars: < joonas> just scale the results.
19:38:10 #corewars: < CoreChild> Yes, I do this.
19:38:26 #corewars: < joonas> having too much of one thing isn't half as bad as having too little of another.  you can scale "too much" into "just right", but not "too little" into "just right".
19:38:43 #corewars: < CoreChild> Sort of, I run 500 rounds for some warriors, 50 rounds for other warriors.
19:38:55 #corewars: < joonas> (the errors in scores get amplified too if you scale "too little" up.)
19:39:41 #corewars: < CoreChild> I have 10 oneshots at 50 rounds to be the equivalent of 1 oneshot at 500 rounds
19:39:58 #corewars: < joonas> yes, that works.
19:40:51 #corewars: < joonas> do you run paper for 500 rounds too?
19:41:55 #corewars: < CoreChild> Yes, though I wouldn't think I need to
19:42:39 #corewars: < joonas> no.  never hurts though. :)
19:42:43 #corewars: < joonas> except your patience.
19:43:12 #corewars: < CoreChild> Hmmm... I just leave it while I go to work.
19:43:34 #corewars:  * CoreChild does not computer optimize very much now.
19:44:17 #corewars: < CoreChild> Also, I will now only computer optimize against computer optimized warriors...  I leave the one's with hand-picked constants alone!
19:45:03 #corewars: < joonas> so you're still maintaining that OoS was "hand picked" even though at least four people have done extensive computer runs trying to find better ones? ;-)
19:45:12 #corewars: < CoreChild> Yes.
19:46:24 #corewars: < joonas> they are incredible steps though, you have to admit.  congratulations once again.
19:47:10 #corewars: < CoreChild> I think I tried a step of 11, 13, 17 and 19.  dist I chose so it was about 1/3 of coresize and the patterns from the two scan pointers fitted well.  Time I worked out on my pocket calculator
19:47:13 #corewars: < CoreChild> Thanks
19:48:21 #corewars: < CoreChild> Easy to add a qscan or decoy-maker to it, but I never wanted to
19:51:08 #corewars: < CoreChild> Two minutes before I have to go
19:51:40 #corewars: < joonas> ok. see you later then.
19:51:56 #corewars: < joonas> I should be off too.  my car got stolen so I've a long ride home.
19:52:08 #corewars: < Mizcu> ouch
19:53:02 #corewars: < CoreChild> :-(
19:54:35 #corewars: < CoreChild> Recon has hand-picked steps.  Can you improve on those? ;-)
19:54:49 #corewars: < joonas> hah.. anything from Moore isn't hand picked.
19:55:02 #corewars: < joonas> I'm almost sure.
19:55:04 #corewars: < CoreChild> Recon's step is
19:55:29 #corewars: < joonas> you're just saying that so you can keep Recon2 in your benchmark. ;)
19:55:29 #corewars: < CoreChild> Right time to go :-(
19:55:33 #corewars: < joonas> yes. me too.
19:55:37 #corewars: < joonas> bye
19:55:41 -!- joonas [jpihlaja@kruuna.helsinki.fi] has quit [ircII2.8.2-EPIC3.004+Kasi --- Bloatware at its finest.]
19:55:45 #corewars:  * CoreChild waves
19:55:49 #corewars: < Mizcu> see'ya then
19:56:10 #corewars: < CoreChild> No, no hand-picked constants in my benchmark
19:56:12 #corewars: < CoreChild> Cheers
19:56:26 -!- CoreChild [~John@217.158.137.194] has quit [mov.i #1,1]
19:58:00 -!- phunk [~daquik@s132.dial2.sne.nac.net] has quit [Ping timeout: 180 seconds]
20:25:56 -!- phunk [~daquik@s125.dial1.sne.nac.net] has joined #corewars
21:50:50 #corewars: < grabek> publishing a blog is harder than one can expect...
22:32:24 -!- phunk [~daquik@s125.dial1.sne.nac.net] has quit [Ping timeout: 180 seconds]
22:41:55 -!- Mizcu [Mizcu@dsl-hkigw4m4a.dial.inet.fi] has quit [gate!]
--- Log closed Wed Feb 18 00:00:37 2004