Re: call for duelmasters help

From: ben@abby.chem.ucla.edu

Date: Fri, 23 Oct 1992 11:58:52 +0000

>In article <1c21loINNe59@gap.caltech.edu>, johnjohn@cco.caltech.edu (John White) writes:
>In article <009625EC.C3F63300@abby.chem.ucla.edu> ben@abby.chem.ucla.edu
>writes:
>
>Oh, you want numbers?  I have a 15-14 (and like 20 tourney fights) PL with
>_69_ skills to his name.  17 wit.
>
>I had a 17 wit PR who died at 20 fights; 40 skills.

These are both average, yah? Even for 17 WT.  I'm HAPPY with a warrior who
learnes 2 skills/fight...  In Khalhums, an ally (Chris Winslow) had an original
convention warrior, Pint, a PL.  With a 14-16 wit he learned <2 skills/fight
for 48 out of 50 fights!
>
>>I don't
>>belive that each warrior has a 'luck stat' that influences his learning.
>
>I do.

IMHO, there are two main pieces of evidence AGAINST the luck stat:

1) You would have no reason to know this, so maybe this will have some
infulence:
AT OrcCon this yar I found a piece of scrap printout with partial info. on a
particular warrior.  This, I am pretty sure, was an ACTUAL screen dump of the
program that RSI uses to 'view' warriors.  FYI, the Warrior was named 'Chaos',
of arena 33, a LU that I think was killed at the last mail-in by a  Khalhums
AB, of all things.  The scrap gave Starting Stats, Current stats (which,
incidentally, were unreadable), Favorite Weapon, (SH), Favorite Activity (5),
Favorite Off. EFfort (also 5), and Endurance number (I think 22), Hit Points
(in the 20's) and a carrying capacity.  I did not see skill ratings, nor a
'luck', 'learning', or 'teaching' rating.  Of course, we know skills exist,
 and I beleive in them...

2) There's absolutly no REASON, programming-wise to include a luck stat.  In
real life, people are 'lucky' or 'unlucky', why not just let a string of
random numbers to the job?  For example, Pint, mentioned above, once broke 2
Scimitars each in two fights in a row at a tourney!  But given stochastics
statistics, I don't believe that this is impossible without a 'luck' stat.
Certainly, some warriors, like some people are luckier than others, but I think
this is just an artifact of the large number of random factors used by RSI in
differentt parts of the DM code.

>
>My understanding of the effect of wit on learns is this:  The higher a
>warriors wit, the more skills it's possible to learn in a fight.  I.E. it is
>impossible for a 3 wit warrior to ever learn 6 skills in one fight, though a
>21 wit warrior can.  One thing I have noticed:  warriors with 17 wits
>AVERAGE around 2 skills a fight.   That's all I can say.

Interesting, I hadn't thought of that.  Though I did see a 6 WT AB (wierd
manager) learn 3 @ tourney...  Do you think It sets a minimum as well?  Do 21
WT warriors EVER learn 0 skills?  What's the most skills anyone ever learned in
a fight?  What WT did they have (prob. 21...)? Obviously, the overall minimum
skill learn is 0.  Is there a maximum, of say, 10?  Help, I need more Data!

>>  Do high levels of Attack in the
>>teacher give you a greater chance of learning one or more Attack skills?
>
>At the face to face, the one of the RSI employees made some comment about
>how she would modify the "house" teams for odd-man-out matchups in AD.  She
>said that all those warriors were maxed out in all skills in order to
>maximize their teaching potential.  However, this might be just how SHE
>understood teaching to work.
>
Well, presumably she's either a programmer or was told by one, so assuming
she's not LYING (no reason to...) Dosn't answere whether it's total NUMBER of
skills, or relative skills in each area...   I just had a thought!  Maybe it's
how MAXXED out the warrior is in skills, rather than just skill levels.  That
is to say a TP with 20 attack skills and an AdEx, would be a better teacher
than an AB with 5 skills and a Master.

>>i.e., there is a correlation between time of fights and nubmer of
>>skills learned, but is it a direct effect.
>
>Hmmmm... I don't understand this example.

I mean to say: Fact: Warriors with longer fights learn more skills.
But do they learn more BECAUSE their fights are longer?  Totally unknown.

>
>Well, you have mine.  Do any of you play in 37?  The Dealer claimed that the
>Dark Circle had program information there a couple turns ago.  Hmmmm.

Dark Circle, huh? Well I have a contact, (Moriarty, a.k.a Prof) I'll see what I
can do.  In the mean time, I think I'm going to write a SimDuelmasters code in
C++.  This program would try to emulate DM , rather than being a stand-alone
game... This way I (or anyone with the source) could try out new theories with
out spending a bazillion dollars running 45 teams.  You could also DIRECTLY
test different aspects of the code...  The other reason for doing this is to
learn how to program in C++...

Eventally, I going to try to collect huge amounts of data from other managers,
but this will primarily be for character generation, i.e., initial skills.  I
have the Bloodgames chart, and have seen the RatPak, but these, IMHO, are far
from complete.  I also need better info on Hit Points, and especially
Endurance... Comments from any Mega-managers out there?

Ben Hitz
hitz@cuhhca.hhmi.columbia.edu

"Avoid Everything"
					-LARD 'The Power of Lard'

References

Re: call for duelmasters help
ben@abby.chem.ucla.edu
Tue, 20 Oct 1992 14:42:13 +0000
Re: call for duelmasters help
johnjohn@cco.caltech.edu (John White)
Tue, 20 Oct 1992 22:36:40 +0000

Referenced By

Re: call for duelmasters help
ben@abby.chem.ucla.edu
Tue, 20 Oct 1992 14:42:13 +0000
Re: call for duelmasters help
johnjohn@cco.caltech.edu (John White)
Tue, 20 Oct 1992 22:36:40 +0000