Path: utzoo!utgpu!watserv1!watmath!att!att!linac!pacific.mps.ohio-state.edu!zaphod.mps.ohio-state.edu!usc!apple!agate!tornado.Berkeley.EDU!felixh
From: felixh@tornado.Berkeley.EDU (Felix Hack)
Newsgroups: comp.sys.amiga.games
Subject: Re: Harpoon...
Message-ID: <1990Nov28.235413.14604@agate.berkeley.edu>
Date: 28 Nov 90 23:54:13 GMT
References: <4036@dftsrv.gsfc.nasa.gov>
Sender: usenet@agate.berkeley.edu (USENET Administrator)
Reply-To: felixh@tornado.Berkeley.EDU (Felix Hack)
Organization: ucb
Lines: 50


  Re:  Harpoon Sub Ops,
    I too have obliterated the computer opponent in every scenario EXCEPT
those involving just subs on my side.  I believe there are two major factors
at work causing the difficulty:
  
  1.  The game data (in this case for sonar effectiveness) is based on the
    Harpoon boardgame data as of last year, whereas this summer a new
    updated data annex was published which strongly downgraded all sonars.
    Where a typical submarine sonar might have a passive detection range
    of 15 miles in the current computer game, the new data downgrades it
    to maybe 6 miles.
      Also, almost all convergence zones were degraded by 1 in the new
      data.  Here's a specific comparison:
	USA BQQ-5 Old Passive Range:  25 miles   Old CZ  2
		  New     "     "  :  10    "    New CZ  1

  As you can see the reductions are large.  Right now the US LA sub
    has a better sonar than the Victor II you're trying to engage, so
    you can hear him at 2 CZ, (with towed array) while he can't.  But
    he CAN hear you at 1 CZ, and has a weapon to do something about it
    (SSN15/16) while you don't.  If you can get inside the 1st CZ you
    should be OK since Mk48 torps greatly outrange Soviet ASW torps
    in the game.  Problem is you have to survive a lot of SSN16s first.

  2.  The ASW system of the computer game is based on that of the board
     game, and it is flawed.  The essential problem is that there are
     detection rolls every half minute, not whenever conditions that
     affect sonar contact probabilities change.  Sure, creeping reduces
     your chances of being detected, but the odds don't drop to zero.
     It may take you a time on the order of tens of minutes to cross
     a hostile CZ at slow speeds.  During that time the enemy gets
     on the order of 50 detection attempts.  Even if the chances are
     only 5% (they're usually higher) simple rules of probability
     dictate that you will be picked up.  After that the chances for
     staying detected rise (at least in the boardgame) so you'll have
     a hard time breaking track.
       The miniatures game Warship Commander seems to have a better ASW
     system in its Sea Command module (long out of print).  Instead of
     fixing detection ranges and playing with the probabilities, it 
     uses a variety of factors to determine a basic detection range for
     some situation (noise, sonar type, etc).  A random element is
     used to modify the range up or down, and it is then compared to
     the true range to see if detection occurred.
       You can see how Harpoon's system has problems, how does it make
     sense to let some detection influencing factors alter the odds of
     detection and others alter the range?  It seems that the Warship
     Commander approach of a unified system makes more sense.

  Felix