[Home] [Groups] - Message: [Prev in Group] [Next in Group]
3021: Re: [MUD-Dev] Virtual Chemistry
[Full Header] [Plain Text]
From: clawrenc@cup.hp.com
Newsgroups: nu.kanga.list.mud-dev
Date: Thu, 14 Aug 97 13:37:22 -0700
References: [1]
Organization: Kanga.Nu
In <Pine.LNX.3.96.970812071341.128B-100000@mpc.dyn.ml.org>, on
08/12/97
at 08:15 AM, Matt Chatterley <root@mpc.dyn.ml.org> said:
>On Mon, 11 Aug 1997 clawrenc@cup.hp.com wrote:
>> at 03:09 PM, Matt Chatterley <root@mpc.dyn.ml.org> said:
>>
>> >On Wed, 30 Jul 1997 clawrenc@cup.hp.com wrote:
>>
>> FWIW I'd strongly recommend using a sparse array and interpolating the
>> missing chunks at runtime.
>Especially if you have a lot of coordinates. I'm probably going to
>set it up so you can leave anything that counts as 'normal'
>undefined, and will be set as 0+/- a small variance for magical flux
>(or somesuch). For instance there is no need to implicitly declare
>neutral Ph, or Nil magical capacitance in a substance.
My first thought was a sparse array where say a particular node might
have value Q, and a node further along the same axis might have a
value R. The system would then linearly interpolate the node values
between. Thus only the nodes which marked points of change in
gradient would have to be defined.
--
J C Lawrence Internet: claw@null.net
(Contractor) Internet: coder@ibm.net
---------------(*) Internet: clawrenc@cup.hp.com
...Honorary Member Clan McFUD -- Teamer's Avenging Monolith...