> In SGII we are introduced to GP ordnance for Artillery. For ships
This got me thinking about how the chits are spread and the odds of drawing a
particular valid chit. Note that Green and Yellow have the exact same number
and distribution of chits.
This is for the drawing of 1 chit. If someone wants to extend these tables to
take into account drawing multiple chits, please do so.
Number of Chits:
G/Y Red G&Y G/Y&R All
T: 20 50 40 70 90 (110 including special)
0: 3 5 6 8 11
1: 10 20 20 30 40
2: 7 15 14 22 29
3: 5 10 10 15 20
Boom: 5 Firer: 2 Targeting: 5 Mobility: 7
Odds of drawing a specific chit:
% G/Y Red G&Y G/Y&R All
T: 18.2 45.5 36.4 63.6 81.8
0: 02.7 02.7 05.5 07.3 10.0
1: 09.0 18.2 18.2 27.3 36.4
2: 06.4 13.6 12.7 20.0 31.8
3: 04.5 09.0 09.0 13.6 18.2
Boom: 04.5% Firer: 01.8% Targeting: 04.5% Mobility: 06.4%
> On Wed, 4 Feb 1998, Brian Bell wrote:
> This is for the drawing of 1 chit. If someone wants to extend these
I tried to do that. If you account for all possible "valid" combinations, the
problem is that without fudging there are simply too many
combinations -- you'd be rolling d10000000 or something.
If you make a separate table for each valid combination, it might be
manageable -- I just never got that far into it. But I suspect it might
still be d10000 or so.
Another choice is to fudge a bit, and group together the combinations with
only marginal probability.
> On Thu, 5 Feb 1998, Mikko Kurki-Suonio wrote:
> On Wed, 4 Feb 1998, Brian Bell wrote:
combinations,
> the problem is that without fudging there are simply too many
I did this a relatively simple way a while back (more than a year).
The way I handled it was to use a combination of a D10(0-9) and a D12.
The D12 counted as the "tens" (I think you have to ignore 12s). I had a chart
to cross reference the die roll with the corresponding damage chit.
To work out higher class weapon hits, you would use five sets of color
coded dice, re-rolling the very very rare duplicates (if you want to
recreate the odds _exactly_). If you aren't worried about duplicates,
(the difference is negligible) then one set of dice is fine.
The problem is that when I made the chart up, I had no opponents, and the
chart got lost in the pile of papers from school;). I suppose I could look for
it, but it's probably easier to start from scratch.
I did post it to the space-marine mailing list (to which I unsubscribed
after I got involved with NetEpic...). Perhaps someone on this list who used
to be on that list still has it handy? Heck, maybe they even used it...
> --
> On Wed, 4 Feb 1998, Tony Christney wrote:
> I did this a relatively simple way a while back (more than a year).
That's easy. A single chit draw fits nicely on a d100 table.
> To work out higher class weapon hits, you would use five sets of color
Erm, that's in effect rolling X times on the 1 chit table, something I
tried to avoid. IMHO, the whole point of rolling dice and table-lookup
is that it has to be faster than picking chits. I don't find rolling 10
different colored d10's very convenient (in effect 1d10000000000).
> The problem is that when I made the chart up, I had no opponents, and
I have the chart on my web pages, it's at
http://www.swob.dna.fi/mini/gzg/
I think.
I think that may be the table I kludged up, but hey, the math is the same.
Gene
> ----------
> On Wed, 4 Feb 1998, Tony Christney wrote:
Not quite, I think that there are ~119(ie. more than 100) different chits.
Hence the D12-D10 combination.
> To work out higher class weapon hits, you would use five sets of
Again, not exactly my intention. You would only have, at most, five
different colors. Each _color_ corresponds to a chit draw. Basically
the dice work in tandem. The major speed increase comes from the fact that you
can "draw" all of the required chits at the same time with a single handful of
dice. Also, it's not the same as rolling X times on the one chit table. Chit
draws are not statistically independent, whereas a d10000000000 is
statistically independent.
Finally, there is another advantage, losing one or more chits changes the
odds. You can't lose one facet of a die;)
Cheers, Tony.
> --
> On Thu, 5 Feb 1998, Mikko Kurki-Suonio wrote:
> The problem is that when I made the chart up, I had no opponents, and
Nope, my chart is _very_ different. I will try to dig it up and post it
for all to see...
> --
> Finally, there is another advantage, losing one or more chits
Speak for yourself Tony. I can personally attest that in some scenarios I've
participated in, I was quite positive I'd lost several facets (the high value
ones) off my dice......
(grin)
T.
/************************************************
> On Thu, 5 Feb 1998, George,Eugene M wrote:
> I think that may be the table I kludged up, but hey, the math is the
It may very well be. I didn't make up that table, I just translated it
into HTML (and managed to lose the credits in process -- I'm really
sorry about that). If yours was an Excel table posted on this list or
somewhere on the web, I'll credit it to you.
> On Thu, 5 Feb 1998, Tony Christney wrote:
> Not quite, I think that there are ~119(ie. more than 100) different
I started work from someone else's (Eugene's?) table with the assumption
that it was correct, so you may be right there. In any case, it's a minor
difference.
Though I do think the chits were on a separate punch-out sheet, so an
even number is more likely.
> Again, not exactly my intention. You would only have, at most, five
So it's 5 d10's and 5 d12's -- not a major difference, IMHO.
> Each _color_ corresponds to a chit draw. Basically
As you said, you can reroll the doubles, but even if you don't, it's almost
the same...
> Finally, there is another advantage, losing one or more chits
That is true... maybe I'll resume work on it and try to fit the results into a
d30 x d30 table (I bought a bunch of d30's from a closeout sale
;-)
I think d100 x d100 is the maximum reasonable...
Yeah it's mine. No worries, the credit really goes to GZG, I just regurgitated
the info.
Gene