RE: Tempest Multigame (behaviour)

On Wed, 26 May 1999, Clay Cowgill wrote:
> You guys must be using some new definition of "skill level" that I'm
> unfamiliar with. Let's say I play to level 23. Now let's have my 62
> year old non-game-playing father step up to the plate and "buy in" at
> the highest level possible. By using several lives and the super-zapper
> he gets past the level and gets the big bonus.

No, he doesn't. Trust me. Everyone has a point at which they top
out in Tempest, and it takes *LOTS* of practice to get through it.

I managed to raise my high score from 350,000 to around 500,000 over
the space of a month or so, by having 2-3 hour sessions on free play.

The first session had me topping out in the 350,000 range. The next,
I made it past the level that had blocked me and held it for a few
games before I got tired and was beaten back to the last level that
had held me back.

There are "hard groups of levels" and "easy groups of levels" - I find
my play better on levels that aren't circular, since you can stay near
one edge of the web and not worry about being surrounded on both sides
by enemies. So I'll have trouble from level 4 (the bow-tie) of a cycle
through level 7 (the four-leaf-clover), but if by some miracle I can
get through to level 8 (the "V"), I can probably fight my way through
to level 12 (the "heart") before running back into trouble again.

Eventually - after a few days of playing at "yellow-heart" level, I
found myself increasingly able to beat the "yellow-bow-tie". Then
the "yellow-infinity", and so on into the low blues.

By turning down the difficulty settings to their minimums, I was able
to last a little longer - but only saw the first "invisible" level
once, because it required clearing level blue-15, blue-16, and
invisible-1. (And in the yellows and blues, there are places where
four or five levels have to be completed in order to advance to the
next bonus, not just the "two levels" in the purples and reds.)

Anyways - invisible-1 was where I've topped out, even with easy
difficulty settings, five lives, and bonuses every 10000. In a
few months, I'll go back to the game and see if I can improve my
skill again.

By the time I ended my "training programme", I could consistently
make it through the reds within one or two credits - something that
was impossible when I started. My high score had improved due to
the large "starting bonuses", as had my skill - levels I'd been
unable to complete when I started seemed easy, but there was always
a level (or series of levels) that took great effort to overcome,
and in the end, the game always won.

(One caveat -- I suspect the greens are easier than the invisibles, as
I found myself *totally* disoriented by not being able to see the web.
I assume that this is made up for by the ever-increasing number of
levels that the player has to defeat on a single credit in order to
get to the next "big starting bonus" level.)

IMHO, the original point that folks are trying to make stands. The
high score on the machine is a valid indication of the skill of the
player. (Corollary: For any given Tempest player, there exists a
level sufficiently difficult that he/she will not be able to beat
it, even with five lives and two credits for a two-player "you get
two shots at it".)

> I'd argue that the to be a "high skill list" it would have to include
> the number of consecutive racks played on one credit; or the number of
> "lives" played per game; or the average time per rack; or hit
> percentage, etc... Not just "I managed to get this many points and you
> don't know how".

The original point notwithstanding, these would be very interesting
statistics to keep. I know I had a lot more *fun* playing from level
9 through to level 32 in a single credit than I did having a whole
series of "got killed before reaching the next big bonus plateau,
try again".

Perhaps the best way to measure skill would be to say "high score reached"
and "number of games continued to get here". Someone who can get to 500K
in the space of 3-4 credits and walk away is arguably a better player than
one who had to play 30 or 40 games to finally "make it through the level
once". (Though the second player could presumably get a comparable score
the next day on less credits, because their skill would have improved due
to their training...)

One fun trick to play with the "time per credit" figure already stored in
the EAROM of the arcade model - you can play indefinitely on the early
levels by letting the flippers reach the edge and sliding beneath them
without shooting them. Oddly enough, it's easier to do this stunt
deliberately at lower levels, when you can watch the flipper jumping and
time your move accordingly - but it happens much more often *accidentally*
on the higher levels. (Although it's more of an "it never happens" in the
reds to a "once in a blue moon" in the yellows to a "twice in a blue moon"
in the blues kind of effect... not something to gamble a life on, but if
you're out of places to run...)

> Thinking back to Al's example I should include EAROM editing just for
> the sake of being able to enter a "beat this" challenge bar at a level
> above what the players have currently been able to achieve...

Also cool - it'd also be nice to avoid the "everyone has to enter
their high scores for the next three hours" if you've just changed
difficulty settings.

Later,
Doug.

-- 
 dougj   |
   @     |
hwcn.org |