roelofs@amelia.nas.nasa.gov (Cave Newt) (07/17/90)
Here, at long last, are the long-awaited (at least, by some of us) results
of my query two weeks ago as to people's best Nyet scores. For those of
you who have been living on another planet for the past few years, Nyet
is an "irritatingly addictive game" of the Tetris genre (but more well-
written than the original). Responses came from five countries on three
continents, so it's clear the game's not simply a Soviet ploy to reduce
*American* minds to quivering Jello.
By far the best scores belong to Brian Hilchie; the next closest contender
(namely, me :) ) isn't even within a thousand points of his lowest score.
(I have heard rumors of someone else who might have scores of his caliber,
but I haven't heard from this person yet.) Unfortunately Brian is the only
person whom I've not been able to reach by mail, so I'm afraid I can't tell
you his secret(s). Oh well.
As to the high-score list itself: as promised, I computed both the mean
(that's "average" to you non-mathemagical types) and the median (the number
for which half the scores lie above and half below). For symmetric distri-
butions these two numbers are the same, but I thought people's scorefiles
might be a little more interesting, particularly since the sample size is
so small. Well, no such luck. The means and medians *are* different, but
not so different that anybody's relative standing changes.
Of course, partly this is due to the fact that only 12 people responded
to my plea (sigh), and as a result the scores that did get sent in are all
over the scale. Maybe if they were clumped a bit more tightly...? Well,
anyway, I'll continue to collect scores if people wish to send them to me
(more on that, below), but I don't necessarily promise to waste time with
statistical nonsense.
One last thing before the scores: if any contributor, past or future, is
NOT using version 1.1 of Nyet, please make that known to me. It probably
makes a difference in the scores.
And without further ado...
Best Mean Median
Score Score Score Name Machine Kbd
________ ________ ________ ______________________ ______________ ___
25810 23980 23625 Brian Hilchie - -
21730 20823 20690 Greg Roelofs 16 MHz 386 y
21600+ ~20160 ~20050 Diana Summers - -
20500- ~19800 ~19700 Tim Tyhurst and wife 8 MHz V30 n
19580 19077 18923 Curt Bennett 9.54 MHz 86 -
18530 16686 16610 Eddie Herrell 12.5 MHz 286 n
17655 15814 15445 Lars Wirzenius - -
15550 14789 14718 Bertil Stenstr|m 8 MHz 286 y
15375 13339 13543 Bryan Holloway - -
11865 10256 10003 David Burd 20 MHz 386 -
Notes:
- While I know how to compute the median of a continuous distribution,
I'm not entirely clear on the method for a discrete set of data, and
I didn't care to waste any time looking it up. :) So for this column
I simply picked the number halfway in between the two middle scores.
- The tildes represent approximate values; they indicate that I was
given incomplete or imprecise scores and had to interpolate. I did
this on the assumption of uniform distribution of high scores, which,
as one can see by comparison of the mean and median columns, is not
a particularly valid assumption: with one exception, the means are
all higher than the medians. Well, as some extra-specially obnoxious
cartoon character is fond of saying, "eat my shorts."
- Not everyone told me what sort of machine(s) they play on, so there
are some gaps there as well.
- The "Kbd" column indicates whether a keyboard enhancer was used; for
example, TURBOKEY, TYPEFAST, or QUICKEYS (the latter two are from PC
Magazine, I believe).
- Folks may continue to send me their scores (all of them, preferably),
along with machine type and speed (e.g., 80286 @ 12 MHz), whether or
not a keyboard enhancer was used (to speed up the typematic rate),
and anything else which might have a bearing on one's scores (it was
reported to me by one player that processor speed made a difference;
this is not supposed to be the case in Nyet v1.1, which is the second
and last version of which I'm aware). If I get any sort of response
I'll post once more to rec.games.misc in a month or so (search for a
subject containing the word "Nyet"). I may also send the results to
the author, David Howorth (71600.521@compuserve.com), although I've
not yet done so. And, in the unlikely event that this thing really
takes off, I might even make the list available for anonymous ftp
somewhere (but don't hold your breath).
- For those who are insatiably curious about all the gory details, my
best score was achieved with the elimination of 101 rows. The second
best was something like 103 to 105 rows, but the screen went away
before I got a good look (life's a bitch). I usually start at level
8, although I often warm up at level 9; I don't use the "peek" func-
tion; and I've been playing off and on for about two years (more or
less continuously since December, though). I do much better if I'm
not anticipating the switch to level 9; I invariably screw up immedi-
ately before getting there if I've peeked at the rows-eliminated box
too recently.
- Follow-ups to rec.games.misc, or via e-mail, please.
-----------------------------
For those *seriously* hard-core Nyet junkies who want the whole ball of wax:
Brian Hilchie, brian@alzabo.uucp or uunet!mitel!sce!cognos!alzabo!brian
25810, 25560, 24465, 23735, 23645, 23605, 23400, 23345, 23135, 23100
Greg Roelofs, roelofs@amelia.nas.nasa.gov
21730, 21485, 21365, 21310, 20830, 20550, 20415, 20250, 20160, 20130
^^^^^ ^^^^^
I should mention that all but these two scores were achieved since I
made the original posting. What can I say? So I happen to be addicted
to the game.... :)
Diana Summers (c/o Peter Summers), U5533129@ucsvc.unimelb.edu.au
"The situation is complicated because she turned the machine off without
quitting the program and hence the high score table wasn't rewritten. She
says the score was 21600 and something, so count it as 21600 if you want
to count it at all. The other scores range from 20500 to 19500."
Tim Tyhurst, tim@looking.on.ca
"Don't have the actual score file handy, but after a year of play, my wife
and I have got to the point where we find it pretty hard to get on the board.
The lowest score is ~19200, while the top three are between 20000 and 20500."
Curt Bennett, curt@zaphod.uchicago.edu
19580, 19475, 19345, 19225, 18940, 18905, 18870, 18835, 18815, 18775
Eddie Herrell, aragorn@titan.tsd.arlut.utexas.edu
18530, 17970, 17610, 17595, 17425, 15795, 15705, 15630, 15380, 15220
Lars Wirzenius, wirzeniu@{cs,cc}.helsinki.fi
17655, 16555, 16505, 15610, 15445, 15445, 15285, 15230, 15210, 15195
Bertil Stenstr|m, stenis@heron.qz.se
15550, 15365, 15180, 15030, 14730, 14705, 14490, 14410, 14250, 14175
Bryan Holloway, holloway@clio.las.uiuc.edu
15375, 15050, 14190, 13975, 13695, 13390, 12820, 11765, 11670, 11455
David Burd, phred!davidb@pilchuck.data-io.com
11865, 10985, 10910, 10690, 10110, 9895, 9625, 9620, 9460, 9395