Hi! |
A|A
(n n)
\_/
I hope this EM gets smaller than the previous one about
segmentation. I'm here, again, to comment several EMs at once.
First subject: Robotz. I implemented a game in BASIC in
last December. I'm not kidding, I have a witness! A friend of
mine saw it working (three robots battled till only one last -
of course, there can be only one!) Yes, the program still
exists, but it is a mess and don't fit the specifications (hey,
give me a break, I implemented it in one week, with no specs
at all, only ideas, just to keep some ghosts of mine away...)
The program didn't send data through a network, but there was
a protocol for communication that simulated it (I made "fair"
robots, of course, what would be the use of I cheating myself...?)
Thus, I could test different robots in a single MSX, as if they were
spread in a network.
But after reading the docs (0.2 and 0.3 - got this ahead, looking
into the latest EMs in my box), wow! Sorry, it's too complicate and I
don't have the time. I'm already engaged into a JavaScript adventure
and won't be able to implement it (the BASIC program did take me a
lot of time, what to say about that!) Yes, it seems to be a very
powerful robot programming language, but I'm not at all sure it can
be done... I'll read it more carefully at home, getting drunk with
grape juice, lustfully eating chocolate cookies...
Okay, next: compression, the smaller is better! Do you know
A.C.T.? It stands for "Archieve Comparision Test". It can be found
at "http://act.by.net". If you, like me, thought RAR was powerful,
you haven't seen 777, UFA, RKive, IMP, BOA Constrictor and others.
Unfortunately, some of them are meant to be only for PC (worst, for
Win95 - they use that stupid long-name structure). Among the others
that run in MS-DOS, some require more than 32MB of RAM to start!
Well, but this is supposed to be about MSX, and it is! The better
the compression rate, more memory the archiving program will require
(there is a wider range of combinations to test). About speed, well,
that's another story. Taking a quick look at A.C.T., it's possible
to notice that some programs are really fast when not trying maximum
compression, and even so can reach really amazing rates. Oh, yes,
some of that programs are free and have the sources available.
The problem with MSX is that it is a small "niche". How many
PC users would be able to handle PMA files? How many types of
files can be read and manipulated in PC and in MSX? Yes, the
proportion of scaring. MSX can read, let me see, about five types
(LHA/LZH, ZIP [seems to be troublesome], PMA, ARJ and softhouse
custom packers). Well, check A.C.T. and try to count the number of
different packers available (it is explicited nowhere in the Page,
but it is for _PC_ compressors). Of course, if you take the problem
to other systems, you go crazy (have any of you stumbled into a
packer named "stuffit", for Mac?) or is completely ignored (how
would a common Mac user read a PMA file?) I think the easier
solution is to adopt one packer, one accepted by as many systems
as possible (LHA, Zip, Arj, RAR etc.) I don't like to admit it,
but Zip is the most widespread (I use Info-ZIP freeware, not the
PKWare original). A good MSX port of it would be nice, but it
requires a lot of C libraries (I'm not sure MSX can deal with it).
It is good to have all that in mind before starting to design a
new and incompatible compressor for MSX (I once made one in BASIC,
using pure Huffman - it took half an hour to compress a 16KB
file...!)
Anime trivia: when I was in Japan in the beginning of 1997,
a new Ghibli "anime" was arriving the local theaters: "Mononoke
Hime" (Princess Mononoke). Only last week, two years later, I
could watch it. Wonderful! If you have the chance to see it, do
it now! The same producers of "Nausicaa", "Tonari no Totoro"
and "The Last Unicorn" (my forever favourite movie...)
... Cyberknight...
<Over>
****
MSX Mailinglist. To unsubscribe, send an email to [EMAIL PROTECTED] and put
in the body (not subject) "unsubscribe msx [EMAIL PROTECTED]" (without the
quotes :-) Problems? contact [EMAIL PROTECTED] (www.stack.nl/~wiebe/mailinglist/)
****
