Game Developer - AprMay_1995.pdf

(513 KB) Pobierz
April/MAy 1995
G A M E
D E V E L O P E R
M A G A Z I N E
G A M E
P L A N
G A ME
G A ME
Delphi is
the Answer
Editor
Larry O’Brien
76702.706@compuserve.com
gdmag@mfi.com
Senior Editor
Nicole Freeman
Managing Editor
Nicole Claro
nclaro@mfi.com
N
ormally I make it a rule not to
recommend any programming
tool that I haven’t used to
develop a program from
scratch. I’m going to break that
rule to recommend Borland’s
Delphi. I’ve only used it to
develop a program that trans-
lated from Celsius to Fahrenheit degrees,
but the only reason I’m not altogether
committed to Delphi is (I’m blushing as I
type this) I’ve never learned Pascal. Well,
I’m going to learn it now. For Windows
developers, and for Windows game devel-
opers especially, Delphi is so far ahead of
the competition it’s embarassing.
The world of Windows development
has always been one of compromise. Tools
such as Visual Basic offer immediate grati-
fication when moving from writing code
to seeing your program run, but the per-
formance has always suffered in compari-
son with compiled code. On the other
hand, compiled C++ offers near-optimal
performance, but all but the most trivial
programs have compilation times mea-
sured in minutes, not seconds. Program-
mer productivity vs. compiled perfor-
mance. Pleasure for the programmer vs.
pleasure for the user. The iron compro-
mise of Windows.
Delphi makes no compromises. For-
get the claims of how many hundreds of
thousands of lines of code are compiled
per minute, the point is that Delphi’s
speed of compilation is such that the
move from editing to debugging is seam-
less; when your screen flickers you think a
background compilation process is being
spawned but, in fact, you’re seeing the
results of the final link.
Delphi is being marketed toward the
current marketing darling, corporate
developers, and there’s no question that it
will cut a swath through the ranks of Visu-
al Basic and PowerBuilder users out there.
However, game developers have even more
to gain by examining Delphi. Why? The
huge, relatively untapped Windows game
market. With Delphi, you can use WinG
to get your images on screen relatively
quickly, and you also can get many more
compile-link-debug cycles in a program-
ming session. Windows game develop-
ment inevitably involves hit-and-miss pro-
gramming challenges (no matter how
experienced you are, the Windows API
can surprise you), so more cycles means
faster development. When you’re done,
your programs run at compiled speeds.
Also, Delphi abstracts the Windows
API. If this abstraction isn’t good enough
for you, feel free to create your own. Del-
phi’s code reuse infrastructure is excellent,
with full object-orientation and the ability
to package groups of objects together as
functional components. Develop an engine
for isometric projection, and you can save
it as its own package and use it as a start-
ing point for future games.
Delphi is based on Object Pascal
which, as I said, is a new language to me.
However, it’s very straightforward—C++
it ain’t. Although I’ve already discovered
some things that I don’t like (the keyword
private
doesn’t act the way I think it
should, and I can never figure out where
semicolons are expected), I’ve also already
come up to speed on the basic structure of
the language. You should have no trouble
picking it up and, just as importantly, you
shouldn’t have any trouble finding new
programmers fluent in it.
You’ll be reading a lot about Delphi
in the corporate programming magazines,
where it will be primarily lauded for its
ability to create database applications. But
I’ll make a bold prediction—the most suc-
cessful Windows game of 1996 will be
written entirely in Delphi. Until that ships,
I’ll be busy learning where the stupid semi-
colons go.
s
Larry O’Brien
Editor
Editorial Assistant
Deborah Sommers
dsommers@mfi.com
Contributing Editors
Alex Dunne
75010.2665@compuserve.com
Chris Hecker
checker@bix.com
David Sieks
dsieks@arnarb.harvard.edu
Wayne Sikes
70733.1562@compuserve.com
Editor-at-Large
Alexander Antoniades
sander@mfi.com
Cover Photography
Charles Ingram Photography
Publisher
Veronica Costanza
Regina Starr Ridley
Group Director
Advertising Sales Staff
West/Southwest
Yvonne Labat (415) 905-2353
ylabat@mfi.com
New England/Midwest
Kristin Morgan (212) 626-2498
kmorgan@mfi.com
Susan McDonald
Advertising Production Coordinator
Denise Temple
Marketing Manager
Andrew A. Mickus
Vice President/Circulation
Jerry M. Okabe
Group Circulation Director
Gina Oh
Director of Production
Circulation Manager
Circulation Assistant
Newsstand Manager
Reprints
Kathy Henry
Phil Payton
Pam Santoro
Stella Valdez (415) 655-4269
MillerFreeman
A MEMBER OF THE UNITED NEWSPAPERS GROUP
Graham J.S. Wilson
President/CEO
Marshall W. Freeman
Executive Vice President/COO
Thomas L. Kemp
Chairman of the Board
Senior Vice Presidents
H. Vern Packer, Donald A.
Pazour, Wini D. Ragus
Vice President/CFO
Warren (Andy) Ambrose
Vice President/Administration
Vice President/Production
Vice President/Circulation
Charles H. Benz
Andrew A. Mickus
Jerry Okabe
Regina
Vice President/Software Development Division
Starr Ridley
4
GAME DEVELOPER • APRIL/MAY 1995
S E Z
U !
More to
the Mystery
by Our Readers
Something on your
mind? A response to an
article, a query about
code, a little poem or
ditty, perhaps? If you
have anything to get
off your chest, please
do. We’ll probably
print it here.
Dear Editor:
would like to offer a few suggestions to sup-
plement Andre LaMothe’s article “The Myste-
rious Mode 13h” (Sept. 1994).
I am a shareware game programmer who has
been using Mode 13h for about two years now. I
can see at least one obvious ways to speed up
the graphics functions Mr. LaMothe described.
The first would be to avoid using the palette
ports (0x3C7/0x3C8) as much as possible. One
blatant way that I can see to do this would be to
eliminate reading the port whenever the pro-
gram needs the value of a specific register.
Instead, simply create a structure:
I
typedef struct {
unsigned char red[256];
unsigned char grn[256];
unsigned char blu[256];
} PaletteStr;
Clear the structure in the beginning of the
program, and also clear the palette (set all
“bucket” values to (0,0,0)) using the write
port. Then, whenever the structure is updated,
simply update the array value as well. That
way, a function can simply look up a value in
an array instead of having to write/read to a
port. This also streamlines the color cycling
function by eliminating the overhead associat-
ed with procedure calls (about five processor
cycles). Simply rotate the array, then push the
whole thing out to the VGA card when the cycle
function is done.
Mason McCuskey
via e-mail
MORE READING MATTER
Dear Editor:
enjoy your magazine, I find it helpful and
interesting. I would like to see the price go
down or the amount of material in it go up,
but I still think that it’s worth it. I think it
would be enormously helpful if after each arti-
cle you gave a brief reading list of good books
that provide more information about a partic-
ular topic. For instance, in the December issue
you featured a great article on Mode C (umm...
excuse me...Mode X) but I would like to know
of a book that would give a beginning-to end
breakdown of the technique. This would help
those who aren’t quite as technically
advanced as some, and keep the folks who are
advanced happy. Plus the market for computer
books is so full of crap that it would be nice to
have someone who knows something suggest
good books.
Gideon Stocek
via e-mail
Editor Larry O’Brien responds:
Good suggestion! We’re running book reviews
regularly to try to cut through the crap (see
Dean Oisboid’s article on page 48). One sugges-
tion: don’t buy a book that calls you stupid. If
the title’s condescending, the text will be as
well.
I
WHERE’S THE CODE?
Dear Editor:
’m looking for the code for Carl Muller’s
article “Hitting on Collision Detection”
(Sept. 1994). The article states that it is
available on CompuServe, but I can’t find it.
Do you have any more detailed information on
exactly where it is, the filename, forum, and so
on? I’d also like to know if it’s available on any
ftp sites.
Leonard Guy
via e-mail
Editor Larry O’Brien responds:
On CompuServe, Go SDFORUM and visit our
library. Or set your browser to the
/pub/gdmag/src ftp directory of whiz.mfi.com.
I
GAME DEVELOPER • APRIL/MAY 1995
7
C R O S S F I R E
Media Vision
Picks Up
The Pieces
Alex Dunne
”We‘ll be quiet, but
we‘ll be here,“
says Robert Brownell,
CEO of Media Vision. The
company took some
recent blows —many
of them self-inflicted.
Now it‘s back with a
kinder, gentler
business sense.
igital entertainment is often
in the media spotlight. Usu-
ally we hear about technolog-
ical breakthroughs, though—
rarely are high-tech compa-
nies involved in controversies.
There have been some
notable exceptions, though,
and surely Fremont, Calif.-based Media
Vision stands out as a Silicon Valley
soap opera. Media Vision scraped bot-
tom last summer when it filed for
Chapter 11, ironically due to the
lack
of
vision that the company—led by former
CEO Paul Jain—had. However, this
year Media Vision emerged from bank-
ruptcy and is attempting a comeback
with the help of new management and
capital. The company has refocused on
its core technology, audio cards, and is
in the process of rebuilding the public’s
confidence and researching new audio
technologies.
The company’s problems began
back in 1993. Media Vision was well
positioned in the red-hot multimedia
market, thanks to its Pro AudioSpec-
trum sound cards and multimedia
upgrade kits, yet ongoing price wars
with rival sound card makers like Cre-
ative Labs and an overextended product
line were taking their toll on the com-
pany. In October of 1993, Media
Vision rolled out a plan to produce
CD-ROM games, instantly making it a
major player in the game industry.
As a complementary product line,
the company’s announcement was well
received by investors. News of the
game venture helped propel the compa-
ny’s stock. Media Vision shares reached
$46 in January 1994—barely 14
D
months after the company went public
at $15. Although the stock was gener-
ally regarded as overvalued, nobody
realized how much so until a few
months later.
Greenberg Blows
The Whistle
Herb Greenberg, a columnist for the
San Francisco Chronicle,
did some inves-
tigating and discovered that Media
Vision had falsified sales records and
engaged in shady business practices to
paint a better picture of the company’s
finances. According to Greenberg,
Media Vision secretly rented ware-
houses to hide returned merchandise,
pressured engineers and sales reps to
get defective products out the door,
offered huge incentives to distributors
and retailers who ordered large quanti-
ties of merchandise (known as “channel
stuffing”), and most egregiously, doc-
tored shipping papers to charge a large
quantity of multimedia upgrade kits to
a previous fiscal quarter.
Jain, who some believe was the
force behind the company’s question-
able practices, was a favorite target in
Greenberg’s columns. One of Green-
berg’s columns even poked into Jain’s
personal life, referencing allegations
that he illegally used Media Vision’s
corporate assets to “court women.” I
bet Jain loathed opening up his morn-
ing paper for fear of what he’d read in
the business section.
When these stories surfaced, the
FBI and the SEC quickly took notice.
On May 9, 1994, the two agencies
began a probe into securities violations
committed by the company. Media
GAME DEVELOPER • APRIL/MAY 1995
9
C R O S S F I R E
Brownell‘s humili-
ty contrasts his
company‘s brazen
past, and he‘s cir-
cumspect about
the hard lessons
the company has
learned with
consumers and
resellers.
Vision’s stock, which had been plum-
meting in recent weeks due to questions
surrounding the company’s financial
status, closed that day at 2
7
/
8
. One
week later, Jain and three other corpo-
rate officers resigned.
In the wake of Jain’s departure,
Robert Brownell, the vice president of
domestic sales, took over as CEO.
Brownell inherited the unenviable posi-
tion of rebuilding the company, and
one of his first actions was selling the
company’s game publishing arm, which
included The Daedalus Encounter and
Critical Path, to Virgin Interactive.
Unfortunately for the company,
Brownell’s move didn’t stave off bank-
ruptcy. Two major creditors balked at
the company’s reorganization plan, so
on July 25th, 1994, Media Vision filed
for Chapter 11. The clincher followed:
the company admitted that its $20 mil-
lion profit for 1993 wasn’t quite accu-
rate. In fact, there was no profit for the
previous year: the company actually lost
$99 million. Media Vision attributed
the revision in numbers to unaccounted
product returns and marketing costs,
improperly recorded sales, and doubtful
accounts.
Since it came clean with investors
last summer, Media Vision has taken
steps to rebuild itself. In August, it
secured a $10 million credit line from
Trust Company of the West (TCW) to
help fill $24 million in backlogged
orders and prepare for heavy demand
during the Christmas shopping season.
On December 30th, Media Vision
emerged from Chapter 11. Its stock,
which just a year earlier had been trad-
ing in the 40s, was canceled and worth-
less. To cover previous debts, the com-
pany issued 20 million shares of new
common stock, of which TCW will
own 43%. Two TCW officers sit on
Media Vision’s board of directors. The
company, once composed of close to
500 employees, is about half that size
now. Media Vision has also entered
into an agreement with the SEC,
according to which the agency will rec-
ommend that no enforcement action be
taken against the company, as long as
Media Vision cooperates with the SEC
and the Dept. of Justice.
From the CEO: Brownell’s
Plans To Rebuild
I recently spoke with Robert Brownell,
and he explained Media Vision’s plans.
His humility contrasts his company’s
brazen past, and he’s circumspect about
the hard lessons the company has
learned with consumers and resellers.
“Our position is more one of con-
trolled growth. We’re not just chasing a
growth curve to see how much in sales
we can ring up. We’re very cautious
when dealing with the reseller channel
that we don’t put too much product in,
[rather] that we put the right product
in. We don’t try to give [resellers] eight
kits when we know only two of them
are going to sell. We’re not in that
mode anymore. I guess some of the
competitors still do that,” Brownell
commented.
Brownell on the practice of chan-
nel stuffing: “That happens everywhere.
[If] you continue to throw out new
products and... you don’t deal with
[your older products] as rapidly as you
should, then you have a problem. We
monitor the sell-through at the retail
level very closely, to make sure that we
don’t put our products in that position.
And we don’t do deals with the reseller
channel that could put us in that posi-
tion. We just say, ‘Buy what you think
you can sell. Here’s our pricing. If this
works for you, then great. We’ll support
you.’”
What are Brownell’s goals for
Media Vision in 1995? “We haven’t
publicly stated our dollar goal, and we
probably won’t until the end of the first
quarter. But [with respect to] my goals
for the company, the first... was to get
through bankruptcy. To get our house
in order. And now we’re in that posi-
tion. We have developed a plan to go
forward, [and] we need to execute on
that plan... 1995 should be a quiet year
for us.”
Brownell has stated that Media
Vision will get back to its core market. I
asked him to elaborate on this plan:
“Our core technology is audio, graphics,
and actually, video. One of the areas
that we’ve been working on over the last
several years is waveguides. We have
one of the licenses out at Stanford and
we think we’re the farthest along, and
we have a working chip at this time.
We hope that this will translate into
some products in the fourth quarter, but
I don’t think it will have a major impact
until next year. We feel that’s our best
shot, and we want to build our company
around it. It will differentiate us in the
market.”
It was encouraging to hear that
waveguides is an integral part of the
10
GAME DEVELOPER • APRIL/MAY 1995

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika:

Zgłoś jeśli naruszono regulamin