textfiles/games/REVIEWS/td3.rev

 TEST DRIVE III

 THE DUEL: TEST DRIVE II has the distinction of being the most popular computer
driving game ever published. It offered challenging gameplay, good graphics for
its time, and lots of sheer fun and enjoyment. So it's only natural that
Accolade, the publisher of the first two games, would want to continue the
tradition by updating the program to current technological standards. (This
review is based on the IBM-PC version.)

 The concept is simple enough: Take the basic premise of the original games
(i.e., driving exotic cars at unsafe speeds on public roads), offer 256-color
VGA graphics, full support for most sound boards, and a true 3-D polygon-fill
game world complete with varying weather conditions. It all sounds terrific, but
something got lost in the execution.

 The IBM-PC version of TD III requires the following: PC, XT, AT, Tandy 1000,
3000, or 4000; VGA, EGA, MCGA, or Tandy 16-color; and 640K of RAM. A clock speed
of 8MHz or faster is recommended. There is sound support for AdLib, CMS, Tandy
3-Voice, Roland MT-32, LAPC-1, and Soundblaster. Copy protection uses a code
wheel. For this review, I played on a 286 at 20MHz with VGA, and Roland and
AdLib sound cards. It should be strongly emphasized that the EGA graphics are
quite poor by today's standards. The digitized images used for the car interiors
(as well as the selection screens) are stored in VGA format only. When displayed
in EGA, the program calculates the images from the VGA format, and the result is
much less than satisfactory.

 At the main menu, the player may select one of three available cars: the
Lamborghini Diablo, Pininfarina Mythos, or Chevrolet CERV III. The selection
process includes some nicely digitized images of the vehicles, as well as basic
performance specs. Different courses can also be selected, although only one
(Pacific to Yosemite) is included with the original program. Accolade plans to
sell data disks of cars and courses in the same manner that they did with TD II,
although it's important to note that data disks from TD II will _not_ work with
TD III. Finally, nine skill levels may be chosen, each introducing more severe
penalties for bad driving, as well as increasing the skill of the computer
opponents. In the same step, the player is asked to choose to race against the
clock, the two computer opponents, or against as many as three human opponents.
Human competitors must take turns driving the course, while the computer
opponents race with you, wheel to wheel.

 Then it's off to the open road. There are a lot of options while driving, such
as three different musical tracks in the background, automatic wheel centering,
sound effects and engine sounds, as well as three "visibility distance" settings
that increase game speed on slower CPUs. The cockpits of all cars are digitized,
though they have so little texture that the 16-color cockpits in TD II are
probably superior; more retouching was in order here. The three-dimensional
polygon filled graphics that make up the game world, however, are rendered quite
nicely. There are bridges, tunnels, railroad crossings, dams, lighthouses, etc.
It is complete and effective.

 The course is divided into five sections that are driven in sequence. Each
section, however, has multiple routes that lead from point A to point B. As you
might expect, some are shorter but more difficult to negotiate, while others are
longer but allow higher speeds. One of the nicest features here is that TD III
actually times and scores each route differently, and keeps records for the
different routes in each section. There are still police patrols to contend
with, and even inclement weather from time to time. Nighttime, rain, fog, and
snow all occur, though they have little or no impact on gameplay (that I can
detect). While this is disappointing, the graphical effect of each of the
conditions is quite good.

 Most all of the preceding seems to imply an impressive, engaging driving
simulation in TD III. Alas, that's not the case, for one issue is yet to be
addressed: playability. TD III fails miserably because it is unplayable. All of
the beautiful graphics and game structure are wasted because of two critical
factors in the game design that were botched: specifically, the control method
and the time base.

 All control, including joystick, is digital. If that weren't bad enough, and it
is, it may just be the poorest digital control I've had the displeasure of
using. It is frustratingly difficult to keep the cars on the tarmac at even
reasonable speeds.

 Even worse, TD III does not run in real time; it's based on CPU clock cycles.
This type of design is common in arcade-style games that don't attempt to
accurately portray real world events. In simulations, however, it's of critical
importance that all movement within the game world occur at consistent rates of
speed, regardless of CPU cycles. The only way to accomplish this is to base all
movement on real time, or at least some percentage of real time, because it is
constant. Once this is done, the speed of the CPU only alters the frame rate, or
the frequency of screen updates in a given time. The reason this is such a
critical issue, aside from sheer realism, is that in a polygon-fill game world,
the number of polygons that must be drawn at any given point can change
radically. If the program is not based on real time, the entire speed of the
game changes according to the world detail from moment to moment. Aside from
feeling totally disjointed and unrealistic, this can render a game unplayable on
high-speed CPUs, because the reaction time may become unreasonable when fewer
polygons are being manipulated. Such is the case with TD III. Even on a 20MHz
286, the program runs about 1.75 to 2 times faster than real time. When played
at 8MHz, it's much, much slower than real time. The end result is that the game
is a great deal easier to play on slower machines. In fact, my experience
resulted in course times being reduced by half when playing at 8MHz as opposed
to 20Mhz. However, as you might guess, playing at 8MHz is anything _but_ the
exhilarating experience this game promises.

 There are a number of less severe gameplay problems as well, not the least of
which is the sound support. The music is the typically mediocre pap, but the
sound effects are truly horrible. After having taken some heat for only offering
musical support in recent releases, Accolade has countered by including sound
effect support, but it still doesn't seem to care enough to do it properly. The
Roland sound is particularly disappointing, considering its capabilities. There
are other problems: Computer opponents can drive through objects, such as
trains, and occasionally other cars; mountainous areas have sections of road
that just drop out of your sight because the pitch (longitudinal rotation) of
the car doesn't change to match the angle of the road surface; collisions with
other cars or being stopped for speeding often result in several subsequent
collisions while trying to dislodge yourself from the scene; the instant replay
feature is hardly worth mentioning because of its limited features and clumsy
interface; the cars spin right up to their top speed without any trouble
whatsoever, in direct contrast to the realistic engine modelling in TD II. The
list goes on, but why bother? These aren't the factors that destroy the game,
but they are a good indicator of the overall sloppiness with which much of TD
III was programmed.

 In short, there is very little to recommend TD III. As was stated up front, the
concepts are solid. That much is even more apparent after having played it. But
a simulation must meet a few basic requirements: It must be fair, it must be
fun, and it must be challenging. TD III meets none of these, unless you consider
challenge and frustration to be one in the same. It's just a shame that Accolade
was seemingly unwilling to go the distance on this one.

 In closing, I wish to note that the designers of TEST DRIVE I AND II,
Distinctive Software, were not responsible for this incarnation of the TEST
DRIVE series. I certainly don't wish to inadvertently tarnish their reputation.
For whatever reasons, Accolade chose to hand TD III over to Tom Loughry, the
designer of STEEL THUNDER and GUNBOAT, both relatively undistinguished and
widely criticized games. It's one of the best examples in recent memory of just
how little a role most publishers play in the quality of software, and how great
a role the design team plays -- this despite the fact that most consumers still
purchase games based on the publisher's reputation.

 TEST DRIVE III is published and distributed by Accolade.

*****DOWNLOADED FROM P-80 SYSTEMS (304) 744-2253