MechWarrior 3 – A Retrospective Review

I have mentioned before the inaccuracy and unrealistic aspects of mecha as they are depicted in all but the most fastidious series and about how they are regularly depicted as being substantially more powerful than physics would dictate in a universe displaying a resemblance to the real world. That doesn’t stop me from enjoying a few series which bring mecha to the fore, the most prominent among these being the BattleTech universe. I first encountered the BattleTech universe with the animated series in the 1990s, which was admittedly poor-quality and inaccurate but was sufficiently exciting when I was eight for me to enjoy it.

Later, I discovered MechWarrior 3, a game set in a similar time period in the universe as the animated series. Despite my inability to get to grips with the controls – not aided by the fact that I didn’t have a joystick, I completed many of the missions but never completed the entire campaign. I left the game aside for several years, partially because of my inability to get the game running on more modern computer hardware and operating systems without the entire operating environment having a seizure. More recently, though, I managed to get the game running on Windows 7 and therefore complete the campaign.

MechWarrior 3 is a first-/third-person mecha simulation game released in 1999, developed by Zipper Interactive and published by MicroProse. The game takes place at the tail-end of the Inner Sphere counter-attack at the Clans following the Clan Invasion and follows the adventures of a lance of MechWarriors from the Eridani Light Horse mercenary company as they fight their way across a planet named Tranquil in search for Brandon Corbett, a high-ranking commander of Clan Smoke Jaguar who threatens to rally the scattered forces of the Clans to fight back against the Inner Sphere invaders.

The mission of the MechWarriors is to destroy several facilities on Tranquil to disrupt the plans of Galaxy Commander Corbett, but their mission gets off to a bad start when the dropship carrying the mechs and their pilots into action is shot down by an array of lasers on the ground. Few of the MechWarriors manage to escape with their lives and those that do are scattered across the planet far from their expected positions. Despite being outnumbered and outgunned by the Clan forces, the MechWarriors of the Eridani Light Horse must rejoin and fight their way against a superior force to find another dropship to retrieve them from the planet, while achieving the objectives designed for a more powerful force.

The game starts you out with a 55-ton Bushwacker, a Medium BattleMech armed with a balanced array of weapons, including a powerful long-range laser, a moderately powerful autocannon and a guided missile launcher. Across the planet, between your allies and enemies, you will encounter seventeen other types of mechs, ranging from 30-ton Firefly mechs with minimal armour and armaments, but which are fast and agile, to 100-ton Annihilator and Daishi (or Dire Wolf if you’re a Clanner) mechs with powerful weapons and plenty of armour, but which move like glaciers compared to the smaller mechs. Altogether, there is a reasonable collection of mechs, although there are substantially fewer than in the sequel, MechWarrior 4. Some of the mechs also seem rather misplaced – the Clan forces having substantially more mechs of Inner Sphere design than you would expect – while some of the more iconic mechs are missing, including the Atlas and Centurion included in the Pirate’s Moon expansion pack.

The game can be played in first-person or third-person, with nice, if generic, cockpit views in the first-person mode. There are multiple axes of movement for the mechs – horizontal movement through the movement of the legs and an adjustable throttle setting, torsos which can be twisted in order to fire in a direction other than that of movement, moveable arms through the freelook setting along with the ability to fit jump-jets to your mech in order to provide a limited amount of jumping ability. With all of these controls, the control schema overwhelms a traditional keyboard-and-mouse layout and I suggest the use of a joystick with either a twisting rudder axis or separate rudder pedals in order to most appropriately play the game.

The game follows a reasonably accurate model of the mechanics of the tabletop game, in particular modelling the build-up of heat through the firing of weapons along with the corresponding dissipation of heat through heat sinks built into or added to the mech, along with the damage system modelled in the tabletop game. With regard to the damage system, this does create some imbalances which only show up in the real-time, fast-paced action of a computer game. Some parts of a mech are quite a bit weaker than others, with the legs of the mech standing out as a particular target to experienced gamers, providing a relatively large target with a large pay-off if they are destroyed. Indeed, apart from long-range shots where the target is out of sight, it is usually a better strategy to aim for the legs than any other part of the mech – a strategy which I hear didn’t go down well in the multiplayer mode.

The heat system, on the other hand, is rather more balanced and provides some sort of balance against the lasers with their absence of ammunition and back towards autocannons which may have limited ammunition but produce little heat. As you approach the heat limit, the mech will automatically shut down, which can be very hazardous when facing off against a large force and if you push your luck too far, your mech will be destroyed as the reactor explodes. A small amount of heat can be vented with the use of a limited supply of coolant, but this will only suffice for emergencies. This system provides some degree of tactical thought, rather than just encouraging an “all-guns-blazing” approach; instead, it encourages one to pick their shots and fire decisively. The cacophony of sirens going off whenever you approach the heat limit is less desirable, on the other hand, which may happen quite a bit with a mech with high-powered energy weapons but high dissipation.

A group of service vehicles called Mobile Field Bases escorts you from mission to mission in the campaign, while their presence is also an option in the instant action and multiplayer modes. During the mission, they provide a place to repair and rearm the damaged mechs of you or your lancemates, while between campaign missions, they carry salvage from your spoils of war. This provides a modicum of logistical support throughout the campaign, carrying ammunition and salvaged weapons and mechs which can be used to enhance your fighting ability. Weapons and ammunition are automatically stripped from fallen mechs and enemy vehicles, while enemy mechs can occasionally be salvaged after a mission if they are destroyed by shooting off a leg or by being hit in the head. With the head being a considerably smaller target and only liable to being hit by luck or by careful shots on an effectively stationary target, the salvaging system promotes the targeting of the legs that was discussed above.

After the first few missions in the campaign or before an instant action or multiplayer mission, you can use acquired weapons, armour and equipment to modify your mechs in the MechLab. This provides a modular interface by which you can replace weapons, ammunition and equipment in your mech based on slots and weight, along with the addition of armour or more powerful engines with any weight remaining below the mech’s weight limit.


Working on the Madcat (Timber Wolf) in the MechLab.

This system does go some way in mitigating the comparative lack of mechs compared to MechWarrior 4, although it does lead to a situation where many of the Light or Medium BattleMechs become obsolete after the first few missions in the campaign, with more focus on Heavy BattleMechs such as the Thor (Summoner) or Madcat (Timber Wolf), which can be heavy-hitting, reasonably well-armoured and not too slow, along with some potential for heavier, slower Assault BattleMechs such as the Supernova, Sunder and Daishi (Dire Wolf). This is further pronounced by the power of missiles in the game, which appear to hit more often than in the tabletop game, giving Light BattleMechs little chance to survive in a protracted battle in open terrain. Nevertheless, the salvage system is a welcome addition to the campaign, giving good players the chance to score some powerful mechs rather early on, which makes going through some of the missions rather rewarding if you play through them right.

The campaign consists of 20 missions which take place in a variety of surroundings. In setting, the missions are varied enough, although most of them consist of destroying Clan facilities and forces, which fits well enough with the story, but except for a few notable missions, most of the campaign takes place in open terrain without any sort of time limit. Since the AI are not especially clever, regularly failing to close up on your lance or at least move in a fashion that isn’t predictable, it can often be most rewarding to stay out of their range and snipe away with hitscan weapons or missiles instead of risking combat at close range.

The plot of the game is reasonably good, although you don’t have to be a dyed-in-the-wool BattleTech aficionado to enjoy the game. There are a few nice references to the wider conflict between the Inner Sphere and the Clans scattered throughout the game, while the interaction between your lancemates is often interesting. Hearing the Clan transmissions between members of the Warrior Caste and their commanders berating them for their poor performances is rather enjoyable as well, along with the frequent cursing of “Inner Sphere surats” or other Clan-specific insults.

Graphically, the game is serviceable for a game more than a decade old, although the limitation of the game to 4:3 resolution ratios with a maximum of 1024×768 in hardware-assisted mode and 640×480 in software mode is somewhat frustrating with the newer breed of high-resolution widescreen monitors. A more pressing matter is the game’s incompatibility in the hardware-assisted mode with newer ATI or AMD graphics cards; this is related to the game’s use of a 16-bit z-buffer and requires a fan-made patch to be applied to the game to stop major graphical glitches which make the game nearly unplayable.


In the middle of battle with some Light BattleMechs.

The sounds are reasonably good, with a good set of sounds to represent the pounding of a mech’s foot on solid ground or the splashing of mechs through water. Weapon sounds are reasonable, with the choppy sound of autocannon fire, the whooshing of missiles or the high-pitched pulse of a laser being fired (disregarding how realistic that might be). Musically, on the other hand, the game isn’t much to speak about; there are two tracks that repeat throughout the game, neither especially long and both more ambient than anything else. If you can’t get the Red Book audio to play, it’s not a massive loss.

There is one major flaw with the game, although it’s not inherent to the game but instead an issue with the time when it was created. The game does not play well with modern operating systems. Installing the game on Windows 7 is a challenge, while trying to get it playing can be frustrating. One issue which I have periodically experienced even with a game patched for the ATI graphics glitches is the tendency for the game to freeze during the loading of a mission. This can be resolved by switching the program out of the foreground, e.g. by opening the Task Manager. However, when returning to the game, the ATI graphics glitches resume, with many textures in the game replaced by rainbow-like smears of red, green and blue, which negates the very solution that the ATI graphics patch was meant to fix. However, when the game does start working, it works smoothly and without particular problems. It’s a pity about the incompatibility of Windows 95/98-era games with more modern versions of Windows, but then again, I’d rather have an operating system that works and worry about the games later.

Bottom Line:MechWarrior 3 is a solid game, with strong fundamental gameplay and a decent plot, but doesn’t play nicely with modern versions of Windows. In terms of gameplay, whether you’ll prefer this game over the more forgiving gameplay of MechWarrior 4 and its expansions all depends on how close you want your computer games to be to the tabletop game.

Recommendation:If you’re interested in mecha games or the BattleTech universe, you will be well-served by this game. Just be ready for a bunch of frustration if you try to run it on anything more modern than Windows 98.

Probing The Inaccuracies: Mecha

There’s something about a gigantic bipedal robot that inspires the imagination. Whether it’s the return to personal, one-on-one combat that many mecha-related series seem to explore, or the idea of a huge humanoid machine kicking ass, it’s pretty easy to see the appeal of mecha. It’s also difficult to dispute that they are, in fact, rather awesome.

Unfortunately, they are also completely pointless.

The first question that needs to be asked is, “What exactly is a mecha good for?” Putting aside clearly improbable designs that can freely fly, as found in the Super Robot genre, it would seem that the mecha would be designed as an analogue to the tank – or alternately, to displace the tank completely. This, to me, seems rather improbable, as the limitations of any of the common designs of mecha – bipedal, tripodal, quadrupedal or spider-shape – far outweigh any advantages conferred upon the machine by that design.

To investigate why this is so, we must examine the general form of the mecha in order to determine its typical characteristics.

The Vincent from Code Geass, a series which I feel gets things very, very wrong.

The first, and probably foremost, problem with this design, which appears to be representative of most mecha designs, is its high centre of gravity combined with only two points of contact with the ground. As anybody who has been pushed over before they have a chance to brace themselves will know, this leads to a considerable amount of instability. As mecha would require an improbable amount of flexibility and speed of movement in order to brace themselves after an impact, this would lead to the design being very easy to topple over, and thus incapable of taking any sort of impact without being rendered immobile and therefore useless.

Of course, an actual mecha design would be fitted with gyroscopes in order to prevent it from falling over when it so much as moved on any sort of surface that wasn’t completely flat, but there’s only so far that one can go with gyroscopic stabilisation, and of course, gyroscopes add weight to the machine. It really outlines the disadvantages of bipedal movement in anything that isn’t biological, humans only being capable of doing it efficiently due to their locking kneecaps and the ability to unconsciously maintain their balance with tiny, almost imperceptible movements.

It isn’t just bipedal mecha which suffer from stability problems and a high centre of gravity. Designs with more than two legs may have a more stable base, which largely negates the need for heavy and cumbersome gyroscopes, but they can be just as easily knocked over with a large enough impact. Once a leg is restrained or destroyed, instant stability problems occur, with the machine being rendered instantly immobile, and most likely falling over because of their inability to redistribute their weight unlike a biological organism. The vulnerability of the legs of these machines means that they are rendered vulnerable to tanks, close-air-support aircraft and even men with portable missile launchers. As it is difficult to distribute armour to the legs of mecha without making their movement cumbersome, it would appear that mecha would be limited immediately by the weakness of their legs.

The AT-AT from Star Wars, a series which may not have been realistic, but which outlined the ease of knocking a big mech over.

This isn’t the only weakness of a design based on legs. Leg movement is a form of reciprocating movement, where a piece of machinery repeats a back-and-forth (or up-and-down) movement. While this has proven to be the only successful form of ground movement in animals, reciprocating motion is not considered to be desirable for machinery which is used for propulsion. In engine design, a reciprocating engine requires far more components and usually wears out more quickly than an engine utilising circular motion, and attempts at replacing the piston engine in cars, planes and ships have been common ever since the development of the electric engine and gas turbine.

The gas turbine has displaced the reciprocating engine in all but the smallest aeroplanes since the 1960s, either in the form of the turboprop or the jet engine, while larger ships commonly use turbine engines in order to propel them instead of more complicated, more difficult-to-maintain piston engines. Only in cars and motorcycles has the piston engine persisted; the superior fuel consumption of such engines at that size compared to gas turbines and Wankel engines has allowed them to carve out that niche. However, electric engines utilise circular motion, and with the development of improving battery technology and hydrogen fuel cells, the piston engine will likely be displaced in this market as well.

This has relevance to mecha, because even piston engines convert their reciprocating motion to circular motion at the crank. If one were to directly connect the mech’s legs to the engine, one would either be converting circular motion to reciprocating motion, if a gas turbine or some sort of electric engine were used, or reciprocating motion to circular motion back to reciprocating motion if a piston engine were used. I hope you can see why that would cause apoplexy in many engineers; you’d essentially be transmitting power through another set of complex components, which adds more places for an already complicated machine to fail. If that doesn’t drive the engineers crazy, then it would definitely drive the mechanics that would have to work on it to drink.

It’s unlikely that a direct mechanical linkage to the engine would be used, for not only the reasons outlined above, but also because it would limit the flexibility of the limbs and leave them essentially as simple, crude metal struts. A far more likely system to be used is a hydraulic system, similar to the digging implements found on bulldozers. This would allow for movement of the legs more closely related to the movement of human legs, but would still be considerably less efficient than the movement of actual human legs. As discussed above, the locking kneecaps and ability to quickly change one’s balance lead to efficient bipedal movement in humans, but what would distinguish us from mecha capable of doing the same thing is that human muscles work on the microscopic scale, with nanoscale particles involved in the molecular biochemical activation of muscles. Ultimately, this scale allows humans and other animals to have impressive strength for their size, using a lot less energy than a comparative hydraulic system would use.

The Cauldron Born from the BattleTech series, a series which at least does things a little better than most mecha series. A little.

Returning to the general form of the mecha, apart from the disadvantages conferred by the instability of such a top-heavy design, the height of such machines leads to another obvious disadvantage: It leads to them being very noticeable. For something that purports to be an analogue to the tank, that is rather a significant weakness. Some people seem to forget that tanks are hardly invulnerable themselves; their tracks are potential targets to even outdated anti-tank launchers, while tank-on-tank combat can lead to the destruction of one of the tanks with just one lucky shot. As such, tanks attempt to decrease their profile and the amount of area to target by running hull-down, using terrain to disguise and cover themselves. This is not a luxury afforded to mecha.

The weaknesses of mecha versus tanks continues with mobility. By virtue of independent driving of both tracks, a tank can turn on its axis, while this is difficult, if not impossible for mecha to do. In order to turn the legs of a mech, one requires a complex series of components which far outstrip the complexity of comparative tank steering systems. As with the difficulties posed with reciprocating motion, these complex systems are useless for anything except making engineers and mechanics very angry.

Even then, the movement will be awkward, which would be especially dangerous in urban combat. Tanks are hardly the most appropriate weapon system in that sort of warfare either, to be fair; they are particularly vulnerable to improvised explosive devices and anti-tank launchers fired by people concealed in buildings, but mecha are even worse in these environments, with problems pursuing or retreating, which is rather problematic.

Just when you thought that there couldn’t be any more mechanical problems with mecha, physics comes and bites the idea in the arse again. Mecha are typically very large machines, and with increasing size comes an interesting correlation. For every squaring of surface area of an object, its mass goes up by the cube of the original object’s mass. While a human male may be on average 70kg, when that same humanoid shape is scaled up to several times that of a human, the mass increases correspondingly, such that mecha end up extremely heavy. A small increase in the height of a mech can necessitate the use of far more powerful servo systems and hydraulics, which is expensive not only in energy but also in cost.

The excessive weight of these machines can cause problems in other ways as well. A heavy machine resting on supports with limited surface area in contact with the ground leads to high pressure underneath. Tanks require wide tracks in order to prevent themselves from sinking into soft ground, but unless a mech had ridiculously wide feet, it would be likely to get stuck very easily in anything softer than concrete or baked soil, and to break up roads in urban terrain. Not particularly useful when you already have mobility problems.

Having discussed the weaknesses of mecha design, let me reiterate that I can still accept the inclusion of such machines in certain series, subject to some rules. I think the most important rule is that the series doesn’t take itself too seriously about the realistic use of mecha, unless there is a very good reason for their inclusion.

The second rule is that mecha in a series really have to “belong” – a criticism that I level quite heavily at Code Geass, as I don’t believe that an empire fundamentally deriving from the British would focus their efforts on huge mecha, as there is little in British tradition to suggest a significant interest in such developments. Ultimately, I think that the alternate history angle of this show, which actually could lead to a very interesting setting, is somewhat let down by the inclusion of something that doesn’t really fit. It might be said that I would need to watch the show with a certain amount of suspension of disbelief, but as the other instalments in this series of articles may suggest, that’s something I can’t always do.

On the other hand, BattleTech can be taken as an example of a mech-related series that I do enjoy. The mecha seem to fit better into the series than some other series involving such machines, and although there is a significant disparity between their portrayal of mecha and designs which would work in real life (insofar as such designs could work), they at least don’t portray the machines as invulnerable, giving it the heat-venting problems which add a bit of extra tactics to the series. I think it’s a good example of how to do mecha correctly without necessarily making them realistic.