Most discussions of mecha viability seem to give advice in the vein of handwaving laws of nature or making the audience (readers, players etc.) not look too closely at the justifications. I'd like to take a different approach: face the problem head-on and declare the existence of a series of superscience inventions that we know would be impossible in real life - inventions that are of use for making mecha viable, mitigating their usual disadvantages relative to tanks, but can't be applied to improve tanks to a similar degree.

I'd like to request some help in identifying what would be the minimum required set of such inventions and technologies (in terms of what flaws they mitigate).

For example: mecha are commonly criticised for having worse armour coverage and thickness than boxy tanks; a force field that can resist the same amount of damage no matter the shape, is generally more helpful for survival than conventional armour is, and doesn't upscale or downscale too far from the desired mecha size, would likely (as far as I understand) mitigate the armour-related disadvantages of mecha. We also know that a shield like that would be blatantly breaking some physical laws. But such if such breaks are strictly limited, this can be a good solution. Of course, that's just a single example for one of the mecha flaws that need to be addressed, and there are many.

Additional notes/constraints:

  • The solutions are for use in 'near future' settings or settings that are technologically divergent but comparable to the former. E.g.: settings similar to those found in RoboCop, Deus Ex (1, 3 and 4, but not 2), Super Force, Tiberian Sun, Transhuman Space, Infinite Warfare, Choice of Robots etc. If the solution can be fitted into a less technologically advanced setting (WWI at the earliest), that's a bonus but not a priority.
  • Mecha size is assumed to be in the range between ED-209, RoboCop 2 and a Titan in Titanfall. Hopefully that also means that the mass would be on an order of magnitude comparable to at least some of the real tanks.
  • Shape doesn't need to be constrained to humanoid, though it's a big plus if the solutions are applicable to humanoid mecha.
  • Ideally, solutions should be applicable to mecha piloted by biologicals.
  • Support for over-the-top tropes (such as swordfights or flying transformer mechs) is not a priority. If a minimum set of inventions happens to enable something like that, it's a plus but not necessary.
  • While obviously answers will involve a lot of pseudoscience (so this question probably doesn't qualify for the tag), it is best to treat said pseudoscience with a degree of scientific-like rigour and respect, at least when it comes to applying it and its products.
  • How the technologies work isn't important (and would be idea generation anyway), the important thing is what set of flaws they must mitigate (at a minimum).
  • An answer is good if it shows a minimum required set of such inventions that makes fielding mechas in most of the sorts of scenarios where conventional armoured vehicles are normally fielded. They don't need to completely obsolete tanks, just become worthwhile enough in similar scenarios that they can viably fight side by side, or one against the other. It is OK if the scenario overlap is incomplete and/or if one or the other serve better in various subset of armoured warfare scenarios.

closed as primarily opinion-based by Renan, Morris The Cat, CaM, Starfish Prime, Ash Jul 9 at 14:17

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 4
    $\begingroup$ I think the only way to make mechas compatible is to not focus on them being better tanks than tanks. More of a shift in the perspective - mechas wouldn't be better tanks but tanks can't be better mecha. You move the battlefield to where tanks are not effective and then mecha will be the better option. $\endgroup$ – VLAZ Jul 9 at 11:04
  • 2
    $\begingroup$ There are several questions on this site about the feasibility of a variety of walking mechs. Not sure any of them are duplicates but they could certainly be useful to you. $\endgroup$ – Joe Bloggs Jul 9 at 11:04
  • 1
    $\begingroup$ Said topics also heavily points to what VLAZ said: mechas are awful tanks. From experience, most of the answers you'll get will point out that, no matter the techonolgical improvement you'll make up, tanks are just better at being tanks than mechas. It is simply due to the base form you want to give them. It's like asking "what kind of biological improvements could we engineer to make humans better wolfs than wolfs". Answer: none. Shift the fight to another field and you'll fare way better. $\endgroup$ – Nyakouai Jul 9 at 11:26
  • $\begingroup$ @JoeBloggs I have searched for those prior questions, and I found them to be approaching the topic from other angles, which is why I posted my own question. This is why my question is focused on 'what makes them bad and what, at a minimum, must be mitigated to make them competitive in this area'. $\endgroup$ – vicky_molokh Jul 9 at 11:38
  • $\begingroup$ @nyakouai that Mechs are aweful is like the idea that lactate is bad for your body. The more we look into it the more advantages lactate gives. While its true we dont have mechs now, I have no doubt that we would see multi-legged mechs assuming the tank remains a viable combat method and we arent relegated to insect-sized and shaped cameras that guide artillery shells to the target (or micro-sattelites in the atmosphere or computer warfare etc). Just like tracks did not make wheeled vehicles disappear because of unique advantages the legged mech would have its own advantages. $\endgroup$ – Demigan Jul 9 at 11:47

There are a variety of areas that your mecha needs to address. You might can overlook some of these, but the more you cover, the better your mecha will fair against tanks...

Balance and Computing Power

You need faster computers. The current state of the art for balancing robots isn't sufficient to meet the demands of combat and moving at speed. These demands are going to require computers that can react to sensor readings to provide necessary responses to shifting balance in real-time.

Sensors and data

You need more sensors that can send data more accurately and faster than today's systems. Your robot must have massive amounts of data about the 3D layout of the terrain, the robot's own body positioning, ground material (hard surfaces require different balancing techniques from soft; wet surfaces balances differently from dry; slick surfaces react differently than high-friction...), weather, and oh, yeah, what the enemy and ally forces are doing... And all of that data has to be perfectly reliable, 100% of the time. If you have one sensor array for everything, and a sniper hits it, your robot is done. So that means redundancies.


You have to minimize weight. Armor is heavy. Engines are heavy. Fuel is heavy. Munitions are heavy. You need to reduce your weight as much as possible for a variety of reasons. Lower weight means more mobility: higher speeds, faster changes of direction, and fewer restrictions on passable terrain (try taking a 70 ton main battle tank through a shallow lake). Lower weight means you have a system that is less likely to be top-heavy, which causes balance problems. Lower weight means your two-to-ten legs are less likely to sink into soft, wet, soil and get stuck.


You need power. The robots you see in those really cool videos doing balancing tricks, etc. usually have cables running to an external power source. The M1 Abrams main battle tank uses a gas turbine engine that gets about 0.6 miles per gallon of fuel and has a 500 gallon (1900 liter) fuel tank.

Your mecha needs a power plant that is more fuel efficient and far lighter. Lighter because otherwise the whole thing is top heavy and therefore harder to balance and more prone to falling over. Batteries won't pack enough power and will take too long to charge, so you're going to need a more exotic source of energy. Fusion reactors? Can you miniaturize those enough AND keep them stable under combat conditions? And this power plant needs to deliver more power than your tank's engines. Because wheels and tracks are more energy efficient than lifting and moving limbs, in general, so your mecha requires more power to move the same distance at the same speed relative to a tank of the same weight.


You need some sort of exotic armor. The Abrams tank uses steel, ceramics, and depleted uranium for its armor because of the material's density. You need a material that's lighter and thinner. Lighter because weight matters (see above). Thinner, because you don't have a big box to armor. You have comparatively spindly legs to armor. Spindly legs that need to move around, not just be hauled by tracks.


You need lightweight weapons, to whatever extent possible. This reduces the overall weight of the vehicle. And it would be best to avoid massive cannon like a main battle tank's main gun. These weigh in at about 1200 to 3300 kg and require cooling systems, recoil absorption systems, and other ancillary (and heavy) sub-systems. The recoil of a main battle gun might topple over your mecha. So you need lighter, faster, munitions. Lasers maybe? They won't require you to pack around tons of depleted-uranium ammunition, at least! And they might produce less heat and less recoil, maybe? The Abrams carries 42 rounds for the main gun, plus machine gun ammunition. You need more than this to outperform a tank.


The Abrams tank crew is four: a commander/machine gunner, a gunner, a loader, and a driver. Your mecha needs space for its entire crew. With better weapons and computers, maybe you can reduce the crew requirement to two, a driver and a gunner/commander? But they require space, air, food and water for campaigns, and safety systems. They are piloting a machine that has not just typical tank motions to deal with (driving fwd/back/steering plus turret controls) but also balance-related issues. Maybe your computer system can automate some of that, but you still have to account for the differences in complexity and comfort. Please provide some padding to absorb the shocks from foot-falls at speed that your crew will experience.

Target profile

Your mecha stands tall and proud. Typical trope-approved mecha are at least 30 feet (9.1 meters) tall. For comparison the Abrams tank is 8 feet (2.44 meters) in height. Guess which one is easier to spot coming towards your defensive position? So your mecha has to have some technology(ies) to counter their target profile. They are easier to see at a distance, so you need stronger defenses against targeting systems. Heat shielding to block night vision and heat-seeking missiles, for example. And better armor, because they're going to take incoming fire sooner and therefore longer than a comparable tank. And higher maneuverability, because a fast-moving target is harder to hit. Camouflage may be pointless for something so tall, but it might help? I am not sure, but if you don't want tanks to be superior, this will need to be dealt with, somehow.


The published operational range of an Abrams tank is 265 miles (426 km). Your mecha should be able to go further between refuels (see Engines, above), and, to outperform tanks, needs to carry more firepower and/or greater ammunition reserves than that tank (see munitions, above). All while packing enough supplies for the crew. Because otherwise, your mecha is going to spend more time waiting on the supply vehicles to catch up than actually fighting.


How often is your mecha shut down for routine maintenance? How easily/quickly is it repaired? What percentage of "common" maintenance issues can be handled by the crew while under fire or otherwise cut off from specialized maintenance personnel and tools? Can they do any kind of improvised maintenance in the field, or is your mecha so super-advanced that it requires specialized equipment to do anything on it? How likely is it to suffer damage that necessitates taking the unit out of service? Your mecha must be more dependable than a tank or other comparable weapons platforms.


While not a critical requirement, one of the complaints against the Abrams is that it was designed from the ground up as a tank-killer. That's what it does best and everything else is basically just a condiment on that all-American hamburger patty.

If your mecha platform can be configured for different roles without compromising any of the above concerns (roles like scout/recon, heavy weapons, urban combat, anti-aircraft, etc. etc. etc.) then it has an advantage over tanks. Maybe the weapons systems are modular and can easily be swapped out between heavy tank-killer systems and lighter, faster, anti-aircraft platforms for example. Or maybe the lighter weapons give it greater speed and range for scouting purposes?


This is where the feet (or treads) meets the road. Is your mecha more maneuverable than a tank or other weapons platform, across the same or a wider array of terrain types? I've alluded to this in several of the above points, but this is the make-or-break one. Tanks have their problems with terrain -- they can be stopped with highway barriers. Any mecha worth the name will step over such paltry defenses with ease. Tanks can't cross water without serious and complex adaptations (or other vehicles). They can get mired in a few feet of mud.

But your mecha can get stuck too. Sandy soil, or extremely wet soil will cause problems for a mecha. They can sink into wet soil easily, since their full weight rides on anywhere from two to ten or so legs. That's a small contact area compared to the full treads of a tank. I've seen tracked construction vehicles sink to midway up the cab in soft soil and they're far lighter than a tank. What will your mecha do, on those comparatively tiny feet?

You might can hand-wave some of the various problems with mecha, but this one cannot be ignored. Otherwise, your mecha will be a non-moving weapons platform, waiting to be destroyed via aerial bombardment or an infantry soldier with a portable rocket launcher of some sort.

  • 1
    $\begingroup$ This is a good answer that addresses what was being asked, and that points out the sorts of improvements that would be needed. It is a candidate for becoming the accepted answer. $\endgroup$ – vicky_molokh Jul 9 at 14:57
  • 1
    $\begingroup$ I think you and I were basically writing the same ideas at the same time. =P $\endgroup$ – Morris The Cat Jul 9 at 15:06
  • 1
    $\begingroup$ I would argue that for weapons a bipedal mech would borrow most of its weapons from things like Helicopters that already had to solve the weight and recoil problems to remain mobile and airborne. For the energy requirements I wouldnt use the calculations on a childscooter but on skates as they mimic the walking motion more. Also you have to consider discontinuous terrain because comparing tanks and mechs on a flat slope is unfair to the mech. without considering things like potholes, outcroppings, shifting sand or rocks where a well designed mech would take less effort is unfair. $\endgroup$ – Demigan Jul 9 at 15:50
  • $\begingroup$ And I'd say that a mech's target is far more likely to be places where a tank and a mech are on relatively equal footings, (no pun intended) terrain wise. Cities, factories. Military bases. These are only rarely placed in unreachable-by-tank terrain. $\endgroup$ – CaM Jul 9 at 15:56
  • 1
    $\begingroup$ @Demigan there's a problem there... if your mechs are using the same weapons as attack helicopters, why would you use one INSTEAD of a helicopter, which will be much faster no matter what kind of terrain you're talking about. If you have defenses that work on a mech, they work even better on a VTOL. If you have the power to make a bipedal mech move around with speed and agility, a VTOL will be even faster and more agile. $\endgroup$ – Morris The Cat Jul 9 at 18:58

If you want to make mecha viable, you need to address the root causes of what makes them very NOT viable right now.

If you look at the cutting edge of robotic design these days, they all rely on a musculature made out of a system of hydraulic rams running off a set of pumps. This more or less mimics the mechanical design of actual living things, but the problem is that this is catastrophically inefficient in terms of power usage.

Consider the Boston Dynamics BigDog, which is about as close as it gets to State-of-the-art. BigDog runs off a 15hp engine, weighs 240lbs, and can carry a further 340lbs at four miles an hour.

By comparison, a 125cc dirt bike uses the same amount of horsepower to carry the same payload over the same terrain, but weighs half as much, can carry that load much, MUCH faster and further.

This is Critical Problem One:You need something that can turn power into mechanical leverage at least an order of magnitude more efficiently than anything we have currently.

Even then, as you point out, a mecha will NEVER be able to make use of armor as efficiently as a tank can, so heavily armored mechs taking direct hits from tank cannons is right out. This means your mechs need to be fast and agile enough to either avoid being SEEN or avoid being HIT by heavy cannon fire.

It's worth pointing out here that a major design consideration for armored vehicles is keeping them as FLAT as possible. Closer to the ground means harder to see means harder to hit. A bipedal mech is directly opposed to this design consideration, and while a 70 ton tank can keep a pretty low profile at only ~8feet tall, a bipedal mech the same size would be more like twenty five to thirty feet tall and thus a MUCH easier target to find on the battlefield and hit.

This is Critical Problem Two: You need a battlefield environment where being a Big Freaking Target isn't a critical disadvantage. The simplest solution I can think of is a technological environment where active defenses have outstripped offensive weaponry. If you have stuff like the Israeli Trophy System that's efficient enough, and if you DO NOT have directed-energy weapons, then the only way to ensure a kill is to either hit your target with more projectiles than its Point-Defense can handle, or fire from close enough that the Point-Defense doesn't have time to react. This scenario makes a taller combat platform more useful because putting your sensor and point-defense systems as high up as possible makes sure you can acquire and engage threats that would otherwise be concealed behind obstacles.

Finally, you need a very sophisticated control system. The idea of controlling a bipedal humanoid machine with levers and joystics is, frankly, ludicrous. In order to realize the full benefit of having a human-shaped vehicle, you need the pilot to be able to control it as efficiently as he controls his own body.

Critical Problem Three: You either need an incredibly sophisticated control system that either takes all fine motor control out of the pilot's hands, or gives all fine motor control to the pilot in an intuitive way. Option one would work sort of the way mech video games do where the pilot just tells the vehicle where to go and an onboard computer system handles all the complexities of moving the legs around. This works best for non-humanoid mechs that are more like a walking gun turret. If you actually have arms and hands, then you want something a lot closer to the stuff that Masamune Shirow draws, where you have a force-feedback waldo system that directly connects the entire musculature of the mech to the musculature of the pilot OR, alternatively a system where you link the control system of the mech directly into the pilot's nervous system.

Important Disclaimer: I think humanoid mechs are stupid and find the idea of them EVER being a viable option on the battlefield laughable, but this was still a useful intellectual exercise.

  • $\begingroup$ Could you elaborate on why you think that directed energy weapons would be bad for mecha? $\endgroup$ – Ruther Rendommeleigh Jul 9 at 14:53
  • $\begingroup$ @RutherRendommeleigh directed energy weapons make being a Big Freaking Target a critical disadvantage because you can't dodge them or shoot them down. Mechs are Big Freaking Targets compared to conventional armored vehicles because they're much, much taller. $\endgroup$ – Morris The Cat Jul 9 at 15:04
  • $\begingroup$ DEWs (as we know them) are also not very effective at longer range and probably have to hit the same spot for a while to do real damage to anything big, so a nimble vehicle that can turn quickly and weave in and out of cover might be at an advantage. $\endgroup$ – Ruther Rendommeleigh Jul 9 at 15:11
  • $\begingroup$ DEWs as we currently know them deliver their energy in fractions of a second. Nobody is using the kind of continuous beams you're thinking about, even today. $\endgroup$ – Morris The Cat Jul 9 at 15:27
  • $\begingroup$ This is a good answer that addresses what was being asked, and that points out the sorts of improvements that would be needed. It is a candidate for becoming the accepted answer. $\endgroup$ – vicky_molokh Jul 9 at 15:38

Bipedal mechs are the worst. Mechs with 6+ legs is where its at, as those can fill a niche in climbing steep terrain tanks cant cross and have less vulnerability when a leg is lost compared to a tank losing a track. Bowever since you want bipedals...

Minimum requirements are:

  • cost. Replacing a track is relatively cheap compared to a (piece of a) leg. The loss of a leg should either have little effect on combat effectiveness at first (spider mechs!) Or become so cheap and easy that the price and time difference becomes neglegible.

  • Efficient leg design. The closest and most active field would be prostetics that have to mimic the body. While people are currently in awe of things like those carbon "spoons" that can replace legs, they are in effect barely two steps above a simple wooden leg. A real biological leg is a million times more complex and capable than the most effective electrical prostetic today. Just the ability to use up to 90% of the energy of the previous step into the next one is a huge advantage that would equalize energy requirements between tracks and legs.

  • efficient leg control. While we have incredibly complex and capable computers we still have trouble making them balance or walk in a way that doesnt take up a lot of energy. This gets worse if a part of the leg is damaged. While it takes humans often all of 3 steps to figure out the most efficient way to walk despite randomized damage many computers cant do this yet if it isnt pre-programmed. And a combat-ready mech should be able to take damage and adapt if a joint locks up or loses some actuators and strength.

  • Efficient feet. If your Mech is bipedal and as heavy as a tank you got problems crossing any soft terrain without breaking it (again, spider mechs!). Design of the feet would need to solve problems with grip on the ground and each step should not have the leg sag into the ground (unless portions of it are designed as a way to gain grip, like spikes).

  • lightweight high strength materials, preferably cheap ones. Graphene would potentially offer a great combination of armor, support structure and be lightweight enough that the Mech remains useable.

  • new conventional warfare. If for example Graphene armors become so strong that kinetic penetrators fall in favor and you have to use chemicals that destroy the armor or cause shockwaves to rupture components inside the tank then the "box on wheels" configuration loses value as armor is less important than say mobility.

  • 1
    $\begingroup$ You're missing the biggest issue with mechs as opposed to tanks: Power efficiency. The drivetrain for a mecha with ANY number of legs is going to be larger, heavier, and an order of magnitude more complicated than what would be required to move something the same mass on treads or wheels. This means that without a COMPLETELY new means of power and mechanical transmission, your mecha will always be MUCH slower with weaker armor and a smaller payload than a conventional armored vehicle of the same mass. $\endgroup$ – Morris The Cat Jul 9 at 12:59
  • $\begingroup$ @morristhecat an order of magnitude? where does this come from? And did you miss my point about efficient leg design to regain a large portion of the energy of each step to use in the next one? $\endgroup$ – Demigan Jul 9 at 15:25
  • $\begingroup$ see my answer below regarding the comparative mobility of a BigDog robot and a 125cc dirt bike with the same horsepower. Order of Magnitude is technically correct in this case. $\endgroup$ – Morris The Cat Jul 9 at 15:29
  • $\begingroup$ @Morristhecat but that is because the technology isnt there yet. Bigdog does not regain any energy with each step for example. Its like comparing aircraft technology of the wright brother's first aircraft to the trains of that age and saying "and therefore planes will never outmatch trains". But technology advances, and each vehicle got their own place in society based on their advantages and disadvantages. $\endgroup$ – Demigan Jul 9 at 16:23

Mechas are the only tech there is.

No superscience needed. Quoting Donald Rumsfeld:

As you know, you go to war with the army you have, not the army you might want or wish to have at a later time.

Your people have acquired mechas of whatever type you like for your story. They found them - a lot of them, in good working order. These mechas are not products of the people who found them and use them or products of anyone those people know - they are either from a far distant time, or a forgotten alien cache, or something of the sort. Maybe they suddenly arrive; some dimensional transport with a wrong address. In any case the mechas are there, and there are lots of them. What is not there is tanks, or particle beams, or any of the other high tech weaponry one would expect from a culture able to build mechas.

Your characters are resource poor and their technology is not advanced. But they are not stupid, and they can operate the mechas which were apparently build for creatures similar to themselves - although many features are being discovered on the fly, and the principles by which they operate are so far beyond your civilization as to nearly be magic. They can be repaired because they are modular and parts can be modified from one Mecha to the next.

In a world like this where the tech was early industrial revolution and they found mechas, they would use mechas because that is what they have.

  • 1
    $\begingroup$ This post seems to be answering a very different question than the one that's been asked. In fact I'd say it outright ignores the content of my question. $\endgroup$ – vicky_molokh Jul 9 at 13:51
  • $\begingroup$ You consider your question too narrowly. You are in a contest, you are committed to your favored approach and are faced with an opponent who routinely beats your favorite approach. The shortcoming with mechas is that tanks are better. You think the only answer is to get into the weeds and improve your favorite approach; that is legitimate but it is not the only answer. The other option is not to compete against the superior opponent. Without the superior opponent the shortcomings of your favorite approach are no longer shortcomings. $\endgroup$ – Willk Jul 9 at 14:42
  • $\begingroup$ The question is scoped the way it is because it is not being asked in a vacuum, but because its solutions are intended to be applied to specific modifications of specific sorts of settings (but going into those additional factors would bloat the already long question). Also consider that too-broad questions are already not very well stackable. $\endgroup$ – vicky_molokh Jul 9 at 14:49

The only advantage of mechs over tanks is mobility. In all other aspects at any given technology level tanks would be superior to mechs.

But still lets try some assumptions:

Tanks are good en masse. If technology somehow restricts number of mashines that can be produced - that may give mechs a chance (if not for technology - see BattleTech verse).

  • shards of some alien artifact (say meteorite made from unobtanium) wich can be used to produce extrimely poweful alternating magnetic fields (can be used as power plant, weapon platform (gauss gun, wave-motion lasers), metal-deflecting shields). Argument to use mechs - this mashines would be deadly to any nearby humans (and electronics) due to high microwave EM radiation and have to act alone, without direct suport from other troops.

  • enhancing human brain with some some sort of computer and integrating it into the machine. Only few can survive the process and stay sane. Tank "body" would be too limiting and unnatural to stay sane, it means we have to use mechs. This will reque some supporting technology - compact powerplants, advanced weapons wich benefits from this cybertech (ex., mind-controlled missiles and counter missiles, drones - think of light version of "Suprime Commander" vers).

  • $\begingroup$ This answer seems to be approaching the question from adding limitations, which is contrary to what was being asked. $\endgroup$ – vicky_molokh Jul 9 at 11:39
  • $\begingroup$ Without adding limitation any, I repeat ANY technology, with wich you want to "improve" mechs would improve tanks twice of that. Yo need to nerf tanks somehow for your idea to be logically coherent $\endgroup$ – ksbes Jul 9 at 14:25
  • $\begingroup$ Unless it's a technology that cannot be applied to tank, or that isn't as drastic when applied to them. Thus all the talk of superscience/pseudoscience in the question. $\endgroup$ – vicky_molokh Jul 9 at 14:50
  • $\begingroup$ Both mech and tank are large and complex chuncks of metal, plastic and so on. They share a lot. And thus I can hardly imagine any realistic non-magical near-future techlogy that can be applied to mech but not to tank. To be more precise all I can imagine is in my answer. $\endgroup$ – ksbes Jul 9 at 15:01
  • $\begingroup$ The question explicitly has 'superscience' right there in the title (as opposed to, say, merely 'advanced science'), and there's mention of pseudoscience too. $\endgroup$ – vicky_molokh Jul 9 at 15:42

Not the answer you're looking for? Browse other questions tagged or ask your own question.