Mecha design; From cube to humanoid

The previous post about mecha design was all about the basic ideas that yours truly tends to use when it comes to transforming or shape changing robots. As mentioned, they are not definitive and many would probably contest them, but they work just as well. However, all transforming mecha follow one essential thought pattern most of the time; from inhuman shape to humanoid shape. This shape can be whatever. Cars, planes, guns, dinosaurs, trains… pretty much everything has been turned into a robot. Hell, there used to be a saying on imageboards that the Japanese can transform anything into a mecha if they just want to. Of course, there are those that simply change utility shape between modes and never become humanoid. These are relatively rarer in scale of things, but the overall discussion follows the same pattern overall. You have a shape that you want to force into another.

The title of this post is misleading. The term that I should be using is cuboid. However, I am going to break any and all good language practices and keep mixing cube and cuboid to label any cuboid shapes. This would an example of marketing of sorts when you get down to it, as many companies want to use cube in a similar sense. Nintendo’s Game Cube being one, with it being a cuboid even when the Game Boy player is attached.

As with any matter like this, there is no one correct way to do anything. The examples here are simply just for the sake of examples and being as simple as possible. Expanding on basics and building on them is really the only way to get around.

The core idea is to take a cube and “spread” it to the similar breakdown as human would be, if we’d draw human with simple geometric shapes.

Continue reading “Mecha design; From cube to humanoid”

Three approaches to transforming mecha designs

Unfolding, folding, opening, twisting, turning, exposing areas and revealing hidden parts is basically what mecha transformation is all about. There is no one way to do it, and the sheer amount of examples there exists eclipses the scope I’m willing to work for free. To tackle transformation schemes in general requires part problem solving and part puzzle making in a nice balance, where a irregular shapes can be turned into e.g. a humanoid and vice versa. By first introducing this sort of base idea of categorizing transforming mechas into will give some foresight how I’ll tackle the subject down the line.

Much like Three approaches in mecha design (which will be rewritten at some point this year,) I tend to employ a similar template for transforming mechas specifically. These three are not necessarily connected to the three initial approaches as some sort of rule, but they do work under them if you’d wish to make a transforming mecha. These might help you to lock down your approach better. This post can barely scratch the surface of it all with the given limit I’ve set to myself.

The three approaches in transforming mecha design are Fantastic, Toyetic and Realistic. As with previous, there are overlapping elements with each of the three and can be even split into sub-categories if necessary. Examples of Fantastic transforming robots are all the outright impossible ones in any form outside animation and movies. Getter Robo and Gurren Lagann are probably the best examples, where thing just fall into their place and morph into new shapes. Mass shifting is nothing short of expected and even mandatory.

Continue reading “Three approaches to transforming mecha designs”

Music of the month; Lilia ~Winter Version~

Let’s dedicate this post to the changes that I need to make things viable again and what that means for my own time use and this blog. First, I won’t be dropping the two posts per week pace, that’s something I won’t back out on, unless something significant keeps me from doing it. The reason for this is that realistically I can’t make a living in my current profession. Craftsmen are not valued to any significant extent and their craft or skills are face the same end. The same tends to go towards designers across the board, and if you can’t make the right connections, there’s not much you can do. As such, I’ve taken a drastic decision to re-educate myself for a profession where I can utilise my previous experiences. To what exactly is something I will leave for the time being.

This means I don’t have much time in my hands. The aim is to go through three to four years of studies in one. That is stupidly fast pace, which requests me to concentrate my efforts and resources elsewhere. However, the nature of this blog won’t change too much if any because of this. Rather, I expect it to add further depth as I get more familiar with certain aspects of… well, that’s the open bit for you.

This is also the reason why there has been no new podcast for some time now. Not only the translator staff is busy at their own with both Muv-Luv related matters but also with their personal stuff. Juggling the schedules together has become exponentially more difficult, and sudden changes in what happens and when will become a daily thing to yours truly, at least. ARG is not killed, it’s just biding its time. The same thing really applies to the idea of my voice blogs, as I noticed that producing those in the way I’d like them to takes about four times longer than just writing. Maybe I should just do a stream of thought without a script, but how that would come together nobody knows.

Winter’s arrived here, meaning that while snow is still a scarce, cold weather has arrived and things slow down to take things with certain sure and safe pace. It also means Schwarzesmarken‘s second VN has been released, which means I can read both VNs in one go and watch the animated series. I’ve pushed the whole review thing back for almost a year now because I want to have a proper perspective on both of them without being influenced by hype or other views. Needless to say, both the VN and animation needs to stand on their own two feet, and comparisons between the two can be made. However, it should be noted that the two were made based on the Light Novels, which essentially served as a base script more than anything else. The animation changes things around to fit in the allotted time, while the VN has a lot more time and space just to dwell into things. That’s just the nature of the mediums.

There was no Monthly Three last month  as those take a lot of reading and planning. It may not seem like that, but they really take their sweet time to come together, and I usually plan all three parts in one go. Exceptions happen, of course. The same applies to the whole mecha design things. I do intend to write a TSF comparison this month, which will also serve as the month’s mecha design post. I haven’t decided which one, I need to check what images I have in stash and what I can get. However, for the time being, I do not intend to force myself to do a Monthly Three, unless a subject pops up towards me. Of course, I could use that for the mecha design stuff. Speaking of mecha posts, the post Three Different Approaches in mecha design will get a complete rewrite at some point in the future, and the old one will be replaced with that. However, I will archive that older version for future.

I will most likely insert few personal posts about games on smart phones. This is because my old Nokia finally went bust and I had to purchase a new one. This post, or posts if I end up making multiple, will be observations about mobile gaming in contrast to e.g. handheld console gaming.

I admit that lately this blog has not been up to the standard I’d like to think it has stayed at for a long time now. A lot of news and events that I wanted to write about have come and gone, but my time and simple stamina have been used to a more pressing matters. As said, if I were paid to write, I’d take this more seriously. This is more or less a hobby. Sometimes it stresses, sometimes it feel almost cathartic.

For now, I’ll have to leave you with this, despite it leaving me with a lacklustre feeling. I need to fix my tyres, somebody had slashed them the other night along with seven other’s.

Mecha design; manipulators

Consider your hand. You control all those 27 bones through muscles and tendons. The nerves give you feedback and send your commands down the like, commands that you are not even conscious of. Twist your hand, and you see it twisting. The large muscles come through the skin, but all the fine motion is lost unless we specifically look for it. It can grab and hold things in a wide variety of positions and ways, some that we don’t even know before someone else teaches that. These hands can build and destroy in equal amounts, they are our the tools of our creations.

Transferring that to a giant robot is a bit of a hassle.

Much like with a lot of other direct transfer elements with human body and giant robots, adapting hands 1:1 is an easy concept for sure. The idea of similar multi-use manipulator is attractive from the get go, but depending on the setting, human-like hands might not be the best option. A human-like hand requires far more parts, development, maintenance and simple tech than a say a pincer or more simple manipulator. Of course, the main argument for having a hand for a giant robot is its versatility, especially when it comes to weapons. However, that’s something that could be easily done with hardpoints where weapon is being mounted. We should also question how versatile does the hand of a giant mecha be, especially for a war machine.

Broadly speaking, all human-like hands with mecha follow the same basic idea, there isn’t much deviation. It’s either smooth or cubic. Using this example from a VF-19 serves as a good showcase.

VF-19 hand

While it looks complex, it’s more about the layered elements that make it look complex. Inner functions are of course barely thought, they’re not important. The fact that it looks like it could work and has plausible design elements, like the knuckle guard and fingers’ segments layer on top of each other when bent, is more than enough. Studio Nue has always preferred rounder elements to their design (sometimes dubbed as Bubble hands), especially with their older works. In Gundam, Sunrise and Bandai have preferred using more cubic hands, although exceptions are aplenty.

Gundam MS fed manipulator

The above generic Mobile Suit manipulator was designed for the models, but seeing how Bandai and Sunrise design their mechas models in mind nowadays, it’s a good example of a hand that’s more or less designed for wielding a gun and a beam sabre. It’s a bit more straightforward than VF-19’s, less well-rounded. The question of course is, if this hand is largely made for weapon carrying, why isn’t it designed as such?

The answer is, of course, because of Rule of Cool. When mechas are designed as characters, they’ve almost always given large amount of human characteristics in order to showcase dramatic events. Hands are no different in this. Beam sabre battles would be less dramatic and interesting if the manipulator would be specifically designed holder than a hand.

Controlling a hand like this has basically three options, direct 1:1 input, control macros or brain wave input. Variations and combinations do apply. While a “glove controller” would be idea, that’s pretty much what you do then with that arm. It’ dedicated for that arm, and the rest of the controls are either automatic or left other arm or legs. We discussed control macros previously, and this is most likely the best option overall, if brain wave scanning tech is not available in your setting.

Designing mecha’s hand really isn’t anything hard; just look at your own and mechanise it. Give it details for something to grab attention and some panels for easy access.

Giant robots don’t really have a need for similar level of sophistication when it comes to their hands, a simple grasping arm should be enough with some level of modification to suit the needed purpose. Hardpoints add a lot of versatility as well.

These take less maintenance and production costs would be lower too
These take less maintenance and production costs would be lower too

Of course, fiction doesn’t need to play by the rules of reality all that much, and if technology is advanced enough in a fiction to produce these things, why not? They could of course build better and simpler manipulators, but sometimes you do seek more complex solution for the sake of all the options it could give you. A gripping manipulator above doesn’t really offer many ways to grasp a thing.

Some franchises mix human-like hands with specifically designed manipulators, Muv-Luv popping to my mind foremost.

To be fair, this is complex for the sake of being complex, some of these steps could be dropped
To be fair, this is complex for the sake of being complex, some of these steps could be dropped. It’s a pretty good example of a very specific manipulator arm that works in junction of the main hand, something that I personally would like to see done more

Another one would Mobile Suit Z Gundam‘s The O with its assisting manipulators underneath its skirt. These manipulators question why would The O even need human-like hands, when the three-prong manipulator does everything they do. The answer to this is, of course, because the human design does not use that sort of hand. In a way, mecha in general should always be contrasted to armoured knights of legends, but that’s another topic.

Hands are ultimately something that Japanese inspired mecha design does. For giant robots, America has always preferred more built-in options. MegaBot’s Mark II is a good example of this.


American vision usually attached the weaponry onto a pre-fixed arm that may have some freedom of motion to it, but is always more dependent on the movements of the main body. Compare this to Suidobashi Heavy Industry’s Kuratas and the difference in approach is notable.

The idea of having this built-in approach and lack of manipulators is just as valid.  While it lessens on-the-fly options and puts some limitations, it eliminates loads of moving parts that would require maintenance. The most prominent film example of this sort of thing would be our good old friend, ED-209.

I should probably write a whole entry on ED-209.
I should probably write a whole entry on ED-209.

Unlike with mechas with arms and manipulators, you can see ED-209 guns are its arms with no manipulators, as it needs none. It’s a robust little connector that looks sturdy and serves only to take the beating from the cannon’s recoil and swivel enough to shoot whoever full of holes.

Keep an eye to hands you see in mecha films and shows. Take notice how they are portrayed and how they function. Rarely you will see them doing things outside the capabilities of human hands, and showcasing how they are actually controlled is even rarer. Sometimes they take advantage of what a machine hand can do, like how Gundam washes clothes by rotating its wrist 360-degrees in repetition.

Washing machine Gundam

Mecha design; organic vs industrial

While organic vs industrial design is relatively dry-cut most of the time, I do feel that it’s a subject that needs to be touched upon to give further idea how to put some twists to whatever giant robot you are making.

Most Japanese mechas have a level of organic in them in form of general humanoid shape, but organic design is more about bio, about the smoother, naturalistic lines. This is of course contrasted with harsh corner and straight lines in the design that are almost innately machine-like, produced by human industrial forces instead of nature. Something like a tank or a car can be fully inorganic in design, but more often than not, elements from nature are applied to make it more pleasing to the eye.

There is a constant middle ground between the two, but it’s not exactly biomechanical. Gradual change between organic and mechanic design in giant robots has more to do with the base of general visual, while biomechanical is straight up combination of the two in harmony. There’s also techno-organic design, but that’s sort the same thing. It just has slightly stronger emphasize on the technological side rather than having the two governing together.

To use actual art an example of organic design, Art Nouveau is by its core is tied to naturalistic growth and style. It’s a direct contrast to the industrial style. For example, Gustav Gurschner’s Lampe Nautile, Vers 1899, exhibits the basic loose roles for organic style.


The rules are not hard, and I’ve effectively already mentioned them; curved, flowing, natural. There are no real harsh corners anywhere on the lamp and no visible connection points. Instead of steel gray, earthly bronze was used combined with the pearly look of a shell. From visual side of thing, go check Alfons Mucha, my personal all-time favourite.

That’s the stuff

To directly to Art Nouveau, Bauhaus’ had many core industrial designs that still affect how things are made, produced and designed. An industrial design is rather the opposite to organic, leaving less room for the organic growth and cutting the chase.


This 1930’s Bauhaus desk lamp shows some of the core elements in the rules; unapologetic in simplicity, not hiding joints or the fact that the form follows function and not the other way around. Rather than an earthly bronze, steel shine is applied to the piece with a brass joint at the base with a white baccelite switch. Even the switch is emphasized with a slight raiser from the base.

The two lamps both would serve in their function as a light giver, but the other fits for more moodier lighting, while the other is more a tool for office use. This relative idea is apparent in mecha design as well.

Aura Battler Dunbines Aura Battlers most likely is the well-known organic mecha from the 1980’s, based on Yoshiyuki Tomino’s work and ideas.


Dunbine’s appearance is based on a humanoid insect. It has a largely curvy body with visible bone white claws. While its colouring isn’t anything out of the ordinary, considering the time, but one of the main points it has for it are the yellow insect wings it has on its back. All that gives is a distinct feeling from previous Tomino’s works, all of which largely used industrial cubic shapes.

Another 80’s mecha that is more or less organic in design is Iczer-Robo.


Iczer-Robo is a relatively early example of a bionic being, composed of both mechanical and biological components. It’s outer appearance has flowing smooth lines, but do carry certain industrial vibes. It is between the two, but inside it is very much organic. We even see Iczer Sigma’s birth in the series in a giant tank without any of its armour, basically saying that Iczer-1’s robot are not as much build as they are grown. In many ways, Evangelion’s concept of having an organic being in an armour restraints controlled by a human inside a cockpit surrounded by a liquid is nothing new, as Iczer-Robo did it first.

To directly contrast Dunbine, let’s go with the King of GMs, Ideon.


Ideon is such a strange design at first, especially when you consider it is formed of three separate units. At first, it’s not particularly pretty mecha to look at, but it grows on you. It’s follows the archetype of a blocky mecha as its body can be broken down to cubic geometrical shapes very easily. Drawing a very rough sketch with just boxes is very easy and good practice. It’s completely opposite to Dunbine’s shapes. Some years later, Makoto Kobayashi actually designed and built an organic Ideon model, and while that is more directly organic being, the contrast is striking.


The contrast between the two is striking, but both are the same core design, just in different style. Not only did it lose pretty much all of its boxiness, but also lost mass here and there. Some elements were changed to fit primordial god theme slightly more, which is evident of its slightly grotesque appearance. Plain red was replaced with broken, earth red instead with the occasional blob of light grey and bright red at joints.

Maybe one of the most famous industrial looking robot in sub-culture is the one that was designed to look like an American car.


Robocop‘s ED-209 is an exceptional example of industrial designed mecha. It’s form follows the function and nothing is really unnecessary. The joints look robust and strong, mechanical. It’s colour is largely that dull gray with a blue hue with black governing top of the main body, red are spared for wires and weapon bits. Steel grey is evident from bits that require to look like bare steel, and you have that yellow-black striping showing what parts to be wary of. It’s a hulking beast that doesn’t have softness to it despite having curves. They’re all cold and designed, rather than organically stemming from the body. It’s a terrific, iconic design.

Of course, there are a lot of things you can do between organic and mechanical design, not just in looks but how the mecha act and move. Zoids are largely designed to look like industrial machines, but their organic nature comes from them acting like animals instead of machines. Shield Liger for example moves like a real big cat and all the joints and the like are designed to accommodate this despite it’s overall industrial look. Just look this PV of MasterPiece-01 Shield Liger and how they made the model itself move.

I’m rather impressed on how they got the side parts to move like it was breathing or moving muscles around

As you can see, you can mix organic and mechanical together rather large degree for various kinds of effects. Just like with every other post in this mecha design series, the best way to look into this is to study actual existing examples. For organic, it’s the body structure and shapes of real life creatures that you could use to make a giant robot. For industrial design it’s much easier, as there are numerous books going over that topic.

Real world is a very good source for examples to learn from, instead of looking into existing mecha design. Modelling a mecha after something real and giving it a mechanical twist, but perhaps in an organic fashion, can lead to interesting and great designs. Or just nab a fighter jet plane and use its elements to make a whole tech three of giant robots.

Your giant robot’s controls

For the sake of simplicity, mecha controls can be divided into three category; control jokes, direct trace input and mental control. Often these are mixed and matched with each other to produce a more sensible and maybe even a more plausible way of controlling your giant robot, but in the end all of them are just as bullshit. The level of sophistication that’s gone in the design can always be appreciated, but in the end there’s nothing much else to it.

Joysticks or similar handles are likely the most common form of controls. You can find these from pretty much any Gundam outside G Gundam, in Muv-Luv’s TSFs, Macross and so on. These controls rely on the basic idea of any control jokes, and I’m sure most people have played some sort of flight simulator with a joystick to understand the basic functions. If not, get yourself two joystics, one for each hand, and boot up Descent in dual-stick control mode. That, or pick up Twin-Sticks for Virtual-On for some TSF gameplay.

Continue reading “Your giant robot’s controls”

Plane Elements in Tactical Surface Attackers; A-10 Thunderbolt II

The Fairchild Republic A-10 Thunderbolt II, also known by its nickname Warthog, is an iconic piece of aviation. It entered service in 1976, it is still the US Air Force’s primary low-altitude close air support aircraft. It was designed to counter enemy (i.e. Soviet) armoured units and artillery, nothing less and nothing more. Its core design was to allow it to fly low, take hits and litter the battlefield with bullets. Its high-lift wings have large control surfaces, making the A-10 very manoeuvrable during its flight. It also helps the A-10 has a short take-off and landing allows it to function near the frontlines in rougher environment. These wings also cover the craft’s engines from down below, adding an extra layer of protection. Its ease of control allows pilots to do night missions with just a pair of night-vision goggles.

These engines are General Electric TF34-GE-100 turbofans with 4 115kg of thrust. Turbofans were selected over conventional jet engines due to the fact that they gives off less heat, thus making them less vulnerable to heat-seeking weaponry. Their high position gives them an extra layer of protection from ground fire.

In addition, the A-10 was designed to be largely symmetrical. Many of the parts are interchangeable between sides, including the engines and main landing gear, making it easy to be operated from austere bases with limited facilities.

A-10’s primary weapon is the GAU-8 Avenger, a 30mm gatling gun that takes most of its internal space. Somebody once told me they designed a cannon awesome and big enough that they needed to bolt an aircraft around it. It is the largest cannon ever fitted to an aircraft, and uses both depleted uranium armour-piercing and high explosive incendiary rounds, firing either one 35 rounds per second. In addition, the A-10 can carry a large range of general bombs, cluster bombs, rockets and missiles, including the Maverick anti-armour and Sidewinder anti-aircraft missiles. It can carry up to respectable 7 264kg of additional weapons weight.


This BRRRRRRRRRRRT is a well spread meme

Essentially, the Thunderbolt II flies in, shoots the ever-living shit out of everything, makes the battlefield radioactive with depleted uranium and leaves metal wreckage in its wake, possibly with a distinct smell of napalm.

A-10 Thunderbolt II saw numerous little upgrades to it over the years, like the Pave Penny laser receiver pod in 1978, inertial navigation system in 1980 and the like. GPS systems were installed in 1999, and in 2005 the A-10 fleet began to receive the Precision Engagement upgrades to improve it to the new electronic warfare. This included better fire controlling system, electronic countermeasures, digital stores management, LITENING, Sniper advanced targeting pod integration SADL, VMF, GPS-guided weapons and upgraded to DC power among other upgrades. Now, the A-10 fleet carriers the A-10C designation.

The main difference between a Tactical Surface Fighter and Attacker is their role. Whereas TSFs are all about mobility and Hive infiltration, the TSAs are all about ranged combat with overwhelming fire power. They are, without a doubt, the shield to TSFs’ sword. The A-6 Intruder proved itself in beach landing operations, but due to the lack of Jump Units their role would always be limited. The answer to this was the A-10 Thunderbolt II, designed to litter the field with bullets like its real world counterpart.

The A-10 is essentially designed on F-4’s frame, and it shares its problems with increased bulk and weight. The A-10 balances these out adding even more armour (some optional!), superior Jump Units and sheer amount of fire power it carries. On its shoulders the A-10 carries two GAU-8 Avenger gatling guns in addition whatever weapon they can carry in their arms, like the WS-16 Assault Gun. It doesn’t have any Mount Pylons, but it needs none. In addition, the A-10 has Javelin CIDS Mk. 1 system installed all over its body. These are essentially explosive spikes that can be shot out in an explosive manner to get Tank-Class BETA off its surface.


The additional armour makes it look like it's hiding from its sempai. Just add blushing lines in there
The additional armour makes it look like it’s hiding from its sempai. Just add blushing lines in there

The role A-10 serves on the field is simple; massive crowd control. Its main role is to keep the smaller BETA strain under control, mainly the Tank-Class, while the TSFs can concentrate on the larger strains. The two comp each other, as A-10 is not terribly good against the larger strains due to its lack of mobility and melee weapons. Their main moment of glory was in late 1983, when Attack Squadron Pit Masters defended Hamburg from BETA invasion. Despite 50% losses, the invasion was halted, and the German civilians gave the craft an affectionate nickname Kanonenvogel after the Ju-87 Bomber.

Much like the real world version, the TSA A-10 gained some upgrades throughout the tears, mainly upgraded to use Operation by Light controls, newer and lighter armour plating as well as access to the Mk.57 Squad Support Gun and AMWS-21 Assault Gun. Much like with the real life version, the TSA has overall better performance after the upgrades.

The only officially coloured lineart of A-10 can be found in Muv-Luv Alternative Total Eclipse World Guidance’s upgraded tech tree in a very miniscule size. There’s a fan render out there that’s using wrong colours too. Technically, the one used here is a A-10C, but the two share the same lineart

This TSA has the least of lines out of any piece I’ve done thus far. The main reason for this is the same as with MiG-21; it adheres to the in-universe logic that F-4 was the starting point and directly ascending it or using its frame use its main form. The TSA A-10 follows more the idea of field littering support unit than the form of the craft. This is applied to the armouring as well. Even the Jump Units are unique in that they replicate only the latter part of the craft and one of the sides.

There is very little A-10 in A-10 in terms of clear visual cues. The additional armour gives it more curved surface resembling the aircraft, but outside that it’s very stripped down. I would have preferred to see a more direct adoption of sleek curves from the plane itself instead of opting to follow the F-4 TSFs. At least the legs could’ve used some elements from the landing gears.

Of silhouettes and robots

First, as a side note, I’ve put up a separate page that lists all mecha and robot related posts I’ve made. You can access it from the list of pages above the changing header image.

Silhouettes are important and overly visited point in character design. To go directly to the point, a silhouette needs to be uniquely recognizable. This has gone to the point that we all recognize a ball with two smaller balls on top of it side by side as Mickey Mouse’s head, and that silhouette cannot be replicated and sold. This applies to giant robots as well, and if you’re into robots, the following ones should look familiar.


It’s not hard to decipher the distinct look from the three above. Roundy, blocky and organicy. We recognize Mazinger Z because of its overall body shape, but it’s head and Breast Fire panels is what makes it stand out the most. For RX-72-2 we see the usual Gundam V-Fin, the shield and that rifle. Proportions, beam sabres and legs also give it away. A lot of Gundams share a very similar silhouette and people can make a mistake, but that’s one thing that makes them a Gundam. For EVA-01 the overall shape stands apart from the previous. Those shoulder pylons are a dead give away, as are the legs and the overall lanky pose. The horn is also another element that gives it away, even thou the overall head shape would be a better signifier, but this image hides it into the left shoulder pylon.

To hardcore robophilist, recognizing silhouettes across the genre is not too difficult. Some are head scratchers. To a person who is just glancing at these, RX-78-2 looks like a Transformer.


There are thousands of giant robot designs out there, Transformers hitting several thousands. All franchises with numerous designs and have run for years, like Gundam, most likely are hitting well over thousand. I’m just throwing these as guesstimates, but it illustrates a problem; not all designs can be completely unique from each other, and often within a series there is a pre-existing elements that dictate certain elements of a design that makes it instantly recognizable. For Gundam, it would be the face or the V-fin, and exceptions do exist. This is also why TSFs look so similar to most people, as they see the silhouettes better than the detailing. A Gundam has colours to make it look different because of their toyetic aesthetics, but a TSF is very mundane in colours in comparison, and due to many factors a lot of them share a similar silhouette by design.

An idea and purpose can dictate the look of the design just fine, but that’s just one initial approach. A method I’ve seen car designers use is start with a scribbled blob of non-descriptive nothing and see what’s in there for them.


Facing left, with a long horn swept towards back. Can you spot what looks like a Robot Bunny and an Orbital Frame to me? There’s also a skater, I think, and what I can only call a mechanical hummigbird

I’m not terribly good at this myself, but it does give some good ideas. A friend of mine showed me this some years back, and he can do some nice sketched renders. Do check his Twitter for neat stuff. Out from all those blobs, I lined out one that could be a neat starting point for a head design. This may seem stupidly easy and nonsense, but it does not negate the points of learning how mechanics works. This is a very useful method to test out shapes, and while I didn’t have no rhyme or reason to these, you can make sharper corners, more cubic or whatever tickles your fancy kind of shapes. Whatever suits your needs. Essentially, this is sidestepping the need to look for a shape, when you allow your subconscious to vomit out everything, and after that you just see what you have on paper. Of course, everything from this would need a large amount of detailing, but that’s later when you’ve locked down what way you want your piece to look in overall terms.

You can apply this to one part alone, or the whole damn thing you want to make. However, do keep in mind that this is just the very barest of starting points, as you’d still need to collect the shapes together into a cohesive whole and make them look right. That head design, if I were to create a whole linework just based on that, it would have swooped main curves with sharper angles to accent it. You can do as many shapes as you want, and often only a handful can give you some idea what you may want to go with. Much like everything else, you train this as you do it, and you can see I’m not the master of this approach due to preference of scribbling lines from whatever visual image I in my head. However, I do see this a more useful and easier way to approach of How do I get shapes?  I guess I’ll use Gundam as an example how to approach a design where there are set rules, thou you could just read the rules in TSF design posts for that.

Try this out if you’re in a block and can’t find the right shape. Sometimes what you need isn’t strict shape and form, but splattered scribble to give you a hand. Y’know, see the forest from the trees.

Mecha design; 10 points to consider

I’ll be aiming to do a mecha design post once per month. These are nothing major in their nature, as mecha design is really just really industrial design applied for fictional machines. I’ll be tagging all posts as mecha-design, and I’ll go back and tag the old ones as well.

This time I’d like to introduce ten factors that may affect your mecha design, or at least something you should consider about while doodling. Most of these posts will mostly touch on bipedal mechs, but non-humanoid designs should also consider the points in this post.

1; Silhouette size and lowering

Mechas tend to be rather sizable objects. In most cases they are few stories high and making their visible silhouette as small as possible is something you need consider about. Kneeling down often lowers height and silhouette size. Sometimes a transformation is done to lower the mech down and drop its frontal silhouette as much as possible. Lowering your mecha is also important when utilizing large weaponry.

In 08th MS Team we see a Gm Sniper kneeling down to stabilise the shot ans well as to minimize the effects of the recoil
In 08th MS Team we see a Gm Sniper kneeling down to stabilise the shot as well as to minimize the effects of the recoil

Minimizing the profile of your mecha is not too similar to tank warfare. Certain tanks can depress their cannons ten degrees, and these ten degrees allow them to climb a hill a little bit for further protection, minimizing their visible silhouette from enemy tanks other side of the hill. Having a weapon that can be shot around head height may be a good idea when it comes to shooting from cover. These can range from Guncannon’s shoulder cannons to TSF’s Type-87 Assault Cannon sitting on a pylon.

Speaking of size of your mecha, remember to put up some

2; Sensors

Giant robots are a good platform for all sorts of sensor clusters all around. Often these are not incorporated into the visuals of the mecha itself. For example, a 360-degree view requires multiple cameras and sensors to give that visual, for e.g. Gundam often have nothing else but their main camera, “eyes” and second camera in the back of their heads. It’s not too uncommon see a large camera cluster in a mech’s head, but rarely there’s anything that would resemble a sensor anywhere else. However, they are required to be there, and perhaps using certain kind of protective design for them can yield you relatively unique look. Of course, you can go more archaic and have a cockpit that doesn’t have a 360-degrees view. In case of cockpits with a glass dome, like in fighter jets, you may be able to go away with visible cameras altogether.

RX-78GP01-Fb's head doesn't have extra sensors outside the mentioned. You could have cameras on both sides just fine
RX-78GP01-Fb’s head doesn’t have extra sensors outside the mentioned. You could have cameras on both sides just fine, if you want to showcase sensors further

Having sensors also mean you need instruments in the cockpit to showcase them, from normal camera view to specialized views like IR. Mechas need a mix of instruments to show level of the horizon, energy levels and pressure levels and so on. Warfare units also require ammo count and damage charts visible alongside with numerous tactical views.

Speaking of cockpits, you need to think of

3; Cockpit placement

Where the cockpit is in your mecha changes its nature. Most popular places are in the middle of the chest and in the head. Chest area offers most protection as it is the centre of the mass while head gives smaller pin-point target and supposedly better view. Whatever the placement is, the cockpit needs to accommodate its pilot/s. Often you see cockpits that have a rather straight seat that reminds a fighter jet cockpit to an extent. Fighter cockpits are a good comparison point with mecha in general terms, but seeing how a mecha needs to be quick on its motions, the cockpit needs to have some sort of extra suspension to cushion the shocks. Be it sliding seat that dampens the trashing or suspend the whole cockpit somehow. Evangelion uses LCL to damped shocks and to protect the pilot, as well as give pure oxygen to the pilot. Life support system is important element a well, especially in space, and an emergency ejection system would be a nice thing to have, preferably with a powered armour of some sorts.

Of course, you can have your pilot suspended in a strange space without showing much. GaoGaiGar has a sort of direct interface system, where all Guy pretty much controls GaoGaiGar in 1:1 with those devices. Probably.
Of course, you can have your pilot suspended in a strange space without showing much. GaoGaiGar has a sort of direct interface system, where all Guy pretty much controls GaoGaiGar in 1:1 with those devices. Probably. We’ll talk about mecha control interfaces later

Speaking of shocks,

4; Joint reliance

Most mecha have basic metallic joints. Bandai has essentially engineered their designs to the point of replicating their functions in plastic. This is not all that impressing once you start reading on industrial designs and realise that you can design very intricate joints when you don’t need to actually give two shits about reality. Turn A Gundam has beautiful joints that are both well protected and function incredibly well.

However, in-universe you still need to give a reason why your mecha’s joints are not buckling and crackling under all the weight and strain. Having them sturdy material is one thing, something Gundam does almost every time. Another is to have biological component to it and design your mecha to be at least partially organic. Iczer-Robo is mostly an organic mecha, and thus its joints more or less look like pieces of armour. Underneath there is muscle and some sort of super strong skeletal structure underneath. EVA-units do this as well. You can also use artificial muscles that are made of complex composite materials, plastics and rubber to simulate functions of biological components while giving them better shock absorption. One example of this sort of artificial muscle structure can be found in TSFs.

TSF's joints do not conform to what normal human joints can do, and there's no reason to limit yourself in that manner. Mechanical joints can be more versatile when designed properly
TSF’s joints do not conform to what normal human joints can do, and there’s no reason to limit yourself in that manner. Mechanical joints can be more versatile when designed properly

Whatever you decide to do, remember that it all needs to have

5; Stable distribution

While the joints are there to keep your units standing and moving, one thing you need to consider in your design is how stable the design is. Mass is a bitch, and whatever design you have, it requires careful thinking how your mech will be able to stand up. Multiple legs are always an option, and e.g. Ligers from Zoids are very stable because of their four limbs and ability to shift their pose very widely. Bipedal mechs don’t really have this luxury, and this is why you need to consider how much mass can you pack, e.g. into backpacks of your units. If the centre of gravity is too far from the centre of you mecha, it needs to compensate it somehow, either leaning to an opposite direction, to have supports on the extended piece touch ground or opposite weights.

The sensor clusters come into play here as well, as those combined with automatic balancing system should keep the mecha straight without the pilot adjusting it manually. While many say that driving manual car is like piloting a mecha, driving an automatic is far closer analogue because a mecha requires large amounts of automated systems in order to have maximum efficiency. Our walking and running is mostly automated by subconscious, and automated systems streamline the operation to a similar level.

Aestivalis Ground Frame from Martian Successor Nadeshiko is a good example of good mass distribution. Note the large feet that are supposed to carry all the weight above and the additional drive system in them
Aestivalis Ground Frame from Martian Successor Nadeshiko is a good example of good mass distribution. Note the large feet that are supposed to carry all the weight above and the additional drive system in them

This applies in space scenarios as well, as a motion requires equal or higher countering motion to stop it. In Gundam you have AMBAC, or Active Mass Balance-Control. This system allows the Mobile Suits to shift their limbs and other points of mass to act out as intended. Similarly, your mecha may need some

6; Propulsion

Most mechas designed have some sort of propulsion system outside their limbs. Some have a secondary mode for wheeled drive, whereas others have thrusters to throw them around. Whereas AMBAC basically allows mecha to act in a zero-G, it can’t move unless something is pushing it forwards.

Attaching a variety of thrusters should allow your designed mecha to do some nice acrobatics. Larger thrusters allow jumps and flight, whereas smaller thrusters can be used to direct the unit better. For example, a small thruster on the front side of the left shoulder would push back at that point. With the help of other thrusters, it can do a faster turn or complete spin than what it would be able to do with just its basic joints. This effect is doubled in air and especially in space, where three-dimensional fighting requires additional abilities. Secondly, a propulsion system also allows your mecha to get on its feed faster and safer. Attitude control on any design is important, however it is realised.

Tallgeese has minor thrusters on its legs (just under the knee) and few bits elsewhere. The main thrusters are sitting on its back, hips and in the groin, moving at the centre of the mass more than at the extremeties
Tallgeese has minor thrusters on its legs (just under the knee) and few bits elsewhere. The main thrusters are sitting on its back, hips and in the groin, moving at the centre of the mass more than at the extremities

Positioning of your thrusters is important. To push the centre of the mass carefully requires thrusters in the main body of your mecha. Gundam W’s Leos are good examples of mechas with thrusters in their groins, as this is one of the best places to have a thruster to soften a landing.

Going overboard with the thrusters may be a bad idea, as your design still needs to get some

7; Power

Be it GN-Particles, G-Stone or any other form of bullshittium, no mecha will move without proper explanation how it gets its power. As a mechanical design at giant robot scale, limbs are very inefficient when it comes to power consumption. Whatever power source you have for them, it requires to be strong in order to move them at a reasonable rate. This can be a crux in your design overall, like it sometimes is in Gundam. Minovsky Particles allow large production of power that can be used in many ways and has some side effects. GN-Particles effectively are magic pixie dust that can be used to power things up as well as create anti-gravity.

Minovsky Ultracompact Reactor is the cornerstone that allowed Universal Century's technological evolution simply by giving enough power to do giant walking robots
Minovsky Ultracompact Reactor is the cornerstone that allowed Universal Century’s technological evolution simply by giving enough power to do giant walking robots

You also need to consider why these things are used just for your mecha. Are they hard to produce, do they require certain size that hasn’t been miniaturised, is it an alien tech that normal people don’t have their hands on or is it just power of the soul? Whatever it is, consider how well such energy source could be used in more conventional vehicles, or rather, how would a conventional vehicle act with such a source and where it would be located. Don’t forget about the lubrication and other fuel for thrusters and such, if needed.

All this of course needs to have

8; Protection

Outside superweapons, mechas are large targets. Having a mixed amount of protective systems in your design is a good idea. These range from such simple things as wielding a shield to anti-personal weaponry to active anti-missile targeting systems. There are designs that are naked in this sense, but often they have sturdy armour to compensate, are fast enough to dodge things or have some sort of beam shields surrounding them. Depending on the role of the mecha you’ve designed, you might want to give their design some level of visible protection, even if it ends up being active layer that blows outwards.

These elements can be also made into weaponry or assist in other ways. TSFs’ Type-92 Multipurpose Supplemental Armour, i.e. shields, have a top part that can turn 90-degrees and contains hexagonal reactive armour plating, which can be punched into a BETA’s face and explode it. Sometimes shield have serrated edges to cut things, or house missiles or beam sabres. Spaced armour can be another option.

Frame Arms line of models has this pretty in there. Cannons are set high, and both arms are mostly for shielding. Legs have extra plating as well, and those plates drop down to add more support. The shields have notches in them where the cannons can rest for more stability, and the slightly smaller top allow the head to see through them when pushed together. note that the chest is also angled to direct incoming projectiles away from the cockpit
Frame Arms line of models has this pretty in there. Cannons are set high, and both arms are mostly for shielding. Legs have extra plating as well, and those plates drop down to add more support. The shields have notches in them where the cannons can rest for more stability, and the slightly smaller top allow the head to see through them when pushed together. note that the chest is also angled to direct incoming projectiles away from the cockpit

Whatever protection you have, you also need to consider

9; Maintenance

This isn’t a huge concern in fiction, unless you are aiming to some level of realism. Having the most complex design may not be the best of idea, as the more complex something becomes, the harder it is to maintain.

Consider old cars. They are rather straight in their approach how they can be fixed, there’s not much high technology in their engines or other systems. They are rather simple things to drive. Modern cars on the other hand have a large amount of tech thrown into them that a normal street walking mook can’t even lift the engine cover off anymore.

The same applies to mecha. The more complex systems, the more time and effort it will take to maintain it. Shape may not necessarily make the maintenance harder, but production of spare parts and the like may be affected. Thus, considering in-universe how certain elements are used and developed may be necessary. Armour panel lining may also showcase maintenance access hatches and the like, which you may have in your design. It’s been a fashion for some time just to fill the surface with all sorts of lines and have them lit up, even thou there’s absolutely no goddamn reason to have them.

Fictional schematics are mostly bullshit, but you might want to consider what goes under the surface at some point. It may make a good design point and even better storypoint.
Fictional schematics are mostly bullshit, but you might want to consider what goes under the surface at some point. It may make a good design point and even better plot point. Notice the landing gear

Maintenance of course is easier if the mecha has a well-defined

10; Role

Have a clear role in mind for the mecha. To use a real world example, the F-35 Lightning II was to be a multi-purpose fighter, but it really sucks in every field. It can’t turn well enough, it can’t climb, it can’t run away, it’s special shape and coating doesn’t make it all that stealthy, it’s heavy as hell, its thrust-to-weight ratio is lower because of this and the 20 tons of thrust puts an extreme stress on the engine components. Its fundamental design flaws keeps it being better than last generation of fighters. I love the TSF design, but the real fighter is slightly too fat for my taste.

The same applies to mecha design. Having too many elements to cover on one design will make it a clusterfuck and an eyesore. A transformation elements may give it an edge, but only if the transformation is smooth and well thought out, and we’re not going to touch transforming mecha designs anytime soon, because people have hang me from my balls when they hear me saying how Macross has essentially milked the exact same transformation scheme for thirty years now with slight changes here and there.

Fast mechas tend to have aerodynamic shape, supporting mechas have big guns and defensive ones are fat in armour. It’s like basic rock-paper-scissors. Role should be your starting point with the basic idea what you want, because all design ultimately stems from a need, to find something that fulfils a needed niche.

This is something that needs to be emphasized; a good mecha is design starts from an idea of something. A character like robot, a hero, a villain, the sniper or the like. These starting points give you a direction you to go, and when you have its role clear, then you can start thinking of the details.

Ialdabaoth is a close combat martial arts mecha. To define role, you most likely need to apply something like one of the three approaches in mecha design I tend to employ
Ialdabaoth is a close combat martial arts mecha. To define role, you most likely need to apply something like one of the three approaches in mecha design I tend to employ

Review of the Month; design progression in Metal Gear franchise

[Update, 25.08.2015] This post has been awarded a Review of the Month status. It’s original title was Metal Gear; where old designs are more advanced than new ones

With Metal Gear Solid V coming to our way sometime in few weeks, I’m taking a topic from the backburner that I haven’t been doing even if I have planned this for some time. Let’s give a look at the designs and the design progression Metal Gear franchise has.

Let’s straight about this; the design flow in Metal Gear is screwed up. This is due to two things; Kojima’s own lack of desire to follow his established continuity and Shinkawa’s REX becoming a fetish that permeated the whole franchise, essentially making it stagnate to one design and its variants.

If you were to play the franchise’s main series in production order, the design progression would be valid and make sense. However, the moment you take into notion the canon order of events, things just sink into a black hole. This is due to apparent weird technological progression, where the 70’s and 80’s have better technology than in the 90’s. In-universe, that is.

An argument that has been thrown around to explain this weird technological schizophrenic schism is that all this tech has been for black ops, and thus never seen in the daylight. This is more or less bullshit, as black ops technology has always seeped into real world, especially those that have made world a much easier place to live or have made incredible leaps in sciences. Like creating wormholes that are safe to travel for a human being without extra gear, or perfect holographic projectors, or batteries that can power aforementioned piece for immense amounts of time. These examples would not just vanish as they are beneficial not only to the people, but to governments everywhere. An example of this can be seen in-universe as well, with cyborgs becoming an everyday thing around MGS4 and further in MG Rising.

Speaking of cyborgs, it’s comedy gold in retrospect that Snake’s Revenge got panned by Solid fans because it turned Big Boss into a cyborg, a thing Kojima just went with Gray Fox anyway and other characters in the series. An example of duality, if nothing else.

In order to properly showcase how screwed the whole design deal is, we’ll use the production order of the games rather than the canon order of the story.

TX-55 Metal Gear, the first one until retcons
TX-55 Metal Gear, the first one until retcons

As Kojima based Metal Gear on western movies, so does the first Metal Gear reflect the American sensibilities when it comes to mecha. Without a doubt there’s a strong hint of Ed-209 from Robocop, just with less gasoline guzzling elements. As it was designed by the Japanese who had more or less no technical knowledge, the first Metal Gear looks like it could topple over if somebody shoved it. Top heaviness is nothing new to Japanese designs thou, and the more industrial look they were going with doesn’t make it more realistic.

Some do argue that Metal Gears are one of the more realistic mechas out there, but the best you can do to this is to laugh. This was the first of its kind, and everything that would come down the line would base themselves on it. Things that would carry over would be the placement of the main weapon on its right shoulder, the basic tri-pointed feet design where you have one toe back, two forwards, additional weaponry near the groin and armless design.

Metal Gear D, a solid evolution
Metal Gear D, a solid evolution

Metal Gear D is a perfect successor in terms of visuals. It takes what was iconic to the first one and gives it a wholly new flavour. It’s not a unique departure, but that’s fine for a direct sequel. Similar bi-pedal positioning, weapon placement is roughly the same and the overall silhouette is similar enough to be recognized to be in the same line. It’s a more refined, a more detailed design that, most importantly, gives an idea of progression.

Metal Gear D does have a stronger Japanese flavour to it, but it’s a bit more downplayed quite a lot thanks to the attention to the realism the wanted to have with it. It still looks top heavy, and the scrawny chicken legs don’t seem too trustworthy.

There was a mass-produced model in Metal Gear 2, Metal Gear G, but it never appeared in-game so we’re going to skip that.

Metal Gear Fetish Fuel
Metal Gear Fetish Fuel

Metal Gear REX was designed around the limitations of the PlayStation. It’s far blockier than either of the previous ones because of this, with more sleeker flats all around. While it shares bi-pedal similarities with the two previous Metal Gears, REX is a departure. The familiar silhouette is gone and replaced with a more bestial visage. This is not necessarily a bad thing at all, and the design does carry certain recognizable elements. REX is a departure from some of the ideas of how Metal Gear should look like, and shows that it doesn’t need to look the same as long as its gets the job done. Which is funny, as no Metal Gear has ever done their intended job properly.

Unlike the predecessors, REX’s body is more balanced. The large, wide and low key legs carry the two-segment main body in a far more ideal fashion. The feet change the toes a little bit, where it has two spikes in the front to give traction to said legs. This can be regarded as a solution to how the predecessors were a more rigid in design. Indeed, the way REX has been animated in the games makes it act and function like an animal to the extent on having it open and close its ‘mouth’ for roaring. It’s stupid really, and a sign how certain sentimentalities from Japanese pop-culture had already seeped in. Otacon’s line about things looking like his Japanese animes does apply the rest of the designs in the franchise.

Perhaps it should be mentioned that Metal Gear Solid was the first Metal Gear game many people played, not either version of Metal Gear.

Metal Gear Ray, a further departure

Metal Gear RAY’s design was to reflects that what a Metal Gear is became a fluid thing in Kojima’s head. The thing Metal Gear was supposed to be was a bi-pedal tank with nuclear capabilities, but authors can’t twist and give excuses to name anything however they want. As such, RAY lacks all of the established qualities it takes to be a Metal Gear, a thing that you can over analyse to fit Kojima’s intention for the narrative Metal Gear Solid 2 carries.

RAY’s organic and sleek nature reflects it amphibious nature, and is an antithesis to REX’s blocky and sleek design, further playing into its whole anti-Metal Gear nature. It’s a bit over-designed, with the whole four-jointed legs and circles everywhere on its red-brown parts. It has much more relation to Zone of the Enders in this sense, and it wouldn’t take much to convert RAY into an Orbital Frame. The organic elements still had their function, as RAY had more nanotech in form of nanopaste that would fix its artificial muscles it used. There is a mass-produced model for the US Navy with complete reddish brown colour, dull point knees and far smaller tail. Otherwise the overall design is the same.

There’s Arsenal Gear too, which has nuclear capabilities, but it’s a goddamn underwater battleship and not a bi-ped. It doesn’t play in the frame what a Metal Gear is, however author intent tries to force the idea of changing warfare in with it. However, we will touch upon the Shagohod, because its design does play in how Metal Gears and their approximate mechanical designs formed.

Not really two legged, but a step towards it
Not really two legged, but a step towards it

The Shagohod, in many sense, fits as the first step towards what would become the basics of a Metal Gear. While Big Boss MGS games are basically just big retcons, the Shagohod’s existence doesn’t break anything in the design line as such. It’s a bit more traditional tank than a Metal Gear despite it being a screw propelled. Screw-propelled tanks were a thing, but overall their usage is relatively limited, but they can make decent amphibious vehicles that should be able to traverse difficult terrain.

Despite Shagohod not being a Metal Gear, it the same base elements that the first one established, as well as showing the first signs of REX fetishism. You can find the nuclear launcher on the right shoulder as usual, secondary weapons between legs, or screws in this case, with the radar dome placement on the left, which is a carryover from REX’s design. That’s not the only thing that Shagohod carries over REX thou. The two-segment design makes a return, and for whatever reason Shagohod can sort of stand up so that the tip of the screws are digging in to the ground , raising the level of the first segment. While the overall design looks neat, you are able to see how the first segment also resembles REX’s head segment.

Whether or not the Shagohod’s design represents the mid-1960’s is a good question. The angularity the first segment represents wouldn’t be widely seen until T-55 in real world tanks, but the second segment is almost too bulbous and shapely to fit any tank overall. Outside these the Shagohod almost could fit the era very well, if we take these very loose definitions and terms of general military science fiction into consideration. I would almost say that the Shagohod is almost the most sensible design that has come from the Metal Gear franchise.

Screenshots because RAXA lacks any properly well done renderitions,
Screenshots because RAXA lacks any properly well done render

With Metal Gear jumping to PSP, a mainline game would come down the line sooner or later. Like it or not MGS Portable Ops is a valid entry in the series. Metal Gear RAXA was a prototype for Intercontinental Ballistic Metal Gear, which we never see in action in the game itself, but we can assume the two look essentially the same. You would think RAXA would be free of influence from REX, but if you look close enough towards its middle section, you see the same T-shape head from REX repeated. At this point one thing is clear; the design progression in Metal Gear is not valid in its own universe. It only makes sense on a meta-level like this.

RAXA is once again a weird thing for a Metal Gear. The ICBMG is capable of nuclear strike for sure, but both of them move on four legs. Or rather, float in the air and fly. It’s better just to ignore what a Metal Gear is supposed to be and just go whatever is named as one. Then again, we could also ask why were all the powers so keen in developing a Metal Gear when every single one had failed before. Anyway, there’s not much to be said about RAXA, outside that outside its wings, it does follow a sensible continuation from Shagahod, outside the whole REX elements. There’s few bits that are a bit too advanced for its time even, but this being SF story, we’ll let that slide.

Moonlight walker
Moonlight walker

The Gekkou isn’t a Metal Gear, but as it follows the same design continuity as REX and RAY, we’ll handle it relatively shortly. The idea for Gekkous was to take REX’s head and modify it with RAY’s leg parts. MGS2 was the moment where Metal Gear as a franchise took itself towards more 2000’s science fiction with nanotechnology and semi-biological components as opposed to hard hardware machinery based ideas, and Gekkou juxtaposes the two into one being.

While there are few variations of them, like suicide Gekkou, the above example is as simple as it gets. It still carries some elements standard elements, like the radar dome on its right shoulder, thou it is far more in the centre than previously. There is also a sensor unit between its legs. Speaking of legs, there’s like six joints in there, which Kojima put in relatively heavy use on how the Gekkou’s move. Three toes also make a return here, just in different order than usual. Unlike the Shagodog, the elements from REX and RAY are very prominent, but that can be coined to normal technological progression. However, even then something more unique could’ve come from Gekkou’s than recycle and mash REX and RAY together like this. There is an idea in there, thou it’s up to individuals if Gekkou’s should’ve been more distinctively unique.

Metal Gear MkII and MkIII will be omitted because they can be summed as Metal Gear’s head as body, screen as launcher.

From left to right; Pupa, Chrysalis, Cocoon and Peace Walker
From left to right; Pupa, Chrysalis, Cocoon and Peace Walker’s two modes. Not to scale

While the AI Weapons are not counted as a Metal Gear, they are a design step. However, they all run on a relatively sophisticated Artificial intelligence, breaking whatever illusion you had about the design and technological continuity the series had. Still, they carry certain elements from past designs that make more or less sense, outside Cocoon. Cocoon’s just a big damn dumb tank with absolutely no redeeming value in itself. Its design is messy, overblown and something a five year old would like to see. It’s all around awful and they should’ve disregard this from the get go. Pupa, however, is basically a jacked up Shagohod. Now that makes sense, as it’s more mass produced version with a bit more streamlined with tracks rather than screw-drives. Extra weapons and the stupid boosters fir the overall design just fine.

Chrysalis can be seen as an evolution of RAXA, but it makes little sense how it flies. It also carries a rail gun, something that was supposed to be completely new and ground breaking for REX. At least this time they went away with the REX elements with both of these, resorting to a far more basic industrial look. It works and sets itself apart among the AI Weapons, and not just because it can fly. It may be a small thing, but switching the rail gun’s and radar’s placements is a change that gives it slightly more characters. Well, as long as you disregard that both of them are lifted almost directly from REX.

And oh, they all have Hatsune Miku singing as their voice. I don’t have anything against Miku, but vocaloid has no place in Metal Gear. Then again, the franchise has gone more or less full anime since MGS, so it’s easy to explain. Plus Miku was at the height of popularity at the time, and it seems Peace Walker was developed to cater to Kojima’s son. Metal Gear had gone from being a view of western action movies to a view of western action movies filtered heavily through Japanese pop-culture.

Peace Walker on the other hand is just weird. Carrying what essentially is Shagahod’s launcher on four legs that are resemble strongly armourless REX legs. The other box is a more traditional missile launcher. Initially it was supposed to look like REX and have a similar hangar, but that was reserved for ZEKE. Peace Walker is a strange mixed of bags. On one hand it’s pretty neat, but on  the other hand it’s absolutely retarded. The big ball there is basically its head and the AI pod seen on other AI Weapons reside next to it. The overall design is a lanky transformer, as the four-legged mode had to act like some dog or similar mammal, whereas the two legged mode acts like a theropod dinosaur, again making a connection to REX.

Oh lord...
Oh lord…

REX’s elements had been popping up everywhere after its initial appearance in MGS. Metal Gear ZEKE is basically where Kojima tells that games before that doesn’t matter, as ZEKE’s design was to look like on of the future Metal Gears, but older and rougher on the edges. ZEKE ended up being essentially a stripped REX. The design differences in details are plenty, but the silhouette could be mistaken by an unrehearsed eye. Weaponry placements, radar, the flat head minus the leg of the normal T-shape, legs, they’re all from REX. In meta this makes sense, but in continuity it makes zero sense. This Metal Gear we have here is essentially at the same level as REX in overall technological status, with more or less functioning AI and a goddamn rail gun. While the player has to destroy AI weapons over and over again in order to produce ZEKE. As such, some parts can be switched around to make it look less like REX, but the basic frame will always be the same. While that basic frame is balanced, it should just tip over because of that rail gun.

amiami wonfest
Metal Gear REX 3.0. Or 4.0, depends how you want to count it


With The Phantom Pain being released soonish, I’m not going to pretend this variant of REX doesn’t exist. This isn’t even funny, just sad. Much like ZEKE before it, this Metal Gear ST-84 is technologically more advanced that its two successors will be. It’s even got more technical design than REX had with vents on the sides of its head and relatively smaller silhouette, thou the changes make it more frontal heavy. Unlike REX, the ST-84 is unbalanced, but it will be animated to look like an animal again. Perhaps those two elements on either side of the frontal section are arms, so it can be four-legged again. Then again, perhaps those are arms for its humanoid mode. I’ll be leaving that image linked rather than outright posted if this turns out to be true, but I have my doubts. Why would you have a upright standing Metal Gear? The sheer amount of technology to make it more feasible without artificial muscles as with RAY and Gekkou is immense. It’s a much larger target as it is now with incredibly high profile and useless scraps around its body. If Metal Gear as a franchise tried to take itself seriously before, The Phantom Pain has made itself very schizophrenic with the harsh themes it goes for and… a goddamn humanoid Metal Gear.

With that, the mainline Metal Gear games come to a full circle, as Metal Gear V leads into Metal Gear. I’ve been saying since Metal Gear Solid 2 that there needs to be a remake of Metal Gear 1 & 2 because of how the narrative has changed, new events and technologies have been introduced that do not mesh well with the previously established continuity. Big Boss’ games are essentially big retcons in themselves, and despite MGS3 and MGSPW being pretty decent games, one can’t ignore how laughably bad the continuity is between the older and newer titles. It would have taken a lot from Kojima to stay restricted with what he had done in the past, but as we’ve seen along the years, he is a man who does whatever hell wants without caring one bit how that affects anything else, even if it is a franchise he became famous for.

There are Metal Gears in the franchise that use elements from all the ones seen above. Some are more unique, some are just convoluted. We’ll most likely return to those some other day, starting with Metal Gear 2 from Snake’s Revenge.