This page used to be up on Prometheus' Spawning Grounds, which was my old website. For some reason I couldn't be arsed to clean up this page and upload it to AndroidArts, until now. I think I felt that the project was in limbo, directionless. What could ever become of it? TA Spring material? A project of my own? It just feels very on the fence.
It's now 2015 and I think a freemium game developer has gotten a hold of the rights. I suppose that's where the money is nowadays, and old IPs will end up. However, rumour has it the original TA author might be onboard, so that's nice. I think they are doing the same thing with MoO, involving some of the original authors. Sounds good?
Well, I've noticed that... when a thing has been released into the wild, it takes on an identity of its own in the minds of the fans. The intents and visions of the original author are sort of irrelevant. Sometimes an author returns to an old creation to fix it or prequel it... and it doesn't quite work. For example, Command & Conquer's sprites (which is what the player looks at a lot) have a quite different feel (in my mind) than the more realistic build pics and concept art for example. I never quite liked where CnC went (realistic-goofyness?) with very fragmented/noisy teamcolor patches.
On the other hand, sometimes fans are involved, and we're left with a "great ideas" monstrosity which doesn't work in practice (MoO3?). There's also the occasional developer who are "great fans of the original game" but then proceed to shit all over it with a sequel. Sometimes I think... whether a game works or not is mostly down to luck.
Where were I? I was supposed to be doing something... Ah! Yes, fixing this page up. Total Annihilation is a nice game, and I would still play it if I hadn't sat on my skirmish CD. I've toyed around with TA Spring but never actually installed it on a machine with the power to run it. Haven't played SupCom or PA, but sometimes watch games.
Design wise, I think TA is quite fragmented when it comes to the overall look of the factions, but it's quite understandable given when and how the units were made. I started working on this page in 2005, made some more attempts in 2007, 2009, and 2012. My drive was to make the factions more coherent, but I think my earlier attempts were pretty awful (see the unit sheets below).
Consistency is important to me, so I've established certain guidelines to keep in mind when I draw the units. Also, units can feature certain components which are shared... hatches, weapons, heads, knobs. Aside from giving a coherent look, it also makes the relative scales easier to read. It's a bit like putting a person/coin/matchbox next to something.
|Color preference:||Red, Silver, Black.||Blue, Gun Metal, Biege.|
|Shape preference:||Curved, Symmetry.||Angular, Asymmetry.|
|Weapon preference:||Lasers, Energy.||EMG, Cannons, Mass.|
I think TA did a pretty okay job with team colors. It had decent hues and only allowed the primary Red, Green, Blue and Yellow when there were less than 4 players, preventing colour similarity confusion. In a 2v2v2 or 4v4, similar colors could perhaps be used within the teams.
A lot of RTS games choose color hues which I don't particularly fancy, and to make matters worse, they often color the units using a cheap overlay technique which makes the colors look muddy (they just get darker). With colors like yellow you need to slide the darker shades towards orange a bit to retain vibrancy. TA solved this by having handmade team texture blocks with proper hue shifts in the shading.
However, when trying out the colors on my concepts, I noticed that they clashed a bit with the base colors I had used (gray, biege, etc.). Perhaps there's a need for desaturated unit colors (with more saturation on minimap). Also, dark colors look quite black from a distance so I had to lighten those a bit. These are the colors I could make work:
I'm too old for micromanagement games like StarCraft, and I've always been more of a spectator player, setting things up, adjusting strategies. For some reason RTS games (even macromanagement ones) still insist that I construct my base in realtime. Maybe there are other ways to approach that? Planning out a base (perhaps using block stamps), setting priorities for construction units? High-low priority, and luxury priority (build if rich), hold.
As much as I favour macromanagement and global strategy, I'm slightly annoyed when games like Homeworld and Planetary Annihilation turn into dot-wars. I'd rather see an effective minimap and fixed camera view solution (original TA in hirez mode feels about right).
Of course, for a macromanagement game to work, there has to be choices, especially early ones. Perhaps the player can upgrade factories to mass production certain units more feasibly. There could be a limited amount of units to upload in the commander's "gene" database, preventing the endgame from being "same choice of any/all units". I'm thinking each unit takes up X bytes of memory, and there's a budget of how much data a commander can bring to the field. The player can create several commanders (named, with profiles). A new commander gets the default units, but the player can add and subtract units. To make things more spicy, unit data can be compressed, but then the commander has to spend game time uncompressing to "unlock". I like the idea of a game session being themed and often impose restrictions on myself, building only certain types of units.
Imagine a situation where a player wants to bring a Krogoth, and uncompressed so he can attack earlier. However, now he has much less space for regular units and won't be able to counter all enemy strategies effectively. Another player brings many types of tanks, preparing for a mid game land battle, but her budget only allows for a general purpose navy ship and sea plane.
I like asymmetrical map layouts, resource distributions and factions. I'm not much for games designed to be played competitively. So many interesting concepts become available once you let go of symmetry and fairness.
It's interesting to be forced to adapt to new circumstances. Perhaps one commander starts in the desert, using harvesters to collect metal instead of building mexes. Another commander starts on a volcano and gets tons of geothermal and a good position, and the third commander is hidden somewhere in the ocean in a ship graveyard. Perhaps you know none of this in advance.
I think air and long range artillery often ruins the strategy aspect of RTS games. Being able to quickly deal damage anywhere... it's a bit like playing whack-a-mole on a featureless plane. A strategy game should be about more than that. IIRC, TA added a deadly flak turret and artillery walls early on in a response to certain strategies. Of course, a game should not be the same through all of the tiers with only unit size escalating, but I think stuff like bombers and nukes degrades the game just to bring it to a randomly determined close.
Utopia for the Amiga is one of my favourite RTS-SimCity games, but unfortunately many of the mechanics from the latter are ill suited for TA, which of course has no people. SupCom did something interesting with building adjacency bonuses though, and as a turtle, I'd sponsor any idea which makes for more mindful base building.
Something which would increase readability is to have certain fixed types of weapons which can be recognised visually. Games Workshop does this well with their Lasguns, Meltaguns, etc. Here I've tried to support this approach by color (and size) coding the lasers and some other stuff. For example, a small red laser would always have the same range, damage and maybe reload time. If you see three green lasers mounted in a gatlin formation, you know that you're dealing with at just a glance [a HLT :)].
Lasers would be accurate things which fly straight and don't do much structural damage. They just stop things from working by destroying internal components, like a malicious surgeon. Like so, they'd be bad at destroying wrecks. Mortars could always be inaccurate, but lightweight since they don't have a long barrel, etc. Basically, all the weapon types could have certain characteristics in which you could trust, and they'd be visually recognisable, and come in certain grades which tell you something about their range and damage.
Actually, I don't quite like what I've got here so far, art-wise. Some of the detailing is haphazard and non-figurative. However, I do think it's a good idea to design RTS units from the top. They should be like symbols or icons gliding around. The placement of colour masses is more important than details when it comes to readability (something which is essential in... well, most games, I think).
These re-designs are pretty liberal. I could probably stay a lot closer to the originals (and I do on the more recent sheets). As for modelling and texturing (if that's ever going to happen), I think the texture should do more work than the model. To exaggerate, units could almost be a flat 64*64px pictures gliding around and people would never notice unless they move down with the camera to the ground for eye candy shots. Again, I think readability is far more important than detail noise smacked over a 1000 poly model with no distinct features. I'm a failure as a modeller, but I think maybe 100-300 polies would be enough for most things. More polies just means more work to make those damn UV maps, and less units done. Maybe I'll draw some orthos one day.
2015 note: This art was made from a topdown perspective and mostly consists of random lines and boxes with fancy rendering. I think it's despicable, looking at it now. Good lord, this page is image heavy, I'll leave these as links.
Arm PeeWee launcher - 2015 note: I remember doing this before SupCom came out ('05), posting it on their dev forums, and SupCom did have a unit launcher. Who knows what happened...
Side and front views WIP - It's troublesome to try and design stuff seen from ground level when it's going to be viewed from the top ingame. Perspective drawings (35-10 degrees top) would be better, but it's still fun to do side view line-ups every now and then. It's one of those things I enjoyed looking at as a kid (and I'm still a kid in that sense). The car and tank are human scale. One thing that's relevant to the TA designs (CORE especially) is that they don't need to house humans, so there's no hatches, doors, windows, etc. Interior tank space would be used for other things, and 'spalling' damage wouldn't be an issue. Aircraft would be able to pull more than 10g, etc. Actually, the ARM might still use pilots, but they're probably superhuman and use neural interfaces to become one with their vehicles.
A "PiratePoo" made this 3D model from a FlashTank doodle I had. Unsure about poly count, but maybe this is the right magnetude.
2015 note: Isometrical doodles from '09-'10. Experimented with PeeWees carrying weapons like many japanese mechas do.
Factories could have these little nano-spray mandibles, busy at work.
2015 note: Informing the player about the game state, past, present and future trajectory.
2015 note: Morty Kbot, somewhat faithful to the original unit.
Now some concepts made in 2012 and 2015. These were drawn in perspective because it's more descriptive and effective as concept art. Overall a lot more faithful than the 2005-2007 stuff. I change minor details but keep colour placement and general shapes. I do smaller ankles though. It makes no sense to put a lot of mass at the end of limbs and I haven't liked the look of Megaman feet since I was 14. My forms are chunky and doesn't have much of realistic greeble, but at the same time I don't want to get too close to cartony-goofy. Core and Arm I imagine builds chunky utilitarian stuff, simple panels. There's no need to strap stuff onto the hulls like tow cables, spare fuel tanks or shovels.
I don't like "Unit Veterancy" much. It doesn't make sense in TA where the units are based on cloned/copies veterans already. However, it would be cute if units with many kills get a name so you can follow them around.
Core Vehicles. These are often done with a 0.5 mechanical pencil, scanned, then scaled around in Photoshop. The colored stuff has gotten a second digital line art iteration/pass. I try to not go into detail early, because I want to have the freedom to explore the larger shapes easily. These units will be seen from quite a distance, communicating their identity. The Krogoth, if made much larger, needs to be more... like a pillar I think... rigid? It's superheavy and not just a scaled up KBot.
With these paraphrases (from Z and Godzilla) I'm not trying to make straight copies. Instead I'm borrowing the general shapes. There could be several Krogoth counter units (all armed with masers to take down Krogzilla of course), but with a limited blueprint space the Arm commander might have to choose which counters, if any, to bring.
More paraphrases. Dune II was my first RTS. Yessir-Acknowledged-Yessir-Acknowledged. Maybe the Arm and Core stole designs from whatever species got in their way during the 4000 year war?
More paraphrases. I was never a Command & Conquer fan. The game seemed... stale somehow, and the little crawling men who wouldn't die, or died in droves. Some sort of RPS abstraction there which I didn't care for. Looking at Red Alert, I sort of like the sprites though, with the solid colour tanks. The actual concepts were more towards plain trucks and tanks. Later the CnC style migrated towards goofy-cool with greebles on top. The problem with big guns and swords is that they feel fake, swinging around like frigolite, doing trickle damage.
I'll take this opportunity to talk about what I feel is an appropriate design philosophy for this project in general.
>>> I keep most of the gun barrels on the thin side, like in TA (Bulldog, Reaper). This gives the units a more realistic feel even though I don't go full realism in terms of caliber.
>>> I try to design from a distance, mostly avoiding hull greeble (which would be noisy on a small sprite, 3D model or 6mm scale miniature). Besides, I think greebling is a sign of design failure, a bit like padding sentences with um... like, um... eh... like, I mean, it's like space is, um, being wasted that could... um... be used to express something valuable instead. I once heard an advice for giving speeches. if you feel an "um..." or somesuch coming, it's better to make a silent pause than to annoy the audience.
2015 note: Design notes and chibification. Arm uses female jockeys (clones of skilled pilots, iirc).
The Arm and Core have decimated a million worlds, we're told. It's clear that there's no real other threat. Some of the planets they ravaged might have been inhabited though, so I think it would be interesting to show this in a matter-of-fact way. Natives would be little more than decoration, mobile resource lumps. They are not announced, and no judgement is passed on the player in how he treats them. A small army of 1960's unit might be a threat to a lone AK, but that's as far as it goes. If a small village gets in the way of a construction unit on patrol, it might simply be absorbed without fanfare.
The exceptions are the Con and the Ext in my story. The Con are based on Core/Arm technology and have less of a technological disadvantage. The Ext is a "hard mode" race which had some time to develop technological prowess during the Core/Arm downtime, but it's not much of a threat.
Needs to be storyboarded [bracketed text]. Everything subject to change. As for compability with the Core Contingency ending. Maybe the Core 'imploded' the galaxy and a strange effect took them to a paralell universe. Or the bomb didn't work. I never played CC so I don't know much about it. Also, my writing is made of cheese.
[A metal floor.]
[Shadow of a flying bug.]
[It lands on a plant sticking out of a crack.]
[Another bug takes off.]
[Camera pulls back to reveal a carpet of lave-grass.]
Over 6000 years have passed since the Arm and the Core assured their mutual destruction.
[Overgrown ruins, partial grass on Core Prime]
Or, so it had seemed.
It would come to pass that the Ext, race of alien crustaceans,
[Giant green crustaceans tinkering with their ship controls and sensors]
[Sensor style jerky zoom in on a dot becoming the commander wreck]
finds the remnants of a Core commander adrift in the bleak vastness of space.
[Ext ship closing in]
The Ext's attempt to restore the commander to use it as a weapon against their enemies would prove catastrophic.
[They're fixing it up, keeping it restrained in a contraption. Actually, it's probably reparing itself.]
[The eye starts to glow fiercly]
(It needs to be better established that the Con is the enemy of the Ext)
The Commander broke free from its prison to once again live the only way it could.
[Commander D-guns his way out through the hull,]
[...leaving the ship to explode/burn as a neat backdrop.]
[A nearby planet (or moon, or dwarf planet...)]
[The commander burns (safely) through the atmosphere, crashing on the surface.]
[The commander gets out of the sizzling crater to set its eyes upon an Ext or Con city]
[Mid story intermission - black out ]
[A small fleet of Core ships leaves (and/or prepares to leave) the now destroyed city...]
[...and the planet's atmosphere.]
After having patiently probed the corners of the galaxy,
[Core ships (including a sensor disc ship) traveling though nebulas...
[...alien star systems,]
[...with a mysterious probe ship(s) (one jammer) following it]
it found was it was hoping not to find.
[Core ships approaching old Arm outpost.]
The last surviving Arm are perhaps to be found laying dormant in a large temple structure.
But, the door has already been breached, hmmm.
[Arm Seed Ziggurath, needs a shot of the commander inspecting a busted door.]
The Arm would have been an easy prey if it hadn't been for the Con, a race of tenacious hostile machines.
[Con had followed Core and set an ambush (using large dropship cubes hanging in the sky).]
Appearing out of nowhere, the Con thwarted the plans of the Core, unwittingly freeing the Arm clones from their slumber.
[It appears the Con had gotten inside the building already. Something inside explodes, sending a bunch of Con units flying out.]
And something else, something very familiar.
[The Arm commander appears out of the rubble and smoke. Note that it would be suicide for commanders to shoot each other, they need grunts to do the work. Unsure how this is resolved. They can safely D-Gun the Con though. Might be a reluctant temporary ally? Arm uses Emergency gate out device, leaving Core to its fate (it has incoming reinforcements)?]
2015 note: I think this story board/synopsis from '07 still sort of works... but it needs iteration like everything.
It has been speculated that the Con were the result of a rogue Core construction K-Bot which spent thousands years patrolling repairing itself and the remaining buildings of its outpost. Maybe there was a timer overflow and its program got corrupted, and a personality sprung forth from the garbled behaviour data (perhaps they have unused data like our latent genes, or old game cartridges). The Con doesn't have a lot of diversity in their arsenal, yet. They are simple minded and completely irresponsible. Maybe a bit wacky, but not in a clowny way.
The Ext are a young race who have been forced into war. This means that many of their units are experimental and unconventional. They're inferior to the other factions (not balanced). The Core and Arm have polished their military proficiency for millennia.
In many RTS games, old and modern, the units have a small selection of predefined behaviours, such as "Attack-Move", "Guard" or "Patrol". Dark Reign offered a bit more control. Units had thresholds for automatically retreating for repairs, and they could roam around on their own. I'll attempt to take unit programming further, breaking behaviour up into its base components ("what happens when"). There's a risk that it'll make the game play itself (micro-wise), but I think it could be interesting to try as I'm more of a macro-guy anyways.
Standby: Stay, Area, Roam, Hermit or Social. Relevant when not having move orders. Stay means it'll just sit there. Roam means it will drive around all over the map. Area roam means it will move around but stay nearby. Social means it will move close to another friendly if any nearby. Hermit means it will move away from others.
Move to: Exact or Near. When a unit has arrived it will always go into Idle orders. Exact means it'll keep trying to get to an exact spot. Near means it'll settle for anywhere nearby.
Move type: Straight, Drunk or Diversion. Drunk is heading mostly towards the target, sort of zig zag. Diversion means it tries to get to the target from a different angle (a new attack vector) Useful for avoiding front defences and keeping point of origin hidden.
Attacked: Engage, Ignore, Retreat, Avoid, Standby. Engage means moving into range and attack. Ignore means it will not move into range, but possibly return fire. Retreat means heading back to point of origin. Avoid means still trying to get to target, but going sideways for a while. Standby means it'll go into its standby mode.
Target destroyed: Resume or Standby. Resume means the unit will resume its movement mission or return to it's previous location or movement target. Standby means it'll go into its standby mode.
Enemy in range:
Engage, Driveby, Standby.
Engage means the unit will follow the enemy target if it moves out of range.
In Driveby mode the unit will continue moving towards its movement target, but do a driveby shooting, if it has a turret that is.
Standby means it'll go into its standby mode.
! Warning, feature creep ideas ahead! I'll leave these here just to have them on paper.
Formation speed match: Match slowest or Ignore. Relevant if apart of a group selection. Useful for making units with different movement speeds arrive well grouped.
Formation position: Rear, Middle, Side or Front. Where a unit goes in a formation. Might want keep the artillery in the back and flame throwers in the front.
Avoid, Ignore, and Stalk list: Perhaps it's possible to make units less suicidal (infantry getting close to harvesters which run them over, columns of tanks always taking the path through artillery-valley). Harvesters of course could talk enemy infantry. Set lists on a unit-type wide basis rather than individual units. An alternative solution is to allow the player to put danger flags on the map. Here be dragons. This affects the pathfinding weight map.
Control: Player or AI, allows the player to let an AI handle the unit. There could be different 'AI generals' and they use their resources differently (Head butt, Sneaky, Coward, Pacifist). Maybe a late game building, because then there are more units around which the player won't have time to manage.