Sunday, March 27, 2005

Game Post Mortem: Hard Vacuum

Mining a 12-year old game design for innovative game mechanics.
(Plus the complete set of 8-bit graphics available for download

In 1993, I worked on an unreleased RTS game design called Hard Vacuum. Dune 2, the father of the modern RTS had just come out in 1992, revolutionizing the the gaming community's perception of the possibilities of the strategy game genre. We believed that RTS games were the future and that we were the inspired game developers who were going to popularize this hitherto ignored genre.

Some historical perspective is in order. Warcraft wouldn't come out until late 1994. Command and Conquer wouldn't arrive until 1995. What many PC gamers today consider to be one of the most burnt out genres was at that time new, fresh and completely unknown to most gamers. The term 'RTS' hadn't even been invented as far as I know.

As I write this in 2005, I'm bemused by the path the RTS game genre has taken from stunning innovation to stagnant maturity. The latest Age of Empires is certainly a gorgeous game, but the core mechanics of the genre have changed little since it's peak in the late 1990s.

From this perspective, I revisit a 12-year old game design conceived near the very birth of a genre. In particular I'm interested in the unexpected solutions to common problems. What sort of bizarre innovation occurs when a game designer can't steal from the latest best seller?

Hard Vacuum: The Basics
Hard Vacuum was set in a science fiction world run by large corporations. You were a mercenary battle team, out for hire to the highest bidder and ready to kick ass. Admittedly, a rather generic plot, but then again entries into a new genre are rarely concerned with plot. Hard Vacuum also had most of the basics of your typical real-time strategy game.

  • Resource Gathering: There were mineral deposits that you could build mining towers on to gather resources.
  • Base Building: You could build a variety of building for generating units, new buildings, and additional resource gatherers.
  • Fighting: There were a variety of units that could be deployed to attack enemies, destroy bases, etc. You could select them with a rubber band and send them towards the enemy.

Hard Vacuum: Innovations
There were several interesting systems in Hard Vacuum. Some made it into future RTS games and some did not.

  • Supply Lines: When you built a mining tower on a resource deposit, there was no need to manually build and manage drones to carry the minerals back and forth. Instead, a road was built from your base to the mine. Drones were automatically created when the mine had a full load and sent along the road to your base. Enemies could blast your supply line and interrupt your flow of resources. So protecting fixed supply lines became a bit part of the strategy.
  • Variable Height Terrain: This was a 2D tile-based game intended to run on 386 and 486 machines. We had a full system of variable height terrain. Units on higher ground would have targeting advantages over units on lower grow
  • Deformable terrain with flowing water and lava: You could blow huge holes in terrain with artillery in order to divert streams and lava. Drop a water bomb on a city and watch it wash away an entire troop. Drop a dirt bomb to create a mountain.
  • Walls: Walls played a huge part in the game to give large defensive barriers to attack. These existed in Dune2, but were extended in HV
  • Player created landscapes: Walls, base building, supply lines, and advanced terrain modification all contributed to a game where players built intricate maps during game play. We couldn't afford to spend lots of money building maps so instead we had to create fun systems that let players create unique and interesting combat situations.

What We Lost as Game Players
All all the mechanics in HV, the one I fail to see in new games is the focus on player-created environment. With our bloated budgets, we can afford to spend a two dozen man years creating intricate content that players use once and then throw away. Oh, you spent and hour playing map 3b in the single player campaign? Blizzard spent 5 months designing that. Now you are never going to play it ever again. What a waste.

There is a big lesson here. Without a successful game to promote the concept of player-focused creativity in the RTS genre, this intriguing game mechanic was stillborn. The concept certainly has some validity since we can see that it blossomed in the sim genre in the aftermath of SimCity's success. But without clear examples of successful application in the RTS genre, designer never really had a chance to steal it, implement in their 'next generation' games, and evolve it into something more polished.

I believe strongly that genres evolve in simple steps based on modifications to previously successful games. Games like C&C and Warcraft solidified the major conventions of RTS games. Unfortunately, in the game industry, massive success breeds design stagnation. Once a successful formula for creating a RTS was discovered, gamer designer copy the basics and then focus on polishing the basic formula. Better cut scenes, improved graphics, tweaked unit control for expert users, etc. As game designers, we only look back as far as the last successful title.

Let's play "What If"
Imagine for a moment, if a game like Hard Vacuum had achieved popularity. The ripples of a successful early title have a profound effect on future titles. Would it be obvious the most casual game player that terrain modification is a fundamental part of the basic RTS formula? Would things like supply lines and dynamic player created environments be the standard?

An even crazier thought branches off from here. If these design mechanics had become standard, what would a modern RTS game complete with physics, 3D graphics, and king-of-the-genre polish look like today?

When we look at the evolution of genres, these are intriguing questions. Game designers are too complacent in their acceptance of genre norms. It is up to us to shake things up a bit.

An Exercise for the Reader
Pick up and play an ancient copy of a game that fathered a genre. The further back you can go, the better. Clear your mind of all expectations and knowledge of what the genre evolved into. That cool thing that Half Life did with conversations. Forget it.

Now reinvent the genre. What are the core primitive concepts and where can you take them that would result in addictive player experiences? You have the opportunity to reinvent an entire decade of evolutionary game design in your head. Chances are you will spawn a few original ideas.

If you can get the final project through the miserable distribution system that plagues our industry, you might even have a new genre buster on your hands.

The Hard Vacuum Team
The team that made Hard Vacuum was typical of many groups in the early 90s. We were small, self funded, still in school and willing to work crazy hours. A team like this today can not even hope to create a major block buster game.

Yet, in that time period, this seemed to be the standard. Id took the world by storm with it's first person shooter Doom in 1993. Tim Sweeney and James Schmalz were busy starting up Epic with a titles like Epic Pinball. Most teams were still below 10 people and a large number of successful games where released with core teams of 4 or 5.

I've only been able to keep track a few of the folks involved.

  • Daniel Cook: I did the artwork and a bit of the HV design. I went to work with Epic and created a game called Tyrian that paid my bills for a long while. After that I designed a program called Anark Studio, which is used to create thousands of interactive 3D project. We even started selling it to the game developers this past year.
  • Ryan Geithman: Ryan was our rocking lead programmer. He went on to do some amazing work on Baldur's Gate: Dark Alliance and the new Bard's Tale game. An industry man.
  • Patch: Patch was another programmer. I have no idea what happened to him. He would call me up on pay phones to talk about game design. Every few minutes he would have to play a series of tones from his little blue box to get another few free minutes of talk time. Conference calls through phone phreaking. Those were the days. :-)

Free Artwork
I don't have the design documents any longer, but I do have a large number of the game tiles. These were done primarily in DPaint and Brilliance using an Amiga 1000. We didn't have no stinking Wacom tablets so these were all drawn with a mouse. Pixel art at its finest.

  • ~60 vehicles
  • ~120 terrain tile sets
  • ~30 misc graphic (bullets, blood, clouds, etc)
  • ~30 buildings
  • Interface screens

Use them, abuse them. If you make anything interesting, let me know.

take care


  1. You probably got sick of making all these games... remember when you made penguins? They were somewhere in this game... Hehehe... I wish I had my old graphics from Generic Quest (remember all those Ultimate VI style sprites I did?), or Iron Plague... but I doubt I could ever find them. You definitely got shafted with these things, you did so many sets of graphics... and oh... how the coders bragged up their abilities to do a game... and oh how we had no clue just how much work was required to FINISH a game... Hehehe...


    PS> You should post your old Star Trader graphics, or whatever that game was... the very first one... Heheh...

  2. *Grin* I can't help myself. I'll keep making game designs and trying to get people to make them.

    The truth is that programmers are inherently unreliable, because they need to do so much more upfront work than the designer or artist. ;-)

    take care

  3. Re: But RTS do ship with player created content tools.
    Yeah, that's something I need to clarify. When I say 'player created content', I do *not* mean mods. Mods generally require specialized skills and are geeky as all hell.

    What I'm talking about is a game that is allows you to build an intricate map as a fundamental part of the game. Examples include:
    - Populous
    - SimCity
    - Tetris
    - Will Wright's new game Spore.

    RTS games started to go down that route with base building and unit building. Every game could be radically different from one another based off where you built and when you built. But then they started to go the Starcraft route with intricate maps, highly scripted elements and limits on where you could modify things. The result is a set of disposable maps instead of a system that allows extensive player modification of their world.

    Game balancing
    I think terrain modification is only difficult to balance if it is an 'extra' feature. As in "We need plot, scripted maps, balanced units, AND let's toss in terrain modification at the very end. If you are building a game like Populous from the ground up, it should become part of the core gameplay and undergo months of in depth balancing.

    take care

  4. Ray, I've got some of the GQ gfx at home, I can mail you if you want.

  5. Cool, the origin of datacubes :D
    I think the only RTS games I've played extensively are Dune 2, Warcraft and Warcraft 2. I've also tried Z: Steel Soldiers and Command & Conquer. I got Age of Empires in a pack of cereal a while ago and just started playing it. I know it's a popular game but I can't figure out why. It's exactly the same as Warcraft 2. I can't see any significant differences at all. Warcraft 2 is Warcraft with boats. Warcraft is Dune 2 with two types of resources. I'm only four games into the genre and it's already stale to me. Even the winning tactics are the same for each game.
    Z: Steel Soldiers seemed pretty different. Factories and stuff were already on the map and you had to control their sector to use them. You have to send your troops out around the map instead of building up a huge army at your base to smash your enemy. You could blow up all the rock walls which shaped the map too.
    Age of Empire's maps are too static. All the historical fluff in the game makes a big deal out of Egypt's yearly floods that bring new fertility to the delta. But a game can span thousands of years and it never actually happens. There's a set amount of resources on the map and that's it. You can't even plant trees or fish sustainably. You can't import stone or build harbours. I keep wanting to build canals and flatten hills and set fire to trees. Fire is another thing rarely seen since Populous. The mecha strategy game Titans of Steel lets you set fire to forest which makes mechs walking through them overheat.
    The only RTS games I've seen with flowing water are the Clonk series. I've only played the first few. They're a bit like Lemmings with resource gathering and fighting. At least one cave flier has rain and flowing water too. It's from the same perspective as Clonk. In both games channeling the water into big pools and dumping them on stuff is fun.
    I guess Magic Carpet is also a RTS. You could do a lot of the Populous style stuff to the landscape in that but it didn't have as much of an impact seeing as the creatures and people were pretty much only there to be blasted and turned into mana.

  6. Oops that's just "Z." Steel Soldiers was the 3D sequel.

  7. Yeah, I remember that terrain creating was my favorite part of Warcraft 2, I would spend hours trying to create fun balanced map, or something new and bizarre. When the limited Map Maker couldn't do everything I wanted I turned to fan created tools that allowed control beyond the bounds of the original game, such as altering the tiles of land to allow boats to cross through it or conversly allowing land units to walk on water, oh and my favorite, overpowered peons that could slay dragons, of course I was in high school at the time, so I was a bit more easily amused but lightning throwing peons.

    This thing about terrain blasting in Hard Vucuum sounds like alot of fun, and I have always loved walls in RTSs even though most games implement them poorly, making it imposiple to properly use them to enclose a base such as not letting them butt up agianst other terrain leaving gaps between the mountain at your back and wall in front. Ranting aside, I would love to play Hard Vacuum, I'm curious how far along was it when it was aborted. Would there be any chance that the existing code could be turned over to a fan group or otherwise revived? Perhaps even something as crazy as being recoded as a N64 game to be played on the new Revolution virtual console. (I have a pet theory that the Revolution controller will allow new RTS innovation, as the mouse like control is much more natural for RTSs)

  8. I think the Hard Vacuum graphics look excellent, and I would like to use them in the open source game OpenRTS

    This is an open source RTS-game released under the GNU General Public License.
    Would it be possible to use these graphics with this license?

    Please contact me at or so I can get your permission to use the graphics.

    - Andreas R.

  9. Just so you know, I'm using the Hard Vacuum tileset in my game "Killer Steam-Powered Robots from Dimension X".

  10. Have no idea if you still check this msg list. But anyway... I can't find even a refrance to the Dos/Windows version of this game. Did it get released?

    I am interested in what you mean by Flowing water and lava. As is the flow of water in a river could be interupted?

    Anyway thanks for the help and the article.

    grgory.ewing you can guess the drill At univie dot ac dot at

  11. How can i open the game?

  12. Nice graphics, i might use them for a free webbased game, will tell you about that.

    thanks for sharing :)

    Snowstyle from France

  13. I like the idea of going back to a genre's roots and goofing around to see what you can come up with, but I disagree with you on the player-creativity part.

    You said that games evolve in small steps based on modifications of previously successful games. That's true (and if we ever get some independent developing without publisher-like financing, we'll get games based on ideas that were in financially unsuccessful games like ICO or Okami). But the thing is, I don't really think that giving a player the freedom to create an environment would ensue with much fun. Or much creativity, actually, because either the whole thing would involve customization or it would involve trial and error, which are the raw materials for frustration in way too many cases (I know this because of my experience with games like Worms: World Party).

    What I mean is, if you handed someone a game where they would be able to create, like, a weapon, for example, you wouldn't always get very satisfying results. I know we can't judge now because no-one's bothered to do it and I'm pretty sure someone would come-up with an idea to switch the concept around, but if it were successful we'd be at a stall again because everyone would copy, and if it weren't successful no-one would try to improve. I guess that player-creativity just has too much potential for a dead-end.

    But I do want to point-out that I would like to see developers actually trying to do something with player-created environments/weaponry/anything than nothing at all. I'm really kinda sick of RTS nowadays because the only "improvements" we're getting are stuff like "emotional AI" (The Lord of the Rings: The Battle for Middle-Earth) or better graphics (every game, I guess). And besides: "We couldn't afford to spend lots of money building maps so instead we had to create fun systems that let players create unique and interesting combat situations," so creating maps costs money, and I'm guessing that publishers would stifle any creativity or riskiness when there's a lot of money involved.

    I think going with more unpredictability would be better. Every time I re-play a scene in Metal Gear Solid 3 I end-up doing something different and it seems like everything I poke has been done, ya know? But then again that would cost money, and I see a Deus Ex-VS-Half-Life design-approach-argument (re-playability VS give-your-audience-the-entire-meal-at-once approach) on the horizon so I'll stop right here.

    Sorry if this is spam.


  14. * WITH publisher-like financing ... sorry, typo.

    Definitely spam, this time. Sorry :/

  15. Tyrian 2000, what a game :D

    I seem to remember Earth 2150 having deformable landscape. You could dig trenches, flatten terrain, tunnel underground, and so on. Good times!

  16. This comment has been removed by the author.

  17. Hi, I'd really like to use your graphics in a game I'm making, but I really need some clearance on which buildings did what. For example, what was Factory2 used for? Lol. As well as where were the units created? The Base? or mothership? Ect ect. So if you could please email me with description of each building and what they did, I'd really really appreciate that. Also, if I ever get this done, or have a beta release at least, I'll be sure to post it. ^_^

    My email is:

  18. "Game designers are too complacent in their acceptance of genre norms. It is up to us to shake things up a bit."

    Given the quality of the rest of your posts I'm surprised that you came to this conclusion. It's been my experience that the designers were not the problem but the publishers. Those dreaded suits who came and made sweeping changes to any design that didn't fit in w/ the current hot game mold.
    I do agree with you that open source and 'cloud design' will be the wave of the future, simply because it's gotten to expensive to make a triple A title these days. (Massive layoffs everywhere hello?)

    Love the site!

  19. I'm developing a console-style RPG for XNA Community Games, and the HV graphics are the foundation for my world map. I spent forever trying to find some game rips that I could use for prototyping, but these are perfect as-is. My only modifications are going to be for color and day/night variation tilesets.

    First quadrant of world map (almost done):

    Forum (includes links to art and story):

  20. Hi Danc! My name is José Alberto and I'm from Spain. I took these tiles a couple of weeks ago and I'm trying to make an open source RTS in C++(it's for my final career project). I think i will show my game in a few weeks but it won't be completed until a few months :P

    i just want to thank you for these resources, they are awesome!!!

    You made me a huge favor :)

    Kind Regards

  21. i was just wondering what's in the HVGFx folder, when i extracted it, it turned into some strange file type and i tried to figure out how to see tehm but i'm clueless. Do you still remember what program they were supposed to be used for?

  22. .lbm is a dpaint format. you can use pictview to see them.

  23. My own "Hard Vacuum Tile" game (beta)

    1 of 3:
    2 of 3:
    3 of 3:

    spanish chat sorry :P

    Kind regards

  24. Hey there :) Quite enjoyed your post, and enjoyed the pack of game art even more. I'm updating (as always x.x) my current game to use some of these tiles, and I'll let you know how it turns out!

    Thanks for the release; it's always good to see more people releasing open content :)

  25. Hi Danc.

    Thanks for sharing your awesome work.

    I'm developping a space trade and fight strategy game for android mobile.

    Some screenshots of the first release :
    Space Trader


  26. I cleaned up some of the nature tiles for use in Tiled (

  27. I just found out about these tiles, very nice. I got inspired to do a resource gathering RTS with a few different takes. I'm not sure if I'll go to the ambitions of the original concept, but I have different ideas.

    Partly to blame is me playing Dune 2000 lately, renewed my interest in these sort of games.

    Anyways, I compile the tiles into a handy tile sheet and started work on a map editor, check it out here.

    Contains the compiled tileset as a PNG. If you want to know more of my plans let me know.

  28. This is excellent classic work and a great history lesson. Thanks for the share.

    Rodney Miller
    (a.k.a. life DigOurGame™

  29. Just to say thanks so much for making these graphics available!

    We used them in our game "Shyft" that is currently in beta.

    Here is a link if you are interested:

  30. Using your graphics I created a top-down style helicopter shooter in HTML5.

    Thanks for providing the artwork!

  31. Why do Tank4.bmp, TankBase.bmp and TnkTurt1.bmp have different number of tiles in vertical columns?
    Do larger tanks only move in diagonals? This doesn't make sense to me.

    See a preview of what I'm working on at
    code is on github, pull requests are welcome.