GalCiv IV Dev Journal #1

Published on Monday, May 10, 2021 By Brad Wardell In GalCiv IV Dev Journals

And so it begins…

Make sure you check out, and in particular the FAQ and Game pages, which will give you a lot of details on what’s coming.

A little background

Before we start, I should introduce myself. I’m Brad Wardell. I designed and programmed the very first Galactic Civilizations game back in 1993 for OS/2. I literally programmed it out of my college dorm room after picking up “Teach Yourself C in 21 days”. 

While OS/2 didn’t take off like IBM thought, it gave me the opportunity to make a game that focused on really good AI and a unique style of gameplay. For the past 30 years, I’ve been making space strategy games, albeit with more resources than back in 1993 when I was hand drawing space ships with an icon editor.

Two philosophies

While I was programming on my 386SX and talking on Usenet, the guys at Simtex  were making a game called “Star Lords,” which was eventually released as Master of Orion. These games represented two main philosophies on how to do a space strategy game – the free form movement style of Galactic Civilizations, and the phase-lane/star to star method of Master of Orion.

image  image

Most space games seem to have taken the MOO route of point to point. It’s not hard to understand why. Every tile, even in space, uses RAM. Think how small Civ maps were back in the 90s. Since GalCiv was on OS/2, we had entire MEGA-bytes of memory to work with.  Even in GalCiv III, those huge maps consume a lot of RAM.

Modern Space Games 

In the early 2000s, we became friends with a company called Paradox.  They were our European distributor for Galactic Civilizations II.  The guys at Paradox and Stardock worked like peas in a pod. In 2012, the Master of Orion IP went up for auction. Both Stardock and Paradox were eager to get this IP. However, we were both narrowly outbid by, who later went on to make a new Master of Orion.

Instead of a Stardock or Paradox Master of Orion game, we ended up with Galactic Civilizations III and Stellaris.  While one can speculate how things might have gone if either of us had acquired the Master of Orion IP, I think most people are glad with how both games turned out.

Galactic Civilizations III preludes

After Stardock sold its Impulse platform, I decided to focus my energies into co-founding a couple of start-ups. The first, Oxide, was made up of the lead Civ devs over at Firaxis.  They had recently finished Civilization V and some of them had interviewed over at Stardock to investigate what came next. We ended up hiring Jon Shafer, who was the lead designer of Civilization V - he did the original design for Galactic Civilizations III.  A whole bunch of our internal terminology is based on some of his UX innovations (such as the “Shafer button”).  

The Oxide team was focused on making a next-generation game engine.  Stardock had tried to make a 4th generation engine for Elemental and it was a disaster. Oxide developed Nitrous, which powers Ashes of the Singularity.  Today they’re working on a big secret project.

Meanwhile, Soren Johnson and I were putting together another studio made up of some Firaxis vets called Mohawk to create Offworld Trading Company.  Today they’re about to release their second game, Old World.

Being the CEO of Oxide and the President of Mohawk (and President and CEO of Stardock) meant I wasn’t available to work on Galactic Civilizations III. Early on, Jon left to make At the Gates and Cari, the lead developer of GalCiv I and II (for Windows), was on extended maternity leave. So, GalCiv III was quite a challenge to develop.


The GalCiv III that launched in 2015 is a very different game than the one in 2021, as the two screenshots (launch and current) make clear.



At release, Galactic Civilizations III got great reviews and was a good game at launch.  But it wasn’t a great game.  We had our work cut out for us.

Lessons from GalCiv III

To understand why some people felt GalCiv III was a step back from GalCiv II, we need to look at GalCiv II.

GalCiv II was filled with story-driven events.  I hard-coded these in C++, but made a lot of them and they could be very in-depth and interesting.  This meant that every game of GalCiv II could end up feeling like an epic story.


But it wasn’t just the events, it was hundreds of tiny touches that increased immersion. For example, the player could look on any species ship and get a readout of its equipment with race-themed named for these components. The Altarian weapon names were always super passive aggressively named like “Not necessarily friendship giver Mark IV”

Plus the stats. The endless, unnecessary stats.


Even the combat seemed more interesting.


But, Galactic Civilizations III was a design of its time.  Designed in 2012 and released in 2015 the goal was to make it more mainstream.  "Streamlining" was the word of the day. GalCiv III wasn’t the only casualty of this line of thinking.  Elemental: War of Magic might have been buggy, but it had depth. So many details. 

But by 2016 we were making Sorcerer King, which had streamlined all the “rough edges” out. In a pre-Unity world, this strategy made sense. We wanted to make sure these games would appeal both to hardcore gamers and also more mainstream gamers (not “casual,” but people who might not appreciate a half dozen modifiers on a weapon).


Once Unity games started to flood the market, GalCiv III found itself to be too complicated for the casual market, but too light for the hardcore gamer who now had options like Stellaris.

New Directions

If you ask someone if you should get Galactic Civilizations III today they will say “YES But you have to make sure you get Crusade”.   After Ashes of the Singularity and Offworld Trading Company shipped, I was able to come back to GalCiv.   I had had my own design document for GalCiv III back from 2010 which focused heavily on the concept of citizens and civil wars. Some of these ideas went into GalCiv III: Crusade.  We were able to begin adapting GalCiv III for the new market reality.

Updating GalCiv III’s design via expansion packs, however, is a bit like trying to find new missions for jet aircraft whose designed mission has become obsolete. This is where GalCiv IV comes in.

Where we want to go

With GalCiv IV we now have enough memory and processing power available to build what amounts to a simulator behind the scenes while presenting it in a nice, easy to understand, turn-based strategy game UI.  What this means is that we want a game of GalCiv IV to feel like you’re actually running a space faring civilization filled with interesting characters. Rather than having an AI just for each alien player, we want an AI behind every single character in the game – and your civilization is made up by a lot of characters. And every character has a potential story to tell.

This means, from a gameplay point of view, that the player is still in charge of a vast, interstellar empire that is exploring, expanding, exploiting and exterminating things, but the galaxy is a livelier place than it was in the past. There are many more mechanisms in play that can affect things, a lot more moments of “Well crap, in hindsight, I feel like I should have seen that coming..” which results in players feeling like they keep getting better and better at the game each time they play.

The gang is back together

So Cari is back from maternity leave, I’m back from managing Oxide and Mohawk, we have Derek (Kael of Fall from Heaven fame) as the lead designer. Paul is back to being able to focus on UI and space ship making, Jesse is back to make sure our graphics are amazing, Sarah is back to make sure our underlying UI system is insanely powerful and useful, and we have new people on the team who previously worked on games ranging from Star Control to Sins of the Prophets. 

It’s going to be a good time!


Galactic Civilizations IV Dev Journals