Wednesday, November 6, 2019

Game 106: Galaxy

Galaxy concludes a very odd chapter in Data Driven Gamer. B-1 Nuclear Bomber, Midway Campaign, Tanktics, and finally Galaxy were all published by Avalon Hill, which is the thread tying them together, but unlike other selections of games grouped by company, Avalon Hill had little to do with their creation, save possibly for Midway Campaign whose origins are obscure. In the years following Galaxy, Avalon Hill would release games more closely tied to their own board game products, such as Computer Diplomacy, Incunabula, and Under Fire, and reviews by Computer Gaming World rank these titles higher than any of the games I’ve covered. They would then withdraw from the video game market until 1994, and re-enter it with more direct adaptations of their board game products, and in this period produced their greatest hit, Achtung Spitfire! None of these games attain whale status, and I won’t be visiting them, so it’s just possible that by my own methodology, I am completely missing out on the big picture of what “real” Avalon Hill computer games were like.

I’ve just played Galaxy’s predecessor, Galactic Empires, and I won’t cover familiar ground in this post. According to its author Tom Cleaver, he sent it to Avalon Hill on a whim, and was surprised when they commissioned a rewrite, and asked him to supervise ports to the other computers of the day.

Unlike Galactic Empires (and all of the other Avalon Hill games I’ve played so far), Galaxy was available on both cassette and floppy disk, so AppleWin is finally suitable for an authentic emulation experience. It also has a manual, which I reviewed before playing.

There are two pages worth of instructions, most of them things I already knew, but a few things were new, either things I didn’t realize before, or that changed in Galaxy.
  • Players’ homeworlds will start with more than 100 ships if they are far away from neutral worlds and/or close to opponents, and less than 100 ships if the opposite.
  • Worlds are now identified by letters rather than numbers. In a game with more than 26 worlds, worlds 27 onward are identified by punctuation marks or symbols.
  • There is an options menu, accessible by pressing the Esc key, with these options:
    • View starmap
    • Open time/distance calculator
    • Enable/disable sounds
    • Save game
    • Quit game, converting your worlds and ships to neutral, and allowing the other players to keep playing
    • End game, causing the computer to declare a winner
    • Change game duration
  • Combat works as before, with the addition of a random “gunnery factor” for each side, which determines shooting accuracy.

The manual gives a number of gameplay suggestions too:
  • If a player mistakenly cancels another’s turn by pressing Enter during theirs, the offender may be punished by allowing the affected player to enter one ship movement order on the offender’s turn.
  • When there are three or more players, conduct truces, alliances, and deals, and feel free to lie, backstab, and retaliate.

The box also contains a log-pad to assist players in their multi-turn strategy.

Lastly, the manual has a number of scenario suggestions, which I find quite welcome in any game that depends so much on parameter customization for replayability.

The solitaire scenarios are especially welcome. Galactic Empires’ only suggestion was setting the turns and worlds to be equal, which was impossible with five worlds, much too easy with 40 worlds, and ridiculous when the neutrals are allowed to build. The ultimate challenge here, though, of beating 40 worlds in only 50 turns, seems to be quite impossible indeed when considering that in Galactic Empires it took me nearly 100 turns just to conquer 39.

“R” and I played Galaxy, this time using the suggested two-player game rules of 20 worlds, 20 turns, and no neutral builders.

One immediate difference from Galactic Empires is that there are no ingame instructions. I don’t know why they went to the effort of removing them, but evidently they were thought redundant with the inclusion of paper instructions.

The options menu at the start looks identical to the one before, with one addition – you can enter your name!

Finally, I can be Admiral Ahab!

The four character maximum may be limiting for most players, but it suits me fine.

The star map now has colons delineating spaces between worlds. It’s less aesthetically pleasing than the star map of Galactic Empires, but more useful for measuring distances.

There’s no longer any screen showing you the planet coordinates, but with the ability to look at a starmap any time, it isn’t needed. The new Time/Distance calculator is quite useful too.

This can be misleading, though. “Travel Time” is the number of turns needed to travel AFTER the end of the current turn! And it will arrive at the END of the "turn of arrival," and be available to you for more orders on the NEXT turn. E.g. – a value of 0 means your fleet will arrive at the end of the current turn, and be available at the start of the next turn. I would mentally add "1" to these values.

The screen hides whatever letters you key in, which more or less prevents the calculator from leaking your battle plans to any spying opponents.

The main screen isn’t radically different from the old one; it still looks and feels like a BASIC program, but the text is laid out a bit more attractively, with good use of highlighting to indicate the header and column separator line.

I started off by spreading out about half of my ships to take some nearby planets, leaving the other half behind to defend.

The combat screen has been completely overhauled. Instead of feeding you results line-by-line teletype-style, there’s a dynamic head-to-head combat screen showing you the fleet statuses as they take turns firing, with laser gun sound effects and a crude explosion animation whenever one side scores a hit. The number of remaining ships on each side steadily decreases with each hit until one side reaches zero.

Another addition is the gunnery factor, which is completely random, different for each side, and partly determines the fleets’ shooting accuracies. The result makes attacking more difficult; if the defender has even a small slight gunnery factor advantage, then because defenders fire first, they will eliminate a large number of your attackers during the first volley. To win an invasion, you must either have the advantage in both numbers and in gunnery factor, or have a very large numbers advantage.

A big gunnery factor advantage doesn’t benefit the attacker as much as it benefits the defender.

R found this out the hard way when he tried to invade my homeworld with 80 attackers, which outnumbered the 56 I left behind, but the gunnery factor was in my favor, and his fleet was decimated, leaving his gambit unsuccessful and his fleet size diminished.

My strike fleets tended to be more successful; without neutral ship building, the defenses on neutral planets were weak and most were taken with 10 or so ships. Had R been trying to do this too, I’m sure that this would have been a more challenging and therefore more interesting contest, with both of us trying to guess who will send which ships where and how many at once, but his ability was diminished by his failed invasion. He did manage to conquer one neutral planet, world “K” nearby his homeworld “B,” and declared his realm to be the Burger King Empire.

We did clash a bit over a planet “E” which produced a luxurious 11 ships per turn, but with my higher ship count, reinforced by industry from other planets, I won out, and went on to conquer his Burger King system.

R wanted a rematch, but we didn’t really have time. I’m sure we will play again sometime in the future, but I’m not sure that I’ll write about it. I can see this being an interesting strategy game about picking and choosing which planets to send your ships to. My early expansion left my defenses spread pretty thin, and an underdog would always have a chance to conquer something; as there’s no way to adequately defend every planet.

Compared to Galactic Empires, the biggest difference in our game of Galaxy was the smaller defense fleets on the neutral planets, and this was an option in both games. Apart from that, all Galaxy really introduced was admiral names (Ahab/Trig rather than Admiral 1 vs. Admiral 2) and the random gunnery factors, which played a significant gameplay role, as I’m reasonably confident that R’s 80-ship fleet would have conquered my homeworld under Galactic Empire’s rules.

I played more in solo mode, using the suggested scenarios as difficulty guidelines. I conquered a 10-world galaxy in 10 turns on my first try, using similar tactics as in Galactic Empires.

The next step up was 20 planets in 13 turns, which sounded tricky, but again, I beat it on the first try, with similar tactics as before. Most of the planets fell to my initial 10-ship fleets, and a surprisingly large number had no defenders at all. A few better defended ones held out, but there was just enough time left to marshal my surviving ships, supplemented by new ones, and conquer them.

Scenario 3: 30 planets, 15 turns

For this scenario, I sent my starting fleet to the nearest planets first, in order to start building up my fleet as quickly as possible. Starting on turn 2, I sent my new ships to the farther planets.

The closest planets, as before, put up little to no resistance, and supplanted my fleet. By turn 10, I had most of the galaxy under my command, with just a few holdouts within striking distance of various planets.

This was also the round where I reached the orders per-turn limit.

Wouldn’t be the last time Ion Storm prevented a release this year.

As my strike fleets got closer to the farthest neutral planets, I moved my reserves close behind to the nearest friendly planets, ready to mop-up any holdouts. Despite some logistical errors, I won with a turn to spare.

Scenario 4: 40 planets, 17 turns

In Galactic Empires, I managed this in 22 turns, so doing it in only 17 seemed a bit intimidating, but Galaxy also, so far, seemed to have smaller neutral fleets.

The map, with 14 symbols representing worlds, was starting to look ridiculous, though.

At the end of it, despite making a serious strategic error where I sent lots of small fleets to ‘X’ despite not owning it (wasting a great deal of time and ships), and more than once accidentally sending ships to the wrong planet thanks to getting the symbols confused, I conquered all 40 with three turns to spare.

That was it for the idle-neutral scenarios. The next level would be 5 worlds, with neutrals that build ships, with a 20 turn limit.

Scenario 5: 5 worlds, 20 turns, neutral builders

This map turned out to be basically impossible. I conquered D and E quickly, and scouted B and C, to find out by turn 7 that B had more than 100 ships and was building 13 per turn, while C lacked any industry at all. My entire combined industry from A, D, and E was only 13 per turn, and I’d therefore never be able to out-produce B. If I had sent my entire starting fleet to B right away, then this would have been almost ridiculously simple, but there’s just no way to know that ahead of time.

Trying again, I got this map:

And it was almost gloriously simple. I sent scouts toward B, D, and E, and then steamrolled C and moved onward to B. The worlds B and E turned out to be completely undefended, so my scouts conquered them, and when my strike fleet arrived at and took B, I redirected toward D, conquering it by turn 10.

What a difference a little bit of luck made! One game I got a setup that would have been impossible unless I had sent blindly my entire starting fleet to the most industrious planet from the first turn. In the very next game I had conquered the universe in half of the allotted time.

Scenario 6: 10 worlds, 35 turns, neutral builders

My experience was once again luck of the draw. Few, if any neutral worlds would ever be close to my starting world, so if I didn’t conquer any high-industry worlds in the first few turns, then I’d inevitably encounter some later in my campaign, and by then they would be so well defended that I wouldn’t have enough time to build up a big enough fleet to conquer them all in the allotted time.

Then I played a game where my fleets conquered two moderately industrious worlds, and with them was able to eventually conquer the rest, none of which had any more industry than those two. I conquered the galaxy in only 21 turns.

Scenario 7: 20 worlds, 40 turns, neutral builders

Here, there were enough worlds in the galaxy that some were close enough to reach in a few turns, and I found it necessary to use classic Galactic Empires 40/100 tactics – spread out to these planets aggressively from the start, send newly built ships to a strategic point on the assumption that you will conquer it, and restart if any of your fleets fail to conquer, or if any turn out to lack industry.

The gunnery factor played a role here, and a big difference can turn a sure victory where you vastly outnumber the defenders into a defeat, or it can turn a narrow victory into a crushing defeat. It can work in your favor too, when your gunnery factor is much higher than theirs, but doesn’t often turn defeat into victory – usually it would just turn a narrow victory into a somewhat more decisive one.

An easy victory became a costly one.

I did eventually beat this scenario with time to spare, but only after several false starts where my initial scatter fleet failed to meet expectations, and the wasted time amplified as the neutrals kept building more and more ships while I built more of my own to re-take their ever growing numbers.

Scenario 8: 30 worlds, 45 turns, neutral builders

I gave up on this penultimate scenario, when I began with an oversized fleet of 182 ships, sent out nine fleets to nine different planets, and every single one of them either turned up empty-handed or got decimated by a smaller defender fleet with superior gunnery.

I can’t help but feel a bit underwhelmed by Galaxy. I’m sure a lot of this has to do with the fact that I spent so much time with its predecessor, and was expecting more changes than this. According to Tom Cleaver’s notes in a 1982 issue of CGW, he re-wrote Galactic Empires for Avalon Hill over the course of four months with the help of University of Louisville engineers. I just don't see anything that nearly amounts to a total rewrite. All I can really see for the effort is a slight upgrade in presentation and a few new modules, such as the time/distance calculator. It’s still a BASIC game, and by comparing the code, it’s obvious that this isn’t a complete re-write; most of the lines are identical. Galactic Empires’ code even has more comments and less use of torturedly condensed multi-statement lines.

That said, the inclusion of playtested scenarios in the manual are a very welcome addition. The only significant gameplay change from code, as far as I can tell, is the gunnery factor, and I’m not sure that it really improves the game, and in the case of singleplayer scenarios, I found it made an already frustratingly random game experience even worse.


  1. Is the gunnery factor something that is re-rolled every time you attack a planet? Ie. your fleet might reach D and attack with factor, say, 83 and then the same fleet moves on to E and attacks only with factor 52?

    1. Yep. Attacker is a random number 0-1 square-rooted (meaning an average of 0.71, and a standard deviation of 0.21). Defender is calculated, so that Attacker^2 + Defender^2 = 1.

  2. Not sure I like that. Especially considering the to-hit formula of [Gunnery Factor]/100 * SQRT(Targets/(Shooters + Targets))

    The more attackers you have the less effective each are and the more effective the defenders are. (Clearly, this game follows the Law of Conservation of Ninjitsu: :P)

    And if you roll low for your gunnery factor, you're guaranteed to be against enemy that's always high. So there's no grace chance of enemy also having rolled bad accuracy factor to even it back. That makes the luck factor much higher, since 50% of time you will roll lower than 0.5. If both rolls were independent, you'd have very coarsely approximating 25% chance to roll high(er than 0.5) and opponent rolls low, 50% chance both rolls high or both rolls low, and only 25% chance that you roll low and opponent rolls high. This would make extreme swings of luck much rarer.

    1. The formula reminds me of Lanchester's Laws, though this could easily be coincidental. The more shooters, the more likely that multiple shooters will aim for the same targets, which reduces each individual shooter's odds of scoring a kill... but still overall increases the group's number of expected kills. And the more targets, the less likely that multiple shooters will aim for the same targets (or perhaps make it more likely that a miss will become a hit), increasing each individal's accuracy, but still leaving the group with a larger number of remaining targets.

      "50% of time you will roll lower than 0.5"
      Correction - 50% of the time you will roll lower than 0.71. But the rest of your paragraph still holds, because if you roll below than the mean, the enemy will be above the mean. About a sixth of the time, the enemy will be 50 points higher than you.

      The power differential likelihood can be expressed and graphed as:
      Y = sqrt(X) - sqrt(1 - X)

      where X is a roll between 0-1, with fair odds for all possible values.


Most popular posts