Friday, October 25, 2019

Game 104: Tanktics

As I noted in my “road to Galaxy” post, Chris Crawford’s Tanktics is the earliest computer game with a direct relation to Avalon Hill, first written in 1976, self-published in 1978, and distributed by Avalon Hill in 1981. Mobygames lists these as the releases:
  • 1976 – IBM FORTRAN version, by Crawford
  • 1977 – KIM-1 version, by Crawford
  • 1978 – Commodore PET BASIC version, self-published by Crawford
  • 1981 – Commodore PET re-release by Avalon Hill, coded by Crawford
    • Apple II port by Tod Frye
    • Atari 8-bit port by Crawford
    • TRS-80 port by Bob Smith
  • 1983 – FM-7 port by CRI Middleware

Preceding this post, Crawford and I exchanged some emails, and his very valuable notes cleared up multiple points of ambiguity.

[Tanktics] at Wargy I was my first attempt, and had many weaknesses. I completely rewrote it from scratch to get the version for the KIM-1, and then completely rewrote that from scratch to get the version I self-published. I then made some revisions for the Avalon-Hill version. […] All of the games in the series used a map with cardboard counters. The computer printed out coordinates of tanks and the player used those coordinates to locate the tanks on the map. Before I self-published, the maps were handmade.

On the Wargy I game:
“Wargy” referred to the meeting; the game, I believe, was called Tanktics, but I’m not sure.

On the KIM-1 version:
The shift from FORTRAN to 6502 assembler required major changes. I preserved the basic concept of a group of tanks fighting over a simplified terrain with just four types of terrain. The map had only 256 hexes.

On the self-published release:
This version was closer to its predecessor, but it used a larger map. […] The version I sold had a black-and-white printed, folded map.

Crawford also sent photos of some of his old maps, with additional notes.

Tanktics II was the first game I did for my KIM-1 system. It had only 256 hexes and a single tank on each side. The two players actually split the little keyboard down the middle. Each one had his own map.

[Tanktics III] was the full game for my expanded KIM-1 system with 4K of RAM and my “tiny terminals”. Again, the two players could not see each other’s maps. Each side had 8 tanks, but different tank types could be set.

Tanktics IV [was] the commercial version I made for the PET. This was printed and folded twice for shipping down to 8.5” by 11”. It used the same map as I used in Tanktics III, only black and white.

None of these versions remain available anywhere now. Avalon Hill’s extant 1981 releases, which use a different map layout from these previous versions, can be thought of as Tanktics V, and I’ve opted to play the PET version as the most “original” available version.

The game’s most distinguishing feature, unchanged from the earlier versions, is that you must have a map to play this game, showing a kind of designer’s mindset that I haven’t really seen since the Magnavox Odyssey. We take it for granted that a computer game will be self-contained in the computer program itself, needing only input from the user, and displaying all of the necessary information to the user on-screen. Even in 1962, SpaceWar! had on-screen graphics to represent everything, only relying on the players to manually keep score, and before long it was capable of doing that on its own too. Here, Crawford views Tanktics as more of a computer-aided board game than a computer game, though unlike Odyssey the computer does keep track of the complete game state – it is the player’s responsibility to manually move the physical game pieces as instructed, not as a means of changing the game state, but as a means of providing himself a visual aid, since the computer program has no graphics.

You use the map to plan your tanks’ movements, and key them into the computer, which will calculate the results of your actions, decide your opponent’s actions, and display the results; your tanks’ new positions, enemy sightings, and hits and misses. This necessity will make playing Tanktics a hard pass for most players, but I found it oddly immersive, like being in a command center, isolated from the action and making life and death decisions blind except for the terse radio updates coming from the tank commanders (“Panzer VIb-A arrived at coordinates BB23, spotted JS-II-G at coordinates Z15”), and a map that I had trust myself to read and update accurately.

Avalon Hill provided a full color map board in the box, but the scans available are low resolution and the printed coordinates are difficult to read.

Eventually, I found something even better than a legible scan; a module for a board game simulator called Vassal, which includes a high resolution board scan, and partial rules implementation. I’d still need the computer program to play, but could use Vassal to keep track of the tanks’ movements and positions instead of needing to use Gimp as I had originally planned.

Download the module and Vassal engine here:

Program screenshot

Vassal board initial status

The manual is nearly twice as long as Midway Campaign’s, and just as dense, with one page of unit stats and five pages of gameplay instructions.

There are five scenarios to play:
  • Meeting engagement – Meet the Russians at an objective hex and eliminate them.
  • Hedgehog defense – Your tanks start at the hex and the Russians will come to you.
  • Armored assault – The Russians have stationary guns placed at the hex and you must attack.
  • Line defense – Your tanks are positioned in a line in front of the hex and the Russians will try to breach you.
  • Line assault – The Russians have stationary guns placed in a line in front of the hex and you must try to breach them.

You may then decide the tanks for each side – you may pick between one and eight for yourself, and however many you pick, the Russians will get twice as many. You must also decide which tanks each side will get, and here I feel this poses a design problem. How is the player supposed to know what tank layouts will provide a fun challenge? The manual suggests giving yourself weaker tanks to increase the challenge, but would it be reasonable to defeat 16 of the strongest Russian tanks with 8 of the weakest Panzers? And if not, where is the line drawn between a reasonable challenge and an unreasonable one? The manual gives no concrete suggestions. It also paradoxically suggests giving yourself a mix of strong and fast tanks for both mobility and firepower, and this creates a conflict between making a sound tactical decision and creating a challenging scenario.

I decided my first game would be Scenario #1, with 8 vs. 16 tanks, and the most powerful available to each side.

Stat tables

Setup menu

Starting positions

Right off the bat, my company could see three of the Russians. Tanks can fire or move, but may not do both on the same turn. As part of a move, they may rotate, the sole purpose of which is to orient the tank’s front, where armor is the thickest.

Russian tank ‘j’ was the most vulnerable, seen by my tanks d, f, and g. My tank ‘d’ could also see g and i.

I had tanks ‘f’ and ‘g’ fire on ‘j,’ knocking it out.

My ‘d’ fired at ‘i’ but missed.

I moved my remaining tanks a few spaces toward the objective, seeking cover in dense terrain when possible. Three Russians fired on me on their turn, scoring one hit.

Turn 2 start

An oddity in Tanktics is that rotating in place costs 2 movement points per edge. So, for example, rotating 180 degrees costs 6 movement points. However, when a tank moves, it automatically orients itself in the direction of travel at no additional movement cost. This leads to an odd situation where rotating a north-facing tank 180 degrees so that it faces south costs 6 points, but moving it south into a clear hex only costs 2 and still causes it to face south.

My three front-facing tanks fired. My g knocked out their i, my e knocked out their g, and my d missed. I kept moving my other tanks, and on the computer’s turn, they did not fire.

Turn 3 start

Only e had a shot here, and missed. The rest inched toward the objective. Careless movement cost me; two of the enemy’s tanks blew up two of mine.

Turn 4 start

I had e fire again, and missed again, and the rest of the surviving tanks continued moving up, more careful this time to stick to dense terrain.

It paid off. The Russians didn’t see me, and at the start of the next turn, I had three in my sights.

Turn 5 start

Four of my five tanks could fire, but they all missed. The fifth, b, moved north into the depression.Once again, the Russians did not return fire, and on the next turn, I saw even more of them.

Turn 6 start

Here I noticed a tank in a place where I had left wreck in the last turn. Unlike the Germans, Russian wrecks get removed from the map, probably because the AI doesn’t know how to move around them.

My tanks all missed again, but the Russians didn’t return fire. This continued for a few turns, my undetected tanks either moving toward the objective or firing and missing, and the Russians inching closer as well.

The Russian numbers dwindled, but as the tanks got ever closer to the objective, and therefore to each other, they started spotting me and firing, knocking out another one of my Panzers. By turn 15, it was down to 4 vs. 4.

Turn 15 start

Panzers a, b, and f knocked out d and h, and e inched ever so closer.

Turn 16 start

4 vs. 2. I blew up n, missed o, and then o circled around the objective hex, where my tanks blew it up as it reached the southern-adjacent hex were l’s wreck was.

The scoring system doesn’t make a lot of sense. Points are awarded and deducted as follows:
  • 50 pts for owning the tank closest to the objective
  • 1 pt per hex moved toward the objective per tank
  • -1 pt per hex moved away from the objective per tank
  • -3 pts per turn
  • Variable points for destroying tanks

But… the game only ends in three possible ways, regardless of scenario. You lose all your tanks, you destroy all their tanks, or you quit in the middle. Only destruction of the enemy is a satisfying conclusion, which guarantees you the 50 points for having the objective, and the points from destroying tanks dwarf the points gained from hex advancement. The three-point penalty per turn is the only factor with an appreciable score difference, and in two of the scenarios, where you are defending the objective from a Russian advance, it makes even less sense that you’d be penalized if the enemy took a long time to reach the objective, while you waited to destroy them. Perhaps in a game with fewer tanks, the scoring system would have more variance from the little things, but playing with fewer tanks didn’t appeal.

In fact, I found replays of the other scenarios disappointing. In the “defend” scenarios, you’re encouraged to wait for the Russians to come to you, since you don’t know what direction they’ll come from, only that they are headed for the objective, and moving too far from the objective will cost you points. In the “attack” scenarios, even if you can punch a hole in a weak spot of the Russian gun line and break through to the objective, you’ll still need to search for and destroy the rest of them before the game ends, which is tedious to do safely and reckless to do quickly.

I also would have also liked just a bit more transparency in how the sighting and shooting mechanics work. The manual states that it’s purposefully obscure, and that they expect you to learn the same way real armor commanders did – by experience, trials, errors, and improvement. Fair enough intentions, but I’d still prefer to be met halfway. Terrain and distance clearly make a big difference, but I was never sure whether it was better to aim for the tank in the forest or the tank in the rough a bit further away. I tried to look at the BASIC code for clues, but found it mostly incomprehensible, despite only being a little over 200 lines long.

In spite of some strange design choices, and AI that can be charitably called unsophisticated, this is a very functional and complete personal computer wargame, clearly way ahead of its time in 1978, though a retrospective capsule review by CGW suggests it was a bit long in the tooth by 1981. Multiple contemporary disk-based wargames by SSI offered greater depth and on-screen graphics, as did Crawford’s own Eastern Front 1941, published the same year by Atari. I’m not scheduled to visit these games until my 1985 phase, when Phantasie and Balance of Power will show up on my whaling list, but I’m looking forward to it.


  1. Heya !

    I started my own blog on wargames and the first game had to be Tanktics :)

    A few comments :
    - I agree the point system does not make sense indeed, but there is something else you missed. The number of points given by tank destruction depends on the number of tanks destroyed (eg the first tank destroyed is only worth 10 points, the second 30 if I remember correctly). This means that for scenarios with very few tanks (let's say up to 4), it is very easy to find yourself in a situation where you killed every single hostile with minimal losses, but STILL lost on points because destroying the enemy tanks does not compensate the 50 points for the objective plus the 3points by turn for just being the human.

    - I haven't been able to find the BASIC code of the game, or maybe I did not understand how to open it. How did you do this ? On old magazines I found proposals to "improve" the code and I would like to test a couple for an interactive game I want to test

    1. Your wargame blog is something I've wished existed! Having just played through a bunch of SSI's early titles without really knowing what I was doing, you can bet I'll be following, and I'd love to hear your comments.

      Aboout the BASIC code... for the PET version, you can type "list" after loading the PRG file into VICE (PET). I can't remember if there's a good way to extract from the PRG, sorry. For the Apple version, though, get CiderPress. It can open DSK format images and read the BAS files in a modern UTF-friendly text format.

    2. Well, you covered a lof of the early wargames already.
      Some of them are 2 players, for instance Computer Bismark as you know. If you are interested, once I reach those games we can try two players and report each on our blog. It could be ... interesting.

      So far I have only 2 games, Computer Bismarck is coming next and could be a good game ; I don't believe Starfleet Orion is too much your cup of tea.

    3. It's a neat idea, but I think I've had my fill of these games. But good luck covering CB - I'm eager to read it!


Most popular posts