Prehistory XVI: Gamnim (Nim)

Release Date: September 3, 1970

Platform: PDP-10

Genre: Strategy

Developer(s): Unknown

Publisher(s): DECUS

Next up, we have Gamnim, or Nim as the game calls itself internally. This is another of the 3rd of September crew of games (of which there is a second Nim game). I've already covered one Nim game, the Polish Marienbad, though I couldn't play it, making this the first Nim video game to be played on OGC (and it definitely won't be the last).

Once again, the author is unknown. BASIC Computer Games suggests the game originated at Dartmouth College, but provides no author. A modified version would appear on the Sol-20 microcomputer in 1977, once again courtesy of Ralph E. Hopkins. That version appears to be where the Gamnim title comes from.

Completely free to set the game up the way you want.

I think this might be the first game with options to modify gameplay - the game setup is fully customisable! It asks first for the win condition you want to play by: either as taking the last object, or forcing your opponent into taking the last object. It might be a small detail taken for granted nowadays, but I think that's pretty neat compared to the text games I've played so far. Afterwards, it asks for the number of object piles you want to play with, and how many objects to be in each pile. Even better. BASIC Computer Games recommends the traditional setup of Nim, which is 4 piles with 7, 5, 3 and 1 objects respectively. So that's what I go with.

It's been a little while since I played Nim - I think the last time would've been the Fairchild Channel F's version, hunting RetroAchievements. Vaguely, I remember there being some trick to win concerning odd and even numbers - can't quite remember specifically.

Just a bit rusty... yeah, let's go with that. Rusty.

I lost the first game. I think I'm remembering the trick being to make sure the pile is on an odd number if you play with the traditional win condition (taking the last object). Anyway, I play a few more times, losing every single time while I try to remember how to win. 

Eventually I had to go to the internet to refresh myself on the Nim strategy. There's this whole thing about the "Nim sum" and making sure that it equals zero at the end of your turn. I found the article that I used previously, and it makes decent sense of it. Other articles I looked at were a bunch of gobbledygook that overcomplicated things wildly, bringing binary multiplication and all sorts of other weird mathematical stuff in that I, frankly, don't care to read.

The most important thing that linked article notes is that, with the traditional setup, the player going first is disadvantaged. If you're playing a computer that is programmed to play perfectly, this means you'll always lose going first. It's the reverse of Tic-Tac-Toe, in that sense.

Having reminded myself of these facts about Nim, I won a game with the traditional settings going second. The "Nim sum" rule works the same way regardless of win condition, so going second will always net a win in this version if you play perfectly. I know because I tried the "avoid last object" win condition, thinking this would reverse the rule, but I quickly found out it doesn't. 

It took me about 30 minutes until I decided to refresh myself on how to win Nim.

Winning a game with each win condition was good enough for me. So, I decided it was time to push the limits. I was wondering - how many piles can I have? And how high can the numbers go?

You'd have to be really keen on Nim to try this...

I discovered that you can have up to 100 piles of objects, with a maximum of 2000 objects in each. I cannot even begin to fathom why you'd want to play a game of Nim that massive, or how long it would take to complete (and I'm not intending to find out, either). It's there if you're an absolute die-hard Nim fan, though, and have a few hours to kill. That's all there is to Gamnim, so on to the scores.

Time Played: 40 minutes. I did something a little differently, recording my playthrough while simultaneously note-taking and researching Nim strategy, hence why this number is a little higher than previous games.

Difficulty: 5

It's really tricky to judge, since it's so heavily user-defined. Putting it in the middle is fair, based on that. Plus, you're going to have a hard time winning, since the computer always plays perfectly.

Game Design: 3

It's quite standard as a Nim simulation, and doesn't do anything apart from allow you to simulate various Nim scenarios. But, on that point, I do have to give the game props for allowing you so much choice in how you set up the game. No other game up to this point has allowed for customisation, so it gets bonus points for that. The other issue is that you can only play against a computer that plays perfectly; this would be perfect as a two-player game. It essentially means that, to make use of the customisation, you have to really enjoy Nim as a game already.

Controls: 9

Can't complain much, has those comma inputs, but is otherwise very simple.

Visual: 3

That paragraph of instructions is quite an ungainly mess. The game itself is presented quite plainly.

Functionality: 5

I had no issues with stability, nor were there any glitches.

Accessibility: 3

Fairly standard for a text-based game; helps that it's based on a simple game concept.

Fun Factor: 2

To be honest, it's not that enjoyable for me. I don't really care for Nim as it is, especially when I have to play against a perfect computer opponent. While the customisation options are neat, I don't see myself ever coming back to try them out. This would work much better as a two-player game.

That's a total score of 25/70 (35.71%), placing Gamnim in the E tier, between 1Queen and PDP 10 Timesharing Basketball. Sounds about right.

[Add.] Upon completing the rescore project, Gamnim took a bit of a hit to its score, losing 4 points down to 21/70 (30%)

Next up is Horserace. Joy. I love horse racing.

Don't forget - if you enjoy my blog, be sure to leave a comment and follow so you don't miss any updates!

Prehistory XV: Digits

Release Date: September 3, 1970

Platform: Unspecified mainframe

Genre: Simulation

Developer(s): Unknown

Publisher(s): DECUS

Telling you right now, this is going to be a quick one. This barely counts as a game, to be honest with you.

I had to skip over Apawam and declare it lost, as the only seemingly surviving version of the game is one modified by Steve North for the 1978 edition of BASIC Computer Games, and from what I read from the descriptions of both games, North's version isn't so much a modification as it is a complete overhaul. I'll be treating that as a separate game.

Digits is one of about half-a-dozen titles with the same release date - 3rd of September, 1970. I don't know where MobyGames gets these dates from, but I'd love to know (possibly the DECUS program library?). The original author is unknown; all that is known is that the program originated at Dartmouth College - specifically the Kiewit Computation Center at the college (a project spearheaded by John Kemeny and Thomas Kurtz - the inventors of BASIC, which Digits was programmed for).

Digits presents a rather odd concept: you are to come up with a random string of 30 numbers between 0 and 2, and the computer will try to guess your numbers using something BASIC Computer Games called "pattern recognition techniques." These numbers are inputted 10 at a time, and to win, the computer must make less than 10 correct guesses (which the game claims it should be able to do "by pure luck"). That's the game. It's beyond simple.

The game's instructions. It's cute that it asks you to write on paper. Who does that these days?

I find it amusing that the author of the game writes the computer to speak in first person. It's endearing, somewhat. For my first attempt, I thought I'd have a laugh. I got my phone's notepad out, and randomly smashed the 0, 1 and 2 keys until I had at least 30 numbers. Inputting them was actually a bit annoying, as the game doesn't tell you how it wants the numbers inputted. The required syntax is to type the numbers in, with a comma after each number, but no spaces.

I confused the poor thing.

Eventually, after looking at the game's page in BASIC Computer Games, I managed to get the numbers in correctly.

There we go.

So it managed 4 correct guesses on that pass. Not the best start when I want it to guess less than 10.

Ugh, and there's still one set to go.

It got 6 right on the second set, so there goes that game.

My last set was an improvement, at least.

As a consolation, it only scored 2 correct guesses on the third and final set. I noticed that it didn't select 0 as a guess at all that round. I could've selected 0 30 times in a row and it wouldn't have got one right...

I decided I had to win, so another round was required. Taking what I learned from that first game, I took a different approach to choosing my numbers - I selected longer strings of the same number, interspersed with solitary numbers along the way. It seems to me like the computer prefers to guess long strings of one number.

I'll take that as a small victory.

 My strategy appears to be paying off. It only guessed 3 right in the first round this time.

Ha! Some "pattern recognition techniques" these are...

Oh dear, the computer had a bit of a disaster on the second set. Only 1 correct. I noticed from the first round that it seemed to be preferring 2 this game, so I tried to avoid overusing it.

Maybe I should've done the all zeros thing.

I went a little silly on the last set, doing zero 5 times in a row. You'd think it'd catch on after maybe the third one, but it missed the whole lot! It got 4 of the last 5 right, but that only got it up to 8 correct guesses, so I win. There's nothing else to say, so score time.

Time Played: 6 minutes. That's how long the two rounds took.

Difficulty: 1

It's a guessing game, do you really expect any challenge? You just plonk in numbers and hope to computer doesn't guess them. All luck, no skill.

Game Design: 0

Barely a game at all; it's just a lucky guess simulator, so there is literally nothing to say about it.

Controls: 7

I'm knocking off points for the weird syntax stuff with commas, and because the game doesn't tell you that that's what you need to do. Otherwise, it's easy.

Visual: 5

It's presented decently. All the text is neatly formatted, no typos, and the guess spreadsheet is clear. The game is very polite, always saying please when asking for my numbers, so I can at least give it that.

Functionality: 4

Again, that syntax thing is silly, but it works fine otherwise.

Accessibility: 3

It's one of the more accessible text games thus far, due to its simpler-than-simple concept.

Fun Factor: 1

Yeah, there's nothing here for me. Plonk in some numbers and hope for the best. Next.

This gives Digits a score of 20. Percentage wise, it's - miraculously - not the worst game on the tier list. At 28.57%, it's above Qubic in the E-tier. While I hated Qubic, I'm not sure I'd agree with that result. Although I'd rather play neither, I think I would play Qubic ahead of this any day of the week. Perhaps a revision of Qubic may be needed in the future?

[Add.] Upon completing the rescore project, this game managed to get even worse, dropping to a 19/70 (27.14%).

In any case, I'm glad this one's over and done with. Gamnim is up next - one of many, many, many versions of Nim released across the 70s.

Don't forget - if you enjoy my blog, be sure to leave a comment and follow so you don't miss any updates!