A Brief History of... Gomoku

For the next A Brief History Of article, we're actually taking our first real adventure over to Japan. Bet you didn't think the first Japanese game I'd be covering was a board game, eh? It's funny how things play out when tackling gaming chronologically.

Of course, Japan is going to be a massive player - nay, the defining player in the video game industry come the late 1970s-onward, thanks largely in part to Taito, Namco, and Nintendo. But, as of writing this, we're in 1973. Taito and Sega are busy copying Pong, Namco is merely a distributor of Atari games in Japan, and Nintendo is barely on the map. Really, all the action is taking place in the United States.

But, as we can see from this article, and my most recent Gomoko article, that didn't stop Japan from being an influence prior to their earliest arcade hits. Developers just had to look to different source material, and they found it in the form of board games. As we saw in the Gomoko article, the earliest of the Far-Eastern board games to get the computer game treatment that we know of was Go. That was in 1968, written by Albert Lindsey Zobrist. Zobrist is an important figure in computer science, as he invented a hashing system, suitably named after him (Zobrist hashing), that is used in computerised board games like Go and Chess. We're not going to go too in depth on information relating to Go in this article, however - I'll save that for a separate Go article.

That being said, it's extremely difficult to separate the history of Go from the history of Gomoku, as the games are inextricably linked. For Gomoku is simply a variant of Go that gained a whole life unto itself after arriving in Japan. Go was created in ancient China, with sources dating it being between 2,500 and 4,000 years old! It's still being played today, which makes it one of the oldest - if not the oldest - extant board game still actively played today.

Go-Go-Gomoku

It's entirely unclear how and when Gomoku developed out of Go as a separate variant. Nobody really seems to know, from what I can tell. The best sources I could find merely suggest that the two developed alongside each other, with Gomoku - or Wuziqi, as it's known in Chinese - being a simpler game using the same board and pieces that was designed to be more easily played by children. Other sources imply that it might've actually been invented in Japan... What is clear to me is that the history of Gomoku is a complete mess.

Fortunately, the core difference in rules between Go and Gomoku is abundantly clear. Go is about territory control and capturing enemy pieces, whereas Gomoku is an alignment game - basically a larger scale version of Tic-Tac-Toe or Connect Four (and, consequently, far more interesting.) Instead of three or four in a row, Gomoku goes for five in a row. *Connect Five, if you will. The name "Gomoku" itself refers to the game's five-in-a-row alignment rule. The full name of the game is Gomoku-Narabe (五目並べ) - Go（五）meaning "five," Moku (目) meaning "item", and Narabe (並べ) meaning "line up;" literally translated "five items line up."

Back to the mess that is Gomoku's history, we now turn to how the game arrived in Japan. This isn't much clearer than the development of Gomoku out of Go. In fact, there are two dates floating around for the arrival of Go and Gomoku in Japan. Great. Reminds me of the dating for the Biblical Exodus, as we have an "early" date and a "late" date to consider. For reference, when I say Gomoku, I'm referring to both Go and Gomoku. The two don't really become separate entities in Japan until much later on. Plus, it saves me the hassle of constantly writing "Go and Gomoku" all the time.

We'll start with the early date. These sources state that Gomoku arrived in Japan via Chinese emigrants in ~270 BC. That's a damn long time ago, especially compared to the late date. In this chronology, Gomoku arrived under a different name, Kakugo (覚悟), a word that translates to "readiness." However, there's no actual evidence I could find to support this early date. One source cites a book from ~100 AD that references the game, but there's no evidence I could find of a text like that actually existing.

The far more likely-to-be-true date for the arrival of Gomoku in Japan is the late date of 754 AD. The story goes that it was brought back to Japan from China by a Japanese ambassador named Kibi no Makibi, otherwise known as Kibi Daijin (Minister Kibi). Born into a noble family, he was sent as part of an envoy to Japan in 717, staying there for 18 years, absorbing all the he could from the China of the Tang Dynasty. After returning, he was heavily involved in the governance of Japan under several emperors, before being sent back to China in 752. It was from this second trip, concluding in 754, that Kibi brought Gomoku over to Japan. It's actually from Japan where we get the name Go from, despite the game's Chinese origin. It's derived from the word igo (囲碁). The second of those Kanji is quite literally the Kanji for Go. It became such a popular game in Japan that it got its own Kanji all for itself. The Chinese name for Go is Weiqi (literally "surrounding game"), and Gomoku's Chinese name is Wuziqi ("five piece game").

Artist's rendition of Kibi no Makibi.

Before I continue on with Japan's history with Gomoku, I do want to briefly mention that it also spread to Korea, likely a century or two, at least, before it got to Japan. Go in Korea is known as Baduk, with the Gomoku variant being known as Omok ("five stones"). It was present in Korea in the 7th century AD at least, and one story purports it to have been brought to Korea by a Chinese sage contemporaneous with Confucius (5th-6th century AD). Again, there's blurring of lines here, as whether it was just Go that was brought over, or if Gomoku was included, or if Gomoku was brought to Korea from Japan at a later date... thinking about it that way turns it all into a circular mess that makes my head hurt. Moving on.

Back to Japan, we are going to be jumping ahead a bit in time to pick up the story of Gomoku.

How much is "a bit?" Oh... about a thousand years.

This is unfortunately where we really run hard into the problem I mentioned earlier about the inextricable link between Go and Gomoku. For Japan, for the next 1,000 years, Gomoku's history is Go's history. All the attention is on Go, and the first story to even mention Gomoku as its own separate entity takes place in the 18th century. That's right - no writings, no mentions, not one peep about Gomoku for a thousand years.

Breaking Free

As I don't want to dive deep into Go's history here, I'll provide a brief overview, as the context is needed to know how Gomoku got to the point of unhooking itself from Go.

For about 500 years, Go found a comfortable little niche of prestige in the halls of royalty and nobility in Japan. It became a highly regarded pastime that also became quite competitive. One story, recounted by Edward Lasker in his 1960 book Go and Go-Moku: The Oriental Board Games, tells of a Japanese prince visiting China, about 100 years after the game's introduction to Japan, for the sole purpose of challenging China's best Go players.

Eventually, around the 13th-14th century, Go managed to escape the halls of nobility, and found popularity first amongst the military classes, and then the general public. Again, nothing is said about Gomoku. Go's popularity continued to boom, though. So much so, that once the Edo period commenced under the Tokugawa Shogunate, Go academies were being created for the purpose of training players in the now-ancient and revered art of Go.

Our first story of Gomoku comes now in the 18th century, as previously mentioned. The story goes that the spread of Gomoku as a separate game from Go can be attributed to a merchant named Kuwanaya Takeemon, who was famous for his skill at the game. He was a 10th generation oil merchant who regularly visited the noble Nijo family, and it's reported that he spread the knowledge of Gomoku through his business travels.

By the end of the 19th century, Gomoku had finally attained its own identity separate from its parent game. It had a book dedicated to it, the Gomoku Jōseki Collection, dated to 1856. It also began to go global, attracting the attention of the British Empire, with the Brits writing of the game later into the century. However, they didn't call it Gomoku - they instead called it Go-Bang or Goban. Eventually it would be distributed across the West in the 20th century under the names Spoil Five and Five in a Row.

However, we've yet to address the most significant event in Gomoku's history - Renju.

Renju

As Gomoku increased in popularity over the 19th century, many efforts were undertaken to try and "balance" the game. It had been discovered that, in what is called "freestyle Gomoku" - starting with a blank board - the player going first (black) is significantly advantaged. It's at the point of being a "solved game," if you knew how to play perfectly, you would never lose. So, various starting setups and rules were experimented with to try and find a nice balance where both black and white had about an equal chance of winning.

The problem was, that this version of Gomoku didn't have a name. There were a whole bunch of names being used, and it got to the point where nobody really seemed to like any of them very much.

Enter Ruikou Kuroiwa, a journalist and founder of the Japanese newspaper, Yorozu Chouhou. He decided to take it upon himself to find a new, respectable name for Gomoku that would remove the stigma attached to the old names. What he came up with was Renju, which literally means "connected pearls." The name was announced in Yorozu Chouhou on December 6, 1899, and has stuck ever since as the name for the globally recognised, professional variant of Gomoku.

Renju began to be played competitively further across the world as the 20th century progressed. Eventually, in 1988, the Renju International Federation (RIF) was founded, for the purpose of holding world championships. The inaugural championship was held in 1989, appropriately, in Kyoto. Since 2009, the championships have been run on a biannual basis. While there are other noteworthy variants of Gomoku played all around the world, none took off quite like Renju.

Video Game Gomoku

Looking into the digital world of Gomoku, it makes for reading quite similar to Hangman, in all honesty. A good amount of action in the 70s and 80s, and dropping off in interest afterwards, with few games being produced.

The first computer Gomoku appeared in 1972 as a BASIC, text-based game, written by Greg Thagard. This rendition of Gomoku plays a 9x9 board. In 1981, it was ported to the Sol-20 microcomputer, and is still playable today. The computer is slow, taking up to a minute to decide its moves, but is competent at the game...

Thagard's Gomoku, Sol-20 port.

...unlike 1973's Gomoko, by Peter Sessions. I've already waxed lyrical at the massive failure that that game's computer opponent is, so I'll spare you the details on that here. What this game introduces as an improvement on Thagard's game is a choice of board size, from the smallest 7x7 board, up to the Go-standard 19x19 board.

Quick reminder of how incompetent Gomoko is.

1976 saw three new Gomoku programs appear. However, only one of them was actually a game. That one ran on the Intel 8080 microprocessor, and was published by RBB Software Products. No credited author is known of. As for the other two... well, those were Gomoku-playing programs that participated in computer Gomoku competitions. Yes, just like how there were computer Chess competitions, where the brainiest boffins and computer scientists tried to develop the smartest, bestest Chess-playing program they could, there were likewise those who did the same with Gomoku. One of these, developed by D. Walden, managed third place in the 1976 championship.

1979 was the year where Gomoku saw the most attention from game developers, with a whopping five games releasing that year for all the brand-spanking new microcomputers that arrived only a year or two prior. The Apple II, Atari 8-bit line, TRS-80, Altair 8800 and Commodore PET all received versions of the game.

The 1980s saw the first Renju computer game appear - this one from a Japanese developer, released for the Apple II. 1980-83 saw several more Gomoku and Renju games appear - this time not limited to home computers, but there was also an arcade game, Gomoku Narabe, released in 1982, and Gomoku Narabe Renju by Nintendo for the Famicom/NES in 1983.

Of course, Nintendo had to make an appearance.

After that, the interest in video game Gomoku fell off hard. Only one game in 1988, and a mere handful across the 1990s and 2000s. And that brings us to the state of Gomoku video games today. Mind you, it's not much of a state at all. More of a ghost town, really. The most recent Gomoku game, Let's Go Gobang was, according to MobyGames, released in 2021 as a digital download on Windows and Nintendo Switch. Nothing else in five years (at the time of writing this article.) But, there are at least a handful of browser-based versions available to freely play that a quick Google search will bring up.

Endgame

Here ends the lesson on the history of Gomoku. As you've seen, it was so deeply intertwined with Go for much of its existence that, despite the evident antiquity of Gomoku, it has little historical evidence to show for it prior to the 18th century.

But, it finally managed to break free from the shadow of Go and become a internationally-regarded, competitive board game in its own right. Especially after the invention of Renju. The most recent world championship from 2025 was fully live-streamed on YouTube, for those curious on what the state of Gomoku as a sport looks like today. They stream six games simultaneously, so it can be a little tricky to keep track of everything, but it's all there.

The troubles with the historical evidence, and working across sources in two languages made this a more difficult assignment than this series' debut article on Hangman. On the flipside, I got to learn about a whole new competitive sport that I never knew existed before. That's the beauty of doing these broader, historical articles - I always learn something new. That makes the work tremendously rewarding to do.

What's next for this series depends on whatever project I complete next. Could be Tic-Tac-Toe, could be Hexapawn, or it could be something completely different. My workflow is still in a state of flux at the moment, with all the changes going on around here. So it could be a while.

#042 - Gomoko: Japanese Connect Four (Five)

[Ed. I've spent an awful lot of effort in changing up the style of my videos into something with a bit more production value. Well, as much "production value" as my self-taught video editing skills can provide. This video is more of a "mini-documentary," and mixes produced content with live gameplay. Hence why it's taken so long to complete. Let me know what you think!]

Release Date: July 1973

Platform: Mainframe (BASIC type-in)

Genre: Board Game

Developer(s): Peter Sessions

Publisher(s): Digital Equipment Corporation

Here's a curious game - one from the People's Computer Company, yet is exclusive to DEC and David Ahl's 101 BASIC Computer Games. I scoured the first several years of PCC magazines, and no mention of this game was to be found anywhere.

History

There were very few early video games based on distinctly Eastern board games around up to 1973. The only ones I can see on my list are "missing" games. One is a version of Go from 1968, and another is an earlier version of this game - Gomoku (Sessions misspelled the title), from 1972. I'm unsure whether these are recoverable in their earliest format (the '72 Gomoku was ported to the Sol-20 in 1981, so we'll get to see it then, at least.) We also have Awari, which I've already covered, but that game's origins are centred more around the Near East and Africa, not the Orient. This, for now, makes Gomoko the earliest extant video game based on a game from the Far East.

By quick way of reminder, Peter Sessions is a game author we've seen a couple of times before in my 1973 coverage. He also wrote Reverse and The Game of Chomp for PCC, before leaving the publication the same year (perhaps why Gomoko never featured in the magazines...) Gomoko here is his final credited work, according to MobyGames. As far as specific information is concerned about Gomoko, there unfortunately isn't anything beyond what we're given in 101 BASIC Games.

These cartoons really are something, aren't they?

As for Peter Sessions, after he completed his final game, he remained in the computing industry for many years. He also delved into his other chief interest: music, writing a book on music theory, entitled The Functions of Chords: For Pop, Jazz, and Modern Styles. Very nice.

The Game

Sessions' final game comes with a welcome message, "Welcome to the Oriental game of Gomoko. What is Gomoko, or, correctly spelled, Gomoku? And why is it Oriental? Well, I won't go into too much detail here - I'll save that for my Brief History of Gomoku - but what I will tell you is that it's a Japanese board game that dates back several centuries, and also spread to neighbouring China and Korea. The objective is simple - line up five stones of your colour (black or white) in a row (horizontally, vertically or diagonally) on the board to win. Why it's not four-in-a-row like, say, Connect Four relates directly to the Japanese superstition surrounding the number four. In the language, one word for the number four is the same word for death. So instead we have what could be basically considered Connect Five.

The rules.

Typically, Gomoku is either played on a 15x15 or a 19x19 board, with the black and white stone pieces of Go, a related Oriental board game. Sessions' game allows you to choose a board size between 7x7 and 19x19. According to 101 BASIC Games' notes, 7x7 is the smallest size "that can be used for a meaningful game." It would be rather difficult to make five in a row on a 4x4 board, wouldn't it? It also warns that the computer isn't very good, nor does it keep track of who has won. My playing experience can confirm that the computer is quite terrible.

All seems fairly regular here...

I played my first round with a standard 15x15 board, and the computer's moves were... well, they were certainly moves, I can say that. The game notes do say that the computer's moves may surprise, though I'm not sure if randomly placing pieces on the outskirts of the board is the surprise Peter Sessions had in mind.

...until the computer pulls out this.

Interestingly, when I played a second game on the smallest board (7x7), I played almost the same strategy, and the computer responded in almost identical manner. It seems to not know what to do if you start your pieces in the centre of the board. For a third round, I tried developing my pieces in a different direction, and the computer responded in a slightly more rational manner - although it was still highly incompetent and didn't understand the concept of blocking your opponent's progress. I find this rather disappointing, as there have been some computer opponents that have been relatively decent in recent times, and ones that even learn from their human opponents.

Scores

You'll likely be expecting a low score for Gomoko, and you'd be correct in that expectation.

Time Played: 5 minutes

Difficulty: 1/10 (Brain-dead)
You don't have to think a whole lot when the computer makes nonsensical moves that allow for you to easily win every time.

Gameplay: 1
I can only give it a 1 for gameplay because it's rather broken in its execution. Yes, there is a range of board sizes to choose from, but with an AI so incompetent it hardly matters what you do.

Controls: 5
It's very standard for the type of game. The co-ordinate system is simple and as expected.

Visual: 1
There's nothing necessarily wrong with it visually - it does what you'd expect it to for the type of game it is. I do find the larger board size a little overwhelming and harsh on the eyes with how many zeroes are present on screen at once.

Functionality: 5
I think the AI is broken, but that's a design choice, not an error in the code (or just poor programming.) The game itself runs well, and I didn't run into any problems other than it not being fun.

Accessibility: 3
This rendition of Gomoku is relatively easy to understand for a text-based adaptation.

Fun Factor: 1
The times when the computer isn't being completely broken provide the faintest glimmers of rewarding gameplay, but those are very few and very far between. It isn't fun to always win.

Overall: 16

Gomoko earns a pitiful score of 16. This puts it in league with some less-than-stellar company in the F tier, alongside the likes of Button, Button, Who's Got the Button? and Stars. It's not a game I'd recommend playing, not even out of historical interest. Some bad games are entertaining on account of the hilarity of their terribleness, while others are just bad and hold no value at all. Gomoko is one of the latter.

You may notice of you check the Master List, that I've updated the Tier List. I've changed the score lines and tier thresholds to make it work with a 100-point scale. Very few actual position changes, but the F-tier is a whole lot smaller now. Gomoko still manages to find itself there, even after the changes.

Up Next: A Brief History of Gomoku.

A Personal Note

I know I said that the article on the Cathode-Ray Tube Amusement Device was coming up next. It's since been delayed by a couple of weeks. I want to quickly apologise for that, and simply admit that I heavily underestimated the amount of work that project required. Even a couple of years into this blog, and I'm still learning things - particularly about work flow and ensuring a consistent stream of content. This article is quite heavy on research, and the accompanying video is quite ambitious for me, and in a very different style to what I've been doing. A lot of growing pains.

On a personal level, things have not been well either, and that has impacted my motivation to work on blogs and videos. It becomes increasingly hard to not feel depressed and guilty about performance or not doing the right things when one is surrounded by people that disapprove of everything you do and constantly question your character and faith. I don't typically like to share a whole lot about my personal life, but I'll admit that this year has been one of the most difficult of my entire life. But, be reassured that I am making changes, and to surround myself with people who will have a positive impact on me going forward.

#035: Checkers - King Me! (Forever)

Realised thumbnails might help the videos... bear with me as I figure out a style I like.

Release Date: July 1973

Platform: Mainframe (BASIC Type-In)

Genre: Board Game

Developer(s): Alan J. Segal

Publisher(s): Digital Equipment Corporation

It's been a while since since I talked about the board game Checkers... or Draughts, if you prefer. In fact, it was one of the first games I ever wrote about - what I consider to be the first true video game ever created - back when computers were the size of houses and programming then was akin to the dark arts.

#029: Awari - Mainframe Mancala

Release Date: July 1973

Platform: Mainframe (BASIC type-in)

Genre: Board Game

Developer(s): Geoff Wyvill

Publisher(s): Digital Equipment Corporation

Sorry that it's been a while since the last article. I haven't had the best of times since then, and I think I've experienced significant burn-out. I don't want to go into more detail than that, as I prefer to keep personal matters just that - personal. Just know that the blog has still been on my mind, but with all that's happened recently, I just haven't had the energy to work on it until now. I have a couple of weeks off of work, and some of that time and energy I'll direct towards getting back into the blog.

Now, on with the show.

It's fascinating how some of the world's most well-known and popular board games have ancient roots. Chess has been around for centuries, and Backgammon for millennia. Another ancient game still played today is Mancala. It's a board game with many, many regional variations played all across the world. One of the more modern variants is called Kalah. This version was created in the USA back in the 1950s.

If you've ever seen a game board like this, you've played Mancala. Source: https://www.thesprucecrafts.com

The basic objective of Mancala and all its variants is that two players compete on a special board to collect more stones (or seeds) than the other. Each player has six "houses" on their side of the board, which are circular indentations, with two larger elliptical indentations on the edges, called "stores". Players take turns "sowing" their stones into their houses and, based on the variant rules, players can capture their opponents stones. In Kalah specifically, a player can capture the stones in the house immediately opposite if their last stone is sown in an empty house. They also can get an extra move if the last stone lands in their store. The game ends when one player runs out of stones.

Why am I bringing all this up? And why am I talking about Kalah specifically? Well, that's what today's game is about. Awari is a digital, text-based implementation of the Kalah variant of Mancala. It's a bit odd that's it's named Awari, which is an alternative name for Oware, the variant played across Africa and the Caribbean, which has very different rules from Kalah. 101 BASIC Games ('78 edition) even describes Awari as "an ancient African game," despite using the Kalah rules. I did read up on those rules, but I'll be honest in saying that I really don't understand them at all. Oware is far more complex than Kalah. Perhaps just because I'm more of a visual/doing learner, simply reading the rules often doesn't get the point across to me. All that to say that I won't be explaining the rule differences between Kalah and Oware here because I don't understand them.

Technically, this isn't Awari, nor is it ancient (unless you think the 1950s are "ancient.")

Now, Kalah also made its way from the United States over to the United Kingdom, which is important because that's where Awari was created. It's the first British-made game that's appeared for quite some time, too. The most recent one on my Master List prior to Awari is a missing version of Bulls & Cows called Moo, from 1968. The last British game I actually wrote about is the second game ever made - OXO, from 1952! That was in only my second-ever article from March last year!

Awari's location can be further centred, courtesy of the information provided in 101 BASIC Computer Games. Bradford University is Awari's origin point, up north in Yorkshire, England. A single individual authored the game, that being Geoff Wyvill. Awari is the first of his games we've come across; he has a couple more games on my list, and even his brother, Brian, has one. Wyvill's information is surprisingly easy to come across online, as he's currently Professor Emeritus of the University of Otago's (that's in New Zealand, for those who don't know) School of Computing. During his time at Bradford, he earned a PhD in Artificial Intelligence, and also possesses a BA in Physics from Oxford. Wyvill notes on his self-maintained website that he completed his PhD in 1978, and - given his majoring in AI - it's likely that Awari, featuring a computer opponent, was developed as part of his studies.

Wyvill's attempt at explaining the rules, part 1.

I came at this game with a bit of trepidation, as I wasn't fully understanding how the game worked from the explanation given in 101 BASIC Games. Once I played a full round, though, the general game flow began to clear up for me. 

Part 2. I think my explanation is better. But I'm biased, after all.

It goes like this - you have the field set up as standard for Mancala and all its variants: 3 stones (apparently 4 is the standard, according to Wikipedia) in each of the 6 pits for each player. Each turn, the player will select which pit they want to sow from. The stones from that pit will then be sown, one stone per pit, in a counter-clockwise direction from the selected pit. The selected pit is left empty, a stone is not sown in it. If the last stone sown lands in the player's store (in-game called "home"), then they get another turn.

For taking stones, the way that works is that, if the last stone is sown in an empty pit, then it and all the stones from the pit opposite it are taken an placed in the player's home. Once a stone is in the player's home, it cannot be removed. This loop continues until one player has no more stones on their side of the board, and whoever has the most stones in their home at that stage wins. Part of the reason I write out an explanation is to try and explain the game to myself in a way that makes sense to me. Hopefully it also makes sense to you.

As for how you select a pit in game, this is rather simple. Each pit on your side is numbered from 1-6, in that order, just like on your keyboard. That actually makes it really easy to remember which pit is which.

No instructions. Figure it out yourself. Ouch.

I didn't record my first attempt at Awari, as I just wanted to get to grips with the rules before I had a proper go at it. Wyvill was quite merciless to the new player, bucking the usual trend and providing no in-game instructions. For a game like this, it could really use them, too. Somehow, despite having no idea what I was doing, I won this first attempt, 15-14. The only strategic idea I had was to get two moves at the start, and then I just winged it after that.

My thoughts regarding strategy after that were that it seemed to make sense to try and keep my number of stones per pit low, and to sow the pits closer to my home first, giving me options to take stones and react to whatever the computer does. I know there's a "mathematical best" way to win, like with Nim variants, but I preferred to remain unspoiled on that for now. I think it's more rewarding in most cases to figure out strategies on my own. Also like Nim, the player going first is at an advantage, and can be unbeatable if they know what they're doing.

Action shot. It's hard to do play-by-play commentary for Awari.

The second game was far more enjoying and satisfying, having had more clarity around my strategy. I found a useful ploy to build up to 3 stones in pit 5, all of which could be claimed in one turn thanks to how the game rules work. I used that several times through the game, and ended up winning the round 19-15, after several sharp swings and lead changes.

Victory. You can also sort-of-see my pit 5 strategy at work.

Apparently, according to Wyvill, the computer is designed to learn to play better as you go on. This does appear to be the case, as I played several games in a row to test, and that ended up with a 2-2 draw, with the computer beating me by 1 point, then soundly by 9, only for me to learn and come back to throttle him by 10 in the last game. All that's included in the video up top. I think near the end I realised that it's important to not only count your own moves, but what your opponent might do in response. Focusing on getting double moves and blocking let me have such a strong win in that last game.

And with that, let's move onto the scores.

Time Played: 20 minutes

Difficulty: 4 (Mild)

This is one of the simpler variants of Mancala, after all. The computer also evidently isn't too hard to beat, as I beat him having no earthly idea what I was doing. The computer does learn and get more difficult to beat with each passing game, though in my experience it didn't become overbearingly challenging.

Gameplay: 5

Awari is decent in the sense that it retains all the distinctives of Kalah. I wouldn't say it's a terribly in-depth game, but there's a small amount of strategic depth available. One has to consider not only their own moves, but their opponent's, and there's also limited opportunities to work both offensively and defensively, with the extra move being an incentive and a tool for further strategic play.

Controls: 6

It's set up quite nicely, having the pits' numbers lined up the same as a regular keyboard. Makes easy remembering which pit is which.

Visual: 2

I'll give Awari a point for having a fairly clear representation of a Mancala game board.

Functionality: 5

I didn't run into any technical issues.

Accessibility: 2

I must confess, I did procrastinate a little on playing Awari, as I felt confused and put off by the written explanations of the game rules. Didn't make a whole lot of sense, and made approaching the game challenging.

Fun Factor: 5

My subjective thoughts don't stray too far from the gameplay score on this one. I feel kind of indifferent about it, despite there being some strategic depth to enjoy. The computer's learning to play better is an incentive to keep trying, but it's limited in its appeal. I think in this case playing the real thing is preferred.

Tallying up the numbers gives Awari a total of 25. It's a higher E-tier finish, which I think is adequately representative of my thoughts. It's a decent go at making a digital version of a classic board game for the time. "For the time" being the key phrase there. It's still very, very primitive and limited in what it can offer. For reference, as of this article Awari is in 19th spot on the Tier List.

Next time we take to the skies and help with the war effort by blowing stuff up. Sounds like good fun!

Question of the day: Here's something fun for this article. Do you have a version of this game where you live? Have you played it? If so, let us know in the comments how it differs from what I played here.

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

#021: Bagels 2 - Beyond Bagels

Release Date: September, 1973

Platform: Mainframe (BASIC Type-In)

Genre: Puzzle

Developer(s): David "Dave of the Caves" Kaufman [New discovery!]

Publisher(s): People's Computer Company

This is an odd one. I almost didn't want to include it due to the lack of information, and lack of difference to its prequel. As a result, this'll be a very quick article. Which is good, in a sense, as I can work on it and not be behind on content whilst I'm in the process of moving house. Just as a heads up, there might be a slowdown of content over the next week or so.

Beyond Bagels (also called Bagels 2) is kind of a mystery. It appears in a 1974 PCC type-in game compilation simply titled PCC Games. No author is attached to it, nor does it appear in any issue of the main PCC newsletter. The only description given is the program's length: 1015 words. Afterwards it disappears, never to be seen again. I'm not sure where MobyGames gets it September '73 release date from. PCC Games was published after the January '74 edition of the PCC newsletter, so it was likely in between the January and March editions, otherwise March '74 games would've been included.

Beyond Bagels' only appearance for PCC.

The only other bit of information we have on the game's background comes from MobyGames, where it explains that Beyond Bagels is merely a modification of the original Bagels program written by Diane Resek and Pete Rowe back in 1972. Beyond Bagels changes the clue system from the Pico-Fermi-Bagels system to a point-based system. You get a point for each number guessed correctly, and another point if a number is also in the correct space, for a total of 6 points. The only other change is limiting the amount of chances you have to guess the number. The unmodified guess limit is 20, which is quite generous.

[Ed. I've discovered who the author of Beyond Bagels is! Turns out it's Dave "Dave of the Caves" Kaufman. I discovered this while working on the Lost in the Caves article. Kaufman claims authorship of Beyond Bagels in the May 1974 edition of the PCC newsletter.]

Once again, I'm having to turn to the Internet Archive for playing this one. Most of the PCC games are fortunately preserved there through MS-DOS ports. I had to do this one a little differently, as DOSBox has this particularly annoying quirk of not allowing you to scroll up. Beyond Bagels' instructions are longer than the screen, meaning the first line or two get cut off, and I can't scroll up to view them. What I did to circumvent this was download the game and run it through my own DOSBox with the cycle speed set to 10. This forces the game to load far more slowly, giving time to read all the instructions. I know I could use the "modern" version provided, but like I've said before, I'm a purist and want to play games as originally intended. Don't get me started on the state of modern gaming and endless patches.

I managed to get a snap of the full instructions.

I suppose the main question to ask with Beyond Bagels is: do the changes make any difference? Honestly, not really. While the clue system is different, it doesn't change the guessing method. The 123, 234, 345 system still works very well here, and I was able to guess numbers in as few as 6 guesses. [I got down to 5 guesses in the video.]

Sometimes, you just get lucky.

That's all the game has to offer. Like its prequel, there's no alternate game modes, unless you want to modify the code yourself. Oddly, I found myself enjoying this more than Bagels, from what I can remember of it.

Let's see how the two compare.

Time Played: 10 minutes

Difficulty: 2/10 (Easy)

Beyond Bagels still presents the most rudimentary form of Bulls & Cows. I only say this is slightly harder than Bagels is due to the guess limit and a different clue system.

Gameplay: 2

This and Bagels are interchangeable, really. The changes made make no discernable difference in gameplay. It still doesn't tell you that numbers don't repeat unless you try putting in an answer with repeated numbers. One oddity I found is that Beyond Bagels won't let you type a 0 as the first number of a guess. Weird.

Controls: 5

As standard for a text-based game of this ilk. Single digit inputs.

Visual: 5

Also standard. It's formatted neatly, with no typos.

Functionality: 5

Unlike Bagels, this one doesn't have a broken "play again?" question that boots you out of the game, even if you say yes.

Accessibility: 3

Using a point-based system maybe makes this game a little easier to understand; more numbers and less words. If so, the difference is negligible.

Fun Factor: 3

Despite being almost the same game as Bagels, I found Beyond Bagels slightly more enjoyable. Honestly, it could just be that I'm in a better mood when writing this; both games are that similar that it could swing the opposite direction on any given day.

Beyond Bagels does manage to beat out its prequel by two points, scoring a 23. It's only a few places up in the E-tier, managing 19th overall. There's not much reason to play this, in all honestly.

Next up, we have four games in a row by the same author. These ones I'm genuinely curious about and looking forward to trying them out.

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

#019: Solitaire Checker Puzzle - Digital Peg Solitaire

Release Date: June 25, 1973

Platform: Mainframe

Genre: Puzzle

Developer(s): David H. Ahl

Publisher(s): Digital Equipment Corporation

Solitaire Checker Puzzle, or One Check as it's called in the 1978 edition of BASIC Computer Games, actually originates from the original edition of that very same book. It was created by the book's author, David Ahl. I'm going to refer to it as One Check through this article simply because it's less characters I have to type.

I'm not sure where Ahl's idea for the game came from, as there's no information online I can find explaining the origin. His writeup for the game in the book doesn't give much info, either. In fact, it's identical across both editions of the book. 

Apparently playing this game turns you into a wizard.

From what I can observe, the game concept itself bears a lot of resemblance to those peg boards that I used to see around my parents' and grandparents' houses. Peg Solitaire is what it's called. It has the exact same concept of leapfrogging pieces to remove them, only in physical form and with different patterns. My suspicion is that's what Ahl may have been referencing with One Check. This Peg Solitaire rabbit hole goes a lot deeper than you might expect, as we'll soon discover...

Presented for comparison.

The key difference with those, compared to One Check, is that you could leapfrog from any cardinal direction. Left, right, up or down. You can't do that in this game - it only allows diagonal moves, as in regular Checkers. That makes things a bit more tricky. There's a total of 48 checkers on the standard-size Checkers board (64 squares), with the claimed maximum possibility of removing 47 of them. According to Ahl, removing over 40 is a challenge. Challenge accepted.

The game doesn't even bother asking this time - the instructions are slammed right in front of your face straight up. It tells you that each tile of the board is keyed to a number, starting with 1 in the top left and counting up left-to right, top-to-bottom.

No questions asked.

What's great is that you get to see the full 64x64 board, which is represented in binary. 1s represent counters, and 0s represent empty spaces. The board is re-drawn every time you make a move. It's kind of a pain to have to keep scrolling back up to the reference board each time. No way am I memorising the numbers.

The opening state of play.

I managed a respectable 42 counters on my first attempt, which I was fairly pleased with. The strategy I went with was to take all the counters off the border first. What ended up happening is that I cleared out the top-left and bottom-right corners, with all the counters congregating in the opposite corners. Progressing from this point was actually a rather delicate process, with not many moves being available, but all having a significant impact on how the rest of the game would progress. I realised that this might not have been the best strategy to go with.

Nice diagonal symmetry.

Pressing on, I fumbled around, trying to figure things out as I went along. Turns out the strategy you start with makes a huge impact on the rest of the game. I was, at one stage, left with some scattered clumps of counters in the top-right area that I realised I couldn't completely remove. There goes the 47 counter dream for now.

That top right section gave me fits.

I had to make the most of the situation, which resulted in solitary counters spread across the board at the end of the game.

A somewhat disappointing finish.

42's not a bad score, but I'm not fully satisfied with it. Back to the strategy drawing board for this one. I noticed for One Check, that games take much longer to complete than most of the previous games I've played. Part of that, I think, is down to having to constantly reference the numerical board every time I want to make a move. There's also quite a bit of thinking to do. This first game took me about 20 minutes to complete.

So what's my revised strategy? My first thought was to move all the corner pieces. They have no other place to go, so it seemed they were the logical first moves to make. Having done that, I thought that moving in all border pieces from the corners would help further. Turns out I was mistaken on this second point. I ended up with a nice symmetrical pattern, but nothing I could do with it.

Looks nice, but is completely useless.

I tried a few different ideas after this, but the best count I got was 44. This result came from the idea of aiming to take out one corner of counters at a time. I think it was working reasonably well, up until the end where it fell apart.

Best result so far.

I had to take a little step back to reconsider my strategy; reconsider what I actually needed to do to get those 47 counters. I swiftly realised what I needed to do. What needed to be done at endgame was create a situation where you only need two counters to remove the final trail, but one of those two needs to consume the other. Say what you will about the simplistic look of the game - the strategy appears to run deep; it's something I can stick my teeth into, which is a good thing.

I theorised how I could accomplish this "snaking" strategy, as I'm calling it, and came up with a diagram of a theoretical 47 counter situation:

It looks like a random assortment of colours, so allow me to explain what all this means. The yellow squares are the "active" counters. These are the counters being moved. The red squares are the counters the active counters remove. The orange square is where the changeover happens. The first active counter in the top-left moves until it gets to the orange square, then the second active counter takes over, removing the rest of the red counters until it lands on the embossed red square, as the final remaining counter. The lines are there to show the path of travel of the active counters. This probably represents one of a dozen similar scenarios, it just happens to be the one I came up with. Obviously, this is just the endgame - I still have to try and get the game into this scenario, if possible.

So I tried this, and no result. Frustrated at my lack of progress, I decided to do the one very thing that's an unspoken rule of mine: look up a guide. What I discovered truly shocked me: David Ahl's "substantial feat" is actually impossible. 

I also discovered that there's quite a lot of game theory research that's been done on Peg Solitaire over the years. There's far more to it than I realised, and many varieties of boards and setups that are available. A key piece of theory I learned is that there are essentially two types of Solitaire boards: null-class and non null-class. What's the difference between the two? Null-class boards are solvable, while (in most instances), non null-class boards are not. It turns out that the 8x8 board here is not a null-class board, therefore meaning that having a single counter remaining on the board is impossible to achieve. If you want to read more about this, you can get a fuller explanation here.

Unfortunately, that particular site didn't deal with this specific, diagonal setup. There were a few sites that very briefly discussed the diagonal variant of Peg Solitaire, but not with the particular setup, or the detail, that I looking for. Eventually, I came across a page that gave me the exact answers that I was looking for. It calls this variant "8x8 Diagonal Solitaire." It even directly references Ahl's digital rendition! It declares that Ahl's highest tier of 45 - 47 counters is impossible to achieve. The major problem is that the counters will never leave their colour - red stays red, and black stays black, no crossover, meaning that there would have to be at least two counters left. In practice, it actually turns out that the minimum amount of leftover counters is four, for a maximum removable total of 44. I'm actually quite chuffed about that, seeing as I was able to achieve that total without any external assistance. The Zillions of Games site provides a sample solution, which I'll provide in the game video.

Well, this article went in a very different direction to what I was expecting. I found scoring a little bit of a challenge also. It's always tricky with these digital renditions of real-life board games. Do I rate the game based on its source material, or the implementation of it?

Time Played: 2 hours

Difficulty: 6/10 (Challenging)

It requires a lot of thought and planning to get the 44 counters.

Gameplay: 7

I'm somewhat blending the ideas of scoring based off the source, and the implementation thereof. It's a fun game, to be sure, but the implementation has its problems. Constantly having to scroll up and down the screen to reference the number key is a pain.

Controls: 5

Like usual, it's a standard input system. I imagine it could've been a whole lot worse if Ahl chose to use co-ordinates, for example.

Visual: 5

It does what it should. While the formatting is good and all, I'm not going to give brownie points for doing what's expected. That what the 5/10 score is about - doing what's expected, but not going beyond.

Functionality: 5

No issues.

Accessibility: 3

It can be a somewhat intimidating game to learn, but is rewarding once understood. It also helps that, for a text game, it doesn't require a whole lot of reading.

Fun Factor: 8

I found it enjoyable and mentally engaging. If not for the discovery that more than 44 counters was impossible, I'd still be playing, trying to devise new strategies. It really put a damper on my enthusiasm knowing what I now know.

So, Solitaire Checker Puzzle does better than you might think - a score of 33, and a place in the D-tier, just above the original Lunar Lander as the 9th game overall currently. It's a good time, for what it is.

Back to the arcade next time. Going to be racing through space.

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

Prehistory XXII: Battle(Ship)

Release Date: October, 1970

Platform: Unspecified Mainframe

Genre: Puzzle

Developer(s): Ray Westergard

Publisher(s): DECUS

I haven't had to skip a game on my list this time, which is a welcome change to what I've become accustomed to during this series of blog posts. What isn't different here is that it's yet another pre-existing game getting the digital conversion treatment: Battleship. I suspect most people would have played this classic guessing game at least once in their life - I know I did. I think this is the first time I've really had a personal connection to the core idea one of these games is based on, as I have a real fondness for Battleship thanks to my childhood memories of it. I played it quite a lot as a kid, actually. It was one of those games I had a fascination with, I don't know why, but I played many different physical and digital versions of it growing up. I never played any versions of Electronic Battleship, however. It was either fully digital, or fully old-school with the plastic pegs.

Having said all of that, this game, which is probably the first video game adaptation of Battleship, is a bit different. While it still retains the core Battleship gameplay concept, it comes with a twist, one that I'll discuss a little bit later on.

First, as per usual, I'll discuss authorship. The good news here is that we have an undisputed author for Battle: Ray Westergard, of the Lawrence Hall of Science, situated at the Berkeley campus of the University of California. He even wrote the program description found in BASIC Computer Games. I couldn't find much more certain info on Westergard, as he has no other game credits, and there appear to be multiple people of that name: one appeared to be a Unitarian theologian (awkward), and another an abusive university professor (extra awkward.) There's a possibility that neither of these are the Ray Westergard that programmed Battle, considering that the game is near 55 years old, so the Ray Westergard in question may very well have passed away during that time. At least this short dig was an interesting one.

Now, back to the twist I mentioned. It seems that, since this is a singleplayer implementation of Battleship, that Westergard felt that some modifications needed to be made to make it function properly. So, what he decided to do was add some codebreaking into the mix, which transforms the classic strategy board game into more of a puzzle game.

The concept is this: the game provides you with a 6x6 matrix of the enemy fleet, consisting of 6 ships. However, it's a coded matrix, meaning that the supposed positions of the ship on the matrix the game gives you are not the actual locations of the ships. Your job is to simultaneously decode the matrix while sinking the enemy fleet in as few turns as you can. 

Battle's BASIC Games page might do a better job of explaining things.

The instructions of the game expect you to write out your own matrix as you play, which is exciting for me as it's the first time on the blog I get to use spreadsheets. Believe me, I will be making liberal use of the mighty spreadsheet in the future, especially with RPGs... In this game's case, I'll simply use a spreadsheet to write out my own matrix as the game suggests.

You must decode the matrix.

Upon opening the game, you're presented with the coded matrix of the fleet's positions. To make a shot, you input first the X-axis (horizontal co-ordinate) followed by a comma, and then the Y-axis (vertical co-ordinate.) I opt for 1,1, suspecting that the code may involve the 0 spaces. It does not, and I hit open water.

Next, I try 1,6 (top left,) seeing as the matrix says there's a ship there...

It's not a great code if the ship is actually where it says the ship is...

It's a hit! I successfully strike ship 1, one of the destroyers. The matrix says the other spot should be on the right, so I go for 2,6, but that's a miss. That gives me a clue about how this matrix is coded. Perhaps the ship positions are perpendicular to how they're presented on the coded grid?

Armed with this theory, I go for 1,5 next. 

Hurrah for the good guys indeed.

My theory was correct! One ship down, five to go.

I decide to go for the other destroyer next, ship 2. It's highest position is 3,6, so that's what I select. My theory at this moment is that the topmost co-ordinate is correct, but the ship's alignment is perpendicular to what's presented on the grid. Turns out this was incorrect. It's a miss. I try 3,5 instead. 

An unexpected surprise, but a welcome one.

Something unexpected occurs - I get a hit, but not on the ship I was aiming for. I hit ship 6 (a carrier) instead. Interesting...

Next I opt for 2,4, suspecting that the carrier is aligned north-east. Another unexpected occurrence! I hit ship 2! That makes me curious, could ships 2 and 6 have their positions switched? I try 2,3 next, but I miss. 2,5 also accrues a miss.

Here is what my spreadsheet matrix looks like at this point.

At this point, I realise that ship 6 can only be going eastward, as that's the only 4-long space available (assuming that ship 5 is where it says it is). I am correct, and take out ship 6 (while accidentally putting a miss on 6,6.) It's also helpful in ruling out the rest of row 6, as no other ships are small enough to fit in that gap now. At least for ships 2 and 6, my theory of perpendicularity has proven true. In the case of the vertical ship (6), it was also shifted up one space north-east as well.

The thought now is that, perhaps the opposite is true of ships aligned horizontally on the coded matrix? Perhaps they are also shifted, but one space south-west? It proves true in the case of ship 2, as 1,4 is its other position. Halfway there.

Insert Bon Jovi reference here.

I look to ship 4 next, as the other two are diagonals, and I'm not sure how to deal with them yet. If my theory is correct, 4's true position should be 3,4 - 4,4 - 5,4.

Hmm...

Another curveball is thrown, as I hit ship 5 instead of 4. Hm. Well that changes things a bit. There's only two potential locations for 5 - straight west, or straight south. Going south could potentially land me a hit on 3, while confirming the position of 5, so that's what I decide to do, and go for 3,3.

Miss. That's frustrating. The silver lining is that I've confirmed 5's position as westward.

Except I hadn't.

Or so I thought. I go 4,4, and I hit ship 3 instead. What? That's not what I expected at all! That means that 5's position hasn't changed at all... Does that mean 3 has simply been shifted to the opposite side of 5? It has, as 5,3 registers a hit, and 6,2 sees the cruiser fall to the bottom of the ocean.

Where the heck is 4 supposed to be?

So where on earth is 4, then? It can't be anywhere near its original position. There are a few potential spaces for it left on the grid in the south-west section. I come up with the theory that it's in the same spot, but on the perpendicular side. I try 2,1 to test this theory, and am correct. The fleet has been successfully dispatched. My final splash/hit ratio is .3888889. If I'm reading the instructions correctly, that's pretty good, as it's calculated as a fraction (splashes/hits), meaning a smaller ratio is better.

A job well done.

The true positions of the fleet.

Now, having had the first run help me get my bearings, I had to find out whether the code algorithm was consistent or not.

Upon starting my second game (seen in the video up top), I could definitively say the answer to that question is no - the code is different each time. For my second game, the positions of all ships were essentially opposite to what the matrix said. For visual reference, here's my finished spreadsheet for game 2:

It was opposite day for this fleet.

The game matrix is on the left in this image, with my completed game on the right. You can see this "opposite" code most clearly with ships 2, 4, and 5 - with 4 being in the opposite corner, and 2 and 5 also being in their opposing corners, and rotated. 3 is the odd one out in this set, as it doesn't move into its opposing corner, but is just rotated 90 degrees. My splash/hit ratio was the same as the first game, in case you were curious.

I think this is enough evidence to suggest that the game changes the code for each game. I could play a few more rounds to see if it randomises or selects from a limited set of rules, but I don't see much point in doing that. I can't read BASIC, either, so I couldn't tell you from looking at the source code how the game determines the fleet positions and matrix code.

What I can tell you from the code is that there is a little congratulations message programmed if you take out every ship without missing. It simply reads:

"Congratulations -- A direct hit every time."

Nice to see an acknowledgement for a perfect game. I'm not likely to try this, as there's always going to be a large degree of luck involved, but it's certainly possible to achieve.

That does it for Battle. Now to do the scores.

Time Played: 45 minutes

It takes a lot longer to complete a round than almost all of the other games I've played to date. My first run (which I took notes while playing) took roughly 33 minutes, and my second about 8-and-a-half, so I'll round it up to make things easy.

Difficulty: 3/10 (Easy)

It wasn't particularly difficult for me to figure out how to decode the matrixes I was given. The first one was a bit more challenging, as I didn't know how the game was going to code the fleet's position, but once I started to understand how the game did things, it was fairly easy to figure out the code.

Gameplay: 7/20

I'm actually fairly impressed with Battle from a game design standpoint. It's a clever twist to convert Battleship into a code-breaking puzzle game, and it works well. Each round is different from the last, meaning it has some replay value. While still very simple, it's miles better than the glorified RNG simulators that I've been playing recently.

Controls: 10/10

Each move only consists of 3 inputs, with the second always being a comma, and the first and third always being a number between 1 and 6. No issues with inputs on this one, full marks.

Visual: 4/10

It's roughly down-the-middle as far as text-based games are concerned. The fleet matrix is nice and clear, but the text generated upon destroying a ship is a lot, and feels quite cluttered. I think it would've been easier to digest if each type of ship was on its own line, instead of all three being on the one line. There's a little bit of charm and personality in the writing, but not enough to give it another point.

Functionality: 5/5

As with most text-based games of this ilk, everything works perfectly.

Accessibility: 3/5

It helps that Battle is based off a highly recognisable game, so that gets it an extra point, but text games still aren't an easy thing to get into.

Fun Factor: 8/20

I enjoy numbers and patterns quite a bit, so Battle appeals to me quite well as a puzzle game. It's simple but enjoyable, and has genuine replay value.

With these scores, Battle earns the second-highest position on the tier list (at the time of this blog), with a final score of 37/70 (52.85%). It just barely misses out on a C-tier placement, having to settle for a spot at the top of the D-tier. It's nice to see some improvement on the game design front, and something with actual gameplay to talk about instead of yet another RNG simulator.

I'm contemplating some re-scores once the Prehistory series is done, as I need to re-evaluate how I score controls and visuals on text-based games. I'll do a score change round-up blog once I complete the final games from 1970.

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