Prehistory Score Revisions

With the close of the Prehistory of Video Game series, I wanted to take some time to review my process - specifically my game scores. I've felt a little dissatisfied with the scores for text-based games (of which there are a lot in this period), and their resulting positions. I don't believe they accurately reflect my opinions on the games in some cases.

In this post, I'll give an overview of the changes I've made in my scoring philosophy, and will then give a list of all the resulting score changes. Prepare for a short essay on video game review philosophy.

The main place of discontent in my scores for text-based games is the controls metric. This is a difficult metric to grade for text-based games. All inputs are made through the computer keyboard, which is an essential part of everyday computer usage, so how do I judge how good or bad the controls are for a text-based game? There aren't really any ergonomic issues with regards to controller feel, input responsiveness and whatnot, so other factors have to come to the forefront.

I think, for text games, that simplicity is the key factor to consider when it comes to controls. Inputs should be both simple in and of themselves, and simple to remember. For example, if I have to keep referring back to the manual or opening instructions because I can't remember the inputs, that's a problem. However, simplicity can be contextual. For a more advanced command, it's logical to expect the inputs required to be more advanced (and thus also harder to remember.) The inverse is also true - simple inputs for simple commands. If I need to input some long string of numbers, letters or words to do something as simple as "move north," then that is a massive problem.

Inputs for text games should also just make sense. As an extreme example, it would be like a Switch game having the jump command mapped to left on the analog stick. If it doesn't make sense for the game, it's a poor design choice. A text game example I keep going back to is PDP Basketball. The game has two sets of commands - offensive and defensive. The four offensive commands are mapped to numbers 1 - 4. Makes sense, and is simple. The defensive commands, on the other hand, are mapped to 6, 6.5, 7 and 7.5. This does not make sense. There's no need to have decimals here, those commands could be easily mapped to 5 - 8, or 6 - 9. That makes an awful lot more sense and is consistent with the offensive commands. It's also easier to input and remember, so also a good simplicity example.

All of this is to try and make the grading process as objective as possible. However, that also requires a standard by which to judge every game against. Figuring out that standard is the difficult part. What does an average 5/10 game look like? What's a 10? What's a 0? The only way to know these things is through data. I need to obtain more data by playing and reviewing more games, assessing what my current scoring patterns are, and then apply my revised philosophy to adjust scores properly.

Typically I default to giving a text-game a 10 on controls. In most instances, this is because the game requires commands of only one or two inputs (usually numbers.) They make perfect sense and can't possibly be simpler. My problem is, I don't find this particularly satisfying. Yes, those controls are sensible and simple, but they're not enjoyable to input. Perhaps that's another key factor - how fun are the commands to input? Do they have some charm to them? Or do they creatively integrate the inputs into the game's story and lore? Or is it just like inputting data on a spreadsheet?

And so that's where the standard for text-games will begin. From now on, those games that have perfectly sensible and simple, but entirely plain inputs will receive a 5/10 score. With this in mind, here's what the score changes for controls are:

• Starting from the bottom of the tier list, Batnum (1967) is dropping to a 5, which perfectly reflects the new adjustments. It's sensible and simple, but is merely inputting numbers.
• Likewise, Batnum (1970) also drops to a 5. I gave that a 9 originally, but it still runs in line enough with the new standard to not be lower than 5.
• Qubic was initially a 6, which, looking back, I think was quite harsh. Its inputs are very standard and logical, I just kept mixing up the x and y-axis inputs. This'll go to a 5 as well.
• War drops to a standard 5.
• Digits is the first game that diverges from the standard score. I gave it a 7 initially because it's a bit of pain inputting all 10 numbers with commas separating each one. This gets a 4.
• Horserace is now also a 5.
• PDP 10 Timesharing Basketball drops to a 4 for the reasons I stated above when I used it as an example of inputs being inconsistent and not making sense. It doesn't negatively affect the gameplay too badly, but is still a flaw.
• Roulette, Gamnim, 1Queen, Civil War, Lunar, Battle, High Noon and Hamurabi all drop to 5. These games all represent the new standard pretty well.

And that's all the text-based games I've done to this point. However, I'm not done yet.

Visuals is another area that needs refining for text games. I have figured out while playing what I value in text-based games in terms of visuals, so this'll be a much shorter discussion. The key thing I look for is formatting. What I mean when I say formatting is the spacing and layout of the game text. As text-based games require a lot of reading compared to most other styles of game, whatever can be done to improve the ease of reading is going to reflect well on a game. Too much text on one line when it could've been easily spaced out, or large blocks of text make a game feel cluttered, and thus harder to read.

I also consider the writing of the text itself. This will have some crossover with the story metric later on, but I still think is valid to consider in visuals for text-based games. There's two components to this: clarity and creativity. Obviously what I'm reading needs to be clear and understandable. If there's numerous typographical errors, points will be docked, even if what's written still makes sense. It's a presentation issue. 

Creativity concerns if the text has been written with some charm or personality. I appreciate when effort goes into making the writing match the theme of the game, instead of keeping it plain-Jane. Of course, there can be negative creativity when dialogue isn't written well or is needlessly abrasive (many early text games like to insult the player.)

Finding a standard is once again an issue, but dealing with the controls metric helps streamline the process. For lack of a better term, "plain-ness" is probably the best way to define a 5/10 in visuals. If there's not much to comment on, either positive or negative, then it's getting a 5.

With that, let's have a look at the score changes for Visuals. Some games that lost a lot from controls are going to gain back some points here:

• All the plain games that are getting a 5 are both versions of Batnum, War, Horserace, PDP Basketball, Roulette and Gamnim. 
• Qubic is a tricky one. The graphic of the board, while helpful, isn't in the original game, but I considered in in my original score. For consistency, I'll still factor it in to my score, which'll be a 6.
• Digits also gets boosted to a 6. The number guessing table is good and clear, and the game has a little bit of creativity in the writing.
• LEM remains unchanged.
• 1Queen moves up to a 4. I still don't like how the game board goes out of alignment from adding 3-digit numbers in. I think that could've been managed better.
• Civil War remains unchanged, its formatting is excellent.
• Lunar is getting boosted to 5. Its table is actually pretty good, but the instructions are cluttered.
• Battle also goes up to 5. I think I was a little harsh on it.
• Hamurabi also gets up to a 5 for the same reason as Battle.

The last metrics I want to deal with are gameplay and fun factor. These two are usually closely connected, so it makes sense to tackle them together.

I manage gameplay a bit differently to the other metrics I've discussed. For one, I already have one clear standard that I've been consistently applying. That is I assign a zero score to any game that is either completely broken, or completely luck based. Player agency and meaningful choices have become quite valuable to me as design choices, so anything game that invalidates all forms of player choice and influence automatically get a zero. Examples of this from games I've already played are War, which is basically a coin flip simulator masquerading as a card game. You have absolutely no input or influence on the game, which to me totals an absence of gameplay. Those games are the easiest to deal with.

For everything else, it's often much harder. There's a lot more to consider with gameplay - the base gameplay loop, how the mechanics complement and synergise with everything else in the game; difficulty, balance, progression, depth, complexity vs. simplicity, replayability and general enjoyment of everything aforementioned. In a way, it's good that I started with simple text-based games, as there's far less to review and assess with these games.

The process I've decided on is to start every game off at a middle 10/20 score. From there, I'll analyse the gameplay, and consider all the good and bad elements of the game, and try to produce a score that I'm satisfied represents accurately my opinion of the gameplay. To assist in this process, I began taking notes on each game, analysing the gameplay, writing down all the good and bad points of the game so that I have all the data written down in front of me. I did this for most of the games I've already scored (the ones that weren't instant zeros.)

I usually use the gameplay score as the basis for fun factor, and then adjust from there. This is the most subjective metric, and relies simply on how much I enjoyed the game, or how much of the game I enjoyed, and how much I want to return to it. Those games that get a zero score in gameplay often get the same for fun factor. I often become quite hesitant to play games I don't enjoy, which is a helpful emotion for knowing that a game is not fun for me. On the flipside, I tend to get obsessive over games I really enjoy.

Now, here's the score changes for gameplay and fun factor

• Batnum (1967) is going up to a 1 in gameplay. It doesn't have a complete absence of player agency, so it doesn't deserve a zero. There's barely any game, but it is at least a game. Fun Factor remains unchanged.
• Likewise, Batnum (1970) is also getting a score increase. I'm giving this one a 2, because of it allowing you to customise the game setup a fair amount. Fun Factor remains at zero.
• My gameplay score for Qubic was quite harsh, in retrospect. I'm quite biased as I don't like 3D Tic-Tac-Toe. Qubic actually isn't the worst thing in the world to play - the computer is quite competent, but beatable. It's still not a good game, or recommended, so its only getting boosted to a 4 in gameplay. Fun Factor remains unchanged.
• War is unchanged.
• Digits also doesn't have a complete absence of gameplay, so it gets 1 for gameplay. Silly game.
• I'm going to be nice to LEM and boost the gameplay score to a 6. It has stacks more depth than any other game in Prehistory, even if it's exceptionally difficult and not fun.
• Gamnim reduces gameplay to a 2, and fun factor to a 1. It's functionally the same game as Batnum (1970), but you can get really silly with game customisation.
• 1Queen is a dopey little puzzle game, but lacks replayability and isn't a terribly challenging or interesting puzzle to solve. Its gameplay score drops to 3, and so does Fun Factor.
• Civil War goes up to a 7 in gameplay for its resource management and replayability with being able to try different strategies. Fun Factor goes up to 6.
• Lunar is getting upgraded to an 8 for both gameplay and fun factor. It was enjoyable, if quite frustrating and repetitive. I did get moderately obsessed before just giving up because it was too hard.
• Battle gets notable boosts. Gameplay goes up to 9, and fun factor up to 10. It's quite clever, although requires some luck to figure out the patterns. The code being randomised each game is a big plus for replayability.
• High Noon also gets increases, both gameplay and fun factor go up to 7. Probably similar to Rocket in that it doesn't have as much depth as you'd think, but the fun gets boosted from the writing and presentation.

Other miscellaneous changes:

• Increased the difficulty of Qubic to 6. 3D Tic-Tac-Toe is hard, and I got very lucky in finding a winning strategy.

Righto, that's everything. It feels like a lot of work for what's ended up being a rather short article. There's still a bit to go, however, as I need to now summarise the changes to the tier list. I'll start with the earliest game I scored, and move up from there:

• PDP-10 Timesharing Basketball - well, this one was interesting. I realised that I made an error with the initial score, I calculated the percentage incorrectly. The correct percentage was meant to be 32.85%. Upon rescoring, its score has decreased from 23/70 to 21/70 - a percentage of 30.00%, which doesn't change tier. This has it tie with one other game, but this one comes out on top due to earning a 3 for Fun Factor.
• Civil War also has a scoring error - it was listed as out of 80 when it should have been 70. I'll blame my green-ness at blogging for these. Still, the game has seen an increase of 1 point, up to 32/70 (45.71%). It doesn't shift the tier placings - aside from it being placed correctly.
• Hamurabi takes a hit to its score - it has dropped from a 40/70 to a 36/70 (51.42%). This has resulted in it losing its C-tier placement, dropping to the D-tier, and it has also lost its title as the highest rated game to date on the blog. It's not even second. A great shame, indeed, but it would have lost its crown even if its score remained the same.
• Qubic improved a little: up from 19/70 to 22/70 (31.42%), tied with two other games I have yet to get to. It sits in the middle of those games due to its Fun Factor score.
• Lunar gains an extra point, boosting it to 32/70. Doesn't change much for it.
• 1Queen dropped off quite a bit, losing five points down to 22/70 (31.42%). This ties it with Qubic. 1Queen sits above it due to the higher Fun Factor score.
• Digits managed to get even worse, losing a point down to 19/70 (27.14%). What a silly game.
• Gamnim took a hit to its score - losing 4 points down to 21/70 (30.00%). This ties it with PDP Basketball, but that game wins the Fun Factor tiebreaker.
• Horserace is garbage, and that is rightly reflected in it dropping 5 points down to a miserable 17/70 (24.28%). This score disgracefully relegates this game to the depths of the F-tier. It deserves it. It's also tied with a later game, and lost the tiebreaker on Gameplay.
• Roulette lost 4 points down to a 20/70 (28.57%). This ties it with a later game, but Roulette wins the tiebreaker.
• Batnum (1967)'s numbers got shuffled around, but its overall score didn't change. It's tied now with Horserace, but wins that tiebreaker (Gameplay).
• Batnum (1970) bucks the trend, actually gaining a few points! 2, to be precise, up to 20/70 (28.57%), tied with Roulette, but losing the tiebreaker.
• War, which is trash, loses 2 points, down to 18/70 (25.71%), just baaaarely keeping it out of the F-tier.
• Battle only got its numbers reshuffled, its score didn't change. It makes it a winner, however, as it's now currently 2nd on the tier list after Hamurabi dropped.
• High Noon - Are you ready to crown the new king? Despite losing a point from the rescore, down to 42/80 (52.50%), it now sits on top of the D-tier, and the top of the current leaderboard.
• LEM - The scariest, most complex game of this period gained 2 points up to 22/70 (31.42%). This ties it with 1Queen and Qubic, but LEM loses the tiebreaker for being a miserable game to play.
• Rocket - Somehow, this game has managed to avoid all the carnage and come out completely unchanged.

[Add. I actually made a mistake. Battle was actually the highest rated game after the rescore, with a score of 52.85%, just above High Noon's 52.50% My apologies. It's still true that Battle is 2nd on the list now, as I'm seeing this after completing Star Trek [sttr1], which is far superior to all these games.]

And so, this mammoth project comes to an end. At least it felt that way, even if the result is not that long a post. Yet, this project is only a speck in this whole blog's premise. This was only 17 games. Today there'd be at least 17 games coming out every day. Madness. 

I feel like this was a good way to bookend the Prehistory series; looking back at all the games I played, even replaying some, to see how my thinking about games has changed over the period of making this series. I think both my writing and critical thinking have been refined - writing at least is clearly reflected when I look back at my earliest blogs. I still can't believe I thought that doing Prehistory in a single blog post was going to work. More like a PhD thesis. Yikes.

So now, it's definitively time to move out of Genesis and into the Promised Land - 1971 marks the official commercialisation of video games (if only just), and the proceeding rapid spread, innovation and technical advancement thereof. I know I've been away for a while, but just thinking about what's to come gets me very excited to commit some more time to the blog. I hope you, reader, are excited to come along the journey with me.