Posted on Nov-25 2007 by ron
A full game review should help a wide range of readers to come out better informed
about what the game in question is, how it is played, and how well it is executed.
To these ends a review should contain clear descriptions of a game's rules, the activities it demands from the player for progress and the skills it teaches/requires. While taking diligence to avoid spoilers, a review should also offer enough samplings of the game's overall style to allow readers to determine match or mismatch with their individual tastes.
The score should be usable as a quick cross-referencing utility and even summary for those who will not bother to read the whole review, even though imperfections inherent to this approach are obvious to us. The final numerical score should capture purely the value of the gameplay experience while ignoring stylistic expression as much as possible. Style is to be described, possibly even evaluated, but not rated.
Review process rules
We will not score a game unless we either spent at least twelve hours playing it or
finished it. Further, if a game can be completed in any agreeable way, we will only allow ourselves to give
it a positive score if we did so. If it can't be completed, the requirement for a positive
score changes to at least twenty hours of play.
The n-hour clauses allow us to review highscore-based games that lack a clearly defined main campaign or are otherwise too loosely structured to be considered "finished" at any specific point. At the same time we enforce an adequate amount of hands-on experience.
This also means that if we are sure that a game has enough problems in its first twelve hours of play that it can't be scored higher than zero (which is our rating for average games), no matter what comes afterwards, that we can abort the play-through and still do a review under these rules.
We acknowledge that these rules imply a responsibility to not abort review playthroughs of good games in bad faith, just to save time.
Writing on the review may start at the earliest three days after putting the game down. This is to avoid capturing an overly emotional initial response.
Rationale: if a game takes longer than this time to become any good, that's already a serious
flaw, and we are not here to reward those. Positive scores are meant to indicate to our readers that it's
easy to have a good time with the game in question, and a twelve-hour hump of not-so-good time
is incompatible with this idea.
On the other hand, by attaching a completion requirement to positive scores, we can ensure that potentially weaker sections of the game do come to our attention, and we can thus avoid excessive praise for a game that started out great and then fizzled out.
There is no finishing requirement attached to the lower scores, because it limits the amount of pain and suffering we have to go through with busted games. Overrating a flawed game is a much bigger disservice than being a little lukewarm on a game that takes forever to get good (which we take to mean that it isn't really that good to begin with, because the overall experience is already botched at that point).
(NOTE: this here is just the rules. The range of our scoring system and supplemental prose can be found in its own rating system document. Quick hint: -2 to +2.)
A game can receive a 0 or higher score only if it demands some level of skillful adaption from the player. This need not be major multiple simultaneous clockwise button twists by any means, but win-button games may not receive this rating.
A game can receive a +2 score only if there is a belief that it is an adequate entry-point
into the genre, which is mostly an evaluation of the amount of preexisting knowledge the game
assumes, and of the difficulty of its initial challenges. That doesn't mean
the game needs to be easy, but rather that its build-up should be smooth enough to
allows uninitiated players to become initiated before the game gets hard.
Unlike any of the lower scores, the +2 rating also requires a consistent, appealing presentation.
Rationale: the +2 is our highest rating. We want to encourage readers to interpret it
as a signal that there need not be second thoughts, no ifs nor buts, about the quality of the game.
It marks a recommendation to anyone, not just to those who are already sure they like the
specific genre the game happens to represent.
The zero rating is our "decent if limited" rating, which means it marks games that are worth playing. We need to make sure that busted games never receive this rating as to not dilute its value.
Every game ideally should reinforce the very habit of playing games. To game is to try doing something
new for a while, and to foster a healthy games industry, it is necessary that individual games avoid
damaging that willingness.
Our ideal of a game is that of an inconsequential learning experience, where the fun is derived from exploring a set of rules and gaining proficiency in them while the machine provides evaluative feedback. The player should face gradually steeper challenges and be adequately rewarded for good performance. Rewards can take many forms, and even though score or simply progress through a story-line are acceptable as rewards, things that feed back into the gameplay system (such as new abilities, upgrades, consumables etc) are of higher value.
The game should be reliably driven by and responsive to the player's inputs and decisions. Random elements are welcome as a means to provide ever-changing situations to adapt to, but should not have a stronger influence on the result of a challenge than the player's own performance.
A game should have meaningful depth that presents the player with frequent choices and trade-offs, each of which are at least arguably valid in specific circumstances that appear over the course of the game. Hiding one universally correct way to progress in a sea of redundant, inferior alternatives is stronlgy deprecated, on the other hand. It teaches players to ignore parts of the game system, and burdens them with identifying which ones. This is unrewarding and frustrating and runs counter to the behaviours that make a person want to play games.
We expect a game to honor players' investment of time in that it does not diminish or take away traits of the player, or the player avatar(s), that took significant time to gain. Play-skills acquired during the game should remain useful and valid. Customization and specialization accrued by the player over many hours of play should have a lasting benefit.
The game should not just allow, but invite the player to be skillful and efficient. One major requirement of this idea is that the player should not be forced to wait without a strong technical reason. It should be possible to skip cut-scenes, to accelerate or skip dialog, and to enter menu commands quickly, without artificial delays.
Just like we don't wish to cater to gamers who seek nothing but stimulus and maybe an interesting story, we find no purpose in games that fail to deliver anything beyond that. Other media handle these interests so much more effectively that we just can't appreciate any such efforts. A good presentation can of course enhance a game, but a broken game with excellent presentation is still a broken game and will find no place here.
We acknowledge that almost all of the requirements or "should be" clauses are issues of
subjectivity, at least to some extent. Thus collecting them in this document cannot mean
that everything will be judged by a constant objective standard.
The value of this document is rather that of a publicized self-test: adding a formal list of things to look out for and consider during the review process can't make the results perfect in any absolute sense, but it still increases the consistency of reviews. Which is good enough.