Whether you love them or hate them, game reviews (and the writers who write them) hold an important place in the gaming industry for one specific reason: they're the ones the gaming public read to make our final decision: "Do I buy this game or not?" For fans of a given genre or series, the review might not really matter, but for the rest of us gamers, we need good advice on spending our money on good games. But what makes a good game, anyway? Short answer: "One that gets a good review."

But then, what makes a good review? To give us an idea, Omar dedicates an analysis piece over at ProductWiki on the matter.

What makes a good review? Omar cites the following qualities, based on his analysis of four reviews of the Xbox 360's Enchanted Arms RPG.

It should answer the all-important game buyer's question: "Should I buy this game or not?" If reviews seem too long and rambling, it's because they sometimes do more than just recommend whether the game is worth buying. They also serve as fact sheets; sometimes they criticize minor points in the game. Useful stuff for someone who doesn't know the game, but it's just annoying extras for gamers who've followed the game's development.

Since many reviews feature score systems, the scores - or at least the final score - should be related to the recommendation to buy the game or not. This is why metacritic sites like Gamerankings have become popular, says Omar. It's easy to decide on games ranked 9 out of 10 (or 1 out of 10 for that matter), but when the score falls in the "middle range" it leads to more questions than answers for the reader - and the buyer. "Is a "7" Good enough to buy or not?"

When it compares two games, it does so in an effective manner: the games are closely related (genre, platform, age or release date), and compariosns are often used to help illustrate a point. Omar's recommendation is that the game being compared to must be available on the same platform as the title being reviewed, in the same genre, and released no more than two months before or after the review date.

Reviews should be personal. Why? Simple reason: trust. No one can assume what their audience expects them to write, period. But if the reviewer sticks to what he knows best, what he enjoys, then what will happen is that gamers who share his taste in games will start reading his reviews on their favorite games more. And trust what he has to say about a new game.

Reviewers should be known by their audience. The best way, Omar notes, is by the games they like to play. For example, combat flight sim fans will definitely trust the word of an Ace Combat or Combat Flight Simulator veteran. In web reviews, the review should link to the reviewer's biography. Omar rates IGN and Gamespot well on this one because the sites link the reviews to the writer's blog, though he comments that the blogs aren't exactly a precise source of information for what games the writer likes or is good at.

The review should focus on the total experience. Modern game reviews are notorious for "segmentation," Omar says. You know those points where the review would dedicate a couple of paragraphs to lighting and texture, sound effects, nitpicking and highlighting - in other words, "the critique." It's probably fun to read from a technical perspective, but it doesn't answer that all-important question in the first place.

Finally, Omar also rises to the defense of those besieged reviewers, reminding gamers that reviewers - even as they sometimes enjoy the games they play - aren't gamers. At least, they aren't average gamers. There will be important differences in the way they look at games and the gaming public looks at games.

• Reviewers play for long game sessions, to try and complete the game in the shortest amount of time. It makes it hard to "play to enjoy the game" and review it from that perspective - the average gamer's perspective.
  
• In multiplayer games, remember that the pre-release multiplayer the reviewer plays may be different from the post-release multiplayers the public will play.