4 cool things about game development

Hello! This is not a full-fledged post, we just did four translations of blog posts by Tynan Sylvester, the guy who created the Rimworld toy in one hat. He gives out good thoughts about the development of games, for some of which in a decent company he would even be beaten. Below is a short squeeze of each and a link to the translation.

The “ Dream Simulator ” describes the good old history of the world from the Westworld series - when players are offered a complex world with internal conflicts, a masterpiece of art. And they immediately kill and rape everyone. Not necessarily in that order.

After the release of Ultima Online, Richard Garriott said: “We thought it would be something fantastic. We spent a huge amount of time and effort on it. But it happened that the players just walked and killed everyone so quickly that the game did not even go to a new round, and the life simulator did not even start. And this part, which we spent so much time on, was never even noticed by anyone, not once! And in the end, we just removed it from the game. "

And then a completely crazy concept of plot density estimation is introduced - a model in the player’s head. This turned a lot for me.

The bottom line is - you need to look, not what really happens in the model, but how the player perceives it emotionally.

“Consider every interaction in the game - every harvest, every path traveled, every word spoken by the hero. Of all the interactions in the game, what percentage is part of an emotionally significant plot? In a successful game, this percentage is high. "

So where did his AI storytellers come from in Rimworld who slip in unexpected or welcome events! And that's why there is such a complex society - because the model of relations in the player’s head becomes the life of a colony. Not “they broke the breeding link”, but “my characters stopped dating because the major was shot off his nose while he defended the base from the orbital landing” - and this is a completely different attitude to history. And the answer to the question why we love to build houses in games so much is that we are not building them for the characters, but in our heads. In general, it is mandatory to read, there are analyzes on specific examples.

The material is here , the original is here .

Development speed

Further we have “From e-mails: reduction of improvements and non-linear results” - a big spiteful letter, where Tynan violently defends one very cool thought. The main thing in development is speed. The game is not application software, there is no clear technical specification and plan, the whole point is in the samples of hundreds of options. The more I tried, the better the final result. It’s amazing to deal with business - instead of completing 20% ​​of the work 80% of the time, it’s better to plant someone to finish this site, and do the next one yourself, where you can do something serious.

“Most of the things we do are actually not very important, - I really think that the final impression is created by just a few key insights. In the BioShock project, these were Little Sisters and Big Daddies, both of which turned out to be a couple of bright ideas, selected from a giant pile of just ideas. But these special ideas became the foundation of the franchise! When 1% of your work generates more money than the remaining 99% combined, you get a completely non-linear relationship between the time spent and the income received. ”

The material is here , the original is here .

Well, of course, in order to move quickly, you need to prioritize correctly. About how and where to look, he writes in the "Development Landscape".

“The first strategy is to move slowly and constantly check the altimeter readings. This is a test development process, when a developer introduces only small changes at a time and confirms their effectiveness by frequent testing. The advantage of this approach to climbing is that it guarantees finding the nearest peak. The downside is that the nearest peak may not be very high. As a result, we will stand on top of a small hill, although Everest can wait for us in the fog very close, not discovered by us only because of indecision, - this is the specificity of the step-by-step method. You can climb the mountain faster. You can rush forward along all possible options, only testing them slightly, even without really understanding them. This will give us faster progress with less testing. Many developers first begin to move quickly, jumping over mountain ranges, then slow down at the end of development in search of the highest mountain peak in the area. This slowdown process continues even after the release of the game - the release of many updates after the release of the popular game is akin to the developer moving the cobblestones on top of the mountain that he climbed during development to select the largest boulder and climb on it. Sometimes we make giant blind jumps, poorly imagining where we are going. MineCraft Notch got inspiration from Dwarf Fortress ... Not a single developer navigates the landscape alone. Imagine a landscape that boils with activity. Popular areas are literally teeming with developers who map territories with increasing accuracy,

Oddly enough for an indie developer, Tynan talks about how to learn from someone else’s experience and how important it is.

The material is here , the original is here .

Replay value

And finally, a short “Limit of excellence”. The point is very simple: there are two types of games. The former rests on the hardware capabilities of the human body, the latter on the capabilities of the mind. For example, shooting faster than others in shooters is a complication due to the fact that you cannot overtake biology. And it’s better to play chess - by deepening the strategy.

“Imagine a game of tic-tac-toe, but with the condition that you have only a third of a second per move. As a mental exercise, you can remove the element of speed from any game, slowing it down. Imagine the multiplayer Modern Warfare 2, "Starcraft", which is played at a speed of 10% of the original. The second method is the Go method. He has nothing to do with the ability to jerk sharply, but he has many points of contact with the management of integrated strategies and tactical information. The fact is that there are so many options and variable parameters in the game that no one can easily imagine where everything is moving, even if you stare at the game for an arbitrarily long time. Creating such a level of multicomponent content is much more difficult than just making high-speed input. ”

The material is here , the original is here .

That's all for now. Tynan is just cool.