Sun Feb 19, 2012 1:02 am
Help us out: Engine Designer Demo - Scenarios
Challenges in the Engine Designer Demo
Automation's Engine Designer is on the final lap before demo release. Soon everyone
will be able to try this out themselves instead of being limited to drooling over Youtube
videos.
As the Engine Designer by itself does not represent what the game will be like when
it's finished, we want to spice things up a bit for the demo. The aim is to give the
player actual goals instead of just providing a shovel and a big sandbox, which will be
included in the demo as well.
Thus, the Engine Designer demo will feature scenarios / challenges that will put the
player in situations where engines have to be designed with specific limitations or to
certain specifications. Just like in the finished game, you will have to cope with the
ever-changing landscape of the automotive industry.
Your favorite engine-design scenario in the demo
(Remember that it's limited to naturally aspirated Inline-4 engines currently!)
1) Provide a short scenario description. What is the challenge about?
Example: 1973, USA. The oil crisis has a hard grasp on the country, and people start
to feel the pressure to maintain their mobility. More economic car engines sure
would help and turn the crisis into an opportunity!
2) What technical limitations should be imposed on the player in this challenge?
Example: Carburettors only, Cast Iron only, no DOHC, etc.
3) What technical specifications is the engine supposed to have, and how are
those weighted in importance? (Max. 5 different points)
Example: Economy better than x (50%), power more than y (20%),
price below z (20%), MTTF in MRevs better than w (10%)
The respective number in 3) is a hard limit, you need to be better than all of these at
the same time to successfully complete the challenge. If this requirement is fulfilled,
you get a higher score for larger margins you beat the threshold with. Because it is
impossible to have a perfectly balanced and perfectly realistic game at this stage of
development, just use x, y, z, etc. instead of actual numbers. The closed-beta testers
will determine what actual values make for good demo gameplay in the current version
of the Engine Designer.
You are most welcome to make suggestions! Zeussy and Daffyflyer will then be able to
choose the best scenarios being presented here and save time in the process.
Score Calculations
Edit: We (Kubby, T16, and I) have worked out a score system for the demo scenarios.
It is percentage-based in regard to the hard thresholds set in the scenario technical
specifications. It works as follows:
x0 = threshold value for variable x
x1 = achieved value, better than x0
px = statistical weight of variable x as a fraction
Score = int( [ (Abs(x1-x0)/x0 + 1)^2 * px + (Abs(y1-y0)/y0 + 1)^2 * py + ... ] * 1000 )
Features of this system:
> The base score for a beaten challenge is 1000 points.
> Beating a threshold by more than the threshold value yields quadratically more score.
> The score is balanced by the set thresholds and the defined statistical weights.
> The results are easily comparable.
An example with two variables:
Scenario: Engine must have more power than 100 kW (60%) and weigh less than 100 kg. (40%)
Red: You managed to build an engine with 115 kW peak power output which weighs just 80kg.
(Blue: You managed to build an engine with 130 kW peak power output which weighs just 95kg.)
Power score:
Calculate base: (|115 - 100| / 100) + 1 = 0.15 + 1 = 1.15
Square and apply statistical weight: 1.15^2 * 0.6 = 0.7935 (1.014)
Weight score:
Calculate base: (|80 - 100| / 100) + 1 = 0.20 + 1 = 1.20
Square and apply statistical weight: 1.20^2 * 0.4 = 0.576 (0.441)
Total Score:
Add individual scores and multiply by 1000, dropping the decimal places:
(0.7935 + 0.576)*1000 = 1369,5 => 1369 (1455)
So your final challenge score is 1369. Note that with equal weighting, the 20% better
weight of the engine would have been more than 5% more valuable than the 15% higher
power output. That is why the blue engine scored higher - both via a higher statistical
weight and a "higher above threshold"-value.
I hope this example clarifies the suggested scoring system!
Cheers!
/Killrob
Automation's Engine Designer is on the final lap before demo release. Soon everyone
will be able to try this out themselves instead of being limited to drooling over Youtube
videos.
As the Engine Designer by itself does not represent what the game will be like when
it's finished, we want to spice things up a bit for the demo. The aim is to give the
player actual goals instead of just providing a shovel and a big sandbox, which will be
included in the demo as well.
Thus, the Engine Designer demo will feature scenarios / challenges that will put the
player in situations where engines have to be designed with specific limitations or to
certain specifications. Just like in the finished game, you will have to cope with the
ever-changing landscape of the automotive industry.
Your favorite engine-design scenario in the demo
(Remember that it's limited to naturally aspirated Inline-4 engines currently!)
1) Provide a short scenario description. What is the challenge about?
Example: 1973, USA. The oil crisis has a hard grasp on the country, and people start
to feel the pressure to maintain their mobility. More economic car engines sure
would help and turn the crisis into an opportunity!
2) What technical limitations should be imposed on the player in this challenge?
Example: Carburettors only, Cast Iron only, no DOHC, etc.
3) What technical specifications is the engine supposed to have, and how are
those weighted in importance? (Max. 5 different points)
Example: Economy better than x (50%), power more than y (20%),
price below z (20%), MTTF in MRevs better than w (10%)
The respective number in 3) is a hard limit, you need to be better than all of these at
the same time to successfully complete the challenge. If this requirement is fulfilled,
you get a higher score for larger margins you beat the threshold with. Because it is
impossible to have a perfectly balanced and perfectly realistic game at this stage of
development, just use x, y, z, etc. instead of actual numbers. The closed-beta testers
will determine what actual values make for good demo gameplay in the current version
of the Engine Designer.
You are most welcome to make suggestions! Zeussy and Daffyflyer will then be able to
choose the best scenarios being presented here and save time in the process.
Score Calculations
Edit: We (Kubby, T16, and I) have worked out a score system for the demo scenarios.
It is percentage-based in regard to the hard thresholds set in the scenario technical
specifications. It works as follows:
x0 = threshold value for variable x
x1 = achieved value, better than x0
px = statistical weight of variable x as a fraction
Score = int( [ (Abs(x1-x0)/x0 + 1)^2 * px + (Abs(y1-y0)/y0 + 1)^2 * py + ... ] * 1000 )
Features of this system:
> The base score for a beaten challenge is 1000 points.
> Beating a threshold by more than the threshold value yields quadratically more score.
> The score is balanced by the set thresholds and the defined statistical weights.
> The results are easily comparable.
An example with two variables:
Scenario: Engine must have more power than 100 kW (60%) and weigh less than 100 kg. (40%)
Red: You managed to build an engine with 115 kW peak power output which weighs just 80kg.
(Blue: You managed to build an engine with 130 kW peak power output which weighs just 95kg.)
Power score:
Calculate base: (|115 - 100| / 100) + 1 = 0.15 + 1 = 1.15
Square and apply statistical weight: 1.15^2 * 0.6 = 0.7935 (1.014)
Weight score:
Calculate base: (|80 - 100| / 100) + 1 = 0.20 + 1 = 1.20
Square and apply statistical weight: 1.20^2 * 0.4 = 0.576 (0.441)
Total Score:
Add individual scores and multiply by 1000, dropping the decimal places:
(0.7935 + 0.576)*1000 = 1369,5 => 1369 (1455)
So your final challenge score is 1369. Note that with equal weighting, the 20% better
weight of the engine would have been more than 5% more valuable than the 15% higher
power output. That is why the blue engine scored higher - both via a higher statistical
weight and a "higher above threshold"-value.
I hope this example clarifies the suggested scoring system!
Cheers!
/Killrob
