In 2009 I was hired to create a game for advertising my clients products on the Agritechnica exhibition of that year. In the game, called Harvest Mania, the players harvests grain crops using a combine harvester, while benefitting from the products offered by my clients.
On the fair, it was great getting real feedback to the game — above all, kids seemed to have a good time, while parents liked the productivity-topic and the lack of violence. The Wiimote-control we implemented just 2 weeks before the fair also paid off: playing with a steering wheel made the game that much more accessible, and some visitors even wanted to take it home.
HarvestMania on John Deere’s fair booth, at AgriTechnica, Hannover, 2009
At the same time, the feedback revealed some significant problems and imbalances that need to be fixed, not to mention the lack of real content. So after the fair, we couldn’t put it online immediately.
Since then, the game has been on a hiatus due to my personal circumstances. Only recently have I resumed work on the project in order to finally release the game and have a closure.
One of the aspects that I felt needing a redesign is the scoring system. As I’m struggling with it, I decided to use writing to aid the process. Besides this primary purpose, the post will not go much into other aspects of the game for now, so sorry if things don’t make sense to you. I do plan to write a Developer Diary #0 at some time in the future though.
The game mechanics
Yeah, this is the screenshot from our 2009 podcast. It’s that old.
As you can see there is a lot of information on the screen, arguably overwhelmingly so. Back then, I was too deep into the game to see it and come up with a better solution. But it’s also true that the game design simply needs to juggle many aspects in order to let the player experience the benefits of the advertised products.
The product that affects the moment-to-moment harvesting process most is an auto-pilot, called AutoTrac, and the benefits are precision and comfort, allowing higher yields while having less stress.
The stress level is modeled by having a mood indicator that goes down when steering and auto-regenerates when not. It seems to be less problematic for display than the yield.
The challenge of driving a harvester
Unlike typical collectibles in games, simply enumerating the income is not sufficient for Harvest Mania. The quality of harvesting expressed by yield amount per distance has to matter, because AutoTrac allows you to optimize the yield per harvested area by tuning the driving speed to the varying consistence of the crop.
You can’t really do that when you driving, because steering itself is less trivial than one might think: Fields are harvested by driving mostly parallel tracks back and forth. You want to minimize the distance traveled in order to reduce gas usage and soil burden. To do so, you want to maximize the cutter usage, so that each meter driven counts as much as possible.
At the same time, you need to be careful not to go overboard with avoiding the adjacent empty track, lest you end up leaving a thin tracks of unharvested crops. Recovering those would be a really big waste of all resources involved.
So, reducing overlap with harvested areas has number 1 priority. Optimizing the yield quality by speed tuning comes second, and is only viable when you don’t have to focus on precision driving.
This is why the yield quality has to be a playable challenge for the AutoTrac to matter. The game models the yield optimization process by having three qualities depending on how well the driving speed matches the varying speed requirement. The better the match, the more crop you get from the same surface harvested.
The Displays
Click to enlarge.
In order to make the varying yield quality noticeable, the grain is first collected in the the GrainCollector, a smaller container that fills up much quicker than the silo. This way, the player has tiny completions all the time (positive feedback galore: each fill goes “ka-ching!”), and the frequency of the “ka-ching!”s indicates your efficiency.
To further encourage speed optimization, there is a partial pie chart below the silo, showing the percentage of each yield quality.
The lower right corner of the playing field, next to the minimap in the side panel, is the waste display. While harvesting spawns flying crops towards the GrainCollector, overlaps spawn waste particles that go into the hole. This is the primary negative feedback of the game.
Looking at the silo, you’ll also notice the circular icon with an “i”. It is the logo of John Deere’s “i-Solution”, and AutoTrac is part of that product line. The icon there indicates the silo fill required to make the AutoTrac available. I deliberately made it available only after some harvesting, so the player gets to compare manual and automatic driving.
The last addition to the HUD, short before the exhibition, was the collectible “partner logo”, in the upper center of the game view. It was introduced to allow partner companies to plug into the game and brand it for personalized/localized promotion.
There are some other elements, but for score calculation purposes, these are the most noteworthy.
Please ignore the wrong sum. The scoring system is broken — both game design and code.
Partly for showing the performance analysis feature of the advertised product, a report screen concludes each level in Harvest Mania. You see four subscores that add up to a combined score, the Harvest Mania Index, or hmi for the sake of brevity.
Although the subscores were chosen to encourage more efficiency and less stress, the screen is very crowded and it’s hard to get why all the data matter.
Redesign in progress
Gee, look at the time (and the length!). I’m not even at the redesign yet but the above write-up already helped me getting the head back into the game.
So the goal of the redesign is to have a less crowded HUD, a more transparent score system, ideally using a single live number that’s relevant, comparable and understandable without a later summary, all the while retaining frequent completions and constant positive feedback for harvesting. The later summary should still be there but be more readable, which obviously depends on the actual score calculation.
I have some ideas that can be tackled directly in my next session:
Gameplay / HUD
- Use absolute numbers instead of a silo fill as the leading number.
- Use partner logos to activate AutoTrac and save one display.
- Get rid of the huge silo / small GrainSelector separation.
- Use small silos that represent crop units and fill up quickly.
- Get rid of the pie chart and make the fill rate of small silos more noticeable.
- Use a multiplier display for the yield quality and more arcade-style clear feedback.
- Replace the time-limit with gas/distance limit.
Report screen:
- Use a formula that shows the relationships of the game subsystems: total crop amount * (1 – waste %) * average mood level
- Varying crop while using the same gas limit implies efficiency already in part.
- Explicit efficiency is expressed by waste penalty, carried over directly from the HUD for transparency.
- Mood level translates steering stress to a transparent penalty as well. Maybe this should be expressed positively as bonus…
See you soon. (Diary #2 is up with an updated HUD mockup!)
edit: check out diary #4 for yet another, probably final, redesign of the HUD.
