WorldForge Community

Re: Decay - the real reason I think this is not implemented is because if there is decay then people who have all the skills feel like they "need" to go out and hit one mob with every weapon every day, or whatever other mechanic is in place. Either way, if skills decay, there is something players can do to prevent decay of the skills they want to keep. Most players would see this as a required "grind" to keep ALL the skills they've worked so hard for.

Efficiency decreases don't really solve that problem. Players will read "Archery 100 (45%)" as "Archery 45" every time, and will do whatever the game tells them they have to do to keep it at "Archery 100 (100%)". But they won't like the "grind".

Let's appeal to realism again for how this works in the real world. My primary skill is in Computer Programming. I also have a decent level of skill in Singing, since I practice regularly in my church Choir and Praise Team. I have a low, but usable skill in Cooking, Mechanical Work, and General Household Repair. Why don't I train up all these skills so I can make Culinary delights and not just Hamburger Helper, or fix my car or household appliances every time anything breaks and save thousands of dollars? Well, because then I would have to take away from time that I could be advancing (or just using) my skill in Computer Programming.

In reality, skills get easier to learn the more you know about them - to a point. Ask an olympic runner how hard it was to go from a 10-minute mile to a 9-minute mile? They'll tell you they did that in middle school. Then ask them how hard it was to go from a 5-minute mile to a 4-minute mile. They trained for years to reach that. I personally do good to run a 12-minute mile, but if I got in shape enough to get to a 10-minute mile, 9 minutes wouldn't be that far around the corner. I doubt I'd ever make it down to even 5 though.

Apply this to Computer Programming. When I first started learning to program, the concepts were completely foreign to me. I knew nothing of data structures or algorithm design, so my first programs looked like gigantic trees of if statements (apparently the same way the telecom industry's professional programs look today, but I digress). However, my strong skill in Math helped me to learn programming faster, because I had some basis for building on logic and variables. The more I learned to program, the easier it became - at least to increase my breadth. I can now learn a new language or library within a few hours, at least to the point where I can fluently write programs in the language and library as long as I have reference material nearby. But increasing my skill in a particular language and technology platform once I've learned the basics requires months or years of experience. If I want to get any better at, say, Swing programming in Java, I'd need to work with Swing enough that I begin to memorize the library, all of its classes, methods, and even order of arguments. For me, that will take a long time and a lot of experience, or a long, intensive, incredibly boring study session that I'll be less likely to remember than the experience.

We're really beginning to get into learning theory here. Research shows that learning is the process of strengthening synaptic connections. You're brain is always in the process of creating new synapses and destroying old, unused ones. So when you do something new, you use some of those new synapses. The more you do it, the stronger the connection becomes. But you also use synapses that are associated with related topics. When you learn something new by applying knowledge you already have, learning is easier, because you have fewer synapses to strengthen and reinforce - the ones where you are able to say "topic B is like topic A in this way" are already there and fully strengthened. As you gain a more complete understanding of a field and its related concepts, fewer synapses are required to "encode" the understanding of each new concept, to a point. Some research even shows that for some very basic and long lasting associations, your brain can encode this knowledge in a single synapse. For instance, it is often posited that a single neuron fires to associate your mother's face with the word "Mom" (or whatever you call her), even though it may take hundreds or thousands of synapses to associate other names and faces.

But not all learning is achieved by this altering of brain chemistry. For many tasks, there is also a strong physical component to it. Your body must be conditioned to be able to handle what your brain is telling it to do. Your muscles almost have a memory of their own that can only be taught by constant reinforcement. Think of touch typing. If you've been doing it for long enough, you simply say the words in your head and your fingers hit the right keys. You don't even really spell out the words any more. All of that is done somewhere between the language center of your brain and your fingers. But the process of moving tasks down from where you really have to think of them to where they are second nature is where true skill is gained, but where the only true gains come from actual experience.

Bottom Line:

So, with all that said (ok, way too much said), I think a realistic view of skill gains looks something like this:
Every skill has a "learnable" range. Let's say 1-100 can be learned. You can learn this by practicing, by reading a book, or by watching a master do it. It takes minimal investment to get to level 100, say a week or so of casual play, and when you get there, you have a usable understanding of the skill. Don't expect to slay a dragon with Archery 100, but you can handle yourself in the wild.

Also, every skill is related to other skills to some degree. The correct model of this system is an undirected weighted graph. Fishing is weakly related to axe wielding in that both required swinging your arm at the elbow. Chopping is more closely related to axe wielding in that it also deals with using axes and requires similar amounts of strength in each swing and stamina to keep going. The more skills you already know that are related to a given skill, the easier it is to learn. If you already know Fishing, it might only take 6 days to learn Axe Wielding. If you already know Chopping, perhaps only 4. If you know both, you may learn Axe Wielding to 100 in 3 days flat. The intention is, within a few months of play, you have every skill you care about at level 100.

Now, every skill also has the ability to be mastered. Mastery of a skill simply takes its skill level over 100, but is exponentially more difficult to achieve the next level. Furthermore, mastery of a skill decays over time. Level 100 is riding a bike, level 200 is competing in the Tour de France. You never forget how to ride a bike, but if you don't keep working out, you lose the ability to be competitive. So most skills you use on a semi-regular basis will stay in the 100-110 range. Your core playstyle skills will continue to increase until they balance the decay rate. Now here's a question that will have many opinions: how much should the 18-hour a day gamer be rewarded for their investment over the casual 1-hour a day gamer? My personal opinion is that the 18-hour a day gamer should perhaps be able to get 18 skills to level 200 and beyond, but no more quickly than the 1-hour a day gamer could get 1 skill to that level. In other words, cap the skill gains for a day at what you can do in about an hour. Introducing a mechanic that allows obsessed players to be 10x better than casual players because they throw fireballs 18 hours a day doesn't seem healthy, but some games thrive off of requiring this hardcore approach, so the system should allow it. Such a game would need to work harder to prevent people from macroing to 1000 fire - or introduce some sort of automatic offline advancement.