Saturday, 29 September 2012

How to (maybe) Fix (a subset of) Everything

There have been a number of little balance changes recently for the better.  These changes indicate to me that Tiny Speck may be gearing up for release.  Before we get to that point, I figure I should explain how to fix everything that is currently wrong with the game.

I could obviously give some kind of disclaimer to indicate my humility on this subject - you know, I don't claim to even know all the problems let alone the solutions; this is just one idea of an approach that might work.  Instead I'll leave such qualifiers as an exercise for the reader and instead express myself more and more arrogantly as the post goes on.

First, let me state a few goals for the changes I am suggesting.  I believe these goals are consistent with (but not necessarily included in) statements from Tiny Speck about where they would like to see the game:

Encourage Player-to-Player Economy
Tiny Speck has explicitly indicated that this is something they want to do.  In order to accomplish this goal we basically have two sub-goals: 1) to make it easier to conduct trade; and 2) make sure that goods are of different values to different players.

Minimize the Advantage that Beta Players have over New Players
Tiny Speck has never stated that this is their goal, but I think that it is obvious enough that this is important to the health of the game at release.  New players can't feel like they are latecomers to someone else's game.

Give Player Choices Intuitive Results
I am not talking about reducing the whimsy in the game (the choice to Squeeze a chicken produces grain, which is hardly intuitive) but about reducing the hidden complexity and unintended consequences of many game choices.  Buying more energy makes your mood harder to manage up to a point but then unlocks an upgrade that ultimately makes your mood easier to manage.  Choosing Animal Kinship as your first skill is sort of like setting the game one "Easy" while choosing EZ Cooking is a little like setting the game on "Hard" (relatively speaking).  That's the kind of thing we should be avoiding.

With those goals in mind, what sweeping, radical changes could we make to the way everything works to get where we want to be?  First let's look at why we aren't already there.

Problems Here and Now
As I said, Encouraging the Player-to-Player Economy could be seen as having two parts.  Changes to the game in recent months have done a lot to improve player ability to communicate about trades they want to make and to facilitate trades between players.  However, we still have a problem that Glitch is a game where everyone can do everything.  If three different glitchen all have Animal Kinship VII and Arborology V then they have the same relative values for Meat and Eggs.  There is little point in one player offering to trade Meat for Eggs with the other.

Value differs for different players, but mostly it's just a linear thing.  A rich glitch with all the skills values everything more than a new glitch does.  That means that when I want money I sell things to the vendor and when I want things the best way for me to get them is to buy them from other players.  Trade in this economy happens because one player values currants more or less, not because they value goods differently.

That is a natural segue into the discussion of the massive advantage that long time players currently enjoy over newer players.  This advantage manifests in two ways.  First, we have more money and more resources on hand.  But more importantly, we have skills and the ability to generate wealth.  Perhaps most importantly of all is that the long-time players are actually able to do everything by having all the skills.  They had the advantage of learning all the skills when the number of skills was smaller and when the learning penalty was only 3% per skill.  If new skills continue to be added to the game at a rate of one per month then it is possible that new players will never actually be able to catch up and learn everything while older players will have them all.

Lastly, there is a lot of unintuitiveness in the game right now.  If we are fixing things, we should try to make sure the fixes reduce this problem rather than adding to it.

How to Make Everything Perfect Instead
So here are a summary of the radical changes I am recommending:
  1. Base nothing (except for the daily refill) off of maximum energy
  2. Change the effect of being over brain capacity to increase energy consumption rather than increasing learning time
  3. Dramatically reduce the cost of energy, quoin multiplier and brain capacity upgrades up to a moderate level, then have them ramp up to their current values
  4. Rebalance all crafting actions around the concept of value per second crafting rather than percent increase in value of ingredients
  5. Rearrange the skill tree to normalize the value of a skill and remove most prerequisites
First I'll talk about how these changes combine to address the problems above.  Then I'll go into a little more detail about how to make the changes themselves.

Basing nothing off of maximum energy will help to make choices more intuitive and reduce the advantage that older players have over newer ones, it is also critical to implementing the second change, which is by far more important.

Currently the effect of having too many skills is that it takes you longer to learn new skills.  The moral of that story is that your goal is to have all the skills - just that it might take a long time to get there.  It's sort of like a progressive tax system, you move into a higher tax bracket when you pass your brain capacity, but you still want to earn as much as you can.  This, as I noted above, will be a huge source of imbalance between beta- and release- players.  Beta players will be playing an all-skill endgame that new players may never be able to reach.

If instead the penalty for having too many skills actually affected you in the game then this would solve these problems.  The new "end game" state is one where you choose which skills you want to have and how many skills you want to have.  Different players would have different opinions on how many skills was the right number.  New players, after a month or two of learning skills, would be on equal footing in terms of long term potential with beta players (though obviously still behind in terms of current holdings).  But also, different players would actually have different skill sets.  If having all the skills meant a significant drawback within the game then people would choose the skills they liked and not have the others.  As a result there would be different capacity to generate different resources, meaning that there would be a reason for trade.

One of the recent balance changes we saw was a somewhat unsatisfactory change to the "Ticket to" upgrade cards.  There was a big problem that these cards were excessively good for people with 100 quoin multiplier and Tiny Speck had to think of a way to reduce their effectiveness for those with maximum quoin multiplier without making them terrible for people with low quoin multiplier.  The solution wasn't terrible but it was far from ideal.  A much better solution is to change the game so that we don't have glitchen with 100 times as much quoin multiplier as others.

Similarly, whether you are low level or not you can pretty quickly get skills that allow you spend 600 energy at a time.  But at low level it is a lot of work to get your energy tank to 600 or more (it was quite a bit easier when you got energy through levels).  What I advocate is massively reducing the cost of energy, quoin multiplier and brain capacity up to about 1500, 25 and 40 respectively.  Right now, buying up to these quantities would cost nearly a million imagination, which would require being level 28 if you literally bought no other upgrades.  I'd like to see that lowered to around 200k, so that by level 20 more glitchen would have values around those levels.  Energy could then ramp up quickly to it's current price by 2000, quoin multiplier could increase to it's current price by 35 and brain capacity by around 45.  This would provide a kind of baseline for expected values for these vital statistics and define a sort of "mid-game" as being reaching those values.

But doing this, we'd know that quoin multiplier will never be more than about four times "expected" value, instead of being 10 or 20 times.  We would know that glitchen who have played a little while will have enough energy to take on crafting tasks without breaking them up into small batches.  We would know about how many skills someone would have before being expected to fully participate in the economy.

From the recent rebalancing of potions we know that the way Tiny Speck views the return on crafting is in terms of percentages of input materials.  They noted that they increased potions and tinctures to give 25% more value than the input procedures.  This is a flawed way to look at crafting, though.  Instead, a value should be assigned to what a second of crafting gets you.  Using the "percent increase" philosophy, if there was a cooking recipe that required two meat then it would output a good worth 25 currants.  Since all cooking takes one second, that would mean 5 currants are being produced per second.  If, however, there was another recipe that took 100 meat then it would give an output worth 1250, and we would be making 250 currants per second by making it.  The value per time of crafting would be determined by the skill level required to craft.

Which brings us to the last point, rearranging the skill tree.  Right now some skills give a lot more value than others and the prerequisites for skills are all over the map.  Also, in some cases the skill to process materials relies on having the skill to gather those materials (you can't refine without mining, for example).  That reduces the ability of glitchen to work together and reduces the interest in trade.

This post has already gotten very long, so I'll leave the details numerical explanations of what I mean by each thing to future posts.  When I've got all of it done I'll post in the ideas forum to let Tiny Speck know how to make everything better-than-perfect instantly.


  1. Tiny Speck should hire you as their in-house statistician. All of these changes are really well thought out and go a long ways towards fixing this daunting rebalancing problem.

  2. I really worked to get to the full skill list and I would hate to have that taken away. I put a lot of time into the game, so why shouldn't there be a layer of content for the people who've put that sort of time and cash (I'm a moly subscriber for a while now too) into the game. Why does having some people in the game with higher levels cause a problem for newbies? If that's truly the case, then Glitch would always have a decreasing userbase since nobody would want to join after missing the first cohort. I don't think that's a true model of player behavior, but I'm interested in knowing how you came to that conclusion.

  3. I didn't mean to say that it's a simple matter of high level players making new players feel unwelcome. What would be unacceptable would be if players who played in the beta were given some kind of permanent advantage over players who didn't. Skills is a particular area of sensitivity because you get them without playing the game.

    Imagine, to use an extreme and bizarre example, that they decided to get rid of the arborology skills but left those who currently have them with them.

    On the other hand, if the perpetual state of the game is that everyone can have all the skills and all the upgrades then they have a problem with a player-to-player economy. Basically the economy will continue to be as it is now, largely one where the only trades are richer players buying from poorer ones (because we value currants less) or economically exploitative trades - where one player is buying for far less than they feel it is worth and the other player is selling only for the joy of selling. The joy of selling is a legitimate reason to sell but it's hard to expect that it will sustain a meaningful economy in the long run.

    If some people are good at animal kinship and others are good at mining then in makes sense to trade meat for sparkly (well, that will never make sense, but food for EHSP might).

    The solution I'm proposing would mean that you could continue to do everything if you wanted, but you would be less efficient overall than people who chose to really specialize in one thing.