Eidolon Games

I’ve recently been playing Divinity II: Ego Draconis. With the release of Divinity: Original Sin, and the great reviews it’s been getting I thought I’d give another game in the series a try, especially since I already had a copy that I picked up in some bundle sale once. It’s kind of sad that I can’t remember where I picked it up, it was just sitting there in my steam game list, and I had no idea how it got there.

Dubious shopping habits aside, I’ve been enjoying it, and it recently made me revisit a common decision in RPG style games, to loot or not to loot. More specifically, the question is whether to loot or not loot items belonging to non-hostile characters. For example, do you go empty out all of the chests in the local farmer’s house? It makes an appearance in most games, and for the most part there are no consequences for doing it. It’s a common joke to point out the inconsistency of the supposed hero and savior of the world stopping by to go through the houses of everyone in the village and take every last coin, potion, or weapon they are storing.

This is often a moment of character crisis for me. When I play a character through an RPG, I often try to figure out what sort of underlying philosophy drives my character’s actions. This may be a very nerdy thing to do, but in a good RPG there are often decisions between two morally gray options. For example, there’s a man in the village who keeps pigs as pets, and some soldiers have confiscated them to help out in a famine in a nearby city. Do you let the soldiers keep the pigs so that they can help save the lives of people in the famine? Or do you return them to the man because he loves them like children? I find having an overall philosophy for the character helps me make satisfying decisions, and these decisions help shape that philosophy.

Fun tangent, but back to the main point: looting. I often find myself torn on this, on the one hand that potion I’m taking could mean the difference between life and death deep in a dungeon somewhere. But on the other hand should I really be taking stuff that isn’t mine? Like with the moral questions addressed in the last paragraph, I try to find a philosophy to follow regarding looting. What’s funny about it is how I often find myself thinking: “How can I morally justify taking this awesome sword?” I usually end up trying to follow a rule of taking anything that belonged to an enemy or is clearly abandoned, and I can loot anything belonging to a sufficiently rich group or individual. So I won’t take the ten copper from the house of the starving peasant, but I’ll happily take the sword from the king’s armory. Turning my character into a somewhat chaotic neutral person. I do good deeds and try to help people out, but I have little regard for the law and will take what I feel is fair game.

The thing about this is that I really wish it was handled better in gameplay. It often feels like those items are laid out with the assumption that I will take them. If I don’t take the coins, weapons and recovery items that are spread around, then I will have a very difficult time progressing through the game. I wish that games would offer some sort of acknowledgement of your decision to not rob everyone blind, and perhaps offer you something to make up for it. I suppose that would remove a lot of the moral dilemma, the prevailing thought often seems to be that it’s not really a moral decision if it doesn’t cost you anything to take the high road.

I think this has gone on just about long enough, but I did want to share one of my favorite stories about looting, this contains a pretty minor spoiler for Deus Ex: Human Revolution. Right after the first area you find yourself in Sarif headquarters as the head of security. In past Deus Ex games you had the option of sneaking around and pilfering all sorts of stuff from just about any area you found yourself in. So I went to work, sneaking through the gratings, reading people’s personal emails and stealing the candy bars from their drawers. All behavior perfectly in line with the character’s position as head of security right? Anyway, later in the game I returned to Sarif headquarters and checked my email. There was an email talking about how people around the office had recently started having things go missing, and would I please investigate. That was an awesome moment for me, I felt embarrassed about my immediate almost automatic decision to loot the place. It was an awesome demonstration that my earlier actions had consequences, and made me rethink many of my decisions throughout the rest of the game. I wish more games would do things like that.

This past weekend was the EVO Tournament, a large annual fighting game tournament. Coincidentally, as I’ve been working through the Extra Credits backlog I watched this video about how fighting games play out at higher levels of play.  I only watched a few minutes of EVO unfortunately, as interesting as watching these professional level game tournaments sounds, I find I would rather play or make games myself than watch someone else, but all of the coverage got me thinking about fighting games this week.

Fighting games are absolutely fascinating, full of attacks and counter attacks, interesting tradeoffs between risk and reward, as the Extra Credits video says, it’s almost like a chess match. Unfortunately, while I’ve always sensed that that high level game was there, I’ve never been able to tap into it and master it. I’ve bought a fighting game or two for every console generation, and put some time into all of them, but still never been able to figure it out. Part of it probably has to do with the way I play games, I like to play a game to conclusion, then move on to something else, trading a depth of knowledge for a breadth of knowledge.

But the unfortunate thing about my play style is that often the high level game is where things get really rewarding. There have been a couple of games where I finished the story mode, then stuck around to take part in the extra challenges. And those can often be more gripping and exciting than the main story. The tension that I feel as I work my way through an advanced challenge is remarkable, and it just mounts and becomes more intense as I near the end of a very well played run. I can’t even imagine what it must be like for the competitors at EVO.

This all highlights how much hidden depth lies in many games. I play enough to catch a glimpse from time to time, but it takes a lot of time to get that good, and I could easily play several games through at a mediocre level in the same time it would take to get really good at one game. Still, it’s something that I think needs appreciating from time to time, and definitely something that deserves some respect.

For the month of February my one-game-a-month project was Dungeon Sweeper, it’s a dungeon crawler with randomly generated dungeons and minesweeper inspired actions. Basically you start with the dungeon and there are lots of hidden areas. To explore you click on an unexplored square and discover what’s underneath. If it’s a monster, then you fight it and gain experience, if it’s loot you pick it up and grow stronger. If it’s a staircase you can go down it to find another level of the dungeon.

I think it turned out pretty good, although it is sorely in need of a couple of additional features. I was thinking that, like minesweeper, it would be good if you could have some sort of indication as to what was under the unexplored squares. I was thinking that perhaps it could have a number that is within 2 levels of the hidden monster’s level. That would give you a little bit of uncertainty and make the choices a little more interesting. I think it could also use some traps. I was thinking that the squares with monsters would have level numbers next to them, and the squares with traps and loot would have nothing. So you could tell if a square was a monster, but you would never be sure if it was a trap or some loot until you clicked on it. I think it could use some more balancing as well. At first when I was playing it there was basically no way to lose. You started so powerful, and gained levels and equipment so fast that you would quickly outclass any monster you met. I’ve re-tuned it to make it more difficult, but I may have gone too far.

Another thing it could really use is a better combat mechanic, currently you just click a button to attack the monster and it attacks back, and you just kind of keep clicking until one of you dies. You can use potions though, or run away, so there are a couple of choices, even if they aren’t terribly interesting. I think it would be really nice if there were different character classes, and they all had slightly different battle mechanics, it would make things much more interesting.

Still, I think it shows a lot of promise. I think with the new features I’ve listed it could make a very interesting game. I’ll most likely revisit it once I release Flame Warrior. Here’s the link again, in case you missed it the first time.

There isn’t really a lot to show unfortunately. Some streamlining of the power management screen, some new components, greater clarity in the ship management screen, not a lot of new stuff that you could see. There was however a pretty major overhaul of the skirmish scripting and mission scripting systems. The way the mission scripting worked previously is that you would define a mission, and each mission could have objectives. As you moved through the game events would be generated and passed to the mission scripting system. Each of the objectives would check the event and see if the objective was fulfilled. There were also actions you could add to the mission, these would check for the completion of objectives and do their thing when the objectives they depended on were fulfilled. So for example: The player clicks on a planet and the ship flies there, when the ship arrives at the planet a planet arrival event is sent to the scripting system. An objective watches for the planet arrival method and marks itself as completed in response. Then a cash reward object that was watching for the objective completion fires and puts some money into the player’s account. It worked pretty well for simple sorts of scenarios. But if I wanted to do anything remotely complicated, like watch for these two events in tandem, or this event or that one but not both, then I was out of luck.

But, I refactored everything using a brand new approach. Which meant I had to rip apart most of the main systems in the game. It’s nearly all back together and functioning properly, but at the moment everything’s broken and nothing works. The new approach is that you have a mission defined, and each mission has a set of triggers. A trigger has a set of conditions and a set of actions. The game sends events to the mission just like before, and the triggers each check the status of their conditions, then perform actions if all of their conditions are met. They also have a local context that allows them to read and write persistent data. This means that a trigger could look for a set of actions that can complete in any order, then fire when they are all completed, or fire only if one event has happened but not the other. It gives a lot more control than I had previously. Both the skirmishes and the missions use the same system as well, because the requirements for both systems were essentially the same. I’m hoping to have the whole thing buttoned up this week, and start playing around with the greater freedom and seeing what I can come up with.

This also marks a bit of a change in direction for the game. As originally envisioned, the game was sort of like a mashup of Steam Birds, Wing Commander Privateer, and Fire Emblem. Tactical turn based ship-to-ship combat wrapped up in an open world trading simulation with a compelling, ever changing story filled with interesting characters that interacted with each other. Quite a tall order, and early on I realized that the ever-changing story thing was a nice idea, but honestly not something I should focus on if I ever planned on releasing. Chris Crawford has been trying to make that dream a reality for the last thirty years, and still hasn’t found success. Perhaps there are approaches that could yield results, but at the moment my focus is getting something done rather than doing research. So, I realized that an ever changing story with characters that interacted with each other programmatically was right out. So I shifted my focus more into the open world space trading side of things.

Then came an explosion of space trading simulations. I don’t know what’s up, but everyone and their dog has decided that now is the time for making open world space trading games. Kind of funny because I’ve been sitting on this idea for around five or six years now, if I had made it four or five years ago it would have been wholly original. Fortunately none of the other games I’ve seen have taken my approach to combat, so I still have that. But I worry that by trying to make it a space trading game I’m going to be lumped in with all of the other games and get passed over. So I’m pushing it back towards the Fire Emblem side of things.

Fire Emblem, if you’re unfamiliar, is a turn based tactical fantasy strategy roleplaying game, where each of the units is a character with their own personality, back story, and unique interactions with the rest of the cast of characters. You move through the game fighting typical fantasy evil things, and as you do your characters become stronger, and develop relationships with the characters that they are most often positioned next to. It’s a pretty neat series of games and you would be well advised to check it out. That’s what I’m going to be pushing my game towards. I’m not sure about the characters improving their statistics, but I’m going to make it a more story driven and scripted game. Fortunately, this won’t mean a lot of extra development, nor will it mean throwing away big chunks of code.

I’m going to keep the open world map I currently have, the player will move around on it to get to the next story missions, they will also have the option of taking on side missions. These will probably also be heavily scripted and story based, but they will provide an opportunity for the player to earn more money and improve their forces if they feel they need an extra edge. The eventual hope is that I could put together a reasonably good story that would take around four hours to play through without too much trouble. Then I could release another story or two or three. They would use the same engine, same assets, etc. But they would add a lot of value to the game. My thought would be that people who buy the game early on would get the additional stories free, while each release would push the price of the game up for newcomers.

So that’s where it’s at, and I think that I’ve spent more than enough words going on about it. So I’ll end right there. And I really need to find some uniform and clever way of ending these things.

I took a day this week and made a quick game, you can play it here, although it does require the Unity web plugin. I decided to try doing this each month this year, we’ll see how well that goes. I think it’s a valuable thing to do because it gives you experience along the whole pipeline, from inception, to development, to publishing. And it reminds to me that I am capable of finishing and publishing a game. Finally, it gives me a chance to try out some ideas that might be festering in my brain. Also, I was interested in the Candy Jam as a sort of protest against King games trademarking the word ‘Candy’. So the game filled multiple purposes.

I’ve been wondering what would happen if you combined the endless runner genre with the run and gun genre (think Contra or Metal Slug). What came out was kind of interesting I think, and the idea could probably hold up to a more thorough examination.

I also wanted to try and learn a new game making tool. I picked up Multimedia Fusion in a Humble Bundle a little while ago, and gave it a shot. It’s kind of a neat little tool, but after getting a very basic game running, I gave up and switched back to Unity. Mostly because Unity has a much larger set of platforms it will publish to, but there were some things that would just be a lot easier in Unity.

However, I think my time with Multimedia Fusion was instructive. Instead of writing code for the game, you set up a series of events. Each event has a set of conditions that will cause it to run, and a set of actions that will take place when it runs. You can set up an event like: When the player collides with this object kill them. Every frame it runs through all of the events, checks if their conditions were met, and executes them if they were. It was a very different way of approaching the game logic, although it was very similar to how I was trying to set up mission scripting to work in Flame Warrior. So I think it was helpful for me to see how a system like that works when fully developed, and gave me confidence that my approach to the problem is sound.

It was a lot of fun, I forgot how enjoyable making a game in a day can be. I also was amazed at what a difference proper artwork makes. I found a set of 2D sidescrolling tiles that perfectly fit the candy theme, and sticking those into the game made a world of difference.

Come back next week for an update on Flame Warrior’s progress.

Steam is wrapping up their winter holiday sale, with it ends one of my favorite parts of the sale: the trading cards. I love the trading cards because you can sell them and earn enough for a game or two. This year I made roughly $5.75 by selling the trading cards I earned by voting on the Community Choice deals every day.

I sold all of my trading cards as soon as I earned them, figuring that the market value could only go down. But as I was contemplating the end of the sale I began wondering if that was the best strategy for maximizing my earnings from the sale. Would I have been better off holding onto my cards and crafting the badges in hopes of getting a valuable rare item to sell? I decided that I had to know, so I crunched some numbers and came up with some interesting results. I’ll warn you, this is a long post with a lot of numbers an (over) analysis. But if you’re anything like me, and I know I am, then you’ll probably find it very interesting.

If you’re not familiar with Steam or the trading cards, here’s roughly how it works: during the length of the holiday sale certain actions will earn you trading cards. There are ten different cards, if you collect all of them you can redeem them for a ‘Snow Globe’ badge, along with the badge you get a free promotional item randomly chosen from 59 items spread across ten games. These are items you can use in game, and give you status, especially if there aren’t a lot of them. Some of the items are more rare than others, granting more status in the game, and fetching a hire price in the market if you care to sell it.

I did this during my lunch break, as a mental break from programming. My wife laughed at me that a statistical analysis of the trading card economy was a mental break for me. So, all of the numbers listed here are as of 11:30 mountain time on January 2nd, 2014. First I created a table of all of the items you could potentially receive upon crafting a Snow Globe badge, along with how many of them had been crafted. I then divided that number by the sum to figure out how likely you were to receive each item. Additionally I calculated the odds of receiving each item to give a better feel for how rare each item is. So, for example, the chances of getting a Genuine Governer’s Salakot is 1 in 282,015.

[table id=1 /]

With the probability of getting each item, I then looked up the current market price for each item. This was a little bit tricky because there weren’t market prices available for most of the extremely rare items. I’m guessing the people who had them weren’t interested in selling. So, for the items that didn’t have a market price available I multiplied the 2nd most rare item price for the same game by 18. Because the most rare items that did have prices were roughly 18 times more expensive than the second most likely. In the end the rare items were so rare that even large differences in the market price didn’t affect the calculations much. As a final note, Steam has an upper bound on how much you can charge, something like $450. So even though some items have prices higher than $450 you couldn’t actually get their estimated value on the market. Basically I’m saying that you shouldn’t read too much into the prices on the most rare items.

Multiplying the market values by the probability of getting each item gives us a weighted value for that item. Combining all of the item prices weighted by their probabilities gives us an expected value. Essentially the expected value is how much money, on average, you would expect to be able to get for a random item from the list. Pick a random item and sell it, do it a few hundred times and the average price should be pretty close to the expected value. The overall expected value was about $0.25, and this table shows how much each item contributed to that value. As you can see the rare items contributed very little.

[table id=2 /]

So, if your primary interest in the Snow Globes was to make money by selling the items you get from crafting them, like I was, then you would also be interested in the cost of a Snow Globe. I got this information by summing the average market price for each of the cards required to craft a Snow Globe. You’ll notice that there are prices for a standard and a foil globe. There are foil trading cards that are dropped less frequently, I don’t know what the drop rate on them is, if you combine ten foil cards you get a foil Snow Globe. Which, in addition to being shinier, grants you five random items, rather than the one granted by the standard snow globe. The prices have fluctuated over the course of the sale, the prices for the standard cards are about what I was selling my cards for as soon as I got them. The foil cards though have come down in price. I got one foil card at the very beginning of the sale, and I sold it for $3.53, the same foil card today was going for $1.45. So, it seems that selling early/buying late is the way to go. In the end, it would cost you roughly $1.59 to buy the needed cards to get a standard Snow Globe, and $13.78 to get a foil Snow Globe. I also included the expected value of each snow globe. For the standard Globe it is just the expected value of a single item drop, but for the foil Snow Globe it is the expected value of a drop multiplied by five, since you get five items. You’ll notice that the foil Snow Globe costs more than eight times as much as the standard Snow Globe, but is, on average, only worth five times as much. So you’re better of going for the standard Globes.

[table id=4 /]

Finally I have a table showing each item with it’s (estimated) market price. Along with how much you would expect to pay if you were trying to get the item only by crafting either a standard Snow Globe or a foil Snow Globe. It’s pretty clear that it’s much cheaper to get the items from the market.

[table id=5 /]

So, what did my analysis tell me? It told me that my instinct to sell the cards as soon as I got them was right on the money. If your primary goal in collecting cards is to make money, then you are better off selling your cards as soon as you get them. It doesn’t appear that holding onto them improves their value. If you’re into taking big chances though, you could save up your cards and craft them into Snow Globes in an effort to get an item that you could sell for hundreds of dollars. It’s a little like playing the lottery, it could happen, but the chances are extremely low, and on average you’ll lose about $1.25 per globe. Much better to use that money to buy more games that you’ll never get around to playing.