Openbracket

Creating an items table

Today I tried to do a little work on my game but I got stuck on a table which I have been pondering about for a little while now but today it stopped me so I need to address it.

In the game I am working on, players will have items, and will be able to use these items on themselves or others. And these items must be able to have an equation that manipulates the users and the effected players stats. This equation should be able to manipulate multiple stats of each player. I will eventually want to add splash damage but let's start out easy.

So, you have the user who we will call player 1 (p1) and the player that is going to be effected by this item; let's call them player 2 (p2).

Based on this idea I have been able to think of two solutions, one which I ditched almost immediately . But I don't know if the general idea behind this one is any good. So take a look and let me know what you think.

Items table

• id - 1
• name - bardic's staff of unimaginable penetration
• equation - eq{target:10*(p1.str+p1.dex) / p2.def-50}
• target - 0 // 0 = enemy , 1 = allies, 2 = all

As you can see, I am storing the equation in a JSON like object and would simply do a search on the search and replace the bold variables with player 1's and player 2' actually stats.  This method I think would allow me to easily keep the equation dynamic and clean.

Now lets look at something like area effects (splash damage, group shields) and how I would go about store such an equation in the table. This example has 3 equations in the equation object but only would be be used at most. I would either use target and others or all but I put all three is just for example.

• equation - eq{target:10*(p1.str+p1.dex) / p2.def-50, others:10*(p1.str+p1.dex) / p2.def-20, all: 10*(p1.str+p1.dex) / p2.def-20}

With this setup I can have multiple equations and multiple stats in either set up. I did have an option 2 which was to bust out all the stats into their own tables but the more I thought about it in comparison to this it just seemed clunky and overly complex for something that shouldn't be that crazy.

Now, to prove to myself that I could actually do this I wrote some a simple script to do such a thing. Right now the script is messier than it needs to be but I wrote it kinda quick just to prove that I can do that I'm talking about here. I will eventually clean it up when I have some time.

But the gist of the script is it hide's the p1.stats and replaces them with actual numbers. This would naturally all come from a database but for the example I just have everything there.

I'd like to apologize in advanced for the lack of formating. I've tried three different highlighters and they break things worse than how they appear here. So till I find a better... sorry >.<

$eq = 10*5 / p1.str + p1.dex / p1.chr;

function replaceHoldersWithVals($statArr, $str)
{
$pos = 0;
$len = strlen($str);
for($i = 0; $i < count($statArr);$i++)
{
$str = str_ireplace("p1.".key($statArr[$i]),$statArr[$i][key($statArr[$i])],$str);
}
eval("\$e = $str;");
echo $e;
}

Player Controlled Economies and Player Created Items/Quests

Sorry for the long absence. I lost my job and fell into a nasty slump. Its just one thing after another... But anyways, this isn't a blog about my life. It's about PBBGs! And so lets get to it.

As you can guess from the title, I am going to discuss player controlled economies and what would happen if the power to create new items and quest/goals were given to the players. This may sound like a crazy idea, but this level of freedom is not impossible.  Difficult and time consuming, yes, but not impossible. Perhaps through this post and the discuss to follow afterward, we will think of a more efficient way than I may describe in this investigation of the subject.

Player Controlled Economies :

This is something that most often seen in Strats I've noticed but a few RPGs have managed to pull it off beautiful. And the first one that comes to mind is Renaissance Kingdoms. In RK all goods on the market are created by players. Wheat, flour, corn, swords, gold, stone... everything. Excluding beer. Thing only makes a player happy. But the point isn't to describe how great RK is, but instead to see what happens when players control the price of goods.

So what happens when you let people price good at whatever they want? They will naturally try to sell them at the highest they can possibly make a profit at. This will eventually settle down to the perceived value of goods as more people start to sell their wares and figure out the amount of effort need to create the good compared to other goods. In these economies the rarity + time spent dictates the value of the good. So that ring of a fallen Gods son that is plus forty million to every stat and takes a two months to learn the skills and gather the materials will obviously cost one hell of a lot. Whereas the copper pretty ring, that doesn't do anything but look pretty, is going to be very cheap.

Allowing the players to price their own goods gives the community more of a sense of power over what is happening in their world. Making it so that the players create all the goods (or any of the good ones) gives them an level of involvement that you just don't get when dealing with NPCs.  It makes them feel like they have a place in the world and with skills, gives them a sense of progression with their character.

This is something I believe we should all look at closely. As I've stated in the past, player involvement in your game world is a very important feature, and its hard to get more involved than when you are the only baker in your town or the best sword smith in the county.

Player Created Items/Quests

What is one of the biggest limitations when playing an online RPG? In my opinion it's the lack of player actions shaping the world. None of the quests I complete change anything, or none of the items I get are really all that rare. Others will complete the same quests and get the same items.  But what if players could create their own quests, or create unique items? That level of involvement is something most commonly seen in pen and paper games where the players can directly communicate with the Game Master.

But say I wanted to create such a system for a web game. How could I even start with it? What would I need to do to make it work in a fair and balanced way? Wells lets see.

I'll start first with the creation of items since it is the easiest of the two.  I have designed two such systems in the past for games that will likely be a while yet before they get to see the light of day. But the systems work something generally like this: You're character has a list of stats and skills. Lets use generic RP stats:  str, dex, int, wis, con, and char. And lets say my skills are miner, jeweler, crafter, forging, lore, weapon knowledge, armor knowledge,  and imbuing.  As you can guess, each skill is connected to a stat. lore to int/char, miner to str/dex and so on. Now say that to master a skill or two is really difficult and there are 100's of different items that can be used.  So you database all the items, each one's properties using equations based on the players stats  to get it's numbers. But you also have an rather advanced system where each item has a relation to all the other items on how they mix based on a number of variables such as level, stats, skills, qualities, quantities, etc.

With a system of this size and complexity, you have pretty much created a world where even items of the same kind are unique due to the makers skill. And while players will be producing similar some of the time, players will be exploring and constantly working to find the best combination of skills and materials. And if you level all the properties unknown to the player or what the possible out come may be when combining 2, 3 or even 10 items, you will see a lot of player interaction just to get others recipes, or failed experiments.

When I said this was the easiest, I didn't mean it would be an easy thing to achieve or even design. This would skill take ages to design and develop but why not put the time in? I know we are for the most part more hobbyists than professionals making a living at this, but this is the time to start. With advance in technology, and an increase in the number of people using the web, now is the time to make PBBGS from a hobbyist thing to full scale game development. This is a plausible system,  but design and creation are hugely time consuming.

Let's now look at player created quests. This is the hard one. This is not just because of the code, which will be difficult, but because as of now, the best system I can think of to do this requires a human to go over each player submitted quest.

*pauses to go make a rum and coke* it is st. patties day after all. ^^

So, this is a similar to another system I design for another game that will be a while before seeing the light like my other ones.  But I have thought most of the system out. And one of my first thoughts of developing a system like this was how many quests the players would flood me with and how many would be just crap. So I thought, what if I only allowed donators to create quests?  That would cut back the number of quests I would have to filter through. I then thought what if I only let the donators do this twice a month. That would be great! I'd have players create the quests for other players. All that I needed to do was build the system.

*pauses for another drink*

I'll try to remain as coherent as possible through all of this lol.

So what would the system look like to do such a crazy thing? Well let me describe what I had thought of. First I would have the basic type of quests, where a player can pick a NPC by city, give the type of quest, the place the quest would take place, and the reward.  Several different type of quests would have to be offered and many many goals for each quest would need to be available but so its not always go and gather 10 bee wings for the spellcrafter.

The second type of quest that I intended to offer was one that allowed the player to create a boss-like creature and to give a back story. These ones  would offer better experience and better loot, but requires the players to put in more effort into their creation. Uses the same idea, pick a city, then the NPC to give out the quest.

And thirdly, a story arch. This one requires far more effort on the players part to create this quest. First, the quest would need to fit into the worlds story and history, since a quest of this type is meant to be a story, but just a fetch and deliver quest. This type would all the player to create NPC's if need be, a response based dialog and make it several tasks long. Such as one where a player needs to stop a small kobold raiding party that is attack outlaying farms. Upon destroying the kobolds you find that more will be coming, and you are given the quest to go and kill the kobold chief that is controlling them and to return with his head.

You can quickly see why a human is needed to ...

*pauses for another drink*

needed to review each quest before allowing then to be approved and made part of your game. You could of course discourage this by using a strike system or something of the sort where if you purposely create bad quests you get a strike. Too many strikes and you can no longer create quests.

This system, while not as complex code wise as the items, is still difficult because it requires someone to check each quest. Limiting the number of players that can create them and the number those players can create saves the humans much trouble.

So as you can see, while both these methods are difficult to develop and can be time consuming to manage, it is possible to do. I would now like to open the floor to hear your comments. I am very curios to hear your thoughts on these idea's and what you think about the longer development times.

And as stated in my earlier post, we have a new writer here at OpenBracket. Hopefully this will prompt me to write more ^^

The 'tick'

This has been a rather popular post and seems to be a big question and topic of debate among developers. I, for one was confused about how to appoarch this when I first posted about it. But after reading comments made by people smarter than me, and learning more about databasing and scripting I feel I am ready to touch the submit again. Some of this will sound very similar to comments made in my original post on the topic mainly because I feel those people had the right idea.

You may call it a cron job, a timed event, or something totally different. I prefer the term tick. I like it because I think of it has an internal clock where each tick of the clock is defined by the creator of the game. So a tick may be a minute or an hour. But the subject of the this post isn't exactly about what a tick is, but more so how a tick works.

Let's start with an example. Every 20 minutes I want to give the player 10 more energy. If I were to use a cron job, every 20 minutes the server would grind to a halt pretty much as the cron ran through ever users entry updating their energy field. There is also the possibility of using AJAX. When some people hear this they think using a javascript timer that updates the database every few seconds... That's a worse idea than the cron job for bring your server running into a wall.  You could use AJAX to update the players data every 5-10 minutes or so. But I think this is still a a rather inefficient  way of updating the data since it only works while the player in online.

The method I am adopting is the last click. This style of 'faking' a cron is from what I can tell the most efficient on the server, will cut back on lag, and can handle server down time in a method of your choosing.  The gist of it is that every time a player perform an action in game, it will store the a timestamp of that event. Then on the next click you compare the difference between the store timestamp and the current timestamp and perform the according game functions based on the time past.

"But what if I interact with another character?"

Thanks for asking. Say I go to attack a player that has been offline for 5 hours. As soon as I view that player in a way that is more than their name,  it will update their data (you will likely want to use a different method when looking at score lists or ranking so you don't have 200 people updating every 15 seconds). You may think this method would be difficult on the server, but with a properly designed database (at least 2nd normal form) and some well written script, this method will allow for your game to be very scalable. In other words, you could have 2,000 or 40,000 members without the need of rewriting all your crons or updating servers every few months.

"But who would I manage a searching or score lists?"

Say you update the top 100 players in several fields every 20 minutes and the search tables every 5. All you would need to do is simply create a small table containing the last updates timestamp. You compare the two, if it fails to equal for than your set interval, ignore it. You just saved yourself updating all those users unnecessarily. A different method may suit you and your design better, you be you the general idea of it.

The last click way of doing this does take longer to be designed and built but in the end this will pay off. Instead of updating 5,000 members in a row all at once you now update 5 every few minutes. You are performing more updates but the requirements on your side won't be as near as depending, eliminating lag before it occurs.

I hope that was all was clear and addressed some of the questions I have been receiving.

Oh, and by the way; if you disagree with me, please share your method. I write in the blog not just because I want the attention ( ) but also because I want to help the community grow and develop. So please, share your thoughts and comments... unless they are negative. Then you keep those ^^