Wednesday, December 23, 2015

The case for and against a Launcher

Many players do not understand the need for a launcher. Also having done a launcher for SWGChoice I also learned much about what is and is not good for such a launcher.

Given the licensing rules for SWGEmu servers, every legal server needs a mechanism to do due diligence in verifying that the player probably has a legal licensed copy of the game. Also due to how SOE distributed the game you need a mechanism to distribute most of the game client files to the player. So this means that someone wanting to run an independent SWGEmu based server would need to create an installer to do the due diligence and get the needed files.

A server can stop here and let people create a link to the game client and launch it through that mechanism if they have no plans to update the client with new in game items or other changes. In fact creating a launcher may be way more work than they would normally need.

This creates the problem of how do you keep the client files up to date. For those that played on the Bloodfin server before they created a launcher that also updated the client files it was not uncommon for people to log into the server, do something and not see the item in their inventory. This was due to the client having no information on the object so that it could not render the item in the client.

The best resolution for the problem is to modify the installer to also launch the game and before launching the game check to see if the client files need to be updated. Then update them if needed.

Now that you are using a launcher you have the ability to communicate with the players on your server there additional things that you can do. You can open up a page to communicate information to the players like the changes made in updates and if the server is up or not before they actually launch the game. You could even go as far as providing quick links to the server web page and forums. If you are accepting donations for the operation of the server you could even provide a quick link to a page where they could donate.

There is also a subtle thing about having a launcher. Given the correct naming of the launcher and client files, players that use game helpers such as Raptr will find that the SWGEmu is now listed with the rest of their games in one location and their play time will also now be tracked.

So it is very easy to see that anyone operating an SWGEmu based server should also be using a launcher.

Now what makes a good launcher.

I personally feel that it should provide a single location for players to do the following:

  1. Install/verify/update the game client
  2. Provide access to the SWG configuration tool
  3. Since Kodan's talent calculator is freely redistributable, it should be part of the launcher distribution and it should be accessible through the launcher.
  4. Provide communication to the player about the server.
  5. Provide access to the server web page
  6. Provide access to the server release notes
  7. Provide access to the server forums
  8. Provide server up/down status
Now here is what I have learned.

To provide communication to the players I simply distributed the cumulative patch notes. Since a good server should have patch notes this is a very simple thing to add to a launcher. I eventually came to the conclusion that it was better to have a link to the patch notes and to present to the players a small, dedicated page that simply communicated information such as seasonal events or upcoming server updates.

I learned that while downloading an updated file manifest with CRC information and then processing the CRCs would keep the client up to date it could take more time than we wanted. A better system is to version the CRC file and to store on the client system the version number that matches the local files. The Launcher would then get the version number for the client that is on the server and to only download the manifest and check the local files for update if the server has a newer version of the client. This check is considerably faster.

Since there are many personalities who play MMORPGs, I have also decided that a good launcher should have a configurable fast mode. This is how I imagine that it should work. If fast mode is selected when you start the launcher it checks to see if there is a new version of the client to download. If there is no new version, it immediately launches the game client and shuts down. If there is a new version it updates the client and launches to full launcher so that players know that it has made changes and so that they can easily access the release notes or just launch the game. Any launch of the launcher after the client was updated would still use fast mode unless the player changes the configuration. If you code a launcher to work like this you should also set it up so that the player has a way to get to the launcher configuration menu once fast mode is turned on.

An ideal launcher should also be able to check to see if the launcher itself has been updated. Many recent game launchers have the ability to detect a need to update the launcher and do it. This should also be a part of any quality launcher.

At this point everything else associated with a launcher is probably cosmetic. So can you think of any feature that I missed and should be included in a launcher?

Tuesday, December 22, 2015

Item Rarity

MMORPGs tend to use item rarity to encourage mindless grinding of MOBs as an in game activity. They do this by making the item very desirable so players want it so that they do the grind. Then the game developers try to set the rarity such that most players that are farming for the item get it just before they quit in frustration.

There are two reasons why game developers resort to this. The first is to keep people playing. If the game is subscription based this is easy to understand. If the game is funded by micro-transactions this is not very intuitive. But what happens here is that very often the game developers add something to increase the chance of getting what the player is farming for. Often these items boost the look chance or loot quality depending on how the loot system is designed and where it fits into the loot scheme. So keeping people playing boosts income.

The second reason why developers resort to this is because it is very simple to do. You add to the game a way to add MOBs that drop loot and that system allows individual items to have an individually assigned loot chance. Once this foundation is in place it is easy to add the item to existing MOBs or to add a new MOB/Loot combination. There really is no thinking involved and if there is an established guideline for how rare items should be the variable of what to set the chance to get the item has been defined in those guidelines. If a developer was assigned full time to add items like this and test them, it would be possible for that one developer to add a hundred or more of the items in a month. So occasionally assigning a developer to add an item would not be much effort. This would allow a smaller development team to increase profits.

Given the licensing agreement for SWGEmu based servers, maximizing profits is not a consideration since mandatory charges to play the game are prohibited. If anything the income side of the game must be driven through encouraging donations to keep the server running. So how things are done in the game would be influenced very strongly by how they increase enjoyment of the game and making the game so enjoyable that people want to donate to keep the server running and their having fun.

This causes a real problem with item rarity. While players may be willing to grind longer for items to make them feel that their past monthly subscription fees were justifiable, this is not true for micro-transaction and free games. Unfortunately for micro-transaction games this means that the items that you sell have to be so good that people often feel that the game has become a pay-to-win model. (Personally I only use that term when the only way you can get items is to pay real money to get them. Paying to make getting an item less frustrating I find deplorable but not truly Pay-to-win.) For truly free to playt games like SWGEmu this means that what was acceptable in subscription SWG for drop rates would not be acceptable on a SWGEmu server.

There is another problem though and that is making the drop rates way too high. I am not talking about the miscreants who like to jump in and disturb things by claiming that the change is to pander to those who want instant gratification. There is a big difference between removing frustration and improving fun by reducing a grind and handing things out by completely removing the grind. The miscreants simply want to stir up trouble.

The real problem is accidentally doing what the miscreants are afraid of, making getting things way too easy. If getting items in an MMORPG is too easy the MMORPG quickly looses the fun factor. It would be just as much fun as playing Blackjack with all cards face up and where you could see the next few cards to be dealt.

There is one feature of MMORPGs and SWG was one of these that makes it harder to determine when drop rates are too high or too low and that is the ability to hunt for the item AFK. If you want players to get an item after playing for a month there are two number that could do this. If you believe that people play ATK for three hours a day and seven days a week this is roughly 84 hours a month. If the MOB difficulty and spawn rate is such that you get to loot 20 times an hour you want to set the loot chance so that after 1680 they will have looted the item. A very straightforward situation until you consider AFK play. AFK play could allow players to get eight times the chances a day. So if you use the ATK number you will have AFK players getting 8 of the item in a month. Also if the MOB is uncommon you may have the AFK player reducing the chance for ATK players to get their 20 loot chances an hour.

SOE generally would cut the loot chances so that the AFK players would have the one a month odds and ATK players were simply very frustrated. It also developed in SWG a mindset that those that could AFK reliably should be allowed to do so and sell what they looted to the ATK players to make money. There was no need to try to allow ATK players to be anything other than frustrated by trying to get it for themselves. After all getting it themselves was bypassing the player driven economy.

This is why there are discussions in the past SWG forums and in the SWGEmu forums about there being too much AFK play or drop rates being too high on some items. Players are not stupid and at some level understands this.

The real question becomes how do you allow both AFK and ATK play and control item rarity so that things are fair?

What I would probably do with an SWGEmu server is the following. I would create a mission type that could be done that would generate a mission only MOB that could give players the 1680 loot chances a month. This would deal with the ATK play since I have not seen anybody be able to AFK 24x7 mission terminals using the in game AFK tools. For those that want to AFK grind I would place the item on a wild spawn and set the loot chance such that they would get the item once every 15,000 loot chances. This would provide both ATK and AFK players a path to the items that would result is getting about one a month. This would require a little more work but would not take any play style away from either type of player.

Personally I do not like AFK play, but it has become so expected by SWG players to be able to AFK that it is far less problematic to create ways to work around the problems that afk play causes than to remove the ability to play AFK.

Wednesday, December 9, 2015

SWG Nudity Mods

I have to admit that as a person I do have an interest in this. It was one of the early client side mods to do this that got me into using and then making client side mods to the original game. It was the fact that the distributed mod was not fully to my liking and included changes that I had no interest in that caused me to learn how to modify your client and to change existing mods.

I am also a strong believer that when the originator allows that SWGEmu server operators can and should include client side mods that they feel enhance gameplay on their server. For this reason at SWGChoice we included the client side changes to reduce the Hula Hoop effect of tailor crafted belts as part of the .tre files that we downloaded to players.

I was also raised in a household that while being religious never felt that nudity was a good or bad thing. I was raised that the passage on nudity when it comes to Adam and Eve being nude in front of God had little to do with nudity being bad and more to do with Adam and Eve feeling that people should not be so casual in front of God. Being too casual could be seen as a sign of disrespect. After all it is a little hard to be dressed more casually than being nude. Even today where dressing casual has become common for church goers, I still dress up to go to church.

So the question is why then did I not consider shipping this as part of the SWGChoice .tre files and why do I feel that no SWGEmu based server should do it.

The simple answer is that I understand that my views on nudity are not universal. Some people are not comfortable with nudity and others feel that it is inappropriate around children. So the best path is to leave things as SOE did them and for characters that are completely undressed to have underwear on. This is a no work effort and is a reasonable compromise since players who really want to see toons with no clothes on can still place the files in the filesystem to override the .tre files and get this effect.

The more complex answer is why I stopped using nude mods years ago. SWG is not a game that lends itself to characters not wearing things. Because of this there is no natural situation in the game where a nude mod would come into play.

If you do not know what I mean about the game encouraging you to wear things, consider this. In the movies, TV shows and comics you almost never saw anyone wearing a backpack. In SWG you see almost everyone wearing one. The game mechanics force you to wear one or be at a disadvantage. Another similar situation is headwear.  As a human male there is very few pieces of headwear that looks good.  Once again in the movies, TV shows and comics you almost never saw anyone wearing a hat but in SWG players do, even though in most cases it makes them look stupid. Once again it is game mechanics that is forcing them to be worn.

I was overjoyed when SOE added the option to independently hide backpacks and hats. This was almost as good as when SOE added the appearance tab. With the appearance tab your smugglers could wear the armor that they needed to be combat effective under the game mechanics while looking like the smugglers from the movies, TV shows and comics. This of course brings up the question of why I did not start using nude mods when the appearance tab was added. The answer is both simple and complex, it did not fit. As I had matured as a gamer and came to feel that at some level roleplay was always important in MMORPGs, even at a barely there level, I also decided that nudity, outside of Wookies, simply did not fit in well with the portion of the Star Wars universe that SWG portrayed. If this was a Conan the Barbarian, Leisure Suit Larry or even a Sims MMORPG then nudity would readily fit in as a part of it.

So here is the real answer to the question of why I have no nude mod in the server shipped .tre files. That answer is that I do not feel that it fits in with the SWG game. I do know that in general Wookies do not wear clothes in the game and many people try to turn Catina's into nude strip clubs, but in the end I feel that it does not fit the game or improve it for the majority of the people you want to attract to a server.