Showing posts with label pserv. Show all posts
Showing posts with label pserv. Show all posts

26 August 2012

Dead MMOs closing their eyes on private servers?

There are many MMO server emulators out there. Some private servers are run for profit, and when it is the case, the official and legal operator of the game can sue for disgorged profit, as was the case in Blizzard vs Scapegaming in 2009. Interestingly, out of the $88M Scapegaming had to pay, $3M came from disgorged profit (the private server was running micro-transactions), and $85k came from statutory damages for copyright infringement. Some private servers are run for fun and do not even accept donations. In the case of Ragnarok Online, Gravity used to send cease-and-desist letters to the private servers with the largest number of players. Servers with a dozen players did not seem worth the hassle.

Both WoW and RO are still running. However, the policy towards private servers seems to change when the MMO is discontinued by its legal operator. For instance, SWG closed in December 2011. SWG had an emulator called SWGEmu. In August 2012, the president of SOE unofficially supported SWGEmu: well officially I have to say it's bad. but that's officially. SOE or LucasArts could have been copyright trolls, but instead, they have decided to close their eyes on private servers. Maybe we can expect the same from Blizzard when WoW dies?

17 September 2010

Open source MMOG and VW servers

Current MMOG clients are usually thick clients. Hence, programmers just need to emulate the server side of an MMOG. Here are several projects that could be worth looking at.

MMOG Project name Active since Programing language(s) Client Particularities
EverQuest EQclassic (no public sources) Early 2009 C++ Based on EQ Trilogy (second expansion) From scratch
EQEmulator (SVN) August 2003 C++ Latest EQ client The EQEmu test server is one of the biggest private servers
EverQuest 2 EQ2Emulator (SVN) July 2007 C++ ? -
Lineage 2 L2Jserver (SVN) June 2004 Java ? -
Ragnarök Online eAthena (SVN) 2003 C Latest official or custom-hacked (XRAY) RO client -
Helios Emu (SVN) June 2006 Delphi ? inactive?
- Red Dwarf (SVN) 2007 Java ? Forked in February 2010 from Project Darkstar, originally from SUN
Ryzom Ryzom Core (SVN) at least 2004 C++ both client and server became open source in May 2010
Second Life Open Sim (Sources, SVN, forge) 2007? C# Latest SL viewer? allows many different viewers, hypergrid, IBM collaboration
Star Wars Galaxy SWGEmu (SVN) December 2007 C++ ? -
Ultima Online UOX3 (packs) October 1997 (inactive now) C++ ? many forkings
World of Warcraft Arcemu (SVN) June 2008 Lua and C++ Latest WoW client? both branched from AscentEmu, but I can not find its source code
Mangos (SVN) September 2005 C and C++ Latest WoW client?

13 September 2010

Ethnographic play on a 25-rate RO private server

I wanted to play RO for some days. Something quick, but not too easy, and with not too many people to have to talk to while leveling. Kind of a nostalgic try. So I picked a private server which drop and XP rates are 25. That means killing a monster brings 25 times more XP to the avatar than on the official servers, and a 1% chance to drop an item becomes 25% on this server. I conducted an «ethnographic» play: based on my (relatively extended) knowledge of the game I analyzed the server through player avatars in the game but also on the control panel of the website. I found the private server on ratemyserver.net (RMS), but I will not write its name here so that nothing bad happens to its administrator. Launching a private server is an illegal activity after all.


Demographics

At 3pm on Friday, the control panel showed 2 players were connected: T an Assassin and J a Hunter, both around level 80. These are the most solo classes of the game. When I logged in, it was not long before T arrived and, without a word, sent me an invitation to join guild G1, which he was the guild master of. I accepted, and saw in the guild panel that J the Hunter was also in the guild. Three days later, on Sunday night, there were 9 players connected on the server at 7pm. Among them, 4 belonged to guild G1, 2 to guild G2, 2 were not in a guild and 1 was an AFK merchant, used as a mule selling items to other players.

As mentioned before, the guild master of guild G1 was T, an Assassin. He had spent the last 2 days camping equipment on a low-level randomly-appearing monster called Ghostring. He had reached level 98 and had not made any other character. He was very confident and experienced in the game, and gave advice to the other 2 players he had recently recruited in his guild. IRL he was a 19-year-old jobless American. He started playing RO in 2007, had played on iRO Loki (official international server) and 5 other (much more populated) private servers, ranging from 6 to very high rates.
The sub Guild Master was J, the Hunter. He had actually played a lot and was now a lvl-96 Sniper. He had managed to collect a few Field Manuals in fishing, a custom activity of the server. When used by a player, manuals give 3 times more XP for 30 minutes. Although J seemed less confident than T, he said he started playing RO in 2006 on RuRO, the official Russian server. He gave spontaneously several items to other guild members. J and T talked quite much together in the guild chat. IRL, J was a 19-year-old Russian who worked in a call-center. He found his job boring and managed to play at work.
A was recruited after J. He was a lvl-90 Monk, a quite complex class combining support and high damage attacks. He was a basic member of the guild but talked a lot with J and T when he was connected. Each of them gave advice about where to XP or which equipment to get. He said he started playing on several euRO free-trial accounts in 2004, and tried another private server later. He looked quite unexperienced. IRL, he was a 19-year-old German working in a call-center as well (but not playing at work). When they realized through the guild chat that they were all the same age and two of them had the same job, they enthusiastically sent Facebook invitations to each other.
B was a lvl-96 Rogue recruited last. Rogues are a mixed class between Assassin and Hunter. B was not connected at that time.

In the other guild, the guild master was a quite experienced lvl-98 Hunter very in view on the forum. He had, among others, argued that the custom items provided by the administrator were balanced while another player thought they were not. The second player in the guild was a cheerful Bard, a support and/or high-attack version of Hunter. I was enrolled by him in the guild, and he even asked if I needed any piece of equipment. During week days, they were never connected before 5pm, I concluded they worked or went to school. Most of the time they were connected, the guild chat stood silent. When they happened to speak, it was in German to each other and out of the guild chat. Later, the Bard added that the 25-rate server opened following the wipe of a high-rate server. Maybe they knew each other before the low-rate server opened?

Server features

The capital city was Prontera. @go was given to players, but not @storage or @warp; this meant players had to go back to the capital to go where they wanted or use their storage. In my opinion, this increased the frequency of little social events that happened in the capital between players. The custom NPC were those usually found on any private server; they were:

  • The Warpra teleports players wherever they want for free. This solves the problem of long and boring transportation time when there is no player to teleport other players to their destination.
  • The Healer heals players and solves the scarcity of support-class avatars who, most of the time, need a group to XP quickly.
  • Stat and Skill Reseter: resets the avatar skill and stat points. If a player wants to change his character build (eg from a Support Paladin to a Tank Paladin), she can.
  • Job Master saves players the time to do job changing quests.
  • Banker, Identifier, Bonus skill giver (to save time to the player), Universal Renter (for Cart, Falcon or mount)

There was also a Kafra Employee that lets the player save his respawn point when dead or open his storage, but this is an NPC found on official servers as well. A vending zone had been added as well where Merchants could set up their shops, but the server population was too small to have any virtual economy. The guild versus guild event, the WoE, was supposed to happen twice a week but was deactivated by S, the administrator of the server, probably because there were not big enough guilds.

S was German, although his advertising for the server on RMS mentioned the server and forum language was English. He came on the forum once or twice a day to reply to suggestions or technical problems with the game. He did not harass players for votes or donations, and stayed remarkably quiet. In the game, his player account was the Monk A. As a Game Master, He never improvised any live event for the players. Technically, he designed and coded his own website and control panel, manipulated client-side files, added and edited a few small NPCs in the game, tried (and failed...) to balance his custom items (eg the fishing rewards) but never seemed to have modified the server source code.

Conclusion

Certainly, the players from this private server had different origins (Russia, Germany, France, US) and had different gaming expertise. However, all the players were all males of the same age (16-25) and had all tried official servers and/or private servers before. They all played solo but their belonging to a guild showed they wanted to socialize with other players. Alone Together.

This private server had several custom hats. They all required grinding at least a hundred items, and rewarded the player with unbalanced items (eg inflicting +12% damage on other players). Somehow, it was too easy, but players did not seem to see it this way, or at least they did not complain about it - for now... I doubt the administrator had the community management, game design, programming, and graphic skills required to make his server grow as much as he would like to, but I am sure he had the will. Private server administrators are a particular kind of consumer of MMOG company's products - they use the game client. Of course, this is an illegal activity, or maybe a not-yet-channeled activity. MMOG companies could choose to design their games and pay attention to these determined and skilled consumers. They could be included in the product community as content contributors.

18 May 2010

[Literature] A qualitative study of Ragnarök Online private server in-game sociological issues

Back in November 2009, I mentioned the multiplayer game design issues impacting social life inside RO. The paper has been accepted at FDG2010. You can read the final version.

I conducted interviews and observed players' behaviors in an official server (2000 users at peak time) and two private servers (200 and 2000 users at peak time). On the examined private servers, the social environment and some of the game mechanics were more adapted to players' needs. Examples are:

  • a broader choice in avatar customization
  • the ability to observe parts of the game world while not being logged-in (the Control Panel)
  • less constrained groups
  • events and GM taking the player community more into account
  • easier XPing (leveling in RO is the main activity, and it's a lot of grinding) thanks to rates and the Warpra+Healer combo
  • @commands enabled to players so that they can see who is online, sell merchandises with a character while playing with another and compare prices

Clearly, the Warpra+Healer totally changed the game. Champions could camp the respawn of MVP (boss monsters) and kill them relatively quickly with Asura Strike. This left virtually no end-game challenge for guilds other than the bi-weekly War of Emperium (battles between guilds in castles lasting between two to three hours).

20 March 2010

History of RO emulators

This post follows this one about private server emulators. To my mind, understanding the story behind RO emulators can help explain how some servers easily reach 140,000 accounts and 1000 PCU as of March 2010.

This cladogram took me roughly 20 hours. I relied on a very stuffed tree with no dates (made by Tsuyuki before 2008). Other sources are mentioned in the SVG cladogram here. A (google) translation of the Chinese website I gave for AEGIS can be found here. My cladogram ends in 2008 because nothing has changed since then, except maybe a recent ASB battle cry. Many ending dates could not be found, so I sometimes ended them arbitrarily.

I could not find when AEGIS had been stolen from cRO. Given that cRO opened in May 2003, and UniAEGIS and ASB, based on AEGIS, started in early 2003, there seems to be a time problem. Unless cRO was in beta before 2003.

I could not make any link between Athena and AEGIS. Both seems to come from Japanese developers/hackers, but I do not know if Athena was reverse-engineered from scratch (in the pure sense of an emulator) or based on AEGIS.

Cladogram made with gnuclad and converted from SVG to PNG on fileformat.

Edited March 22 to add Cronus, eAPP and other AEGIS leaks and ASB updates. Split of Weiss (delphi/pascal) and Athena (C).

17 March 2010

Private server emulators

Before anything, this post does not (and can not) provide enough sources. And considering the legal issues faced by private servers, many of the links included in this post may become broken at any time. However, I have had some experiences on RO private servers and I have looked at several other MMOG-emulator source codes. Although I believe that for some games, there are more people playing on private servers than on the official ones, private servers (hopefully?) do not receive a lot of coverage. In December 2009, the BBC wrote a somehow sensationalist though decently informed article in which they interviewed a few private server players and administrators as well as official MMOG companies.

General facts and numbers

Although projects such as bnetd or PvPGN are multiplayer gaming service emulators, they do not fit into the category of MMOG emulators mentioned in this post as private server emulators.

Emulators are simulations of the server side of an MMOG. Private servers launch these emulators to provide free versions of mostly any currently popular MMOG: Aion-emu, eAthena (RO), Arcemu (World of Warcraft), L2J (Lineage 2), Clone Wars (SWG) or EQ2emulator (Everquest 2). Some emulator projects such as Mangos try to teach about large-scale C++ projects and claim our software is not intended for running public servers, and we do not support that.

In fact, there might be private servers for both P2P and F2P MMOG (although I could not find any emulator (or private server) for Maple Story, Runes of Magic or any gpotato game for instance). It can seem obvious that some players go on private servers to play for free, but there might be other reasons such as particular custom game design choices making the game easier or with less grind.

I personally believe that private servers could be used efficiently to increase the game longevity at no cost for official MMOG companies. As Celia Pearce wrote in book 1 of Communities of Play, some players even instigated a network of player-run Uru servers to allow players to run the game after its initial closure. Such servers cost nothing, generate more UGC (from better quality) than simple players. This idea is actually close to the notion of abandonware, but with much more players. Playing an old famous game makes potentially players want to know recent games from this company. In an era when Intellectual Properties are golden (see the Final Fantasy, Call of Duty, Fifa, Warcraft, GTA and other Mario series), it may be wise to advertise the most recent sequel of the IP series through the series' former hits. Anyway, whether official MMOG companies want it or not, it is increasingly unlikely that any company, government, or nation can successfully inhibit the near-term and mid-term societal dispersion of FOSS or the FOSS movements (Walt Scacchi, 2007). Better use or influence it than endure it.

Players and administrators

One may wonder how much money do official MMOG companies such as Blizzard, Gravity or NCSoft loose. This lost is hard to estimate for several reasons. First, neither the number of private-servers nor their number of players are known. Until now, I have not heard of any population estimation. Second, it is not obvious that private-server players would pay anything to play on official servers. Moreover, the hardware and network infrastructures needed to host an illegal server can cost less than a hundred dollar per month for several hundred peak concurrent users. Although some servers rely on a few player donations, most of the time the administrators do not need to rely on a RMT system to keep their server afloat. Hence, most private server players are not likely to be sources of revenues if they were to play on official F2P MMOG servers.

Until now, most of the emulator projects I have seen are the results of a forking from an other open-source emulator project. Usually, there was only one main emulator project set up during the alpha or beta version of the MMOG development. The duration of an emulator project depends on the motivation of the open-source team. Update follow-ups, maintenance (quick and efficient fixing of the bugs reported) and technical support (tutorials, guides) mean a lot to an emulator community. But at the same time, they mean a lot more exposure to legal authorities, and official MMOG companies threats.

Emulators

To connect to a private server, most of the time, one has to use a slightly altered version of the official game client. The original client can be downloaded from official websites. When the original client is installed, a text file such as Realmlist.wtf for WoW or sclientinfo.xml for RO can be edited to make the game launcher executable connect to the private server. The whole process is really easy, and I do not understand yet why such files stay accessible, cleartext and human-readable at the root of the game folder.

The server-side is, however, fully developed and maintained in open-source projects. At the beginning, original MMOG servers are either reverse-engineered or stolen. That is why I find that depending on the age of the pirated MMOG, the emulator community relies on different pillars. Emulator projects need sneaky and relatively high-level hackers to start. Then more mainstream coders come along and contribute to the project code base. Official updates are reported on forums by people who play on official servers or directly from database websites. Support is given to novice server administrators, and a community can develop and share the products of its creativity (new items, new quests, etc.).

27 November 2009

Towards another community management feature

I detailed some of the current community management features in one of my previous post. Community management can also be done in an active and «preventive» manner instead of simply being reactive when events happen. For instance, Blizzard has organized its fourth BlizzCon in August 2009. During this, some Developer Panels showed to players how new levels were created. I do not know if there were particularly charismatic developers inside the Developer Panels. But if most of these panelists were unknown of the player audience, either work should be done to improve their visibility to the players or panelists should be people more visible. Also, I think that a close contact of the developers to the players from times to times may boost them because they see the mass of players they are working for and it is really rewarding. Including costumed players from the BlizzCon 2009 in game products like the Diablo III box is a very smart way to get Real-Life UGC and show your players/fans you care about them.

Moreover, I hope there were key players in the audience of the BlizzCon. By key players, I mean Tobold, Ensidia core guild members or Greedy Goblin. As an example, Greedy Goblin writes sarcastically about how to make money in WoW. His blog had 2k subscribers in June 2009 and 3.5k now in November 2009. His latest blog traffic graph shows peaks when he argued with Scott Jennings about layoffs in industry or when he showed his inventory and bank storage. Goblins are not easily one's friends, but if Blizzard manage to bring this Goblin to the BlizzCon in paying for him the flight and the hotel, Blizzard do not only get 1 more friendly player but 3.5k. Of course, each blog subscriber will not be affected the same way. But if the blog owner publishes nice comments concerning Blizzard, they will touch many players.
This point was actually suggested by Nicolas Ducheneaut in a discussion we had some weeks ago. I filled it with examples and included it into my post.

Finally, a crazy idea. Since MMO are attracting more and more people. Real-life player meetings change the way players see the game as they realize that there is someone behind the pixels. Some players even stopped insulting others after they met them IRL. So in a tupperware consultant style, why not organizing metings regularly in big cities? I heard about communities like DS in Paris who organize meetings in Paris Cafés where they play Nintendo DS games together. The Java User Groups (JUG) are monthly meetings where Java developers gather to talk about Java world news, learn new techniques, share knowledge or drink beers. Maybe MMO companies are not wealthy enough at the moment to pay for monthly buffets and refreshments in each of the 188 urban areas of the world counting more than 2 millions inhabitants. However, MMO companies could set up electronic tools on their websites to help their players organize local real-life meetings about their favorite MMO. This sounds like organizing raids, so developing the web app should not be too difficult.
I actually can write about the positive effects such IRL local meetings could provide as this kind of meetings actually happens/happened for some RO French private servers. Some active players or people in the server team sometimes organize real-life meetings called «IRLs» at Asian-culture conferences like the Japan Expo or simply in their home city when there are enough people to meet.

  • The GM team (aka the community management team) knows players faces, discusses IG issues directly and more openly than on a forum
  • The Development Team has direct feedback from players. Also, after the meeting, developers are no longer writing code for the fun of it, they realise that hundreds of people rely on them to have fun.
  • players realise who are the people behind the game, and demystification of the work is sometimes followed by admiration of the people: «they are human, they are doing something fun for me and they spend so much time on it»

So even if no developers can attend the MMO meetings, which is likely to happen very frequently, such events are a good way to promote the MMO company, especially when this company has sponsored/helped organizing the event. This is the way to transform the video of the left into the video on the right. The video on the left was done by an active member of the Alliance-RO French private server community to describe the game and attract new players. This video was done by a fan of the game and the server. The video on the right was taken by an active member of the community in Alger, Algeria. Players say in the video and in the written comment that they have uncovered the secret reason why gentimouton (yes, it was me), a Game Master of the server, does not understand them: he is 70 year old (French quote: nous avons percé le mystere mysterieux du GM:gentimouton ! encore fois °° !!). If you want it, you can ask me for a full translation of the video dialog. This video of a meeting of players in their city brought a precious (but late) feedback to the GM.

26 November 2009

Community Management

Until now, I have heard of 4 kinds of community management. Eric Heimburg explains 2 ways community managers behave with the community:

  • denying the bugs and ignoring the feedbacks of the players: the example of Aion where the American community management team relies on the Korean technical team to do bug fixes. It is actually the same for the WoW European servers which do not benefit from the American development team on their forum.
  • acknowledging their weakness, playing the crowd and feeding the forum trolls to keep them relatively sedated. According to Eric Heimburg, this strategy is followed by Champions Online and WoW on the American community forums.

Eric Heimburg also wrote that developers or support team representing the company to the community can screw up the consistency of the strategy, giving the example of a burning reaction of WoW Lead Systems Designer Ghostcrawler answering a Nerfs: Ghostcrawler, DIAF thread on wow forums.

Another kind of community management is the one followed by LambdaMOO developers, named wizards IG. Wizards have stopped to intervene in game problems since a LambdaMOO takes another direction publication of Pavel Curtis, aka Haakon IG. Julian Dibbell describes how this decision has impacted the community and how community leaders facing holes in the game case law sometimes act independently and in a non-concerted manner, leading to irreparable actions like the toading of a player. In the end, players are now in charge of their own problems in a participatory system where they only sporadically need wizards to enforce a decision the players' democracy has taken.

And finally, the way some efficient and very productive private servers of RO organize their team is interesting. This ideal structure is rarely entirely followed by private server teams. Even using only parts of this structure improves the productivity in giving each person of the team his/her appropriate place where he/she can play his/her role. On a beta server, developers can test their updates with a bunch of trusted players who are eager to test it. On the beta server, game developers are almighty like the Trainman in Matrix, they can create items, kill monsters, etc. with in-game commands. However, on the production server, they are normal players. The forum is cut into sections that enable the work to be shared among the different parts of the server team. Community managers watch the whole forum and report any valuable entry to the game devs or to the admin in a staff section. When community managers do not know how to answer a question, they post the question in the staff section so that a technical guy can answer it in person. Since community managers organize in-game events and have game master powers on the production server, community managers and developers are on an equal footing for the players. The server leader(s) communicate to everyone in the team, recruit new people and have in-game powers on both beta and production servers. They take important decisions like server policies and directions, future updates or the permanent ban of a player. They do not have to be technical or particularly present in the production server, but they must be aware of everything. Instead of spending hours reading the forum, they know what is going on thanks to the community managers. This hierarchy looks a lot like the scrum process of Agile development where the Scrum Master is the server leader, the Product Owner's feedback comes from the Game Masters/community managers and the Team is naturally the Game Developers.

I think which strategy to adopt depends on the size of the community and the skills of the team. As Eric Heimburg wrote, Aion is big enough to indulge a weak English customer service but less polished MMOG have to take care of their players. The RO private server hierarchy might work well for a community of a few thousands of players, but maybe not for bigger communities.

15 November 2009

[literature] A qualitative study of Ragnarök Online private server in-game sociological issues

Edit: this post does not reflect the latest version of my paper. I will know whether my paper has been accepted in April, and I will update this post accordingly. Feel free to contact me if you have any question :-)

This is the summary of the paper I have written. This paper may not explain entirely why I did not post anything in the last 4 weeks, but it still took me much time...

Abstract

position of the paper

In the last decade, online games have raised much attention as more and more players gather on game servers. In parallel, communities of illegal private servers have spread and host numbers of players comparable to the official game server. Based on the Korean online game Ragnarök Online, I conducted unstructured interviews to collect qualitative data on different private servers and on the official French server. This paper explains why Ragnarök Online private servers gather so many players and how private servers could help improve official games.

Plan

  1. Introduction
  2. Playing Ragnarök Online
  3. Methodology
  4. Fulfilling player expectations
    • Rates
    • Avatar customization
    • In-game events
    • Recapitulation
  5. Private server issues
    • Group level-range
    • The Warpra-Healer combo
    • Player information
    • Virtual economy
    • Recapitulation
  6. Issues shared between official and private servers
    • Double Accounts
    • Lack of monster challenge for guilds
    • Donation and subscription
    • Recapitulation
  7. Conclusion

    The wide panel of private servers answers adequately the wide panel of player expectations. Although small private servers intrinsically lack players, their very concerned administrators find interesting ways to get round their problems. More populated private servers benefit from the same technical features as small private servers, but they also share common problems with official servers. Looking at how the most populated private servers deal with player problems might show enlightening.

  8. Future work