Panumbra Newsletter #2 December 3rd 1997

No. 2, December 3, 1997

Dear Panumbra Newsletter Readers,

I hope all of you who live in the US have had a nice Thanksgiving. We certainly had an interesting one, involving the unavoidable turkey, some days off and a fire... so let's get on with the news.

1. Fire, fire... We had a fire in our office building on Sunday, Nov 30th. A transformer blew up right under the windows of the office next door!  Our neighbor's office went up in flames and was completely gutted. We got extremely lucky, the fire department put out the fire very quickly, so it did not spread into our office.
   
   Nothing got damaged in our office, except for heavy smoke damage. Everything was covered with a thick layer of ashy dust and the air smelled horribly from smoke. We could not work for two days, until our office was cleaned. Only today we got access to our computers again. They all started up OK and seem to work fine. We are very grateful that nothing really bad happened, but we lost two days of work.
   
2. Demo release news. All that said about the fire, it is very unlikely that we can release the demo on the scheduled 12th of December (Friday). We will give it our best shot, though. We will try to run all of our system test scripts by the end of this week, so that we know all the bugs and can fix as many as possible.
   
   If we cannot make the 12th of December, the release will be delayed until January. Namely, we will miss our favorite and only server programmer, Dave, who will embark on the journey of blissful marriage and will enjoy his honeymoon for the rest of the year. We wish him all the best! Anyway, if we make December 12th, the demo will probably have quite a few bugs. We will have no on-site support for the demo, but we will check the servers every day during the holidays.
   
3. Difficulty of development. There is a good reason why you do not find many graphical massively multi-player games on the Internet. Any large-scale multi-user system takes years, dollar-millions and specialized expertise to develop. Game companies are wizards in getting the most out of one PC, but typically do not have the expertise to deal with the issues of a large-scale system. Developing breath-taking graphics alone takes years and millions of dollars. You are talking about two large-scale development efforts, a heavy-duty server and a cutting-edge client combined into one. That is why Ultima Online took so long to develop, costs so much to play, and has so many bugs (despite of Origin's best efforts). 
   
   We wanted to develop a graphical RPG that is free for users, and it is easy to access. Add that we did not have money but knew how to develop client/server systems, since Jonathan and me used to be business systems consultants. Picture is clear: the game industry lacks a kind of knowledge we have (large-scale systems), so let's fill in the gap, but do not try to get into the business other people do a lot better than us (graphics).
   
   That logic led us to develop the game engine as a heavy-duty client/server application. The server is designed to be very robust, with sophisticated technologies, which are completely invisible to the user unless they do not work. On the interface side, we chose to develop a lightweight client, and concentrate on the essentials but make it playable on as many different platforms as possible. Java was absolutely suited for that purpose. The lightweight client has serious limitations, but it gives us an tremendous advantage as well: players do not have to have the latest, best equipped machine to be able to play it.  Why am I writing this? It is because it explains why the graphics are the way they are, as described in the next section.
   
4. Graphics. Several people asked for more details about the graphics, so here you go. Our client is in Java, and that brings some serious limitations. First, Java is slow when it comes to handling graphics. This is why we do not have animations. Second, every image must be downloaded from the server. That is why we build the world from a handful of graphics, and avoid using a lot of added bitmaps. Third, every function adds to the size of the Java applet, and as a result, to the initial download time. Therefore we try to avoid any "nice but unimportant" features, so that the client stays slender. Now for the look of the graphics in detail:
   
   a.) World. The player will see a first person view of the world. The graphics are still, and they change when you move or turn. No animations or sounds.
   
   b.) Monsters. A still oval picture of the monsters will be displayed overlaying the graphics. There are four slots for displaying four types of monsters, and a number after their name indicates how many there are. You can see a list of individual monsters in the monster list tab of the information panel. The monsters are displayed all the time when you can see them and they can see you. It looks a little weird that the monsters do not actually stand where they are, but you can get used to it pretty quickly. The reason why we did it, apart from the slowness of Java, is that we wanted players to be able to see all monsters that can see them, all the time. If you place the monsters in the view, you can only see the ones that are in front of you. If you have to turn around to see all monsters, you may loose time due to lag, and they may kill you before you would notice them. In addition, this way we can have as many monsters in the same area as we want, as long as there is no more than four different types of them.
   
   c.) Characters will not be displayed graphically. They will be listed in the "Chars" tab of the information panel. There are two main reasons why we do not have character graphics (yet). First, additional graphics have to be downloaded every time for any new character you encounter, and that would slow down going around the world a lot. Second, and more important, is that this way we have no limits on how many people are in one area or just in one map square.  That is all the graphics. We considered having skill icons, but we decided not to have them since we potentially will have hundreds of skills, at which point it will be hard to come up with new icons which are not confused with the old ones.
   
5. Development news. We received our demo server, and set it up. We continue system testing, and hopefully we can run through all of our test scripts by the end of this week. There are a fair number of client errors, a lot of them due to differences in browsers. It looks like Internet Explorer is significantly faster, but sometimes crashes with no apparent reason. Netscape is more stabile but slower. On the server side the main issue is database performance, we made a lot of progress today, and continue working on it. In the mean time Brian is frantically adding new world areas, races, skills and templates, and balancing data. The mines are currently very unsafe for a first level character (you die almost every time you meet a monster), but balancing continues. 
6. 
   
7. Web site news. The number of subscribers to the Panumbra newsletter reached 58 today. We are listed on the Mudconnector, Geoslums, Yahoo, and the MPOG directory, plus added to a bunch of search engines. We also have two new web pages on our web site, a schedule and a team page. If you joined us before Nov 23rd, please check it out. Soon we will update the web site with a new schedule and new screenshots. A description of races, templates and skills is also in development as well as an online user manual. So keep checking!

That is for today. We keep you updated about the demo release.

Thank you for reading our newsletter!

Cheers,
Marta Gabnai - lead designer