Oneironaut Escher

I make games in SL. SL is awesome for it. Several of the programmers in the audience probably just groaned, and rolled their eyes. I understand other script languages are more versatile and far superior in lots of ways, and believe me, I'm painfully aware of how underpowered SL is as a game-logic/graphics/physics engine. But in lots of ways, most specific to me, I find it superior for game design. And, I'm mostly that - after 30 years, I've finally decided on what I want to be when I grow up - a game Designer. I'm much more focused on game ideas, and gameplay - I hammer out the how and what later. Up until recently when I discovered the wonder that is MicroSoft OneNote, I'd fill notebook after notebook with pictures and ideas, working out specifics of Game Ideas.

SL's real-time immersed building and scripting tools lend themselves well to on the fly testing of game components and mechanics, which is helpful for me as an uneducated amateur game designer and scripter. I can rapidly test lots of different ways to go about something, and more importantly, early on, start 'playing' bits of the game to see if the reality of play is as fun as I thought in Design. This allows for a lot of dynamic back and forth between design concepts and actual construction of the game. It fits the way I work normally, with design changing, infusion execution, and execution shaping aesthetics and gameplay. SL's weaknesses and limitations on the execution side almost invariably end up being a plus for me - in streamlining my scripting and such to work on the TI-85 Programmable that SL seems to be, I end up tightening and improving design concepts along the way.

Okay, that's it for me defending my wacky notion that SL is a good design tool. That's not why we're here.

SL sucks. Well, for marketing the types of games I do end up making in SL. Ironic. Almost all of my 'good' games are high prim, gigantic, sim resource pushers. The hermit who owns a sim and wants to pay what the huge number of hours a project would warrant - is a fairly difficult demoggraphic. I've tested other marketing concepts - ideas like licensing a game to entertainment sims - and not much has worked. Ideally, it would be nice to be able to sell a copy of a game to individuals, and a lower price made possible by higher numbers of sales. Sorta like shareware. But, the market that wants to spend minimally on a game doesn't tend to have an empty sim to even be able to play the game in.

OpenSim has been piquing my interest lately in thinking about this issue. What I'd like to see is a redistrubatle, plug and play OpenSim install that could be hard locked with content like a game. Ideally, with abilities to strip down unneeded SL features and lock down world options, etc. A person purchasing the game would get an executable/install program that would install a barebones OpenSim instance, with the game pre-rezzed and ready to go. SL becomes a game engine.

I say barebones as, I realize lots of people are pushing their comps just to run SL, making running a simulated server on top of that not feasible. Eliminating unnecessary features would allow for more focus on executing the game more smoothly. Actually being able to play a game you buy being a big plus.

My feeling is something like this will come to be, maybe just not yet. I guess my question to you, Dear Reader, would be : "Are people working on such things, does something like this already exist? Is it feasible, or even allowed with licensing and such? Could it work as a marketing avenue?"

Besides games, application and other normally too complex (I'm looking at you, in world Sculptie maker) tools would work as well. I'd like to see, say, a stripped-down modelling/building tool. Less cpu and graphics intensive than a 'Live' instance of SL, purpose built for, um, building - useable even on low end SL machines as a combined sim server/toolset. An offline editor. Stripped of unneeded shiny hindrance.

I think the lengths of my posts tend to be directly proportional to how long I've been without sleep.

Where should I look for more info on what is in the works with OpenSim?

Lucifer Baphomet

Ask Adam

__________________

Oneironaut Escher

Brevity is the. . .

Samantha Fuller

Probably not, the scripting engine for open sim is still much more fragile than SL. Also since OS is modular you could theoretically strip out some unneeded functionality but you would need a very experienced well rounded programmer or you would just end up frustrated, breaking critical stuff. On the other hand you can preload scenes & settings for a very lightly scripted low AV (5 to 10 AVs) RP environment Particularity if you paired the server with a Restrained Life client you might be able to do a simple exploration Tome Raider type game

Adam Zaius

One acronym: MRM.

Read: mrm at Adam Frisby for 'tech specs' (skip first post)
Read: Maxping - Search for examples.

It allows you to do some fairly deep level scripting, without touching the opensim core. It's ideal for games, etc.

My personal suggestion though wouldnt be to throw away the multi-user aspect quite so quickly. Virtual Worlds can do pretty well in that aspect; plus hosting tends to be a lot cheaper - grid connection is often free (eg osgrid.org).

The other aspects you might want to look at is realXtend using their new ModRex (realXtend 0.5 is OpenSim 0.6.5+ModRex); which is a region module that gives you the realXtend functionality (such as meshes, python scripting, etc -- PyScripting is similar to MRM in many ways.)

__________________

Quote:

Originally Posted by Joshua Ignore Adam, he's just a troll.

Joshua Nightshade

No, OpenSim's implementation of LSL is incomplete, not fragile. OpenSim supports more languages in total, most of which are capable of more than LSL is.

Adam Zaius

Oh, it's plenty fragile. It gives too much power to the user, so things like infinite loops can go wrong, even if we do schedule them at low priority -- they can block other LSL scripts from getting VM time.

Adam Zaius

(MRM, etc dont have that flaw - the LSL VM is rather unoptimized, the others dont need it by virtue of design.)

Joshua Nightshade

We're all in agreement that LSL is shit, which is what I believe Samantha was talking about.

Samantha Fuller

No OS's current implementation of LSL of the is much more problematic than SL's and that would most likely be the showstopper for the OP. for example the current rule of thumb for OS LSL scripters is to keep timer events to no more than 1 every 8 seconds. That kind of performance limits scripts to very simple things like fires fountains and doors

Adam Zaius

That shouldn't be right. I personally would avoid rapid-fire events; but they should still run fine.

Joshua Nightshade

My point is that One wouldn't need to code in LSL. He has other choices. Why would anyone restrict themselves to LSL, which is implemented incompletely, when they have other languages to use?

Essence Lumin

Oneironaut, you had some SL game where if you lost you got ejected from the sim. Heavily, you got pushed three sims away. I remember boohooing to you, why did you eject me, because I didn't know it was part of the game.