Guild of Writers

It may be possible to use their technology in Cyan's server side software.

Totally different model. It's just not applicable to URU. I did have a good laugh over how that avatar moves however.

The may be tricks to reducing lag. For instance an avatar way on the other side of the courtyard doesn't need as many polys as a person right next to you. I'm by no means saying that this would be easy to implement, but I'm sure there are others ideas out there.

greendragoon wrote:The may be tricks to reducing lag. For instance an avatar way on the other side of the courtyard doesn't need as many polys as a person right next to you. I'm by no means saying that this would be easy to implement, but I'm sure there are others ideas out there.

Uru already does this. If you notice when you look at the avatar files, there a multiple copies of the mesh, each with less polys. Uru enables and disables each level of detail based on how far the avatar is from the camera. (as a random side note, in MOUL there was an interesting bug where it would sometimes show all of the LOD levels at the same time. Fail. )

Second, in response to Nalates various postings, which are amazingly evenly scattered over half a dozen forums: SL's "technology" is in no way applicable to Uru. Plasma is vastly different from Linden Lab's ... software. I don't know what you've seen in SL that's convinced you that it is in any way compatible with Plasma, or that it can even do everything that Plasma can, but I've seen plenty that says it isn't, and can't. I could go on for pages describing the technical differences that make this impossible, but I won't, because you probably won't read or understand it, and would come back with various arguments on what 'could be done' (given infinite time and monkeys) to resolve them.

Instead, I'll focus on some more ideological differences. You made a post elsewhere about some of your reasoning behind wanting to see all "virtual worlds" integrated under one roof. You mentioned wanting to be able to bring your avatar and "stuff" into Uru. Here's a question for you: what if we don't WANT that coming into Uru? Assuming for a moment that any of this is even remotely possible, do you realize how terrible that would be for immersion? Do we allow people to walk into Uru with their enormous dragon-avatars? Even without allowing attachments, SL avatars look vastly different from all the possible Uru avatars. Uru has a consistent, detailed, somewhat realistic style, and it's very easy to break. Granted, fan ages are going to break it, but those can't just walk into the city. (unless you have one of those funky paging issues)

Nadnerb wrote:

greendragoon wrote:The may be tricks to reducing lag. For instance an avatar way on the other side of the courtyard doesn't need as many polys as a person right next to you. I'm by no means saying that this would be easy to implement, but I'm sure there are others ideas out there.

Uru already does this. If you notice when you look at the avatar files, there a multiple copies of the mesh, each with less polys. Uru enables and disables each level of detail based on how far the avatar is from the camera. (as a random side note, in MOUL there was an interesting bug where it would sometimes show all of the LOD levels at the same time. Fail. )

That being said; there is still room for improvement in URU. As an example: suppose I'm alone by the library and there are 100 explorers down in the Takotah courtyard; the only effects they should have on my client is that their names appear in the age list and their chat on my screen. I can't see them, so my client should care nothing about what they are doing. This isn't currently how it is. If I don't face the courtyard; my client is fine. If I turn to face the direction of those 100 avvies that I can't see; it lags. It shouldn't since they are totally outside my region of influence. I suspect that this is directly related to why removing the cones reduced lag. Whether this is due to location updates, physics calculations or something else is something that needs to be explored.

Well the engine actually has to make calculations in order to determine if it should show you those avatars or not.

Right. What I'm suggesting is that there needs to be a method employed which lets the client know that those avatars are irrelevant. Most likely, it would have to be something server-side like other MMOs.

I think the current problem stems from those avatars being within a defined visible region and because of that physics and/or drawing is actually done. Possibly the use of occlusion could reduce this client-side. I don't think that just checking the location of those 100 avatars should cause the kind of lag that it does.

teedyo wrote:Right. What I'm suggesting is that there needs to be a method employed which lets the client know that those avatars are irrelevant. Most likely, it would have to be something server-side like other MMOs.

This goes against one of the fundamental design principles of plasma. That being that the client is god. The game servers do little more than pass messages between clients. One client cannot 'forget' about an avatar because all the simulation is done on the client, and it will always receive physical updates from the client controlling that avatar, whether it wants to or not. The most that can be done is for the client to stop rendering an avatar when it is outside the visible range, but it will still have to remain entirely loaded in memory so that that calculation can be performed. (speaking of which, occluders should work for this, but it still takes time to determine which objects should be occluded when you are facing a large number of them)

Lots of ideas, problems needing solutions, and solutions needing problems... I really like most of the ideas, and encourage rereading this entire thread. Sarpedon2's ideas are right on, imo. It seems to be a great way to bring newbies into the MYST picture. Perhaps the Village can include 'wall pictographs' of various ages and situations with short infomercials about them to 'tell the URU Story' in an interesting and shortened fashion. So I look and think 'OH WOW' and want to get involved.

My offerings to the alter of creativity are:

1. URU- Ages of MYST as the central idea/theme.
2. Redesign the KI as suggested in many locations. What it needs to do, finally, should be determined by what it needs to
do to make the user more able to enjoy the journey.
3. Enable a 'live voice' chat feature so that a small group (<10) can converse vocally within a small space (1 room). Also
enable individuals in different ages to hold a conference call (vocally), ala Messenger and Skype.
4. Make every puzzle mechanism work readily, or remove it. I still have two MYST ages I can't complete because I can't
get the mechanism to do what I KNOW it needs to do.
5. Keep Ae'gura as the central city where everyone can meet and greet. Open more of the rooms and expand the Library
and the Museum immediately. Maybe even a 'quick connect' from the first Village to/from one part of the Museum,
and maybe the Library as well.
6. Don't set any limits at this time, other than to maintain legality and good relations with CYAN. There's lots of stories
that would be true to URU without changing anythink involving previous characters, including Yeesha and the Bahro.
7. How the 'linking books' work doesn't need to be explained for users. However, I once postulated that the D'ni may have
arrived on Earth1 (our Earth) 10,000 years or so ago from Earth2 (another Earth in a nearby multidimensional
mega-universe) caused by a momentary glitch in dimensional reality. Some D'ni and Bahro, plus a small part of their
total technology, were stranded on Earth1 when the dimensions separated again. Linking books are then a result of their
advanced 'wormhole technology'.
8. I don't know how to resolve the development engine issues, other than to suggest putting together pro and con ideas,
and finally asking someone with CYAN to make a final suggestion/recommendation. This would enable developers to
move forward with advance knowledge that their efforts would be useful in the central effort. Certainly
some 'import/export' software would be a tremendous help so that everyone could use whatever they are comfortable
with.

I just enjoy reading all the ideas being presented, and will need to reread this thread many times, I'm sure.

Nadnerb wrote:[…snip…] SL's "technology" is in no way applicable to Uru. Plasma is vastly different from Linden Lab's ... software. I don't know what you've seen in SL that's convinced you that it is in any way compatible with Plasma, or that it can even do everything that Plasma can, but I've seen plenty that says it isn't, and can't. I could go on for pages describing the technical differences that make this impossible, but I won't, because you probably won't read or understand it, and would come back with various arguments on what 'could be done' (given infinite time and monkeys) to resolve them.

Instead, I'll focus on some more ideological differences. You made a post elsewhere about some of your reasoning behind wanting to see all "virtual worlds" integrated under one roof. You mentioned wanting to be able to bring your avatar and "stuff" into Uru. Here's a question for you: what if we don't WANT that coming into Uru? Assuming for a moment that any of this is even remotely possible, do you realize how terrible that would be for immersion? Do we allow people to walk into Uru with their enormous dragon-avatars? Even without allowing attachments, SL avatars look vastly different from all the possible Uru avatars. Uru has a consistent, detailed, somewhat realistic style, and it's very easy to break. Granted, fan ages are going to break it, but those can't just walk into the city. (unless you have one of those funky paging issues)

Nadnerb, I think you have missed where I am coming from and going to…
The technology I am referring to is up a level or two of abstraction from where I think you are thinking. Plasma and SL code certainly are not going to be compatible. But SL and Open Simulator are advancing their technology and handling larger and larger numbers of players with less lag (the server side meaning) and better client side performance with virtual worlds (VW’s) far more complex than Uru. Using low and high poly AV’s and switching between the two AV models based on distance is something both systems do. That is one way the technology is useful in both and the level of abstraction I am talking about.

I vote to see less lag/better performance using whatever tech is available.

In another thread it is pretty well nailed down that no one outside Cyan seems to know that much about what Plasma can or can’t do or how well. At this point only Cyan has the source code (version 2 – which Chogon says is nothing like v1 ala UU). Whatever either’s limits are, they have to accomplish the same things, render a virtual world and keep track of avies and moving things and get that info to all the players. So, I consider them very similar and the technologies related. Ideas are interchangeable even if coded in different languages.

I vote to use the best features and technology from other games to improve Uru.

Also, whether it is a good thing to move AV’s between VW’s or not, it is the direction that tech is going. I would like to move AV’s between worlds. Many players out there want to also. Governments and corporation have been moving into VW’s and extending the tech. There are good marketing reasons for their doing this and having AV’s travel between VW’s. Whether that is a good thing for Uru is debatable. It is not so much that I want it for Uru as it is this is what is coming and I would like for Uru to be competitive. Whatever one decides, it is a feature of new games that is being developed and that Uru may well have to compete with. There are commercial obstacles to overcome so the whole idea may die, but for now there are several efforts, some open source, trying to make it happen.

I vote to have Uru be as attractive to new players as we can have it and still have it be Uru. Better AV's does not break the IC quality. It is modern day people wondering around in Cavern.

As Uru is now, adding AV’s like SL’s is not possible, adding clothes as SL does is not possible, allowing fans to build things inside Uru as SL does is not possible. Adding any of those things is going to require massive changes to the server and client side code. I don’t see that happening any time soon, if ever.

I also have a hard time imagining Plasma was built to handle a number of features seen in other VW’s. Until the code is out and we know, I’ll continue to think other technologies are likely ahead of Plasma. So, as to proposals for a new Uru, a Plasma upgrade or replacement might be nice, or whatever it takes to improve performance.

As to the immersive quality being broken and Uru’s AV’s being so somewhat realistic… “Uru has a consistent, detailed, somewhat realistic style, and it's very easy to break.” I just disagree. But, no where did I say I want to bring dragons into Uru. I do think that would break the immersion. I also think it would be fun. There are lots of things I think would be fun and make Uru more fun too. But, that does not make them something the entire community wants. There are lots things I would like to see added that will be technically difficult and are, I suspect, unlikely with current Cyan technology i.e, better AV’s, clothes and hair...

I vote for the IC experience to be more current day as the storyline describes. A you-are-you character today would provide a vast array of hair styles... and J'Kla is green...

Nalates wrote:In another thread it is pretty well nailed down that no one outside Cyan seems to know that much about what Plasma can or can’t do or how well. At this point only Cyan has the source code (version 2 – which Chogon says is nothing like v1 ala UU). Whatever either’s limits are, they have to accomplish the same things, render a virtual world and keep track of avies and moving things and get that info to all the players. So, I consider them very similar and the technologies related. Ideas are interchangeable even if coded in different languages.

Firstly, there is a wiki page about Plasma versions, and another page about Plasma on MystLore. Plasma 1 was realMyst. Prologue, UU, PotS, MV, MOUL, and Crowthistle are all Plasma 2.

There are some people in this community (Zrax, Nadnerb, Hoikas, myself and others) who know almost as much about Plasma as the Cyan developers. I have spoken to previous Cyan developers, and I knew more about some of the internal workings of Plasma than they did. Over the past 5 years, we have managed to get a pretty good understanding of Plasma, and we've reverse engineered and refined our code so that it probably looks identical (line-for-line) to Cyan's in some places. libPlasma is an excellent example of our understanding of the engine and its internal class structure.

It is an insult to the work of the community developers to say that after 5 years and all of the various tools, that "nobody outside of Cyan knows anything about Plasma".