I heard of the animated linking panels, however the particle linking effect is something I remember now but not sure how it would be done in Uru.
If I remember correctly, on linkout little orange 'what only can be described as 'kemo-flies'' disintegrate from the persons body and go into the book. It sounds a bit complicated but I assume the same principle as the kemo-flies could be used in conjunction with the linking animation we have now to make this work. And to make the orange particles fly towards the book and disappear on contact?
Visual effects: linking avatars and Linking Books
Moderator: OpenUru.org Moderators
-
- Member
- Posts: 4
- Joined: Sat Dec 13, 2008 11:41 am
Re: Visual effects: linking avatars and Linking Books
Technically the rendering will be done by the client, but the server will require additional computation to work out what objects should be in the panel, their orientation, movement directions, etc, (although you only need an approximation) and there will indeed be some additional bandwidth required to relay that data to the client. My gut feeling is that the additional computation will be a larger contributor to "lag" than the data transfer. The lag will then mainly affect other players, I suppose, as when you're looking at a linking panel you don't really have a frame of reference to perceive the lag. I'd maybe expect some problems when you have a stack of people all trying to link into the same place at the same time and the link point gets congested.realXCV wrote:Not really. But the renderring will probably have to be done by the server then sent to the player. That will increase the bandwidth usage (with more lag).zib_redlektab wrote:I think that realtime panels are going to be very difficult.
Mac_Fife
OpenUru.org wiki wrangler
OpenUru.org wiki wrangler
Re: Visual effects: linking avatars and Linking Books
I believe we talked about this during MOUL, it turned out to be an impossible task as all the Ages withing a given area like Ae'Gura have to load up every Age within them to do flybys in realtimerealXCV wrote:Not really. But the renderring will probably have to be done by the server then sent to the player. That will increase the bandwidth usage (with more lag).zib_redlektab wrote:I think that realtime panels are going to be very difficult.
If we did, it would require significant changes, on second thought still no, if we want to have all books outside of Relto to be publically available, cause then you can't even cheat te process, so its still a no go
Re: Visual effects: linking avatars and Linking Books
The final rendering will be done by the client. But I think the rendering for the content of the linking panel (have you ever heard of a render target ?) will have to be done by the server. Maybe you computer is powerful enough to load every single age (and instance) with all the players in them at the same time but mine is not. On the other hand, I think that even the server won't be powerful enough to handle that.Mac_Fife wrote:Technically the rendering will be done by the client, but the server will require additional computation to work out what objects should be in the panel, their orientation, movement directions, etc, (although you only need an approximation) and there will indeed be some additional bandwidth required to relay that data to the client. My gut feeling is that the additional computation will be a larger contributor to "lag" than the data transfer. The lag will then mainly affect other players, I suppose, as when you're looking at a linking panel you don't really have a frame of reference to perceive the lag. I'd maybe expect some problems when you have a stack of people all trying to link into the same place at the same time and the link point gets congested.realXCV wrote:Not really. But the renderring will probably have to be done by the server then sent to the player. That will increase the bandwidth usage (with more lag).zib_redlektab wrote:I think that realtime panels are going to be very difficult.
A looping of what happen at the linking point would be a better idea. Something like the Myst book in eoa.
Re: Visual effects: linking avatars and Linking Books
It would be possible, and not that difficult.. But I'm not sure it would look really good. Think of it as the effect that happens in the Bahro caves when a pillar disolves into white particles which then fall to the floor. Except that here the particles would fall toward the book.TomahnaGuy wrote:I heard of the animated linking panels, however the particle linking effect is something I remember now but not sure how it would be done in Uru.
If I remember correctly, on linkout little orange 'what only can be described as 'kemo-flies'' disintegrate from the persons body and go into the book. It sounds a bit complicated but I assume the same principle as the kemo-flies could be used in conjunction with the linking animation we have now to make this work. And to make the orange particles fly towards the book and disappear on contact?
Personally I like the simple effect we have in Myst/Riven. They looks a bit like the Yeesha hologram effect in the Cleft, so it'd be very easy to do.
As for the fly-bys linking panels: as Dot mentions it's already possible; all that'd need to be done now would be to actually create the videos (like Andy already did with Teledahn).
-
- Member
- Posts: 15
- Joined: Sat Dec 13, 2008 2:02 pm
Re: Visual effects: linking avatars and Linking Books
Two more thoughts on the realtime panels.
First of all, what would happen with books that are instanced? For example, the Teledahn book in front of the Library, which I believe links to the personal or hood instance. If two people from two different hoods were to look at the book, it would have to show them different things meaning even more computations for the server.
Secondly, how interesting would these panels actually be? I mean yes, a flyby instead of the static image would be cool, but is it really necessary to have the panel be rendered in real time? In a flyby of an age, it would be really difficult to make out people anyway, and if it's a still image being prerendered then it really won't be interesting anyway - how many people hang out around the link-in point of an age? Also, the vast majority of age instances are only visitable by one person, the owner of said age (the exceptions being ages others have been invited to, and even then-how often do invitees visit a shared age?). This means that for the vast majority of the panels, the image would always be the same, and rendering them would be a waste of time and resources.
Just some things to think about.
First of all, what would happen with books that are instanced? For example, the Teledahn book in front of the Library, which I believe links to the personal or hood instance. If two people from two different hoods were to look at the book, it would have to show them different things meaning even more computations for the server.
Secondly, how interesting would these panels actually be? I mean yes, a flyby instead of the static image would be cool, but is it really necessary to have the panel be rendered in real time? In a flyby of an age, it would be really difficult to make out people anyway, and if it's a still image being prerendered then it really won't be interesting anyway - how many people hang out around the link-in point of an age? Also, the vast majority of age instances are only visitable by one person, the owner of said age (the exceptions being ages others have been invited to, and even then-how often do invitees visit a shared age?). This means that for the vast majority of the panels, the image would always be the same, and rendering them would be a waste of time and resources.
Just some things to think about.
Re: Visual effects: linking avatars and Linking Books
That's exactly why I added something about animated static linking panel which is close enough to what linking panel should be.zib_redlektab wrote:Two more thoughts on the realtime panels.
First of all, what would happen with books that are instanced? For example, the Teledahn book in front of the Library, which I believe links to the personal or hood instance. If two people from two different hoods were to look at the book, it would have to show them different things meaning even more computations for the server.
Secondly, how interesting would these panels actually be? I mean yes, a flyby instead of the static image would be cool, but is it really necessary to have the panel be rendered in real time? In a flyby of an age, it would be really difficult to make out people anyway, and if it's a still image being prerendered then it really won't be interesting anyway - how many people hang out around the link-in point of an age? Also, the vast majority of age instances are only visitable by one person, the owner of said age (the exceptions being ages others have been invited to, and even then-how often do invitees visit a shared age?). This means that for the vast majority of the panels, the image would always be the same, and rendering them would be a waste of time and resources.
Just some things to think about.
Re: Visual effects: linking avatars and Linking Books
Zib is right, a real-time flyby wouldn't add much benefit.
It is already possible in a way, and working right now in TunnelDemo. When you link in, you see a flyby ending at a shot of your avatar at link-in point. This is done by grabbing you camera AFTER you link in. Doing that BEFORE loading the Age isn't practical as was said previously.
BTW, animated flybys have been implemented for some time in the off-line userKI, MystV flybys are available for Drizzled MystV Ages.
It is already possible in a way, and working right now in TunnelDemo. When you link in, you see a flyby ending at a shot of your avatar at link-in point. This is done by grabbing you camera AFTER you link in. Doing that BEFORE loading the Age isn't practical as was said previously.
BTW, animated flybys have been implemented for some time in the off-line userKI, MystV flybys are available for Drizzled MystV Ages.
-
- Member
- Posts: 47
- Joined: Sat Dec 13, 2008 10:45 pm
Re: Visual effects: linking avatars and Linking Books
Regarding linking panels, I think a canon linking panel is NOT a realtime view of the Age. It is a view of the Age as it was when the linking book was activated. So a simple prerendered movie is actually more accurate.
Regarding the linking particle effect, it needs to be cool, but more important is that it is feasible with what we've already seen Plasma do (i.e. the Bahro pillars' linking effect).
Here's how I envision it happening. I'm willing to be corrected on all technical implementation details.
EDIT: Oh, and I should mention the timing. Steps 2-5 above happen during the initial "mmmmmmwwoooooowwwww" groan of the linking sound. The particles' moving and disappearing happens with the "Ch ch ch ch" sound at the end of the linking sound.
Regarding the linking particle effect, it needs to be cool, but more important is that it is feasible with what we've already seen Plasma do (i.e. the Bahro pillars' linking effect).
Here's how I envision it happening. I'm willing to be corrected on all technical implementation details.
- Every time the avatar is customzed, for each linking pose the game supports, the game computes a new model of points just "outside" the avatar's polygons along a normal from each polygon.
- Avatar touches linking panel
- Game loads aforementioned "normal points"
- Game calculates an expanding sphere from the linking panel. Whenever a "normal point" is inside the expanding sphere, a big bright shiny gold particle lights up at that point. Each particle should be big and bright enough so that when the avatar is completely inside the sphere, the particles completely obscure the avatar.
- Once the avatar is completely obscured by particles, the actual Avatar polygons immediately disappear.
- The remaining particles follow a path toward the hand touching the linking panel, and disappear as soon as they reach the fingertips.
EDIT: Oh, and I should mention the timing. Steps 2-5 above happen during the initial "mmmmmmwwoooooowwwww" groan of the linking sound. The particles' moving and disappearing happens with the "Ch ch ch ch" sound at the end of the linking sound.
Re: Visual effects: linking avatars and Linking Books
You step 1 implies that if there's 300 linking poses and 30 avatar in an age, you have to compute 9000 models of points.