Uru redefined! Proposal list - add your own thoughts.

[Lontahv]

...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.

Actually, there was just the other day a discussion about having user-uploadable dynamic teeshirt images. This is well within the bounds of reality... I think even without any engine modding at all.

Plasma has the ability of writing prps from a current memory state. So, just exposing some "write" functions to the Python API would mean things like AdminKI parties could be saved. Then, adding functions to change and add data in the vertex-storage-buffer (plBufferGroup stores these buffers) would allow real-time vert-creation and editing. These few functions would make Plasma have the ability to do in-game editing and saving.

The neat thing is that a Headspin developer, in his time at Cyan, wrote a tool where you could run Plasma in 3dsMax and when you changed an objects or applied a texture in Max the same would happen in your Plasma window.

Looking at the way Plasma works I can see how this must've been done. Plasma's pipeline dynamically updates models from the vertex-storage buffers. The key is that it's a mix of static buffers and brains. It updates the model when it needs to (in the case of a vertex morph, vertex animation or any other kind of real-time vert editing).

Thinking of it now I think I understand why avatars are so laggy. Plasma must have to recalculate their vertices every frame because the verts bend when the avatar moves and the verts also have to be dynamic because of the morph-based meshes the models are stored in.

[Nalates]

I could weasel a bit as the full quote is: “As Uru is now, adding AV’s like SL’s is not possible,…” and the intent I may not have made all that clear. But, I take your point, Lontahv, and will stand corrected. I like the post.

You see adding user created t-shirts as not that difficult. Cool. I have thought it would be quite complex. So, I said ‘massive changes’.

If a user can create a t-shirt (in SL they are a UVMap – an image file users upload to the servers) and use it in game, that would seem to imply the image file will live somewhere. The client or the server seem the only candiates. If kept on the client, it will have to upload from the client and then download to all the other players that can see the AV wearing it. Unless these other players keep the shirt indefinitely, they will have to download it each time they see me or at least each time it expires out of their cache. If I email the image to a friend and they add the shirt to their closet then the image may be uploading and downloading many times. This would seem to add lag and consume bandwidth. Or add code to check for duplicates (and if the friend changes the file name...)

While storing it on the server could allow shirts to be shared and reduce uploads, this would seem to imply a flexible server side system for storing user created images. I think it implies someway for players to share things. If t-shirts catch on, there could be thousands of them and that would seem to require additional server side storage. Also when 30 people get together all with different shirts, wouldn’t that create download surge and lag?

If we can do a shirt, why not pants, socks, or shoes? In which case, the requirements for storage, closet space, and bandwidth would increase.

Then there are those pesky people that like see through stuff. So, I would assume there would be some way to control those issues and for ResEng’s to delete items.

It would also seem to imply that the player will have some way to get the shirt into their closet. If I’m tired of the shirt I might want to delete it.

On a scale of 1 to 10 with 1 being a few dozen lines of code server and client side and 10 being a total rewrite of the server and clent code… How would you rate the size of the change?

Do you think we will have fan created t-shirts in-game sometime in 2009?
Can we have pants, shoes, hair, make up… these are just additional textures applied to the AV mesh…

If the AV mesh is part of the performance issue, can that be changed to improve performance?

If the AV is going to be changed, what changes can we make besides efficiency changes?

[Lontahv]

On a scale of 1 to 10 with 1 being a few dozen lines of code server and client side and 10 being a total rewrite of the server and clent code… How would you rate the size of the change?

It all depends on how much you want to and can change. But doing the minimal needed to make it work:

I see nothing standing in the way technically of doing it right now. Changing a current piece of clothing (some prp editing here) and then using the vault to store images like it stores KIImages. Basically, like an imager on your avatar (only it wouldn't rotate though a pile of other pictures on its own ). Since the whole Plasma system already has a way of storing and managing images no modding of server (or even the client code) should be would necessary.

One-time client-side prp editing and additions using the Python API should be the only things needed.

Client or server source-code change: 0

Someone I know (or maybe a few people) at the GoW--the same who came up with this idea in the first place--is (are) currently attempting to do what I just went though.

Maybe it it could be set up so your KI can send shirt patterns into your closet "Device" what shows up. Then go in and change into your customized clothing.

I've personally wanted custom teeshirts for quite a while now in Uru. I think since the avatar is dynamically lit and people do wear custom teeshirts (sometimes with some whacky stuff on them) is RL I they'd look very in-place. In Uru even an avatar missing textures looks pretty good (at least when in a Cyan age) because Cyan does such a good job with lighting.

Going off on a similar thread (not really in response to anyone's post... just talking to myself):

Custom avatar models would sorta burst my bubble. It would mean you have fan-models (they could be anything)... that could get around. Even if you went to the city you couldn't escape them. People would over-customize... It would look like a carnival everywhere you look. Also, even for good 3d-artists it's not always a sure thing they can model a decent human.

The avatars in Uru, IMO are what keeps things feeling pretty IC regardless of their surroundings. They are something you can count on staying the same basic shape (just like with humans).

[Aloys]

People would over-customize... It would look like a carnival everywhere you look. Also, even for good 3d-artists it's not always a sure thing they can model a decent human.

Who talked about humans?..
Seriously though, with custom avatars (even humans) there's the question of the animations. Would it be possible to use the existing male/female anim prp for different models? Say I create a new male avatar that's 1 foot shorter, even if I can somehow use existing animation the avatar hand would miss all its targets by 1 foot..

[Jojon]

Judge me by my size, you do, hmm-m?
Pressing that button with the help of the force, well do I can, from down here. *reaches for a space somehwere below the button*

:P

[Kenguin]

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

Well it already apparently does this. When you look away from a crowd, your "lag" improves, which implies it's at least capable of doing frustum culling. Other simple checks could be implemented, such as distance or a simple visibility tree, without requiring too much processing time.

But what I would like to see is a dynamic system, where the client adjusts how many avatars it renders based on a target framerate. After basic culling as above, it would then start dropping avvies by distance, starting with the farthest away. This would make large crowds much more manageable, albeit with graphical oddities such as avvies fading in and out as the framerate adjusts.

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.

This... doesn't seem like a good design. Especially with the client code being opened. How does the server ensure that a hacked/modified client isn't going to disrupt the other players? Why the heck did they do it this way? I thought "server is god" has been the standard since Quake

[Aloys]

Judge me by my size, you do, hmm-m?
Pressing that button with the help of the force, well do I can, from down here. *reaches for a space somehwere below the button*

lol , that'd be a little excessive, but there are intelligent creatures other than humans in the D'niverse. Like the monkeys people from Channelwood, and there are probably others as well somewhere in the Ages..

[Kenguin]

In terms of ideas for a "redefined uru," I for one would like to do away with arbitrary "awards" for solving puzzles like bahro rings, and instead have every puzzle be rewarded by access to new areas or new story elements. (They way they handled K'veer in MOUL, for example, was excellent -- more stuff like that.) If solving puzzles gives you access to "public" areas (like k'veer) even better -- this integrates the "puzzling" gameplay with the "roleplaying in cool areas" gameplay. If you want to roleplay in more cool areas, you have to solve more puzzles. Also, if those new areas in turn had links to more puzzles to unlock more areas, the game could be organized as a progression of discovery instead of a flat "here's a list of ages to visit."

[Nadnerb]

I guess I can thank Lon for explaining my idea, and also why I'm not sure it's a good one.

Nalates: You (and plenty of other people) seem to consistently and grossly underestimate what plasma can do. (as well as cherry-picking features that offer considerable high ground to SL) I'm not sure if you somehow forgot that KI images are taken locally and then uploaded to the server for storage in the vault, or if you noticed that those images were then downloaded from the vault by all other clients and displayed on in-game objects like imagers.

The only technical change required for custom clothing would be the addition of a clothing item with a plDynamicTextMap on which the texture could be displayed, and some python that would grab the texture images from the player's vault. (uploading said images via the client could require some modifications, but they can already be uploaded using the vault manager. (recall the custom KI images uploaded by the admins during MOUL))

In answer to: "Do you think we will have fan created t-shirts in-game sometime in 2009?", I think it's entirely possible to have such things available within the year, but I don't think it's likely. Priorities may be elsewhere.

This... doesn't seem like a good design. Especially with the client code being opened. How does the server ensure that a hacked/modified client isn't going to disrupt the other players? Why the heck did they do it this way? I thought "server is god" has been the standard since Quake

Well... what you do is... you come up with a story about how if you hack our game, our backer will pull funding and it will be gone for everyone. This serves to turn your entire community into a vigilante organization that ostracizes anyone who might be considered a "hacker" because of course they wouldn't do that if they loved the game like a true fan.

Back in the days of Until Uru, it was possible, with minimal effort (python hacking) to alter the behavior of your client so that you could "disrupt" things pretty effectively. Things like magically teleport objects such as the arch of kerath to the middle of the courtyard and make it the size of a person, link to anyone's age, or broadcast messages to the entire population. It also became evident that a single badly (through malice or ineptitude) hacked client could bring down an entire shard. It's likely that this was still possible during MOUL, but they did take a few additional measures to increase security by hiding the python better.. which wasn't terribly effective. At least not as effective as the funding story.

[Kenguin]

Back in the days of Until Uru, it was possible, with minimal effort (python hacking) to alter the behavior of your client so that you could "disrupt" things pretty effectively. Things like magically teleport objects such as the arch of kerath to the middle of the courtyard and make it the size of a person, link to anyone's age, or broadcast messages to the entire population. It also became evident that a single badly (through malice or ineptitude) hacked client could bring down an entire shard. It's likely that this was still possible during MOUL, but they did take a few additional measures to increase security by hiding the python better.. which wasn't terribly effective. At least not as effective as the funding story.

Well it seems to me that a big priority in open source uru is going to be to increase server security, perhaps by developing some kind of message filter that can detect when a client is trying to do something naughty and blocking those messages or even dropping the client.

I don't know how difficult this would be in the context of Plasma architecture, but considering that Cyan didn't do it, it must be a bit of a challenge.