R-Pi Massively Multiplayer Online Role-Playing Games (MMORPG)

[ikesmasher]

Hi Ikesmasher,

Have you tried the Unity engine? It does the same trick with a single plane, and you can get some really strange effects if it's not tuned properly. You walk up to a tree and it kind of stretches itself, as the plane turns into a 3D object. I imagine they try to avoid jumps by switching from the plane to a squashed 3D model, then gradually filling it out. Fine if it's far enough away that you can't see it, but very odd if you're standing underneath it at the time.

I actually use unity quite often, was actually designing a game or two in it(no idea how far they will go.) i recall exactly what you are saying, although i believe there is a way to prevent it happening when you are nearby. I havent messed around with the tree system a whole lot or in a while.

[doctorwho]

You asked for comments on a user interface for setting up worlds and I was thinking that although you should be able to load projects with large python files and such to make worlds it should also be very easy to make a simple world. Like "world(forest, fantasy)" which when loaded creates a simple sword and sorcery forest world. The great thing is that once you make the basic world building tools you can then easily build up to code like the one I showed above ( I would would like to help do this when the basic world building tools are made).

[specsdude]

I like that idea. So basically you'd give the world generator a seed, and it would make your custom world.
But there would have to be a module or something that mixed worlds...

And would each world come with quests to complement it? So for instance, Pete's dark fantasy world might contain an evil witch who has an army of zombies, and there would be a quest chain which leads to her defeat?

What are other peoples' thoughts on this?

[ikesmasher]

that sounds good. However you would want to make sure things didnt get TOO generic.

[specsdude]

What I personally would like to see in the game is the ability to make cutscenes / cinematics. This could be for a quest, or perhaps when you enter a cave, or travel between planets. You could either use in-game cameras or load your own video / cartoon etc.

I think it would add that extra level of awesomeness which separates mediocre games from good ones. It would also (hopefully) encourages users to make their world more unique and crafted by them, as opposed to a generic world which is similar to the other two hundred with the same configuration.

It wouldn't be that hard, right? . . .

[doctorwho]

The way I currently see it once we have several pages of code that can generate a world (the seed) all you would need to do is modulate it. so you have a world() function which has variables such as atmosphere density, time period, hardness...(all of which has default values) If you wanted to go more in detail you could use the world function to generate the code to your world and then edit it from their. Of course their might be a atmosphere() function which generates a chunk of python code that you can insert into your world that is much more detailed on the atmosphere then the generic world module. People could build their own world modules such as a dungeon() function which lets you go into detail on the number of dungeons, how big they are, what quests are in them...

[PeteX]

Hi All,

I feel you will end up with 'empty worlds' without LOD ...

Perhaps in the end we'll find we can't live without it, for this reason. At the moment, though, I feel the most important thing is to get a product that people can try out. Apart from anything else, more people are likely to volunteer if they can see that we have a real if primitive game, rather than pure vapourware.

If you want to implement LOD, of course, that would be perfect. I'm very happy to be part of a discussion about the way it should work, and to merge changes once they're done.

I have finished a new version of the tree, using the same technique as the Turbosquid tree you linked to.

That does look better, but I still don't feel it's reached the level of the Turbosquid tree. The Turbosquid tree is an excellent model, but I can't help thinking that we need to be aiming for that standard. Whether we like it or not, that's what people expect from games these days.

BTW, while it's hard to get a feel for a model from a single image, I did like the wolf, and the wolf is probably more important than the trees anyway. We don't have any monster models, but we do have some tree models, it's just that they use too many polygons and so limit the size of scene that can be built.

IMO, I'd kinda prefer some sort of user interface.

As you say, it's meant to encourage people to learn programming, so configuring the system through a GUI would rather defeat the point. I also think it would lack a certain amount of flexibility. You would be able to set up simple quests like that, but eventually you would exhaust the possibilities offered by the GUI. The Blender Game Engine, incidentally, is a GUI for writing games, and that's exactly what happens: you eventually start writing bits of Python because the GUI is not flexible enough for complex interactions.

I'll have to take another look at Diaspora sometime. I looked at it ages ago, when it was just starting and wasn't really usable, but it sounds as though it's made a lot of progress since then.

...

Randomly generated worlds sound interesting, but they're probably hard to do well. I don't know how you could come up with quest lines that tell stories, if they are generated based on a random seed provided by the admin. The best you could do, I imagine, is come up with quests like, 'Slay <so many> <kind of monster> and retrieve the <magic item>.'

Has anyone played Daggerfall? From what I've read, a lot of the terrain and simple quests were randomly generated. Did it work well?

I don't think cutscenes would be particularly hard from a programming point of view. Essentially you are just asking for the ability to play a piece of video when a certain trigger occurs. They might be hard to make, though. Were you thinking you would animate them with Blender?

When using cutscenes, you'd also have to be careful that the player wasn't visible to others who weren't seeing the same cutscene. It would be confusing if someone could start a fight with the player while he was away watching a few minutes of video...

It would also (hopefully) encourages users to make their world more unique and crafted by them, as opposed to a generic world which is similar to the other two hundred with the same configuration.

This is an important point, I think. When the text MUD scene went into decline, I don't think it was just because of the appearance of graphical rivals. It was also because it became possible to download pre-written MUDs, so suddenly everyone was running them instead of playing them.

This is perhaps an argument for giving away the kit—the code base plus example models and artwork—but not giving away a large pre-built world. We want to encourage people to develop lots of different worlds, rather than everyone running the same world and then finding that they have no players.

[PeteX]

I've just pushed some code to the repository. You will now see some simple terrain when you start the client. This terrain was generated from a mesh designed in Blender. If you turn right round using the arrow keys, you will see that the cathedral and the trees are still there, too.

(Yes I know the terrain itself needs work; there are some bits where the two textures don't blend together nicely, and the edges of the terrain are a bit wonky.)

[specsdude]

more people are likely to volunteer if they can see that we have a real if primitive game, rather than pure vapourware

Well, I for one am raring to see the game in action...

About the wolf, I think the reason you like it more than the tree is because I based it off real photos of wolves. The tree was more ad lib, so it's a bit shapeless. Right now I'm fleshing out the animation for the run cycle. It's rather difficult. I've had no experience in organic animation, so it's difficult to get it so it looks natural. As a consequence, I've become very good friends with the Dopesheet window!

Hopefully I'll have finished the animation by tonight. I don't what other animations we'll need, any help? Maybe a scratching animation? (no biting, he has no mouth!!)

But there's my update anyway, in case people were wondering if I've actually been doing anything.

[specsdude]

Haha! Finally finished the run cycle. Google found me a nice video of a running dog, which made things a bit easier. The wolf has been retextured a little, and can now run!

Out of interest, this is what the graph of the run animation looks like:



Only 18 frames!

And the wolf:



Please comment and/or give advice!

[ikesmasher]

It looks really good, except i think the tail moves a little too much, and i noticed the center of the wolf's body does not really move at all. But apart from that, it looks really natural.

[specsdude]

the tail moves a little too much, and i noticed the center of the wolf's body does not really move

Yeah, the tail does look a bit floppy. About the centre, that's where the bones all branch out from. The reason I didn't animate it is because I only used the Rotation keyset, and to animate the centre I'd have to actually move the two middle bones, which would need a LocRot keyset. So it would require a bit more work. I most likely will fix it up some time in the foreseeable future, but right now I'd rather concentrate on producing more models for the game.

One thing I was wondering was, what will be the relationship between player's worlds. I remember the discussions about each world having a real location in the universe- is that still on? And will players be able to travel between world easily in game, or will they have to restart the client with a different parameter? Also, will players be able to make whatever they want on their own worlds, and if so, can they take objects into other peoples' worlds?

For instance, I might have a world where I've put a chest full of Legendary High Elven Battleswords. I equip two of these and I'm able to slay every single creature that I've put into my world. Now I want to enter someone else's world. Do I keep the swords?

What I think might be good is some sort of universal point system. Every item is worth a certain number of points. So maybe an Ice Dragon of Krakmar would be worth 23,000 points to the player who slayed it, and a Loaf of Friendlyfruit-bread would be 20. When travelling between worlds, anything you haven't 'earned' has to be paid for in points. So that would stop me taking my epic swords into someone else's world and cheating.

Anyone see any major flaws in this?

[PeteX]

Hi All,

Good news—I've now finished all the downloadable scene stuff. You can now design a scene in Blender, export it, upload it to the Internet, and play it. If you wanted, you could run a server which would allow other people to play your scene, but at the moment there isn't much point because interaction between players is so limited.

Well, I for one am raring to see the game in action...

For people who don't have access to Linux, I've uploaded a video so you can see where we're up to. I'm also now able to start on the website and the Windows version, so hopefully there will be something to see soon on that front.

BTW, does anyone know a good open source program for generating Windows installers? NSIS seems to be dead (last release 2009). WiX looks viable but rather low level, so difficult and time-consuming to use. What would people think if I were to use a tool which is free but not open source, or a fully commercial tool which costs money? Remember, this would only affect the generation of the installer packages for Windows. It would have no impact on anyone's ability to play or create worlds.

I'm tempted to use a commercial tool here, to avoid getting sidetracked. It would be very easy to waste a lot of time messing around with the Windows installer, and that would delay the game.

About the wolf, I think the reason you like it more than the tree is because I based it off real photos of wolves. The tree was more ad lib, so it's a bit shapeless. Right now I'm fleshing out the animation for the run cycle. It's rather difficult. I've had no experience in organic animation, so it's difficult to get it so it looks natural.

Well I certainly find ad-libbing difficult. I find modelling from photographs hard, but things which are entirely made up are much harder. (I was just thinking, personally I think I find this project fun because the difficulty hits roughly the right spot... It's not routine programming, but nor is it the kind of blue-sky research where it would be easy to get completely stuck.)

We'll need some sort of attack animation, I suppose. A scratch should be fine I think. Even in Skyrim the wolves just kind of jump towards you with their mouths open!

I love the running wolf animation. I did think of one thing to check, though. He's running on the spot; if you add some forward motion, do his paws stay still at the times when they are supposed to be in contact with the ground? I struggled for ages with that, when I was trying to do the walk cycle for the player character.

One thing I was wondering was, what will be the relationship between player's worlds. I remember the discussions about each world having a real location in the universe- is that still on?

I'd like to see links between the worlds, but obviously there is a huge amount of work to do before we can even think about it. If I ended up implementing it (rather than Jim or some other volunteer) I'd probably go for magical portals rather than spaceships, but that's a detail.

For instance, I might have a world where I've put a chest full of Legendary High Elven Battleswords. I equip two of these and I'm able to slay every single creature that I've put into my world. Now I want to enter someone else's world. Do I keep the swords?

No.

I think the only way to do this—looking even further ahead now—is to allow trust relationships between worlds. Perhaps we both build a world, and they are roughly comparable in terms of weapons and strength of monsters. We might arrange for our worlds to trust each other, so items, money and character stats are portable between them.

What I think might be good is some sort of universal point system. Every item is worth a certain number of points. So maybe an Ice Dragon of Krakmar would be worth 23,000 points to the player who slayed it, and a Loaf of Friendlyfruit-bread would be 20. When travelling between worlds, anything you haven't 'earned' has to be paid for in points. So that would stop me taking my epic swords into someone else's world and cheating.

Interesting. If I'm understanding you right, you're saying that worlds would have standard building blocks. I can put an ice dragon in my world, and then some central authority notes the fact that it has been slain by player X. As a result it credits him with 23,000 points.

The first thing, I suppose, is that it would standardise the worlds much more. You can take an ice dragon off the shelf and drop it into your world, but you can't implement something that is totally new, like awarding points for solving a puzzle. The other issue, I think, is that a central authority would have to notice the killing of the ice dragon and award the points. I'm not sure how you would make sure that the central authority couldn't be scammed by games that report kills which didn't happen.

[specsdude]

About the installer, there is this:

Inno Setup

Apparently it is open source as well. However I have no experience with this stuff.

I really like the video! It's proof that the game works. I would personally prefer a 3rd person view, but that can be figured out easily, right?
If I installed a Linux distro, would the game work for me once I'd downloaded all the dependencies? Or is there some secret trick that only Pete knows...

About the world-to-world question, I think perhaps you're right. After all, each world is meant to be unique. Players aren't all obliged to run a server, most may just play on a cool/popular world. And all regulating would be done by the server...?

So if I give that wolf an attack, will it be incorporated into the game? Or do you not have any code for enemies yet? I think what I'll start after the wolf is the fabled Ice Dragon of Krakmar. That will be cool to model, if not too difficult.

[craziwolf]

seriously I read this post before any other post and well I haven't even made it the hole way through and my brain has turned to mush! but the long of the short of it is if your accepting help I would like to offer my services! Im a web developer by trade and have recently joined a starting out indy game company as a texture artist / jack of all trades.

In any event if you want all hands on deck im sure I could help somewhere

[specsdude]

Yaayyy!!

A new person!

If you want to help, that is totally awesome. You said you're a web developer, so you could help make the website!

Also, my brain has started turning to mush after seeing all these 3D renders (not part of this project of course!!). I'm going to give it a go, never tried something like that before though.

But anyway, welcome. I'm Dave.

Edit: And awesome signature you have BTW!

[craziwolf]

Thanks! favorite all time quote

Id be happy to do a webpage for the project! I can host it as well if need be

[PeteX]

Thanks for the link to Inno Setup. I've had a quick look and I think it may well do what we want. It doesn't generate the .msi type of installer, but that's only really important if you're writing software that is going to be rolled out automatically to a group of corporate users. (I think that's right, anyway. It's a long time since I've done serious coding for Windows. )

Glad you like the video. As you can see, the main elements of the world are in place: objects, terrain and skybox. I did it as a first person view without really thinking, and I rather agree with you that third person would be better—or perhaps it should be an option that can be set for each world. It's no big deal, probably an hour's work to implement. (There is a more subtle problem with third person views, though, which is that the camera can get tangled up with scenery. That affects all games, and I don't think there is a completely general fix.)

There is a secret trick that only I know, of course. No, seriously, if you installed Ubuntu you should be able to run it. All my changes have been pushed so essentially you'd have all the things I used to make the video.

There is no code for enemies yet, but of course I'll incorporate the wolf when that's done. I'm not actually sure what direction to take next, whether to do player interactions like chat, or whether to go for combat. I feel it makes sense to do the hard stuff first, so we find out about major issues before we're too far down the road. Chat is hard because the server has to send user interface components to the client. Combat is hard because, well, it's complicated—and it's also something that will be different in different worlds, so it forces us to think about the way individual worlds will customise the server.

Craziwolf—thank you for volunteering, and welcome! Your help with the website would be much appreciated. I wouldn't recommend hosting it, though. Given the nature of the project, we're likely to end up hosting one or more of our own worlds, and that will make the hosting requirements more demanding than they might seem at first glance.

I actually started setting up a website earlier today, but I don't think it's a problem because I haven't tackled any of the difficult stuff yet. I've set up Wordpress with the default theme, and created links to various bits of the project like the wiki. I'm part way through setting up a forum.

Of course that's the easy bit. The hard bit is creating a theme, and the harder bit is making the website a useful resource. If you want to tackle either of those, it would be a big help. Thanks! Please let us know if you'd like to take either of those things on.

(Those 3D renders are amazing. I can't believe they're not photographs. Actually that reminds me, I watched Avatar the other day—I never watched it when it was first out—and I was amazed by the quality of the computer-generated scenes. I also realised that I should start turning Blender's Emit slider up. Those things that glow in the dark work really well!)

[specsdude]

I actually started setting up a website earlier today, but I don't think it's a problem because I haven't tackled any of the difficult stuff yet
The hard bit is creating a theme

I'd kinda like it if we had a Friendlyfruit logo. Then we can start rolling out the t-shirts and collectors merchandise! That could be an immediate job for you, craziwolf?

Combat is hard because, well, it's complicated—and it's also something that will be different in different worlds, so it forces us to think about the way individual worlds will customise the server.

About the combat, don't sweat it too much. Remember we discussed having many different importable kits to customise one's world- well, you can just have several different types for combat, such as point and shoot, spellbook casting, etc. What would be awesome is creating them so that they can be used simultaneously. Maybe governed by a higher up kit such as general combat?...
Now I'm getting complicated.

Anyway, the more the merrier! Come on you other guys, you know you wanna join in and have some fun!! The water's great! or is it fruit juice we're in?...

PS: Perhaps I will install a flavour of Linux and try and find out Pete's secret trick. BTW, does anyone know if a Bios bootloader would be installed that would force me to choose between Windows and Linux OSs every time I boot? Cos I HATE that. If I could avoid that, would be really nice. Thanks!

[SN]

OK, so I have available to me:
1 Windows 7 Laptop - I've played with Sketchup and Daz3D (and older 3D creation S/W in the past) so I'm fairly au fait with 3D modelling
1 Fedora Laptop - Is this going to be the best place to try getting the stack running?
3 (yes 3 but soon to be 2) Pi's - Can anything run on a Pi yet?

BTW - I find Blender truly awful to get along with and have binned it.
Also I have (somewhere) some Sketchup export scripts that may sort the Collada/Wavefront save/texture issues you talk about - it was how I was hoovering 3D Warehouse stuff into Daz for back scenery

So where should I start to get the stack running?

[craziwolf]

Craziwolf—thank you for volunteering, and welcome! Your help with the website would be much appreciated. I wouldn't recommend hosting it, though. Given the nature of the project, we're likely to end up hosting one or more of our own worlds, and that will make the hosting requirements more demanding than they might seem at first glance.

I actually started setting up a website earlier today, but I don't think it's a problem because I haven't tackled any of the difficult stuff yet. I've set up Wordpress with the default theme, and created links to various bits of the project like the wiki. I'm part way through setting up a forum.

Of course that's the easy bit. The hard bit is creating a theme, and the harder bit is making the website a useful resource. If you want to tackle either of those, it would be a big help. Thanks! Please let us know if you'd like to take either of those things on.

I can defiantly help out with the theme ill just need an idea of what your after and a few weeks to put it together. how it looks now is good in as much as its simple and easy to navigate. as long as we keep that in mind im sure it will be a useful resource all on its own

[n3tw0rk5]

I co own a print on demand & dropshipping merch company, I'm happy to help on that side as long as profits go to the foundation.

[doctorwho]

I don't know anything about this kind of stuff so I was wondering if the terrain is generated on the fly or if there is a dedicated 3d terrain model that is already in the game. If it is generated on the fly where are the variables (average variance in terrain height and such) kept? It seems that it might be viable to start making simple python scripts to generate worlds now.
I was also wondering if multiple characters can meet in virtual worlds yet?

[specsdude]

I co own a print on demand & dropshipping merch company, I'm happy to help on that side as long as profits go to the foundation.

Oh.

I was actually joking, but I'm pretty certain people would love Raspi T-shirts.

[specsdude]

Ok, I've just installed Ubuntu 12. I'm gathering all the packages needed to run Friendlyfruit. Hopefully I will have a working server by the end of today!

How are other people getting on?