What do you want from the developers moving forward?

[Branan]

Now that Cyan has given us the source, the Guild programmers have lots of options for how to proceed.

As always, our goal is to create the tools and software that you the writers need in order to create your wonderful worlds. Now that we have the CyanWorlds.com Engine sources, I feel it's also our responsibility to help as many people as possible be able to enjoy your worlds, through enhancements to the engine.

Some of the first efforts we have in mind are to port the game to Mac OS X and Linux. After that, other enhancements are possible, and "the sky's the limit" so to speak. I'd like for those enhancements to be driven by the artists. That is, we'll make sure the game can do what you want to create.

One thing we're seriously considering is breaking network and PRP compatibility with Cyan's servers and data. This will allow us a wider range of options in the future. I hope that since many of you will have to re-export your ages for the CWE engine anyway, you'll be willing to work with us when we change things.


Now that that's out of the way, the question in the topic can finally be addressed: What do you want to see from us as we move forward with development?

[Tweek]

Hmm, well, not sure if this is the kind of stuff you're looking for.

In terms of website I would like to see a repository for textures and for models clearly linked to the site so it is easy to find (In regards to the GoW I'm all for centralization and everything being under one roof, however others aren't. There are apparently resources out there for textures but finding them is difficult).

In terms of technology. Back when PRPExplorer was around (before it got replaced by whatever is being used, I never changed as I liked how the former looked and worked) it was possible to load the Age/area up without being in Uru and allowing for a 3D view of the place this was great for map making, something that can do that with the new content would be nice to see not just for us but for the GoCart. Course an updated PRPExplorer would be nice as well for texture hunting/adjustments.

Few years back I was trying to get the Guild of Fine Artists rolled into the GoW and give the artists more of a homely feel here (as it does feel very developer heavy), not sure how to go about making things more comfortable for artists though.

Tutorials, really need to be addressed. More friendly for the less technical minded. Don't automatically assume knowledge of Blender. Number of times I see stuff that says "To do X you must active Y mode now do Z and X is now how it should be". Well nice, I've been using Blender for 3 years now. I don't know how to active Y because I've never had to before.

Hell I've been using Photoshop 7 for something like 8 years and I still find new things I didn't know about.

Aside from that, I'm the wrong person to ask I think, I keep to myself, I build my own stuff (when not using cyans and stuff), build my own textures (when not using cyans) aside from the use of existing tools I'm rather self sustaining in my work style.

[N. Sigismund]

I've said roughly what I'd like to see before, but just to get everything down here...

The content I develop I've always viewed would be used on a server along with lots of other content, some Cyan and some not. I suppose the first thing I should say is that (I know there's hypothetically two versions, cwe-compat and cwe-future or whatever the names are...) moving away from compatibility for Cyan's resources is of significantly less interest to me than the alternative.

Using the example of a similar project - Freespace 2 Source Code Project - their developed stuff has total compatibility for old content and indeed vastly improves that old content. Yet like with us, they don't have the rights to the game assets - only the source code. You still have to own FS2 to play FS2 SCP. I'm not that well rehearsed on that project, but I think that emulating what they did might be wise in a few areas. They did all sorts of stuff, including a great deal of asset replacement. Of course, I'm not suggesting that artists should spend time remodelling Cyan's content for 2011, but certainly I think the idea of doing everything possible to give old content a kick up the backside visually would be a darn good idea.

From top to bottom...

A plugin that works in new versions of 3DS Max - preferably 9 and upwards, but priority for 2010, 2011 and 2012.
A version of the client that runs in the background and works as a really fast level previewer, especially for logic stuff.
Modern real-time lighting and shading that's fit for 2011.
Better implementation of bump, normal and specular maps.
More graphically advanced water. Also, improve implementation of water in plugin so that it doesn't require voodoo magic.
Better physics.
An update to the plugin that does simple stuff - especially journals! - without any need for Python. Ideally as simple as "Open component manager. Select book component. Choose type of book. Choose book font. Choose text file to read book from." And then pow! Anyone who clicks on the object gets exactly that.

EDIT: @Tweek - there have been many attempts to set up resource websites for artists. I tried on on Wikia. The issue is that none are on the front page, as it were. Andy's got models and textures scattered around the place. I've got an entire age which people can steal any model out of whatsoever. I think this is something the GoW could very easily do, but it needs to be in flashing neon and needs to be put at the centre of the GoW's being, as it were. Otherwise it will simply be ignored by the new folk it is designed to help.

[Aloys]

I was about to reply on your OU thread, but I might as well do it here.
Speaking purely from an Age developper point of view, here's what I have in mind:

First of all, I know this is not gonna be a popular opinion; but I think working on updating the Max plug is a misdirected effort. How many people actually use Max? Especially in our community? 4? 5? Blender might not be Max functionnality wise, but it works, it's free, and all current Age writers already know it. And in some ways it's actually easier to use than Max (it's a subjective thing, but I can't stand Max's interface.) What we need is simply a more powerful and simpler interface for PyPRP. (and ideally it would target Blender's redesigned 2.5 GUI)
I hate to say that, because a lot of work went into it; but : we have to find a better tool than Alcscript. It is difficult to learn for artists (yeah I know..) and it makes it simply impossible to manage an Age with more than 20 objects. It is useful, it does a lot of things that cannot be done differently right now, but it is not productive at all.
Cyan's Max plugin is great in that regard; simple to use, intuitive, productive. Let's take inspiration from that, port the thing to Blender 2.5 and call it a day. Easier said than done, I know, it might be a long term project, but really there's nothing that can help Age development more than easy to use tools. PyPRP as it currently stands is simply scary for a new comer.. and even for old timers.

Actually there is something that could help us even more: Uru Blender. Which could actually be easier to do right now with the redesigned python-driven 2.5 GUI. That would be a great tool for Age creators; especially new comers.

Then : Debug client binaries. I haven't seen it mentionned anywhere yet but Cyan's internal debug client should be included in the Client sources, right? We need it. Win32 binaries to be more precise. This might actually come first in the todo list as it should be faster to do than the whole PyPRP mess. That debug Client would tremendously speed up the Age testing process.

There are plenty of other things, better tools etc; but those are the more important and urgent things I see right now.

[Branan]

I absolutely agree with you, Aloys. PyPRP2 targets blender 2.5 and has a goal of writing GUIs when possible. There are still some functionality issues in Blender 2.5 that are making it difficult. PyPRP2 will likely still use HSPlasma, as it will allow targeting both CWE and the older CC version of the engine. That will ultimately be up to Lontahv though. PyPRP2 is his baby.

There's also AgeCreator, which is once again on hold while I work on Client code (though this time CWE instead of PlasmaClient). I'm considering trying to port it on to the CWE libraries, so it will always target the latest version of the game.

Debug clients are easy, and we should be able to build one before the weekend is done. As for anything I wrote or will write in the OU forums... No comment until I can finish ironing out some sort of workflow over there with the OU team. They have very different ideas of how to run a project than the GoW developers traditionally have, and trying to mesh them is going to be exasperating for both sides.


EDIT: For the record, I'm a developer and I absoltely hate alcscript. So there you go.

[Calena]

Being a noob, I'm not qualified to contribute much to this discussion, but there are a few thoughts I had.

1 - As mentioned by Aloys, I'm one that isn't too keen on spending $$$$$$$ on new software to work on ages. I'm happy with Blender, though on my machines it does run better in Linux.

2 - Before updating the graphic rendering capabilities, it might help to re-state the system requirements we're building for. I'm still thinking we're building for 2004 (ok, 2006-8, because who is on the internet with 256Mb RAM?).

3 - Super-simple age-creation software - I'm primarily a techie who struggles with the art. My son is an artist that struggles with the tech. He has played with building in Bryce (quite successfully) and his SO has tried to convince him to build in Second Life. He hates it and gave it up because he feels too limited by what the software can do. The best age creators will always be the rare personalities that excel in both disciplines. The two of us are collaborating on the next age I build. He's going to design it, I'm going to build it. That's how we solved the problem. Still, people like the ability to build and IMHO simple building software would be a huge success and very popular. Heck, even some type of small pre-built ages that people could build up on their own would probably be a winner.

And especially, thank you for whatever you come up with. I realize this is an enormous endeavor that you're expected to do in your spare time. It is appreciated.

[Egon]

A version of the client that runs in the background and works as a really fast level previewer, especially for logic stuff.

I think that this is the most important feature age developers might want right now. Ability to reload age/update age files without restarting whole game.
Most of important stuff have been already mention so here is more of wish-list:
Plugin stuff:
- support for all GUI elements
- ability to implement NPCs (DRA Office team managed to implement NPC but as far I saw they didn't have full control upon it. Like cloths and smooth animation during talking)
- there should not be need for having one set of light for lightmap and one for in-game lights
Engine stuff:
- better support for localizations
- I don't remember terminology now, but as far I can remember Plasma Engine does not use some of modern day standard ways of reducing number of triangles needed to be rendered

2 - Before updating the graphic rendering capabilities, it might help to re-state the system requirements we're building for. I'm still thinking we're building for 2004 (ok, 2006-8, because who is on the internet with 256Mb RAM?).

You might be surprised

[Branan]

Since hardware requirements have been mentioned: I'd like to be able to set the system requirements at a solid DirectX 9.0. That's Radeon 9500+ or GeForce 6000 series, which is approximately 2003. If I had my way I'd target DX10 hardware only (GeForce 8000 / Radeon HD2000, around 2007), but I have a sneaking suspicion Uru players don't keep their computers as up to date as many other gamers . I think targeting DirectX 9 will still allow for moving towards more shaders and lifting some of the limits that are left in the engine from older hardware generations, without being overly restrictive.

(we'll actually be targeting OpenGL 2.0, since the engine will need to run on Mac and Linux, but the hardware levels are the same)

[Luna]

Since hardware requirements have been mentioned: I'd like to be able to set the system requirements at a solid DirectX 9.0. That's Radeon 9500+ or GeForce 6000 series, which is approximately 2003. If I had my way I'd target DX10 hardware only (GeForce 8000 / Radeon HD2000, around 2007), but I have a sneaking suspicion Uru players don't keep their computers as up to date as many other gamers . I think targeting DirectX 9 will still allow for moving towards more shaders and lifting some of the limits that are left in the engine from older hardware generations, without being overly restrictive.

(we'll actually be targeting OpenGL 2.0, since the engine will need to run on Mac and Linux, but the hardware levels are the same)

on a side note, some older cards run Directx9.0 too

@topic: Some kind of pre render/level building or even Urublender would be great.And..actually most things I would want are already mentioned by others

[Agenotfound]

First of all, I know this is not gonna be a popular opinion; but I think working on updating the Max plug is a misdirected effort. How many people actually use Max? Especially in our community? 4? 5? Blender might not be Max functionnality wise, but it works, it's free, and all current Age writers already know it. And in some ways it's actually easier to use than Max (it's a subjective thing, but I can't stand Max's interface.)

Well there is 4 now because of the version required wich is almost impossible to find and so old that no one wants to use it.

Also there is quite a lot of people out there that know how to use max or other more mainstream modelers that do not want to make the effort to learn blender because of how different the interface is.

Look at guys like Loshem (not sure of the name) that made the model of the Guild Hall in the official forums and who wanted to port it in the game, like many of the people that have a lot of experience with modeling he does not want to use either Blender or Max 7 because it's such a pain to completly change your workflow and I would really like to see people like him make ages for uru.

I'm not saying of course that Blender should be ditched, I'm just saying that having a tool for Age Writing that works with a more mainstream modeler will bring more people in and that can only be a good thing.

I for myself use 3ds Max quite frequently at a professional level as well as other modelers and I would love to build some ages for Uru. I tried the Plugin for the Max 7 version but cannot work with something that will no run half of the scripts I use daily and I already have a hard time not mixing up shortcuts and names of things in the softwares I use so I don't want to add blender to the mix.