Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

Dark GDK / DGDK Open Source Project

Author
Message
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 17th May 2010 11:11 Edited at: 17th May 2010 11:15
Quote: "MAUI - Not quite sure of the concept, but to me it seems like a data-driven UI (data being stored in XML-script files). This may be well and good - you would have to have a UI editor (not unlike a resource editor) to pull it off. From what I've seen so far, I think it's rather convoluted and difficult to follow. I like the idea of data-driven UI (for any application - including games) but for S3GE, is should be easy. That's not what I'm seeing so far... I know you've spent alot of time thinking about this tech, but make it easier - we already have WINPROC, dbKeyState() and similar functions to work with. Give us a UI editor and your golden..."


JTK Thanks for your perspective. Yes, MAUI may appear convoluted at this stage, but, its work in progress. I'm refining it as I go and its coming along with better than expected results. MAUI is getting to the point where it can be used to develop Applications and an Editor is the first one planned. This is my focus.

Conceptually, MAUI is a Event-driven UI with Scripted Behaviors/Transitions/Themes/Schemes for 2D/3D Gizmos. Its modeled after a Web-browser using XML and LUA (client-side scripting). Its not your typical Window-like GUI and that's deliberate.

Quote: "Sorry, I had to say something. It seems to me like everyone is "meandering" along their own path - which sounds good, but doesn't move the project forward..."


"meandering"? What about brainstorming? I encourage everyone who contributes to throw as many ideas out there as possible. We will eventually eliminate and select the ones to focus on and develop. As long as someone is working on something and adding it to the project, its moving forward. There is no time line my friend.

haliop
User Banned
Posted: 17th May 2010 11:39 Edited at: 17th May 2010 11:40
JTK i think i can say from all the members in S3GE
Thank you for your honest reply it means alot to us.

and yeah i agree on the part of "development looks like it stopped" sirth of way , but you can't belive what you see , there is much more behind the surface. (well if its a plain staing on the Y axis.. if not.. then beneath).

for example, a few weeks ago , i was able from what i learned here on this thread and from working on S3GE , i was able .. to develop a mini - 3D presentation in about a week, i cant say what it was about for some reasons (dont hate ) , but from what ive learned from Tech and Matty and other users on this forum , in one week i was able to acomplish a decent Presentation.

so its not only stuff to show , it has much more to it, the actuall learning from S3GE project is for itself good enough not talking about the ability to develop some amazing ideas at the end of it and while at it.
just take the demo we are making , i was a huge fan of BattleZone it was the best game i played from that genre it was a huge succes if i remmber correctly , and to think that we are actually building something like that.. i can honestly say. S3GE project, where binary dreams come true.
marlou
15
Years of Service
User Offline
Joined: 17th Jan 2009
Location:
Posted: 17th May 2010 12:22
Quote: "...Unfortunately, we do not know marlou's progress or intentions within the project.. "


Sorry about that..
About the pathfinding..
Progress:
1. Navigation mesh building from darkGdk object - finished
2. Performing A* pathfinding with NavMesh - finished
3. Drawing navmesh polygon in darkGdk - not finished
4. Logging of navmesh building progress - finished
5. Physics integration - not finished

There is one weakness with the pathfinding and that is dynamic changes in navigation meshes cant be handled without lag. This feature is already present in Kynapse and HavokAI. Im still clueless on its implementation.

Intentions on DOSP..
Im having second thoughts on using DarkGDK in making a super game engine. Im trying to write an engine that is a hybrid between Irrlicht and Ogre3D.

Sorry for the inactivity >_<.
Ive been spending my free time reading up on advance techniques and deciphering the Irrlicht and Ogre3D source code to its bare essentials so I can write a Hybrid engine.

We should just be using Windows Form in CSharp for making editors as its the fastest and easiest way to make editors. We just have to bind the C++ codes with Swig. It shouldn't take a long time for that Editor to be done.

Sorry that I haven't been adding anything.

When a person has nothing but a dream, can he dare to dream.
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 17th May 2010 13:07
Quote: "and yeah i agree on the part of "development looks like it stopped" "
Stopped? How so? I'm working on Textboxes for MAUI and will start on client/server code to build up the S3GE App Server (for lack of a better name).

marlou, good to hear from you and thanks for the progress report.

Quote: "Intentions on DOSP..
Im having second thoughts on using DarkGDK in making a super game engine. Im trying to write an engine that is a hybrid between Irrlicht and Ogre3D. "
Sorry to hear that. Perhaps you can contribute what you have developed thus far with Recast for DGDK to DOSP?

Good luck with your hybrid Irrlicht/ORGE3D engine.

haliop
User Banned
Posted: 17th May 2010 13:27
no i didnt ment it stopped
i just said it may appear that way, that we think too much and do less , while in reality we think and build maybe not thinking of the stuff we build meaning new ideas.. or brainstorming as you like..
Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 17th May 2010 15:54
Quote: "Matty, that diagram looks great. I only have some vague ideas for dealing with some procedural techniques. Have been gazing at libraries like Cal3D. We can build on your system."


Cool, this needs some more thought/designing before any code is written, I will post with my progress.


JTK - Thanks for taking time out to share your views.

Quote: "MAUI - Not quite sure of the concept, but to me it seems like a data-driven UI (data being stored in XML-script files). This may be well and good - you would have to have a UI editor (not unlike a resource editor) to pull it off. From what I've seen so far, I think it's rather convoluted and difficult to follow. I like the idea of data-driven UI (for any application - including games) but for S3GE, is should be easy. That's not what I'm seeing so far... I know you've spent alot of time thinking about this tech, but make it easier - we already have WINPROC, dbKeyState() and similar functions to work with. Give us a UI editor and your golden... "

I share some of your concerns here, I don't really understand maui that much and have lots of questions. I am hoping that once finished it will become clear as this seems to be the core system for S3GE.

Quote: "ANIMATION - With Matty's diagram, what I see are various ANIMATION STATES; STATES being the keyword here. According to matty's diagram, animation states should be handled by AI... not an easy task. Similarly, by design (again according to the diagram), states need to be fluid - again not a bad thing... You should focus on the AI aspect!"


I don't see states being handled by ai, obviously ai would control ai character states, but states can be controlled by anything you like, hopefully there will be a small, easy to use command set that will give anyone using the model complete control over what state they want the model in at any time.
That all said, this is a concept at the moment and the more questions/comments the better, I want to get across whats in my head(which I may not be doing well) while at the same time get opinions about the direction I am going with this.

Thanks for comments on Fulcrum, also, lighting was my idea, I'm not going to spend too much time on it but I think its important that any demo demonstrating S3GE takes lighting into account, I want to know that what we have is capable of making any game we start making, before we start making it.


@marlou - great to hear from you, I have an unofficial release of Fulcrum which has raycasting, let me know if its missing anything you need.

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 17th May 2010 17:36 Edited at: 17th May 2010 18:21
Quote: "I share some of your concerns here, I don't really understand maui that much and have lots of questions. I am hoping that once finished it will become clear as this seems to be the core system for S3GE."
Sigh. I hoped my YouTube video did a better job of explaining what MAUI is, how it works, and why I made certain decisions in developing it. I feel as if I've completely missed the mark

MAUI is a ultimately a GUI System like any other. Its modeled after a Web-browser using XML and LUA to create User Interfaces in-game with consideration for 3D, animation, networking, and flexibility. The highlight of the system is Scripted Behavior. Instead of hard-coding Buttons, Sliders, etc, I've put a programmable layer in to define this Behavior so any type of 2D/3D Gizmo can be created. Gizmos can be layered on top of each other to create more complex Gadgets.

What is meant by core system is that MAUI integrates DGDK, Fulcrum, DarkNet and other libs together. It uses the same resources that other game systems and game entities do. In developing MAUI, I've developed code that can be useful for other parts of engine. Of course the MAUI Editor will reduce all the complex inner-workings to simple point 'n click creation.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 17th May 2010 18:26
Quote: "MAUI is a ultimately a GUI System like any other."


Should a GUI not be seperate to everything else? This is how I have been taught in my uni courses and so far it seems to make sense. Also, in what situation would a GUI gadget need physics?
I know that you are trying something different which is great, I just hope it will be intuitive and easy to use.

The video does help alot although what we really need is an example of how to get a simple button on-screen and how do we interact with that button? I took a look at the maui header file last night but that didn't really help. Seems you are doing everything in a LUA file, how do I look at this?

Can I use maui without scripting?


Quote: "Of course the MAUI Editor will reduce all the complex inner-workings to simple point 'n click creation."


I don't really want to have to rely on the editor, the programming interface should be simple for the rest of the dosp team to use, then when we have an editor, that should just speed things up a little(or alot).

Some of my questions and issues may just stem from my lack of familiarity with scripting and networking. My gut just tells me that alot of what is integrated into maui should be seperate systems, take the id handling system, this is really cool but why is it tied up with the GUI stuff.

Anyway, you don't need to really answer all these questions as once I have used it, it may all come together for me and the video will be a good overview I can refer to.

haliop
User Banned
Posted: 17th May 2010 19:23
maybe im wrong , but isnt Maui actually controls everything on screen?

from the Buttons (Ui) and the 3D Character.. Tech is this what Maui aimed to do? control everything inside S3GE using Maui based Lua scripting? am i wrong?

if i am , i think we all need to clearify some stuff , it seems like we have some issues with understanding the current workflow, and i guess we need the Ui editor , even if its simple just to get the hang out of Maui.

for our game Demo , i would recomend inserting a Ui editor inside it, like in the Main Menu - > User Interface Config.

so you have all the buttons , health and ammo , weapon bars , equipment , this way , every player can configure it as he likes it be and even name the headers of the Button Screen like Equipment rename to Stuff.. etc..
one more thing that comes with a Ui Config in the Main Menu is the lack of Dragging the ingame Ui or actually interacting with it when there is no actull need for it like a costumazation in game , why put it in the main - menu as more players would like to have in game , ...
well , if we put it in the In Game session or Loop , then we need to calculate mouse positions and stuff and it might decrease our FPS , if we put it as a stand alone inside the Main Menu , then were not only saving the FPS for the In- Game session but also letting users know ," Mecha Demo , its not just a demo , but an editor that actually shows Maui , A Gui System used by S3GE" , my personnal view on this is that this will intract alot of players same as programmers if they saw how easy S3GE is to handle, like in Spore with the Creature Creator, the more simple it is the more People will enjoy it.

Aem - news, not much in the News department but after writing this , i think i'll like to add atleast the Mecha or any other Player Figure Aem plugin , so players can customize , the results will show on Multi-Player Sessions. i have a lot of work with it , but im getting there .


Mecha Model Animation - Not done yet , since i use an IK (inverse Kinematics) bones stracture i have no idea how it will work or even if it will Export , im thinking about making the entire Animation set then saving as a new file and deleting all the bones not needed for the Export , however i think that will destroy the actuall animation..

im thinking of using old school animation bones , which are just the actuall bones needed for export and i think that will work with DarkGDK ,i have alot of testing to do but im not half way there , it will take some more time and as soon as i'll finish it i'll return to Aem.

goodday.
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 17th May 2010 20:41
Quote: "Should a GUI not be seperate to everything else? This is how I have been taught in my uni courses and so far it seems to make sense. "
Separation in what way? A separate thread? Use separate memory resources? A separate plug-in? I did not have intentions of distributing MAUI 2.0 as a separate plug-in.

Commercial game engines Torque, UDK, others have their own version of a GUI system and use scripting for implementation. We're not doing anything out of ordinary in this regard.

Quote: "Also, in what situation would a GUI gadget need physics?"
MAUI uses the physics engine for high performance collision detection. The simulation is a plus. You could build a point n click game entirely with MAUI alone, if desired.

Quote: "
Can I use maui without scripting?"
We have the source, but, hard-coding a GUI is not flexible, nor recommended. You would have to recompile the entire engine/application when changes are made. This would be equivalent to hard-coding a game level.

MAUI will come with pre-packaged scripts for behaviors, transitions, etc for typical window-like buttons. Users will not have to script this functionality but, have the option to modify or create new ones if desired. I'm developing these scripts as I flush out MAUI features.

Quote: "The video does help alot although what we really need is an example of how to get a simple button on-screen and how do we interact with that button? I took a look at the maui header file last night but that didn't really help. Seems you are doing everything in a LUA file, how do I look at this?"
MAUI loads a XML File (main.maui) to build User Interfaces. Its located in
directory. This is what it currently looks like:


The MAUI Help Docs will explain tag usage and rules. However, you may be able to grasp the flow of things from this file. You define the event states, behaviors, transitions, style, themes, schemes first. Then the base canvas, gizmo, devices. Finally the gizmos. MAUI currently only has a Button Behavior and I'm developing a Textbox now.

Quote: "My gut just tells me that alot of what is integrated into maui should be seperate systems"
I'm not sure what you mean by separate systems. The goal should be to integrate systems together to create a engine. Currently, MAUI is the only core system in the engine.

Quote: " take the id handling system, this is really cool but why is it tied up with the GUI stuff."
Testing. I realize the need for common resource managers that are globally accessible to other systems. I intend to relocate the Resource Managers in MAUI into Common lib when more systems make an appearance. This may also change when Scenergy is built up to handle all game level loading.

Quote: "maybe im wrong , but isnt Maui actually controls everything on screen?"
MAUI will manage all User Interfaces. The current goal is to build up MAUI to provide typical GUI function. A GUI editor will follow.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 17th May 2010 21:16
Thanks for taking all the time to answer my concerns, I have looked at the XML file, this is a great way to get a GUI up and running. Am I right by saying users of this system will only need to concern themselves with the gadget creation near the bottom, at least at first anyway?

If so, will all the transitions, styles, themes etc be in seperate files?

I see the button functionlity is scripted, this is all new to me, if I want my button in-game to call a function, say a function that re-arranges some sprites on screen(random example), do I need to bind that function to LUA? If so, how much work is that and does it slow my program down by much?

Sorry about all the questions, they may be very noobish

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 17th May 2010 23:06 Edited at: 17th May 2010 23:14
Quote: "Thanks for taking all the time to answer my concerns, I have looked at the XML file, this is a great way to get a GUI up and running. Am I right by saying users of this system will only need to concern themselves with the gadget creation near the bottom, at least at first anyway?If so, will all the transitions, styles, themes etc be in seperate files?"
Yes, if they choose to. They can also create new behaviors, transitions, etc. The ability to load separate files is coming soon. When I put documentation together for MAUI XML it will be much easier to understand/use.

Quote: "I see the button functionlity is scripted, this is all new to me, if I want my button in-game to call a function, say a function that re-arranges some sprites on screen(random example), do I need to bind that function to LUA? If so, how much work is that and does it slow my program down by much?"


A behavior only defines how the gizmo mechanically operates (selecting the next event state based on pointer bound/contact collision and a timer). It doesn't execute a Action. You script Actions within the Gizmo definition.

Example:


Function binds are always required for Lua to recognize a C++ function from script. In the scenario, you describe. One could either script the sprite rearrangement in LUA using the DGDK Binds or create one C++ function to do it all, bind it, and call that from LUA. Binding is easy. See the darklua_catalog.cpp (renamed to darklua_bindings.cpp in next update) for examples. Although, I do not desire to bind every freaking method/function in the engine, I will bind as many as needed to get the job done.

I honestly cannot tell you if it will slow your program down. In my opinion, its performance is top notch. LUA is one of the fastest script engines out there, that's one of the reasons why I choose it.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 18th May 2010 00:22
OK thanks, its coming together a lttle bit better in my head, when using windows forms in C#, the editor writes your XML for you and also gives you a button event in code, so you just go to that bit of code and write your button funtionality in there.
I see this is quite similar, only the button event is scripted, is that about right?

This leads me to the question of triggers and events in general for the engine, take my character rigger tool, one job an animation might have is to 'listen' out for certain animation frames to trigger an event, say dbPlaySound(footstepSoundID) when a foot hits the ground.
The only way I know how to do this, and its how button events work in Java, is to have an 'event' class which has a trigger() function, you then create a class derived from the 'event' class and over-ride the trigger() function, inside this function is where you put your event code(dbPlaySound()).
This is the way I would probably do it, but with your button events using scripting I thought I had better get your opinion on it first.

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 18th May 2010 01:47 Edited at: 18th May 2010 01:57
Quote: "OK thanks, its coming together a lttle bit better in my head, when using windows forms in C#, the editor writes your XML for you and also gives you a button event in code, so you just go to that bit of code and write your button funtionality in there.
I see this is quite similar, only the button event is scripted, is that about right?"
Yes and the `button event in code` (script) is stored in the XML.

Quote: "This leads me to the question of triggers and events in general for the engine, take my character rigger tool, one job an animation might have is to 'listen' out for certain animation frames to trigger an event, say dbPlaySound(footstepSoundID) when a foot hits the ground.
The only way I know how to do this, and its how button events work in Java, is to have an 'event' class which has a trigger() function, you then create a class derived from the 'event' class and over-ride the trigger() function, inside this function is where you put your event code(dbPlaySound()).
This is the way I would probably do it, but with your button events using scripting I thought I had better get your opinion on it first."
A `Trigger` in most game engines is usually a type of Game Entity - 3D Bound Objects that are placed in the 3D environment that execute actions (script) when another 3D Game Entities collide. A common example would be Door Switches. Even Physx has a built-in mechanism for managing Triggers.

My implementation for Triggers would be modeled after MAUI's Gizmos. MAUI 3D gizmos could be used as Triggers, and this is where I started to confuse people on MAUI's purpose. So disregard the last sentence.

I'm not sure why a animation would need to `listen` but in the context you describe, I assume `listening` would use some form of collision detection. If this is true, you could very well implement `event triggering` in the same manner I did with MAUI Gizmos.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 18th May 2010 02:02
Quote: "Even Physx has a built-in mechanism for managing Triggers."

Yes, this is true, this is also done with the inherit/over-ride technique I described above, Box2D also uses this trigger(collision) system.


Quote: "I'm not sure why a animation would need to `listen`"
Take these examples, there may be a better way, this is something I've come up with myself.
When a character throws an object, you don't trigger the creation of that object at the same time as you trigger the 'throw' animation, you trigger the 'createProjectile' function when the animation hits a certain frame, being the point in the animation where it looks natural to release the object. I can think of other ways to do this, like put a delay on the projectile but this is not as elegant imo.

Another example is playing a footstep sound as a foot hits the floor, as there is no way of knowing how the player/ai might move at any time its pretty hard to get that right unless you let the animation trigger the sound.

There are lots more examples, mostly involving character models as they are probably the most complex. Maybe there is another way to do this, I'm just making it up as I go.

Quote: "MAUI 3D gizmos could be used as Triggers"

I would need to work with you on this if the animation/trigger thing gets the go-ahead. As I have said before, its just an idea at the moment so if there is a better way, let me know and I'll do a re-think.

haliop
User Banned
Posted: 18th May 2010 13:33
About the model.
im not rigging this, in gets on my nurves tooooooooo much



if someone else want to do it , go ahead.
Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 18th May 2010 14:09
Lol, no problem haliop.
I still don't know what the best tools are to animate stuff and export to DGDK, what do people use?

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 18th May 2010 15:07 Edited at: 18th May 2010 15:55
Quote: "Take these examples, there may be a better way, this is something I've come up with myself.
When a character throws an object, you don't trigger the creation of that object at the same time as you trigger the 'throw' animation, you trigger the 'createProjectile' function when the animation hits a certain frame, being the point in the animation where it looks natural to release the object. I can think of other ways to do this, like put a delay on the projectile but this is not as elegant imo."
I would recommend using the word activate in place of trigger to avoid confusion with previous definition of a Trigger (Game Entity).

Timing Actions with animation for weapons and other animated game entities will be required too. I'm not sure that logic can be stored inside DGDK 3D formats (dbo, 3ds, x) so a high level format built on top of these maybe needed. Good News: this is included as part of Scenergy MECS Format.

My approach would be slightly different, using a system of Scripted Behaviors and Transitions similar whats in MAUI. Lets call this system BOT. BOT would govern all the various entity properties for states, animations frames, etc to execute scripts to call functions for sound production, pathfinding, firing weapons, etc.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 18th May 2010 17:42
Quote: "My approach would be slightly different, using a system of Scripted Behaviors and Transitions similar whats in MAUI. Lets call this system BOT. BOT would govern all the various entity properties for states, animations frames, etc to execute scripts to call functions for sound production, pathfinding, firing weapons, etc."


Its great that you have thought all this through, but I may be best leaving my animation/model rigger for now as it seems BOT may want to handle this.

We have set ourselves the target of putting a FPS game together but it seems like we are missing some basic systems. The first thing I would want to do is get a character up and running, controlled by mouse/keys/joypad, with a basic set of smooth interpolated animations(perhaps with the ability to activate game events). It seems like we may need to wait for MECS before we have any of this, is this right?
Of course we could just hard code all this seperately for now, but that would take time and recources away from S3GE.

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 19th May 2010 00:09 Edited at: 19th May 2010 00:12
Quote: "Its great that you have thought all this through, but I may be best leaving my animation/model rigger for now as it seems BOT may want to handle this."


I've been developing MAUI, DarkLUA, DarkNet Extensions and integrating other libraries into the system for a few months now. I've touched Input, Scripting, Networking, Physics, Graphics, and Sound. It has been nothing but Research and Development (R & D) for me. I can visualize many ways to implement systems based on this experience, however, I also do a lot of internet research on the tech behind these subjects in current game engines.

I'm not insinuating you should change the animation/model rigger. In fact, you may have innovative way for dealing with animation/action syncing. I'm only stating that our perspectives on implementation are slightly different: you visualize the animation controlling the action, I visualize a UI or AI system controlling the animation & action. In MAUI, I've added a programmable layer that controls the mechanical operation of a Gizmo/Gadget. Perhaps, a programmable layer can be added to animation/model rigger to control a animations mechanical operation?

Quote: "We have set ourselves the target of putting a FPS game together but it seems like we are missing some basic systems. The first thing I would want to do is get a character up and running, controlled by mouse/keys/joypad, with a basic set of smooth interpolated animations(perhaps with the ability to activate game events)."


We can do this. MAUI can handle input devices, have you seen my latest MAUI Vid on Widgets? Is the Fulcrum Character controller ready? If so, S3GE will need update. Do you intend to let DGDK handle the animation/interpolation or do you have another solution in mind?

Quote: "It seems like we may need to wait for MECS before we have any of this, is this right?"


We dont need MECS to get started. DarkGDK already has the facilities to load and animate 3D Entities. What it doesent have natively is a system to 1) attach or swap out mesh parts and accessories (such as particle emitters); 2) apply physics to bones/joints; 3) change the color of sections in textures. We'll need to build these subsystems and that's what Modular Entity Construction (MECS) is about.

MECS is S3GE's system for managing Game Entity Assembly and Animation that will integrate DGDK, Physics, other libs. Like MAUI XML, MECS will have a high level XML Format that tells S3GE how to setup/assemble a Game Entity. We will most likely find ourselves creating several XML formats, all part of a suite of Game World Building Formats (Scenergy).

Quote: "Of course we could just hard code all this seperately for now, but that would take time and recources away from S3GE."
Or just analyze the FPS demo that comes with DGDK.

Its all going to require massive amounts of R & D. We may have to develop little tools on the side to test it. We have the option to develop these tools with another GUI Lib or wait for me to flush out MAUI (its getting close to being usable) and develop them with S3GE. The advantage of the latter is we can work together on it, flush out other systems, expand the tools into more power editors.

I recommend we proceed with the major update followed by the Project restructure before we begin this stage of development.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 19th May 2010 01:43
Quote: "Do you intend to let DGDK handle the animation/interpolation or do you have another solution in mind?"

Yeh, GDK can interpolate between animations pretty good, the only snag seems to getting the correct length for the interpolation, you may find a value that works most of the time but looks pretty bad between certain transitions. I would take this into account in any system I design, we may need a different value for every possible pair of animations. This is why we would probably need a model viewer to set it up, it would be pretty tedious finding good values without seeing the effects in real-time.
This is all theory at the moment, I would have to do some testing to see if it would work.


Quote: "Is the Fulcrum Character controller ready? If so, S3GE will need update."


Yeh, that lighting demo uses it, I have been working on Fulcrum tonight, I think I just needed a little break from it.

Quote: "I recommend we proceed with the major update followed by the Project restructure before we begin this stage of development."


I agree, we need to get it to a stage were people can run it without any messing, thats easier said than done though.

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 19th May 2010 02:00 Edited at: 19th May 2010 14:59
Quote: "Yeh, GDK can interpolate between animations pretty good, the only snag seems to getting the correct length for the interpolation, you may find a value that works most of the time but looks pretty bad between certain transitions. I would take this into account in any system I design, we may need a different value for every possible pair of animations. This is why we would probably need a model viewer to set it up, it would be pretty tedious finding good values without seeing the effects in real-time.

This is all theory at the moment, I would have to do some testing to see if it would work."


Its good to hear DGDK can handle the animation job. A model viewer is a necessity and I would expect a simple viewer to be part of S3GEd's Asset Previewer along with preview for Textures, Shaders, Particles, etc. I also anticipate an Animation Editor as part of the S3GEd Suite to create and tweak animations.

Ultimately, S3GE will require it's own suite of Editing Tools and we'll need primitive versions to build the demo. I've anticipated this requirement and have been working frivously to get MAUI to a state where we can do so. Its almost there

So what Editors does S3GE need? I'm taking my ques from UDK and Hero Engine suite of editing tools.

I prefer most (if not all) Editors to be developed with S3GE for a couple of reasons:
1. Co-development of S3GE
2. WYSIWYG Editing
3. Editor Synergy
4. Collaborative Editing via S3GE Server/Client
5. Derived Editors for Games
6. Reduced Workflow

Quote: "Yeh, that lighting demo uses it, I have been working on Fulcrum tonight, I think I just needed a little break from it."
I must of have missed that demo...

Quote: "I agree, we need to get it to a stage were people can run it without any messing, that's easier said than done though."
Let me know when you're ready to update Fulcrum for S3GE. I make my update after some testing.

If all goes well, I will post my proposal for the directory restructure. After we settle on a new structure. I will immediately start working on it. It shouldn't take longer than maybe 3 full days work to setup and test.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 19th May 2010 15:46 Edited at: 19th May 2010 15:47
Quote: "I must of have missed that demo..."

I meant the video I posted on you-tube

Quote: "Let me know when you're ready to update Fulcrum for S3GE. I make my update after some testing.

If all goes well, I will post my proposal for the directory restructure. After we settle on a new structure. I will immediately start working on it. It shouldn't take longer than maybe 3 full days work to setup and test."

I can only do about a hour a day at most until my exams finish mid-June, so its hard to say how long it will take.

If possible, it would be great if people could just put:

Tell the compiler where all our libs, source files and headers are and they are away.
The S3GE.h file would then have all our dependencies linking up the new file structure together, I know its not easy though and we still have that problem with DarkNet.
I may have a little play around with the file structure myself, see if I have any success.

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 19th May 2010 16:58 Edited at: 19th May 2010 16:59
Quote: "If possible, it would be great if people could just put:
Quote: "#include "S3GE.h""


To be honest, double-wrapping the core libs under one header would take very little work. I also see very little advantage in doing so. The main.cpp is most likely the only file that would use such a header and it is the Super 3D Game Engine. S3GE is a Application, not a library in itself. When its all said and done the end user will not have to compile. Just create media/scripts and launch S3GE.exe to run their game.

IMO, the directory restructure offers the most benefit and would promote cleaner SVN updates by putting sub directories and files associated with DOSP under one root directory. I've accepted that this may result in additional copies of libraries on ones local machine, but, it would guarantee no file conflict or corruption.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 19th May 2010 17:34 Edited at: 19th May 2010 17:35
Quote: "To be honest, double-wrapping the core libs under one header would take very little work. I also see very little advantage in doing so."


It is simple and the advantage is not great but when the dosp team want to write samples/demo's etc it saves writing all the #includes out every time, perhaps its just my OCD
I was also thinking that when we had the new file structure the #includes may be more complex eg from:

To


I could be wrong here as I have not really looked at it properly but its about keeping everything in one place but seperate, as in, you can go and look at the headers just for one system, they will be in our one header include folder but then seperated into different systems under their own folders.
I'm not great with this kind of thing but I like things to be organised and intuitive.

JTK
14
Years of Service
User Offline
Joined: 10th Feb 2010
Location:
Posted: 19th May 2010 18:45
@techlord with that last post maui is starting to make sense... Lol

JTK
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 19th May 2010 19:08 Edited at: 19th May 2010 19:25
Quote: "I could be wrong here as I have not really looked at it properly but its about keeping everything in one place but seperate, as in, you can go and look at the headers just for one system, they will be in our one header include folder but then seperated into different systems under their own folders.
I'm not great with this kind of thing but I like things to be organised and intuitive."


I'm getting slightly confused matty. Are you referring to the actual file structure or the VC++ IDE filter structure? The IDE already provides everything in one place but seperate. IMHO, the IDE filter Structure is fairly organized, sorting libs under subsystem filters. There is only one main.cpp to be concerned with. Code a new subsystem, place it in the appropriate filter in the IDE and its actual files in the appropriate directories.

Only the Engine is hard-coded and compiled. Unless your dev'ing a sub system with dependencies on some other libs, one may never even crack open source files to view their contents. I've implemented lua, box2d, etc, and have never viewed any of the source. Most of the time I'm looking at help documentation and relying on Intellisense.

The current hard-code demo is going away and all future Demos/Apps/Games will loaded into the Engine Media & XML/Scripts files. This is the pretty much how most Game Engines work.

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 19th May 2010 19:28
Quote: "I'm getting slightly confused matty. Are you referring to the actual file structure or the VC++ IDE filter structure? The IDE already provides everything in one place but seperate. IMHO, the IDE filter Structure is fairly organized, sorting libs under subsystem filters. "

No, everything is nice and tidy in the IDE, I mean the actual file structure, we have some .cpp files where the main.cpp file is, our include folder has a mix of include/source files, some libs need to be in the solution explorer while others dont, we need to add extra stuff in the project settings etc, then the DarkNet issue.


Quote: "The current hard-code demo is going away and all future Demos/Apps/Games will loaded into the Engine Media & XML/Scripts files. This is the pretty much how most Game Engines work."

O.K, I'm not too familiar with this system, so alot of what I said above might not matter

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 19th May 2010 20:47 Edited at: 19th May 2010 20:48
Quote: "O.K, I'm not too familiar with this system, so alot of what I said above might not matter"
Sure you are. You use this system everyday!!! Your Web-Browser is a Engine. Loads in a file, produces pictures, video, music, accepts input, etc. Just imagine if you had recompile your web-browser's hard code every time to achieve that!

So here's whats going on with MAUI. I've been working on the behaviors that will allow you to manipulate text. This task is equivalent to writing a very primitive version of notepad from scratch. One would probably assume this to be a simple task to accomplish, but, its fairly involved. I took notepad for granted.

WIP Vid
(New Vid Card produces better Vidz)


haliop
User Banned
Posted: 21st May 2010 09:15
Theme / Template
----------------

based on a User - Friendly Scripting idea and some posts where something like this was mentioned (themes).

* this is for later use in the project and i will be more then happy to do it, but for now it may seem a bit futuristic in mind as for what our goals are for the current time *

Objectives: Capture The Flag , Team DeathMatch , Bomb Defusing , Capture a Base, King of the Hill etc..

how does this consirn to Themes?

lets say we are far ahead inside S3GE editor , we just built a cool looking map and now we want to test it (Multi-Player . since Single Player will need to include Timed Events and such..)
but we cant test it right away , we need to insert an Objective of somekind , even if its just a DeathMatch , we still needs that objective.

here the Template / Theme comes in handy.
inside the Editor Menu, i will Add a Script Themplate button.
it will give the User a Start Point and an End Point for he's completed map so he will be able to Test it with a friend or even release it on he's server for players to play it.

why do we need it?
Fasten up our Beta - Testing , while not considering the Objective but taking the Map as the beta testing , where are the glitches, texture problems , player positioning .. physics bugs , balance fixes etc..

more..
any S3GE user , will be able to create a Script Template and save it .. ofcourse nothing new here, he will be able to use it later on or use it on the Online S3GE co-op method of the Editor.

what will it give us? (core team and contributers)
it will represent us with a simple "form" of game editing, you build your map , select who is the enemy who is the ally who is the player , some weapons.. one Template Script or a Few to switch inGame, and you start playing and testing S3GE capabilities , bug tracing , limitations , fps etc.. only that you do it inside a "ready" game, so actually same as beta testing mentioned above , it will fasten development in a way that it will be much easier to accomplish , like have some fun tracing bugs.
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 21st May 2010 09:35 Edited at: 21st May 2010 10:06
I posted a similar concept for Gameplay Templates in my DarkMORG thread (Posted: 4th Sep 2008 10:25)

Quote: "Maps and Quest Building

DarkMORG takes a modular approach to Player Quest Building. Quest are created by interconnecting two or more FPS Tournament Maps by Objectives (also known as Task in RPG circles). Maps are constructed from Pre-Assembled Rooms. Pre-Assembled Rooms are constructed from MEC Constructs.

The idea behind this approach is to allow Players to quickly build Quest from pre constructed maps and objective templates. Each Map is assigned a Game Mode that defines a FPS style of game play in which a Individual or Teams complete an Objective. Upon completion of the Objective, a Reward|Penalty may be a awarded to individual|team Winners/Losers. Winner Individual|Teams are allowed to advance to the next map in the quest.

Note: that in a Team based Goal-based Game Play Mode it is required that all surviving team members arrive to Point B in order to complete and advance to next map.

Tournament Game Modes:
CTF: Carry Object(s) From Point A to Point B, Item can be Dropped (Timed|Untimed) (Reward|Penalty)
Rescue: Protect Object(s) From Destruction as it travels from Point A to Point B, Object movement is automated and the object can be Destroyed|Recovered (Timed|Untimed)(Reward|Penalty)
Race: (First|Last) (Individual|Team) to Travel Point B From Point A (Timed) (Reward|Penalty)
DeathMatch|Arena: (First|Last) (Individual|Team) to (Collect|Destroy) Object x n (Timed|Untimed) (Reward|Penalty)
Stealth: (First|Last) (Individual|Team) to from/to Point B From Point A WITHOUT Triggering Event (Timed/Untimed) (Reward|Penalty)
Hunt: (First|Last) (Individual|Team) to Trigger Event (Timed|Untimed) (Reward|Penalty)
Scavenger: CTF for Point(Timed|Untimed) (Reward|Penalty)
Pub: A Pub is a Non Combative Map in which players can gather to chat, buy stuff, join parties, and select/continue Quest, Enter Tournaments, and Play Mini-games, go home. Each Quest comes with one Pub by default, however, Quest Builders can add more if desired.

Although each Map can have completely different default content and purpose, Quest Builders can provide continuity between maps by using story dialog and comparable content and assets. Each Game Mode can be modified allowing the Quest Builder to set Point Locations, Objects, Timer, Reward|Penality. Maps can also be modified allowing the Quest Builder to add/edit Spawners, Triggers, Platforms, Doors, Push/Pull Props, Warps and Dialog Menus."


I definately want to expand upon this concept for other game genres.

At work and have the need to code, but, no way to do it. So I cracked open TextPad and started outlining headers for Behavioral Rules Artificial Intelligence Node System (BRAINS), Illudium Particle System Manager, Modular Entity Construction Sets (MECS), and Scenergy. These may vaporize into thin air before becoming anything tangible, but, I have to get my ideas on paper.

haliop
User Banned
Posted: 21st May 2010 10:34
yeah! awesome.
altough i dont like the idea of Pre-Assembled Rooms which reminds me of FPS creator or do you meen something diffrent?

about Game Genres altough i wont go for a Genre Devision
i would go for something like this
we can have it in a Button DropDown Selection.

so you have the Quest or Objective Button
something like this:



instead of deving by Genre , let users combine known Objectives from Fps and Rpg with lets say a Strategy Game.. and the opposite.
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 21st May 2010 15:28
Quote: "altough i dont like the idea of Pre-Assembled Rooms which reminds me of FPS creator or do you meen something diffrent?"
The idea was to construct buildings in a modular fashion using prefabs. FPS Creator's interior editing technique is not that bad and I wouldn't rule out using a similar technique, combined with other techniques.

Quote: "instead of deving by Genre , let users combine known Objectives from Fps and Rpg with lets say a Strategy Game.. and the opposite."
The would be the goal but we need a starting point. Perhaps if we breakdown the Objectives into even smaller components, we can identify objective components common to many genres.

We are thinking alike haliop.

haliop
User Banned
Posted: 21st May 2010 19:41
Quote: "The would be the goal but we need a starting point. Perhaps if we breakdown the Objectives into even smaller components, we can identify objective components common to many genres"


can you explain more? i dont understand , sorry.

Quote: "We are thinking alike haliop."

you know what they say about great minds.
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 21st May 2010 20:19 Edited at: 21st May 2010 20:22
Quote: "Quote: "The would be the goal but we need a starting point. Perhaps if we breakdown the Objectives into even smaller components, we can identify objective components common to many genres"

can you explain more? i dont understand , sorry."


My reply was to the following quote:
Quote: "about Game Genres altough i wont go for a Genre Devision
i would go for something like this
we can have it in a Button DropDown Selection."
S3GE is defined in the top thread as: "A Game Engine capable of producing unique 3D Computer Games based on one or more basic game mechanics that define popular computer game genres/sub-genres: FPS, RPG, RTS, Puzzle, Arcade, and many others."

In other words, the primary goal of S3GE is too mix the game mechanics of multiple genres. The Objective Dropdown box is a very good list, however, it may not satisfy all games genres. We may have to break the objectives down further to even smaller components. By manipulating those components, we will be able to create a very flexible Objectives suitable for any genre. New Objectives may spawn from such a system.

haliop
User Banned
Posted: 21st May 2010 21:13
ok i understand, with that said it is nessecry that we keep it "simple" in someway,
enabling users to generate and regenerate tasks/ objectives / missions according to the current enviorment -> for Simple Users

-----------------------------------------------------------------
i keep Simple and Advanced seperate for the sake of Quick or Fast beta testing.
-----------------------------------------------------------------

Advanced :
Quote: "We may have to break the objectives down further to even smaller components. By manipulating those components, we will be able to create a very flexible Objectives suitable for any genre. New Objectives may spawn from such a system.
"



we can even go to an Expert Section where , you click somewhere on the Enviroment or Map , Right Click -> Trigger -> New Script
or Right Click - > Script -> New Objective
inside that , you have all the Default stuff plus the actuall Lua Structure plus PlugIns Connections plus Recompile

Recompile -> the whole section or Plug-in needed to accomplish..
and from here we can even go away and away into the Actuall Mechanics of S3GE .. but do we need to?
my idea of a game engine require no Actuall Scripting at all ,just Buttons,Strings ,Numbers making it fast , secure against bugs and most important Understandable.
Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 22nd May 2010 00:48
Alright guys, I've put a bit more thought into an animation/state solution. I will post more details soon, I'm a bit short on time at the moment.
I will leave out the animation/activation/trigger feature as it may not be needed and it should be easy to add in if ever needed.

When we started I suggested we could give challenges to the community, I think this may be a good candidate for a challenge for the following reasons.
1. I'm short on time.
2. I have a small, clearly defined set of objectives.
3. I have an outline to one possible solution.
4. Its quite a small task, few hours/days.

Any small challenge may bring people into the project, if you don't like the idea of a challenge I could try implementing this myself after exams(mid June), if you're in two minds I could post what a challenge may look like here first and see what you think before it gets its own thread.

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 22nd May 2010 01:55 Edited at: 22nd May 2010 02:02
Quote: "When we started I suggested we could give challenges to the community, I think this may be a good candidate for a challenge for the following reasons.
1. I'm short on time.
2. I have a small, clearly defined set of objectives.
3. I have an outline to one possible solution.
4. Its quite a small task, few hours/days."


Are you (matty) going to organize and manage these challenges? I'm tied up...

Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 22nd May 2010 02:11
Yeh, this one anyway. I'll post a possible format here first over the weekend, see if its do-able.

JTK
14
Years of Service
User Offline
Joined: 10th Feb 2010
Location:
Posted: 22nd May 2010 02:54
I'll be honest with you guys...

As far as S3GE goes, I'm not interested in any challenges yet. I don't think you guys are ready for that. However, if the challenge is meant to invoke ideas on how to achieve the desired effects (outside of S3GE) then I'm all for it as long as they are simple.

Simple meaning a specific example of achieving a means to an end, not as designing an entire sub-feature set of S3GE.

If, for example (since it's a relatively new topic), the challenge involved say "Animation/Activation/Trigger", I would certainly consider participating - however, my participation would be to illustrate "hard-coded" methods that I would put together. You would then have to take my submission, determine what (if anything applies) and carry the appropriate code/design structures into your project...

Furthermore, I would only participate if it involved a small sub-set of features that partly comprise the entire "Animation/Activation/Trigger" mechanism of the feature...

I hope this makes sense to you...

Again, I keep an eye on this thread because I see great ideas forming. I would like to participate, but I'm in no position to take on any "specific" tasks at the time; so if participation involves "challenges" - so be it...

JTK
Matty H
15
Years of Service
User Offline
Joined: 7th Oct 2008
Location: England
Posted: 22nd May 2010 03:23
Quote: "If, for example (since it's a relatively new topic), the challenge involved say "Animation/Activation/Trigger", I would certainly consider participating - however, my participation would be to illustrate "hard-coded" methods that I would put together. You would then have to take my submission, determine what (if anything applies) and carry the appropriate code/design structures into your project..."


The challenge would be small and specific enough to be useful in its own right, whether in S3GE or any other engine/library.

In a nutshell, and I'll be specific for clarity, I want you to be able to change the state of your character model from standing to crouching and the animation system carries on working without any additional effort, all the effort should be in the setup before your game loop.
For instance, if you character model has two 'throw' animations, one when standing and one when crouching, just changing the state from 'standing' to 'crouching' should change all the animations automatically and your throw button still activates a throw animation, but the correct one.
Let me know if that makes sense, I'll probably have to explain it better if I make an official challenge.

TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 22nd May 2010 08:46 Edited at: 22nd May 2010 09:57
Quote: "my idea of a game engine require no Actuall Scripting at all ,just Buttons,Strings ,Numbers making it fast , secure against bugs and most important Understandable."
Visual Editors can minimize the amount manual scripting required. However, the tradeoff is a feature heavy Editor Suite.

I'm an advocate of scripting and believe it or not, sometimes scripting is easier (obviously, I would say this). A programmable layer is a powerful feature to have in any software/hardware application (ie: Shaders).

The LUA scripting language is very close to BASIC syntax and my goal is to keep DarkLUA very simple and user-friendly in S3GE. If you rule out scripting you will end up with another FPS Creator.

haliop
User Banned
Posted: 22nd May 2010 10:56 Edited at: 22nd May 2010 10:58
im not ruling it out , as some features will have to use it.
what im saying is , wait a better expresion of what i ment is:

UDK -> Kimset Scripting
where you dont actually write in a Scripting form of way , just buttons and connectors, you can change however each variable just in a visual way instead of writing code whether its easy on the eye like basic.

i just want to aim S3GE not only to programmers with or without expriance , but for kids and grownups alike.

what i might do , since its an open-source , is to take S3GE and to recompile or more easy to start it with S3GE.exe "+basic 1" sirth of way, so even less Advanced Programmers or ones with no coding abilites what so ever to be able to show high detailed projects or games if you like in a very simple way, not throwing them right away into coding or scripting , making S3GE not only User-Friendly but User-Friendly for Dummies sirth of way.
making a tutorial ,"Build your Own game in a Week using S3GE" after that if you want to conitue working on it then you read "Advanced Games in S3GE" i can actually see it passing High-End Game Engine not with the tech but with the number of people who would use it if it was that simple, so my actually point is , the simpliest it may look dose not really say it dosent have advanced features in it.
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 22nd May 2010 11:35 Edited at: 22nd May 2010 11:37
Kismet is what pop'd in my head when initially read your post, hehe. When its all said and done the end user will not have to compile. Just create media/scripts and launch S3GE.exe to run their game. S3GE could have its own version of Kismet. Ultimately, it will come down to S3GEd's Tools and someone is going to have to design and code them.

haliop
User Banned
Posted: 22nd May 2010 12:08
ofcourse no doubt there.
i have come across some problems with Aem , that i still dont know how to approach, if Aem results will be high textures then the FPS will fall down dramaticly , it was aimed first for 2D Top View Shooter , there it was awesome i created maps in just a few clicks and large maps , but where coming to a terrain data , if i want 2 textures on one terrain , it means i need to use on of these:
1. 2 identical terrain -> each with one textures and alpha chanel
2. 1 large terrain and several small terrains just for the 2nd texture , where it shows.
3. alot of small - meduim - big terrains , where each texture is an actuall terrain part , the problem is that if the texture just takes 1 space then ->
making a small patch with alpha on top of a terrain section like "1."

the 3rd option will work with the best fps , cause it actually combines all other , the problem is , that it will cost us about 5000 object id's , and i do not know if this is possible , plus it will make a lot of CPU calculation ...
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 22nd May 2010 13:03
Quote: "the 3rd option will work with the best fps , cause it actually combines all other , the problem is , that it will cost us about 5000 object id's , and i do not know if this is possible , plus it will make a lot of CPU calculation ..."


5000 object id's, sounds like you need a particle engine. DGDK supports upto 64000 objects ids.

Pondering over a potential S3GEd Editor Suite:

UI Editor
Cut-scene Director
Terraformer (Terrain Editor)
Interior Decorator
Actor Assembler/Skinner/Animator
Physics Visual Debugger
Particle Designer
Shader Designer
Game Play Ochestrator

haliop
User Banned
Posted: 22nd May 2010 14:14
im not sure a particle Engine is the right choise..
i dont think other engines use Particles for the Terrain Data and Textures , ive heard something about shaders with multiple textures save FPS i need to check it.
haliop
User Banned
Posted: 22nd May 2010 14:40
ok forget everything i said about terrains.
i have just found this :
http://www.gamedev.net/reference/articles/article2238.asp
TechLord
21
Years of Service
User Offline
Joined: 19th Dec 2002
Location: TheGameDevStore.com
Posted: 22nd May 2010 14:49 Edited at: 22nd May 2010 15:00
Quote: "im not sure a particle Engine is the right choise..
i dont think other engines use Particles for the Terrain Data and Textures , ive heard something about shaders with multiple textures save FPS i need to check it. "


Just kidding about using Particles for terrain although that may produce a interesting effect. However, I do believe a particle Engine can power flora/grass effects.

haliop
User Banned
Posted: 22nd May 2010 15:31 Edited at: 22nd May 2010 15:39
1st we totaly need Co - Op world building.

wow hero engine is so good its unbelivable.

about grass particles.. no need actually , well maybe if it will be faster then the normal LOD solution.

if you are near a grass model , it shows the full model.
the more you get away from it Tessalation joins in and make it less and less Polygons.
and if you are really far from it , then you have only one Plain with the Model as a texture.

Login to post a reply

Server time is: 2024-03-29 08:10:55
Your offset time is: 2024-03-29 08:10:55