[Louise Andrew]

Animation replacers are just like textures and meshes, they are made by animators, so to answer your question, nope. It's not really in the scope of a bunch of programmers.

Well.. there are smart motion systems such as Euphoria which brings basically AI movement for animations. http://www.naturalmotion.com/products.htm Of course, that is a pipedream, hehe but it is mostly made of programmers. I'd say in a way, animating is a very high level of programming, only the tools are very easy to work with in comparison to writing code for it all...

hehe.

[Joie Perez]

Using mipmaps to store alpha, normals, parralax is an interesting concept. However I know of no function to make use of it in programming.

You know Skydye, some engines today don't really care; some of them make a huge lookup textures which is like packing all textures into one huge file, and have lookup codes to see what to load from the file and where. many do this. One very early example is quake.

For my engine, for my old project we were going to do this per "cell" in game. We'd have something like 128 or so per cell and would preload it into memory once you get close to another cell. With good memory management it could be really effective.

I'm not aware of how this engine works. So these are just suggestions about textures. I'm dyslexic so I'm sure you'll understand why I stay away from codes.

Another idea is a method that take an area that covers 4 meshes and instead of of using 4 tiled texture it will use one large texture to cover all 4. The point being it's easier to load and saves on fps. plus of course it will not have 4 repeated patterns.

Fallout I'm been told by game designers, Instead of using one 1024x 1024 texture on a single meshes area. They are using 2 512x 512 textures. They say it loads faster. 1024 x 1024 is 4 times the file size of 512 x 512.

I feel that both worlds are possible. great resolution and fast fps.

So it all comes down to what you guys feel is best or easiest. :celebration:

Oh. dds uses mipmap. It's standard all anything engine that can use DDS.

[Charlotte Henderson]

I can be wrong but I think the idea is not to edit Bethesda nif as it may not be legal to include edited meshes in another engine.

I also do not like the idea of making changes in original nif for several reasons and legal aspect is only one of them. However, there is a serious difference between "editing the nif format"(Morrowind format) and "having compatibility with all the Nif formats, like OB and Fallout 3 nifs". I wanted to say that MW nif format has an option (NiStringExtraData) to store node specific information in nifs that can be used by programmers for different purposes. It can be used by modders in custom nifs. Therefore OpenMW developers might consider that.

new animations

so to answer your question, nope. It's not really in the scope of a bunch of programmers.

MW animators are simply obliged to follow strict rules established by programmers of the game engine: fixed animation nif structure, hardcoded (limited) number of bones and their names as well as animation sequences (e.g. "Idle", only 39 in base animation nif, compare with unlimited in Oblivion), all-in-one huge animation files, very poor scripting support of animation control functions (especially for PC/NPC/creatures), simplified AI, etc. In case these animation related issues will be not addresses by the project developers OpenMW would be by definition nearly a copy of original MW+MGE.

I understand that the project is in very early stages and such problems like better (more flexible than in MW) animation and AI system has now less priority.

[Jennifer Rose]

As for a wish list... I don't suppose it'd be possible to have locational damage?

I'm hoping we can use Bullet's awesome ragdoll functionality for this aspect of the game.

Are new animations within the scope of this project? Or would that have to be something done as a fan project?

I hope so though a working engine is first on the cards. A1x2e3l is right, new animations would be up to modders but the framework is up to the coders. I think for sure the original format will be superceded by something better. I personally would like to see more animation possibilities available to modders as MW's are dated now.

[Jessica White]

I hope so though a working engine is first on the cards. A1x2e3l is right, new animations would be up to modders but the framework is up to the coders. I think for sure the original format will be superceded by something better. I personally would like to see more animation possibilities available to modders as MW's are dated now.

Oblivion and Fallout 3's system works well, it supports all sorts of things, plus I'm pretty sure you can sequence and control animations using scripts and stuff, I mean Oblivion has http://planetelderscrolls.gamespy.com/View.php?id=1779&view=OblivionMods.Detail, that there shows it's system is better than Morrowind's.

[Miranda Taylor]

There is no compatibility between nif formats e.g. MW nifs (v. 4.0.0.2) cannot be use in Oblivion (v. 20.0.0.5) or in Fallout 3 (v.20.2.0.7), and Oblivion nifs cannot be used in MW. Moreover, every game has specific additions to standard NetImmerse/Gamebryo nif format (e.g. NiBSAnimationNode ? Bethesda specific node).
However, not all MW nif format (v. 4.0.0.2) features were realized in the game e.g. bump maps ? original MW nifs do not have them, but bump maps added in MW nifs (v. 4.0.0.2) by modders have certain interesting effect in the game.
I speculate that changing of PC/NPC skeleton structure to more flexible one e.g. like in MW creatures (no fixed/hardcoded number and names of bones) would require a global modification of original animation files. Moreover, implementation of Havok or PhysX features might also oblige some "additions" to these files.
I think that it is reasonable to stay with MW nif format (v. 4.0.0.2), but it's features should be used so to say more intensively. For instance, gloss maps are not rendered in MW for unknown reason, but they are supported in other v. 4.0.0.2 games, e.g. "Freedom Force".

[George PUluse]

...................

I think that it is reasonable to stay with MW nif format (v. 4.0.0.2), but it's features should be used so to say more intensively. For instance, gloss maps are not rendered in MW for unknown reason, but they are supported in other v. 4.0.0.2 games, e.g. "Freedom Force".

Wouldn't MW have needed a 2.0 or above shader to render spec ?

[butterfly]

Are new animations within the scope of this project? Or would that have to be something done as a fan project?

In a sense, yes. The ability to create and use new animations are in the scope in the project. This functionality is not within the scope of an initial release of the project, as an initial release is supposedly aimed at recreating vanilla Morrowind exactly. However, this functionality may come in the initial release anyways, depending on it's ease of implementation. The actual animations will have to be done by somebody who can animate. Probably not most programmers' area of expertise. And so in the end, the animations themselves will have to be done as a side project.

[Lucky Boy]

Very interesting.

[Jade Payton]

I have another suggestion.

Currently, the Morrowind Engine classifies Light, Medium, and Heavy armor based on several game settings. The GMSTs consist of a numeric value associated with the various armor parts (gauntlet, shields, greaves, boots, cuirass, pauldron). This numeric value represents a weight. An armor part - say, a pauldron - weighing up to 60% of the ipauldronweight value is classified as Light armor. Weights between 61-90% are Medium, and anything over 91% is Heavy.

The request is that OpenMW allow modders to change the % range for armor classification. I, for one, find it very difficult to have a large range of weights for Medium (like I can do for Light and Heavy armors) because of the 61-90% (for a range of about 30% only) restriction.

Secondly, it'd be helpful if the Shield GMST was broken into Shield and Towershield, since TShields necessarily should be heavier. Currently, a problem I run into was that, for instance, a Medium armor TShield would be classified as Heavy armor if I made its weight too much different from the standard size medium-armor shield.

[naana]

Bump, any news?

[Stefanny Cardona]

There seems to be a new landscape video, but I can only guess what big amount of work these guys are busy with.

http://www.youtube.com/watch?v=2wwMZeL74wA

When this will be finished, Oblivion, Fallout 3 and a new Elder Scrolls Game can go home. Alone the modding possibilities in this well thought out world...

[Czar Kahchi]

vardenfell is mirror-inverted;)

[Shelby McDonald]

that video is really old, and that problem was already solved

probably the developer is on holiday

[FLYBOYLEAK]

B limey is this an
U nsubtle attempt to
M ake people aware of this very
P romising mod?

Just a note to say thank you all for your hard work so far. hope all is going well, and am looking forward to seeing the program take shape

[Ana]

Just so everyone knows, the project head, Nicolay (or tbrick as he is known in our illustrious forum), began a much needed vacation 7 days ago. So, unless Yacoby or others have minor updates to the terrain module or whatnot, there probably won't be an "official" update, at least until Nicolay gets back. But I hear that version 0.7 is not so far off, so sit tight folks.

[Dark Mogul]

You can whip up some code in c++ and Nicolay can turn it into D when he gets back.

About the http://openmw.sourceforge.net/wiki/index.php/Contribute page, There are lots of graphical effects and other tidbits on that page that you could quickly code up as a stand-alone C++ program. Then send me the code and I'll integrate it into the engine.

[Sarah Edmunds]

This contribute page... It is a list of things that need contributed?

[Lilit Ager]

This contribute page... It is a list of things that need contributed?

Clicked on it, I linked it.

[Jeneene Hunte]

Just checked it myself (should have before posting, whoops). Anyways... Shaders. I would be happy to work on those (even convert some MW shader to OpenMW), but there's no enough info there to tell how they should be done. I'll download the latest code and look at the render system to see, but is there any more info up somewhere, and is post-processing in the render chain yet?

Edit: It's running OGRE, so nevermind on most of that. I'll take a long look at how OGRE uses shaders and post-processing, then try to whip up bloom, at the very least.

[Milagros Osorio]

Just checked it myself (should have before posting, whoops). Anyways... Shaders. I would be happy to work on those (even convert some MW shader to OpenMW), but there's no enough info there to tell how they should be done. I'll download the latest code and look at the render system to see, but is there any more info up somewhere, and is post-processing in the render chain yet?

Edit: It's running OGRE, so nevermind on most of that. I'll take a long look at how OGRE uses shaders and post-processing, then try to whip up bloom, at the very least.

Yay someone else to help!

[Alexandra walker]

Just checked it myself (should have before posting, whoops). Anyways... Shaders. I would be happy to work on those (even convert some MW shader to OpenMW), but there's no enough info there to tell how they should be done. I'll download the latest code and look at the render system to see, but is there any more info up somewhere, and is post-processing in the render chain yet?

Edit: It's running OGRE, so nevermind on most of that. I'll take a long look at how OGRE uses shaders and post-processing, then try to whip up bloom, at the very least.

Motion blur? That'd be nice .

[HARDHEAD]

Yay someone else to help!

Mhm. I'll have to look how OpenMW itself handles rendering (which means figuring out that nasty D ) but looking over OGRE's docs, I think converting some shaders will be easy. It looks like a paired compositor script and material script (the latter holds the shader info itself). It's a C variant (isn't everything?), so it'll take a few minutes to get the hang of what goes where. I'll undoubtedly crash my copy a few times, then get something halfway decent.

Motion blur? That'd be nice .

Motion blur is insanely hard to do, at least properly. You have to copy and expand and blend and blur each mesh based on velocity. It's possible to do a simple blur, but for real motion blur, it's not a post-processing effects, it's a full render chain deal.
From what it looks like (from a glance at OGRE) bloom will be easy, HDR next, and possibly depth of field after that. Given that OGRE hands out depth buffers in a logical manner, I have almost language-independent DoF shader code laying around.

[Austin Suggs]

Do you think it's possible physics can be implemented?

[The Time Car]

Do you think it's possible physics can be implemented?

Physics can be implemented, and I think is planned. OGRE is compatible with Havok and PhysX, and has wrappers for ODE, Newton, and a few others.