I'm not sure if the NIF format is preventing modders from making good looking models. AFAIK, it still has vertex weighting and normal maps and such, so as long as there isn't a vertex limit it should be fine. But it means it'll be able to use well documented mesh formats, instead of needing to use NifSkope to convert to a reverse-engineered format (such as Ogre's own native .mesh format). Adding new features to well documented formats should also be easier.


I mean a proper 64-bit executable. Original Morrowind is a 32-bit executable, and even when run on a 64-bit OS, it's limited by the 4GB address space, and can't access 64-bit CPU extensions. With OpenMW, a proper 64-bit executable can be built which can access more than 4GB and get at those other CPU features.


Given people who can make the necessary assets and do the proper coding, yeah, it should be possible. Though the biggest hinderence to modding is often the lack of people able to make high quality assets, so without those high quality assets, it may not come out that good.

Will it still pause when loading exterior cells? that's the most bothering thing with MW imo, i can stand the lack of good animations etc, but the pausing every time a new cell needs to load is really immersion breaking imo.

Wow, amen to that!!!

The Nif format is nice but everything I've seen associated with it still looks polygonal and not perfectly smooth. Do the bodies in http://www.youtube.com/watch?v=wgB5DMxKab4 use nif formats for their meshes you think? They look pretty good, although some parts are still not perfectly rounded but I think all video games still have that to some degree. I think with all the tools we have available to use, with this new Ogre3D engine that we'll have access to, modders will be able to take modding to a whole new level. Do you think we'll be able to add in more bones to the bodies? Like the wobble bone used in Oblivion and more?
I like how you said "take 2 steps back, then 300 forward".




You mean TESIII will evetually be able to utilize all that a quad-core or even 8-core machine has to offer? My friend has a 4-core processor and when he runs Morrowind, it's still slow. From what I've heard, Morrowind only uses 1 CPU processor since that's what was available back when the engine was made in the late 90s, I'm assuming. So even if you have 20 processors, 19 of them will be sitting idly.






Kein Schei?! I hate that too. Hopefully, this new 3D Ogre engine will be able to speed up rendering and make it easier on computers to process and load all the exterior cells. Does anybody know if the Ogre3D engine generates all sides of a model? Somebody told me that one of the reasons Gamebryo is a bad engine is because if you have 10 houses lined up in a row, the engine will render all sides & faces of all 10 houses even though you can only see the front of the first one, assuming you're standing in front of it. Game engines nowadays, from what I've heard and read, really only generate what you can see, which is what allows for a boost in their performance. Correct me if I'm wrong. So I'm hoping the Ogre3D engine takes the latter approach.

I don't think that's true for Gambryo. You can easily test it too. Go to a very FPS intensive place where you have maybe 20-30 FPS, then look straight up into the sky (nothing blocking your view, like trees or buildings). The FPS goes from 20 to 200.. So the engine immediately just renders the sky and nothing else even though there's a lot of meshes surrounding you, just not in your view range at the moment.

That's true. Perhaps it just renders everything when you look straight at it.

Probably this, yes. Think of it as if the houses were transparent, or wireframe: the game engine doesn't know that they aren't, so it renders them all, even those that are blocked from view, because it believes they are all within your field of vision. Looking at the sky, there aren't any houses in your FOV, so nothing but sky is rendered.

That makes sense. And when Ogre3D engine is applied to TESIII, will the supermodders (PeachyKeen, TimeSlip, Hrnchamd, Fliggerty, Vality....) be able to tweak the code of the engine so that Morrowind learns how to render only what is in its field-of-view and not every single vertex and polygon like what it does now?

Yes could be that it works like that, would be easy to test too, like sticking 20 or so really hi poly models somewhere you can't see them, but look straight at its placement, and compare FPS results before / after. Might test it later

From my understanding, and please understand that I am not an OpenMW developer, just an enthusiastic follower of the project, this could already be implemented. I'm asking on the official forums: http://openmw.com/forum/viewtopic.php?f=2&t=430

You aren't the only Linux user wanting OpenMW to reach 1.0 status, that's for sure!

I've said before that I'd like to contribute somehow, but having mistakenly thought the project was dead for a while, I backed off. I'm still interested, though.

Drop by our forums. www.openmw.com

Very much so raevol! It's still going, and good man Zini has seemingly volunteered to be the enthusiastic co-ordinator/programming advice go to man. It's not going to reach 1.0 for a while, but it's going at a good clip never the less.

I post there now and again, though it's been a while.