PhysX 3.0

Hi, PhysX 3.0 has recently come out. Is it safe to use Nxogre with this or should I stick with 2.8.4?

I hear the API has drastically changed between the two versions.

I would stick with 2.8.4 if I was you.

I hear the API has drastically changed between the two versions.

I would stick with 2.8.4 if I was you.

If I'm not mistaking, you stopped further development of NxOgre, so 2.8.4 will be the last version supported, and the directory and file structure of 3.0 is completely different (not sure about the API, though).

I hear the API has drastically changed between the two versions.

I would stick with 2.8.4 if I was you.

If I'm not mistaking, you stopped further development of NxOgre


Is this true?

Er, yes?

And is there nobody going to continue the development of NxOgre?

And is there nobody going to continue the development of NxOgre?


That's the problem with projects with basically only one contributor. Once he gets tired of it the project goes dead. Spacegaier might be able to take over but last I heard he's busy with other projects like Ogitor (which IMO will be able to replace NxOgre for most basic purposes once physics are added). Other than that nobody with enough expertise and time has stepped forward yet. Still the current version should be able to handle most needs and even though its inactive it still is the most popular physics package for Ogre.

At the time that 2.8.4 was the recent vesion - NxOgre is/was complete.

Apart from bug fixes, there is no reason to continue regular development on it.

At the time that 2.8.4 was the recent vesion - NxOgre is/was complete.

Apart from bug fixes, there is no reason to continue regular development on it.


Nobody can expect you to continue working on it if you don't want to but there will obviously be a lot of new features for 3.0 and future versions that people would love to have integrated in. Not to mention all the ideas you had for 1.8 and 1.9. So in a way I don't think its complete as long as Ogre and PhysX are still evolving. It has all the standard functionality but not comprehensive. But that is a problem for whoever else wants to tackle it.

At the time that 2.8.4 was the recent vesion - NxOgre is/was complete.

Apart from bug fixes, there is no reason to continue regular development on it.


Nobody can expect you to continue working on it if you don't want to but there will obviously be a lot of new features for 3.0 and future versions that people would love to have integrated in. Not to mention all the ideas you had for 1.8 and 1.9. So in a way I don't think its complete as long as Ogre and PhysX are still evolving. It has all the standard functionality but not comprehensive. But that is a problem for whoever else wants to tackle it.

If you put it that way it will probably never be truly complete (like every open-source library under active development).

At the time that 2.8.4 was the recent vesion - NxOgre is/was complete.

Apart from bug fixes, there is no reason to continue regular development on it.


Nobody can expect you to continue working on it if you don't want to but there will obviously be a lot of new features for 3.0 and future versions that people would love to have integrated in. Not to mention all the ideas you had for 1.8 and 1.9. So in a way I don't think its complete as long as Ogre and PhysX are still evolving. It has all the standard functionality but not comprehensive. But that is a problem for whoever else wants to tackle it.

If you put it that way it will probably never be truly complete (like every open-source library under active development).

Well after all the stuff betajaen planned for 1.8+ I don't think anybody can pretend that there aren't ways in which NxOgre could be refined let alone be expanded. Are all the basics in place? Maybe, but speaking from my limited newbie perspective NxOgre isn't mature in the way for example Word is, where they're basically just rearranging the UI and adding features.

Of course he isn't under any obligation to work on a project he created for free and we should be glad NxOgre is here as is. But I just don't want anyone to walk away with the impression that "oh there's nothing left to do for NXOGRE we don't need or want anything aside from bugfixes anymore". Maybe NxOgre should grow, maybe it should evolve into something different than a basic physics connector library but it should evolve.

The plan for Angua (1.8 or 1.9) was to include serialisation, but very well developed.

In theory it would be capable of doing;

- Load/Saving to Disk through a well thoughtout API system. Where the fileformat could be a number of different kinds, or a custom one for your game/application.
- Load/Saving to different Scenes, i.e. Having a secondary scene which ran at a lower priority and minimal collision detection which is used for special effects.
- Load/Saving to streams; i.e. networking.

The plan for 2.0, was to investigate and start wrapping different physics libraries as well as PhysX. Bullet was going to be the first one to be wrapped.


Of course, I'm not going to do any of that now, I really don't have the interest or need to do it. If any of you think you can do it - don't. It will take you a long time to know the ins and outs of NxOgre and the behaviour of PhysX, as well as learn my coding style and finally implement what I want. By the time you start doing it, you'll be sick of it, and end up being half complete or badly written.

NxOgre is finished. Let's leave it at that.

Of course, I'm not going to do any of that now, I really don't have the interest or need to do it. If any of you think you can do it - don't. It will take you a long time to know the ins and outs of NxOgre and the behaviour of PhysX, as well as learn my coding style and finally implement what I want. By the time you start doing it, you'll be sick of it, and end up being half complete or badly written.

NxOgre is finished. Let's leave it at that.


If nothing else someone will eventually have to update NxOgre for compatibility if they want people to still use it due to the fact that people will gradually move on to newer versions of Ogre and PhysX. There are many projects that were successfully forked or continued with little or no involvement from their original authors. Heck, isn't sinbad basically retired from Ogre itself?

Of course you don't have to really worry about me doing an unauthorized offshoot of NxOgre since I'm generally useless at such things. I guess I just don't understand the discouragement of even the theoretical possibility that NxOgre can continue to grow and improve.

I suppose the next nVidia cards will keep support for applications made on 2.8, right? I'm making a game and I have lot of work to do yet, so it wont be available for a year or more, what do you think of this?

I suppose the next nVidia cards will keep support for applications made on 2.8, right? I'm making a game and I have lot of work to do yet, so it wont be available for a year or more, what do you think of this?


I don't see much of a problem on the user side since standard practice is to keep compatibility as long as possible. Even the most ancient physx versions still run on modern computers.