i'm a little confused about where irrlicht is heading...
i remember prior to the 0.7 release there was a lot of talk in these forums an don the irrlichtNX site about how niko was going to start using CVS and that NX would be the official release form then on or something. but then 0.7 came, and then 0.7.1 came, and no word on whether that was official or NX was. also, the NX guys have been quietly working on this NX++ thing for like a month or so now, and have recently made their first release, though it is still rather lightly documented at this point. not a lot of examples written yet for nx++ as far as i could find.
so it comes down to this question...
Where is irrlicht going, and who is in charge now?
some features I would like to see further developed:
CG and HLSL support-
simplify the existing shader support-
more materials or more easily extendable material system-
simple and offset(parallax) bump mapping-
confused....
confused....
My irrlicht-based projects have gone underground for now, but if you want, check out my webcomic instead! http://brokenboomerang.net
I want to know too. It's very important for me - to know whether Irrlicht and NX++ will be compatible.
There are some additional stuff in NX, like axial billboard, advanced particles properties, dirrectional lights and others. I want to use these things. But NX++ gonna different by me - there are some listener classes /like ogre?/ and if will wrote something for nx++, probably code changing will be needed.
There are some additional stuff in NX, like axial billboard, advanced particles properties, dirrectional lights and others. I want to use these things. But NX++ gonna different by me - there are some listener classes /like ogre?/ and if will wrote something for nx++, probably code changing will be needed.
Whether for good or ill NX++ is not compatible with niko's Irrlicht. At this point it is mostly the high-levelstructure that has been changed. From what little I've seen of Ogre code, I think it is a bit more Ogre like. Unfortunately I am afraid that this split may cause some problems. I personally have not yet tried NX++ because my comp has a fried CPU at the moment and because(last i know) it is still not working with Dev-cpp.
The docs definitely need some more explanations of the new classes in them, and I hope that won't be neglected as Irrlicht's good docs have always been one of its strong points. For the most part I like the new structure, though there are a couple things in it that strike me as odd, perhaps because the docs don't explain them well. Supposedly this first release of NX++ contains most of the significant changes to the Irrlicht interface and while future versions will change things we will hopefully not have to do much recoding. I believe planned revisions include a single flexible vertex type and a unified mesh system with only one or two types (just as many file formats will still be able to be loaded into the engine). IMHO both of these are sorely needed.
For a while after NX came out was reluctant to use it because I disliked the split. In the end, it had too much extra functionality and bug fixes to pass up. NX++ is still more of a split as it increases code incompatibilty, but on the other hand, most of us using NX were already writing incompatible ocde as we were taking advantage of functions and features not in niko's Irrlicht.
Probablly the best thing for most of us would be if niko were to adopt NX++ and work together with the NX++ team. Unfortunately he has not shown much inclination to do this, and I can respect his desire to work alone on his engine.
If niko does not join with NX++ I can foresee the two things moving farther and farther apart. Personally I will use NX++ for two reasons
1) I think we can count on the NX team to quickly merge with NX any new releases niko makes
2) NX++ will be developing much faster than niko's releases. Niko is a very good coder (obviously) but he is one person.
As for who's in charge, I think there are two branches of Irrlicht now. One with niko in charge and the other with the NX/NX++ team in charge.
I believe CG for OpenGL is already in NX++.
The docs definitely need some more explanations of the new classes in them, and I hope that won't be neglected as Irrlicht's good docs have always been one of its strong points. For the most part I like the new structure, though there are a couple things in it that strike me as odd, perhaps because the docs don't explain them well. Supposedly this first release of NX++ contains most of the significant changes to the Irrlicht interface and while future versions will change things we will hopefully not have to do much recoding. I believe planned revisions include a single flexible vertex type and a unified mesh system with only one or two types (just as many file formats will still be able to be loaded into the engine). IMHO both of these are sorely needed.
For a while after NX came out was reluctant to use it because I disliked the split. In the end, it had too much extra functionality and bug fixes to pass up. NX++ is still more of a split as it increases code incompatibilty, but on the other hand, most of us using NX were already writing incompatible ocde as we were taking advantage of functions and features not in niko's Irrlicht.
Probablly the best thing for most of us would be if niko were to adopt NX++ and work together with the NX++ team. Unfortunately he has not shown much inclination to do this, and I can respect his desire to work alone on his engine.
If niko does not join with NX++ I can foresee the two things moving farther and farther apart. Personally I will use NX++ for two reasons
1) I think we can count on the NX team to quickly merge with NX any new releases niko makes
2) NX++ will be developing much faster than niko's releases. Niko is a very good coder (obviously) but he is one person.
As for who's in charge, I think there are two branches of Irrlicht now. One with niko in charge and the other with the NX/NX++ team in charge.
I believe CG for OpenGL is already in NX++.
You do a lot of programming? Really? I try to get some in, but the debugging keeps me pretty busy.
Crucible of Stars
Crucible of Stars
correctElectron wrote:Whether for good or ill NX++ is not compatible with niko's Irrlicht. At this point it is mostly the high-levelstructure that has been changed. From what little I've seen of Ogre code, I think it is a bit more Ogre like. Unfortunately I am afraid that this split may cause some problems. I personally have not yet tried NX++ because my comp has a fried CPU at the moment and because(last i know) it is still not working with Dev-cpp.
The docs definitely need some more explanations of the new classes in them, and I hope that won't be neglected as Irrlicht's good docs have always been one of its strong points. For the most part I like the new structure, though there are a couple things in it that strike me as odd, perhaps because the docs don't explain them well. Supposedly this first release of NX++ contains most of the significant changes to the Irrlicht interface and while future versions will change things we will hopefully not have to do much recoding. I believe planned revisions include a single flexible vertex type and a unified mesh system with only one or two types (just as many file formats will still be able to be loaded into the engine). IMHO both of these are sorely needed.
For a while after NX came out was reluctant to use it because I disliked the split. In the end, it had too much extra functionality and bug fixes to pass up. NX++ is still more of a split as it increases code incompatibilty, but on the other hand, most of us using NX were already writing incompatible ocde as we were taking advantage of functions and features not in niko's Irrlicht.
correctAs for who's in charge, I think there are two branches of Irrlicht now.
not yetI believe CG for OpenGL is already in NX++.
nx++
i have been checking up on NX as ofeten as i get the chance wsince it began, and most recently i have taken a look at nx++. i think i agree that the event system any render pipeline anyway has been made more ogre-like. but i believe they have preserved the best part of irrlicht, which was how easy it is to link in and set up your project. i like the singleton idea, though i am not overly familiar with that technique. hopefully, somebody will re-code the irrlicht examples in NX++ style. that will be a big help 
My irrlicht-based projects have gone underground for now, but if you want, check out my webcomic instead! http://brokenboomerang.net
I've been using NX++ for a few days now and it's great. It's indeed a big improvement. Like the event-system, it's way better (uses listeners). Also the fact that a lot of things work with listeners is nice (Ogre also uses this I think).
In short: it's a big improvement and it should be great seeing a great coder like niko work together with those great guys behind NX++, that should certainly give something ubernice ...
In short: it's a big improvement and it should be great seeing a great coder like niko work together with those great guys behind NX++, that should certainly give something ubernice
..."HLSL support"
way long ago when x was being released for irrlicht, something I read in a Niko's post about adding that with dx9 x format improve over dx8, but as i am not a coder, I never ever understand these things.
[ BTW, my interests in engines is for outputting games for low end machines. trying to get most quality but not too much bleeding edge hardware; at least dynamic-stencil-semitransparent shadows and bones+weights, yep, but not much more...(ie: no need of dot3 or shaders) ...yep, aiming at gf mx...Gimme just that and I'll do wonders....
]
Whatever Niko decides, he's "da man" , hehe
BTW, forks are good. Variety is good. Is richness.
if they both merge somehow, it'd be great. If not, maybe also. As I see advantages in the two possibilities. (stability-good doc/advanced-new features, etc or merging it all to get all benefits )
Anyway, I too think perhaps would be nice in general see efforts joined together....
IE: maybe the irrlicht editor authors could join together to make The Kickass editor , maybe irrlicht and NX may join to make a 'better than both' thing....
maybe is a TOO simplistic view of this freak artist that I am...
way long ago when x was being released for irrlicht, something I read in a Niko's post about adding that with dx9 x format improve over dx8, but as i am not a coder, I never ever understand these things.
[ BTW, my interests in engines is for outputting games for low end machines. trying to get most quality but not too much bleeding edge hardware; at least dynamic-stencil-semitransparent shadows and bones+weights, yep, but not much more...(ie: no need of dot3 or shaders) ...yep, aiming at gf mx...Gimme just that and I'll do wonders....
]Whatever Niko decides, he's "da man" , hehe
BTW, forks are good. Variety is good. Is richness.
if they both merge somehow, it'd be great. If not, maybe also. As I see advantages in the two possibilities. (stability-good doc/advanced-new features, etc or merging it all to get all benefits )
Anyway, I too think perhaps would be nice in general see efforts joined together....
IE: maybe the irrlicht editor authors could join together to make The Kickass editor
, maybe irrlicht and NX may join to make a 'better than both' thing....maybe is a TOO simplistic view of this freak artist that I am...
Finally making games again!
http://www.konekogames.com
http://www.konekogames.com
What will happening if Niko extend Irrlicht event system by different way relation to NX++?
At this time I not use nx++ because of this. Well, maybe Niko will tell us about own plans about it.
@ vermeer, about editors here is also problem - Irrlicht or NX++ tol be supported if code generators are used. Similar problem is with node parameters. In my work I avoid using of high classes, for give possibility for users to works with Irrlicht standart functions.
At this time I not use nx++ because of this. Well, maybe Niko will tell us about own plans about it.
@ vermeer, about editors here is also problem - Irrlicht or NX++ tol be supported if code generators are used. Similar problem is with node parameters. In my work I avoid using of high classes, for give possibility for users to works with Irrlicht standart functions.
About my plans: I currently don't have one. That's because I currently under high load (university, work, ..). And I am really sorry for this, especially at the NX guys. I told them some time ago that after version 7 I'll check in my changes of the engine into their cvs. I didn't until now, simply because I did not make any changes to the engine.
I already took some short looks at the irrlichtnx projects and every time I am really surprised of the high quality and good work the guys are doing. Nx++ is new to me, but as far as I have seen from the API in the last minutes, it seems to be a little bit more incompatible to the original Irrlicht. But it is cool anyway, looks very promising and offers a lot more features.
I already took some short looks at the irrlichtnx projects and every time I am really surprised of the high quality and good work the guys are doing. Nx++ is new to me, but as far as I have seen from the API in the last minutes, it seems to be a little bit more incompatible to the original Irrlicht. But it is cool anyway, looks very promising and offers a lot more features.