Menu
▾
▴
nebuladevice-discuss
|
From: Seika v. O. <sei...@gm...> - 2007-03-13 19:40:55
|
Dear lists, attached please find just 2 small patches. The first patch let the makefile generator generate a correct "clean"-target. The second patch was already mentioned before by marco, it adds an include for algorithm to narray.h. I'm not sure why it wasn't not yet within the SVN but I think something was wrong the way he tried to commit it. Please comment. :-) -- -Seika |
|
From: <ma...@fd...> - 2007-03-13 19:52:02
|
Hi all,
> The second patch was already mentioned before by marco, it adds an
> include for
> algorithm to narray.h. I'm not sure why it wasn't not yet within the
> SVN but I think
> something was wrong the way he tried to commit it.
Actually, I didn't commit the patches I sent, because I have no
user/password in SVN.
That's the reason why I sent the patches to the list, but I had no
answers :'( I hope you have more luck :-p
By the way, I did not mention the error you solve in the makefile.py
patch, but I am agree with it, and I just hope someone commit it to the
repository...
Marco
> Please comment. :-)
>
>
> ------------------------------------------------------------------------
>
> Index: code/nebula2/inc/util/narray.h
> ===================================================================
> --- code/nebula2/inc/util/narray.h (Revision 3233)
> +++ code/nebula2/inc/util/narray.h (Arbeitskopie)
> @@ -17,6 +17,8 @@
> */
> #include "kernel/ntypes.h"
>
> +#include <algorithm>
> +
> //------------------------------------------------------------------------------
> template<class TYPE> class nArray
> {
>
>
> ------------------------------------------------------------------------
>
> Index: buildsys3/generators/makefile.py
> ===================================================================
> --- buildsys3/generators/makefile.py (Revision 3233)
> +++ buildsys3/generators/makefile.py (Arbeitskopie)
> @@ -107,7 +107,7 @@
> # spring clean
> makeFile.write("clean: \n")
> makeFile.write("\t$(RM) $(N_TARGETDIR)*\n")
> - makeFile.write("\t$(RM) $(N_INTERDIR)\n")
> + makeFile.write("\t$(RM) $(N_INTERDIR)*\n")
>
> makeFile.write("default: all\n")
>
>
>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
>
>
> ------------------------------------------------------------------------
>
>
> *** NOTE: To reply to the list use "reply to all", ***
> *** to reply direct to the sender use "reply" ***
> _______________________________________________
> Nebuladevice-discuss mailing list
> Neb...@li...
> https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss
|
|
From: Seika v. O. <sei...@gm...> - 2007-03-14 07:32:15
|
How can we solve the SSE issues? I'm not very familiar with that, but
I'm willing to learn :), unfortunately I can't find any
documentation/tutorial for gcc/g++ to use SSE.
Is it worth to implement a custom SSE implementation, doesn't the
compiler optimize such things correctly?
2007/3/13, Marco Antonio G=F3mez Mart=EDn <ma...@fd...>:
> Hi all,
>
> > The second patch was already mentioned before by marco, it adds an
> > include for
> > algorithm to narray.h. I'm not sure why it wasn't not yet within the
> > SVN but I think
> > something was wrong the way he tried to commit it.
>
> Actually, I didn't commit the patches I sent, because I have no
> user/password in SVN.
> That's the reason why I sent the patches to the list, but I had n=
o
> answers :'( I hope you have more luck :-p
>
> By the way, I did not mention the error you solve in the makefile=
.py
> patch, but I am agree with it, and I just hope someone commit it to the
> repository...
>
> Marco
>
> > Please comment. :-)
> >
> >
> > -----------------------------------------------------------------------=
-
> >
> > Index: code/nebula2/inc/util/narray.h
> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> > --- code/nebula2/inc/util/narray.h (Revision 3233)
> > +++ code/nebula2/inc/util/narray.h (Arbeitskopie)
> > @@ -17,6 +17,8 @@
> > */
> > #include "kernel/ntypes.h"
> >
> > +#include <algorithm>
> > +
> > //--------------------------------------------------------------------=
----------
> > template<class TYPE> class nArray
> > {
> >
> >
> > -----------------------------------------------------------------------=
-
> >
> > Index: buildsys3/generators/makefile.py
> > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> > --- buildsys3/generators/makefile.py (Revision 3233)
> > +++ buildsys3/generators/makefile.py (Arbeitskopie)
> > @@ -107,7 +107,7 @@
> > # spring clean
> > makeFile.write("clean: \n")
> > makeFile.write("\t$(RM) $(N_TARGETDIR)*\n")
> > - makeFile.write("\t$(RM) $(N_INTERDIR)\n")
> > + makeFile.write("\t$(RM) $(N_INTERDIR)*\n")
> >
> > makeFile.write("default: all\n")
> >
> >
> >
> > -----------------------------------------------------------------------=
-
> >
> > -----------------------------------------------------------------------=
--
> > Take Surveys. Earn Cash. Influence the Future of IT
> > Join SourceForge.net's Techsay panel and you'll get the chance to share=
your
> > opinions on IT & business topics through brief surveys-and earn cash
> > http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=
=3DDEVDEV
> >
> >
> > -----------------------------------------------------------------------=
-
> >
> >
> > *** NOTE: To reply to the list use "reply to all", ***
> > *** to reply direct to the sender use "reply" ***
> > _______________________________________________
> > Nebuladevice-discuss mailing list
> > Neb...@li...
> > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share y=
our
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=3D=
DEVDEV
>
> *** NOTE: To reply to the list use "reply to all", ***
> *** to reply direct to the sender use "reply" ***
> _______________________________________________
> Nebuladevice-discuss mailing list
> Neb...@li...
> https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss
>
--=20
-Seika
|
|
From: <ma...@fd...> - 2007-03-14 14:09:31
|
Hi,
I am far from an SSE expert :)
I could see that Nebula implementation used some non-standard aspects,
that are only available in VC++.
However, current version of Nebula _is not using SSE_ (I do not know
the reason), even in Windows. You can check it for example in
mathlib/matrix.h:
#ifndef __USE_SSE__
#include "mathlib/_matrix33.h"
#include "mathlib/_matrix44.h"
typedef _matrix33 matrix33;
typedef _matrix44 matrix44;
#else
#include "mathlib/_matrix33_sse.h"
#include "mathlib/_matrix44_sse.h"
typedef _matrix33_sse matrix33;
typedef _matrix44_sse matrix44;
#endif
where __USE_SSE__ symbol is not defined by default.
As gcc had problems reading sse code, the patch I sent used the
__USE_SSE__ symbol, and only when defined the sse code was compiled.
Marco
Seika van Olstroem escribió:
> How can we solve the SSE issues? I'm not very familiar with that, but
> I'm willing to learn :), unfortunately I can't find any
> documentation/tutorial for gcc/g++ to use SSE.
>
> Is it worth to implement a custom SSE implementation, doesn't the
> compiler optimize such things correctly?
>
> 2007/3/13, Marco Antonio Gómez Martín <ma...@fd...>:
>> Hi all,
>>
>>> The second patch was already mentioned before by marco, it adds an
>>> include for
>>> algorithm to narray.h. I'm not sure why it wasn't not yet within the
>>> SVN but I think
>>> something was wrong the way he tried to commit it.
>> Actually, I didn't commit the patches I sent, because I have no
>> user/password in SVN.
>> That's the reason why I sent the patches to the list, but I had no
>> answers :'( I hope you have more luck :-p
>>
>> By the way, I did not mention the error you solve in the makefile.py
>> patch, but I am agree with it, and I just hope someone commit it to the
>> repository...
>>
>> Marco
>>
>>> Please comment. :-)
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> Index: code/nebula2/inc/util/narray.h
>>> ===================================================================
>>> --- code/nebula2/inc/util/narray.h (Revision 3233)
>>> +++ code/nebula2/inc/util/narray.h (Arbeitskopie)
>>> @@ -17,6 +17,8 @@
>>> */
>>> #include "kernel/ntypes.h"
>>>
>>> +#include <algorithm>
>>> +
>>> //------------------------------------------------------------------------------
>>> template<class TYPE> class nArray
>>> {
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> Index: buildsys3/generators/makefile.py
>>> ===================================================================
>>> --- buildsys3/generators/makefile.py (Revision 3233)
>>> +++ buildsys3/generators/makefile.py (Arbeitskopie)
>>> @@ -107,7 +107,7 @@
>>> # spring clean
>>> makeFile.write("clean: \n")
>>> makeFile.write("\t$(RM) $(N_TARGETDIR)*\n")
>>> - makeFile.write("\t$(RM) $(N_INTERDIR)\n")
>>> + makeFile.write("\t$(RM) $(N_INTERDIR)*\n")
>>>
>>> makeFile.write("default: all\n")
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> -------------------------------------------------------------------------
>>> Take Surveys. Earn Cash. Influence the Future of IT
>>> Join SourceForge.net's Techsay panel and you'll get the chance to share your
>>> opinions on IT & business topics through brief surveys-and earn cash
>>> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>>
>>> *** NOTE: To reply to the list use "reply to all", ***
>>> *** to reply direct to the sender use "reply" ***
>>> _______________________________________________
>>> Nebuladevice-discuss mailing list
>>> Neb...@li...
>>> https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss
>> -------------------------------------------------------------------------
>> Take Surveys. Earn Cash. Influence the Future of IT
>> Join SourceForge.net's Techsay panel and you'll get the chance to share your
>> opinions on IT & business topics through brief surveys-and earn cash
>> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
>>
>> *** NOTE: To reply to the list use "reply to all", ***
>> *** to reply direct to the sender use "reply" ***
>> _______________________________________________
>> Nebuladevice-discuss mailing list
>> Neb...@li...
>> https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss
>>
>
>
|
|
From: Vadim M. <va...@cu...> - 2007-03-15 09:55:11
|
Marco Antonio Gómez Martín wrote: > Hi, > > I am far from an SSE expert :) > I could see that Nebula implementation used some non-standard aspects, > that are only available in VC++. > However, current version of Nebula _is not using SSE_ (I do not know > the reason), even in Windows. You can check it for example in > mathlib/matrix.h: > > #ifndef __USE_SSE__ > #include "mathlib/_matrix33.h" > #include "mathlib/_matrix44.h" > typedef _matrix33 matrix33; > typedef _matrix44 matrix44; > #else > #include "mathlib/_matrix33_sse.h" > #include "mathlib/_matrix44_sse.h" > typedef _matrix33_sse matrix33; > typedef _matrix44_sse matrix44; > #endif > > where __USE_SSE__ symbol is not defined by default. > > As gcc had problems reading sse code, the patch I sent used the > __USE_SSE__ symbol, and only when defined the sse code was compiled. > > Marco I believe that's by design, if you want SSE you should define __USE_SSE__ for your project (in your .bld file I think). -+ enlight +- |
|
From: Vadim M. <va...@cu...> - 2007-03-15 10:28:26
|
Marco Antonio Gómez Martín wrote: > Hi all, > >> The second patch was already mentioned before by marco, it adds an >> include for >> algorithm to narray.h. I'm not sure why it wasn't not yet within the >> SVN but I think >> something was wrong the way he tried to commit it. > > Actually, I didn't commit the patches I sent, because I have no > user/password in SVN. > That's the reason why I sent the patches to the list, but I had no > answers :'( I hope you have more luck :-p > > By the way, I did not mention the error you solve in the makefile.py > patch, but I am agree with it, and I just hope someone commit it to the > repository... > > Marco You didn't get a response because there's pretty much no one left to give you one. Bruce, the last of the great maintainers, retired sometime last year, and right now we really only have one active contributor whose main focus is the 3DSMax Toolkit. Basically the lights are on, but nobody is home. I'll briefly come out of retirement to have a look at the patches over the next few days and see if I can commit some of them. Never tried building Nebula in Linux before but at least now I've got an Ubuntu box to attempt it on. -+ enlight +- |
|
From: Kim H. W. <ki...@ho...> - 2007-03-17 11:03:58
|
Vadim Macagon <vadim <at> cubik.org> writes: > > Marco Antonio Gómez Martín wrote: > > Hi all, > > > >> The second patch was already mentioned before by marco, it adds an > >> include for > >> algorithm to narray.h. I'm not sure why it wasn't not yet within the > >> SVN but I think > >> something was wrong the way he tried to commit it. > > > > Actually, I didn't commit the patches I sent, because I have no > > user/password in SVN. > > That's the reason why I sent the patches to the list, but I had no > > answers :'( I hope you have more luck :-p > > > > By the way, I did not mention the error you solve in the makefile.py > > patch, but I am agree with it, and I just hope someone commit it to the > > repository... > > > > Marco > > You didn't get a response because there's pretty much no one left to > give you one. Bruce, the last of the great maintainers, retired sometime > last year, and right now we really only have one active contributor > whose main focus is the 3DSMax Toolkit. Basically the lights are on, but > nobody is home. > > I'll briefly come out of retirement to have a look at the patches over > the next few days and see if I can commit some of them. Never tried > building Nebula in Linux before but at least now I've got an Ubuntu box > to attempt it on. > > -+ enlight +- > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > Hey, As enlight pointed it out the lights are still on. And I list up some random news: - As it can be seen in the SVN mailing list, 3dsmax toolkit is heavily updated and improved at the moment (but which does not mean that the toolkit is not stable :) due to it is currently used for several commercial projects so it is getting better than ever before. - Some commercial projects are mentioned above, theres are several guys who want to join and contribute but most of them(Korean) are poor at english. That might be one of reason that the commnunity is not so active even the engine is used for commercial projects. - John Smith who mainly merged the last code dump from RL also has some neat stuff of his own which can improve N2 such as removing STL from tynyxml module, supporting IME and unicode and so on. But he recently found his own company so it might take some time till he release that. - For new code dump from RL, I dropped an email about that to Bernd Beyreuther, a leader of RL about a month ago but no response yet. - Floh, the original creater of Nebula Device at RL, now have his blog and there's lots of news you guys want to hear including N3! http://flohofwoe.blogspot.com/ - As enlight pointed out, I currently focus on 3dsmax toolkit and overall production pipeline for N2 and Mangalore. For that, I'm doing a project and willing to open it to anyone who want to participate so let me know that if you have interest on it. Best regards, Kim Hyoun Woo |
|
From: Jack P. <ch...@gm...> - 2007-03-17 20:52:26
|
Hi all, >Rui Paulo Sanguinheira Diogo [r....@gm...] wrote: >It just happened that I get this (quite disturbing) message when I was looking at Nebula again. I used Nebula 1 for some spare time stuff in the past - >though no real game project. I actually loved it because it was very consistent and handy That's exactly my situation too. I also started to play with Nebula 1, the= n after some period of time came back to see great step forward in Nebula2 about 2 years ago. And now as I got some spare time, returned to my project= . I did a lot of research and tried a lot of free engines. In my opinion, Nebula is much superior to all of them, and defenitely better than all "home-made" commercial engines. Of course, I am still a newbee at Nebula an= d do not even yet understand entirely how it works, but things like clean design, obviuos desire to make internal code as simple as possible, usability (WHEN YOU FINALLY UNDERSTAND HOW THE THING YOU NEED SHOULD BE DONE), built-in debug tools & console, and good performance speak for themselves. Thus I am very glad to hear these good news from Kim Hyoun Woo!!! I beleiv= e there are more people interested in this engine, just not all of them write in discussion list. Yesterday I found one example - one guy is frequently publishing Nebula2 tutorials in some eastern language ( http://www.game798.com/html/2007-01/2735.htm for example). I'm not sure though that he isn't writing here, but it looks so ;) Anyway, it shows the interest to the engine. I don't think I can now contribute to merging latest code and bug fixes, as I am not _that_ familiar with Nebula itself and C++ too, but I am defenitely interested in this. I want to use Nebula and probably Mangalore for a simple outdoor game/simulation with nice nature. What I think I can d= o - is writing well-explained tutorials documentation as I make progress, and I will try to do that. I really beleive that lack of detailed well-explained documentation about how Nebula works and what is the right way to do things here is the ONLY reason for it's "unpopularity" and not-so-active community, comparing to OGRE for example. --------- And I want to ask people who understand how the things works with community and RadonLabs check-ins: What actually is the problem with merging RadonLabs code? As I see things, code in the code/Mangalore and code/Nebula2 should not be changed by the community. RadonLabs is maintaining this code on a constantly basis,and I think they find and fix critical bugs faster than community. This way code should be just replaced by the new RadonLabs code, not merged, which can be done even every week if required. Of course, this probably will require som= e refactoring on contributed modules, but I think this is a good point to kee= p Nebula up-to-dated. 2007/3/17, Kim Hyoun Woo <ki...@ho...>: > > Vadim Macagon <vadim <at> cubik.org> writes: > > > > > Marco Antonio G=F3mez Mart=EDn wrote: > > > Hi all, > > > > > >> The second patch was already mentioned before by marco, it adds an > > >> include for > > >> algorithm to narray.h. I'm not sure why it wasn't not yet within the > > >> SVN but I think > > >> something was wrong the way he tried to commit it. > > > > > > Actually, I didn't commit the patches I sent, because I have no > > > user/password in SVN. > > > That's the reason why I sent the patches to the list, but I had n= o > > > answers :'( I hope you have more luck :-p > > > > > > By the way, I did not mention the error you solve in the > makefile.py > > > patch, but I am agree with it, and I just hope someone commit it to > the > > > repository... > > > > > > Marco > > > > You didn't get a response because there's pretty much no one left to > > give you one. Bruce, the last of the great maintainers, retired sometim= e > > last year, and right now we really only have one active contributor > > whose main focus is the 3DSMax Toolkit. Basically the lights are on, bu= t > > nobody is home. > > > > I'll briefly come out of retirement to have a look at the patches over > > the next few days and see if I can commit some of them. Never tried > > building Nebula in Linux before but at least now I've got an Ubuntu box > > to attempt it on. > > > > -+ enlight +- > > > > > ------------------------------------------------------------------------- > > Take Surveys. Earn Cash. Influence the Future of IT > > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > > opinions on IT & business topics through brief surveys-and earn cash > > > http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=3D= DEVDEV > > > > > > Hey, > > As enlight pointed it out the lights are still on. > > And I list up some random news: > > - As it can be seen in the SVN mailing list, 3dsmax toolkit is heavily > updated > and improved at the moment (but which does not mean that the toolkit is > not > stable :) due to it is currently used for several commercial projects so > it is > getting better than ever before. > > - Some commercial projects are mentioned above, theres are several guys > who want > to join and contribute but most of them(Korean) are poor at english. That > might > be one of reason that the commnunity is not so active even the engine is > used > for commercial projects. > > - John Smith who mainly merged the last code dump from RL also has some > neat > stuff of his own which can improve N2 such as removing STL from tynyxml > module, > supporting IME and unicode and so on. But he recently found his own > company so > it might take some time till he release that. > > - For new code dump from RL, I dropped an email about that to Bernd > Beyreuther, > a leader of RL about a month ago but no response yet. > > - Floh, the original creater of Nebula Device at RL, now have his blog an= d > there's lots of news you guys want to hear including N3! > http://flohofwoe.blogspot.com/ > > - As enlight pointed out, I currently focus on 3dsmax toolkit and overall > production pipeline for N2 and Mangalore. For that, I'm doing a project > and > willing to open it to anyone who want to participate so let me know that > if you > have interest on it. > > > Best regards, > > Kim Hyoun Woo > > > > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=3D= DEVDEV > > *** NOTE: To reply to the list use "reply to all", *** > *** to reply direct to the sender use "reply" *** > _______________________________________________ > Nebuladevice-discuss mailing list > Neb...@li... > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss > |
|
From: Vadim M. <va...@cu...> - 2007-03-24 03:42:04
|
Seika van Olstroem wrote: > The first patch let the makefile generator generate a correct > "clean"-target. Just committed it, thanks :) -+ enlight +- |
Thanks for helping keep SourceForge clean.
X