I am documenting the FSAA message to the [gemwin]. It seems to me that it only takes the values 0,2,4,8. There is a check in the C++ code, but its currently a little bit useless:
void gemwin :: fsaaMess(int value) { if (value == 2 || value == 4 || value == 8){ GemMan :: fsaa = value; } else{ GemMan :: fsaa = value; } }
It seems to me that it should issue a warning if 'value' is not a useful value. Also, [FSAA 0( would be an intuitive way to turn off FSAA, it seems to work.
.hc
----------------------------------------------------------------------------
I have the audacity to believe that peoples everywhere can have three meals a day for their bodies, education and culture for their minds, and dignity, equality and freedom for their spirits. - Martin Luther King, Jr.
hi,
i am not sure, but i think FSAA is only for NVIDIA (maybe you should specify this in the documentation). also some time ago, Cyrille sent an email concerning a problem with FSAA:
------------------------------
hello,
with latest (v177 or 180) nvidia proprietary driver under linux, anti-aliasing did not work.
i don't really understand why, but removing the line 144 : setenv("__GL_FSAA_MODE", svalue, 1); from Base/GemWinCreateXWin.cpp does solve the problem : it allow nvidia-setting to control the application aliasing.
it look like a bug in nvidia driver, but it would be nice if Gem could have a workaround. i propose changing : setenv("__GL_FSAA_MODE", svalue, 1); to : if (fsaa!=0) setenv("__GL_FSAA_MODE", svalue, 1);
that should not harm anyone.
any objection?
Cyrille
------------------------------
still without answer. pat
hello,
after few month without antialiasing, i realized that this as changed on new nvidia driver. FSAA 4 is no more valid. you should use FSAA 5 for the same effect (really can't understand nvidia logic).
So, no check / no warning is the best. C
Hans-Christoph Steiner a écrit :
I am documenting the FSAA message to the [gemwin]. It seems to me that it only takes the values 0,2,4,8. There is a check in the C++ code, but its currently a little bit useless:
void gemwin :: fsaaMess(int value) { if (value == 2 || value == 4 || value == 8){ GemMan :: fsaa = value; } else{ GemMan :: fsaa = value; } }
It seems to me that it should issue a warning if 'value' is not a useful value. Also, [FSAA 0( would be an intuitive way to turn off FSAA, it seems to work.
.hc
I have the audacity to believe that peoples everywhere can have three meals a day for their bodies, education and culture for their minds, and dignity, equality and freedom for their spirits. - Martin Luther King, Jr.
GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
Then what about [FSAA 3(, [FSAA -1(, etc. I think a warning wouldn't hurt, you could add in 5 as a supported value.
Another option is to have a [antialiasing 0/1( message with just gives a default AA level for people who don't want to think about it.
.hc
On Mar 24, 2009, at 3:47 AM, cyrille henry wrote:
hello,
after few month without antialiasing, i realized that this as changed on new nvidia driver. FSAA 4 is no more valid. you should use FSAA 5 for the same effect (really can't understand nvidia logic).
So, no check / no warning is the best. C
Hans-Christoph Steiner a écrit :
I am documenting the FSAA message to the [gemwin]. It seems to me that it only takes the values 0,2,4,8. There is a check in the C++ code, but its currently a little bit useless: void gemwin :: fsaaMess(int value) { if (value == 2 || value == 4 || value == 8){ GemMan :: fsaa = value; } else{ GemMan :: fsaa = value; } } It seems to me that it should issue a warning if 'value' is not a useful value. Also, [FSAA 0( would be an intuitive way to turn off FSAA, it seems to work. .hc ---------------------------------------------------------------------------- I have the audacity to believe that peoples everywhere can have three meals a day for their bodies, education and culture for their minds, and dignity, equality and freedom for their spirits. - Martin Luther King, Jr. _______________________________________________ GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
----------------------------------------------------------------------------
As we enjoy great advantages from inventions of others, we should be glad of an opportunity to serve others by any invention of ours; and this we should do freely and generously. - Benjamin Franklin
Hans-Christoph Steiner wrote:
Then what about [FSAA 3(, [FSAA -1(, etc. I think a warning wouldn't hurt, you could add in 5 as a supported value.
Another option is to have a [antialiasing 0/1( message with just gives a default AA level for people who don't want to think about it.
ok, here is a list of valid values with the old nvidia drivers (96.73)
GeForce, GeForce2, Quadro, and Quadro2 Pro 0 FSAA disabled 1 FSAA disabled 2 FSAA disabled 3 1.5 x 1.5 Supersampling 4 2 x 2 Supersampling 5 FSAA disabled 6 FSAA disabled 7 FSAA disabled
GeForce4 MX, GeForce4 4xx Go, Quadro4 380,550,580 XGL, and Quadro4 NVS 0 FSAA disabled 1 2x Bilinear Multisampling 2 2x Quincunx Multisampling 3 FSAA disabled 4 2 x 2 Supersampling 5 FSAA disabled 6 FSAA disabled 7 FSAA disabled
GeForce3, Quadro DCC, GeForce4 Ti, GeForce4 4200 Go, and Quadro4 700,750,780,900,980 XGL 0 FSAA disabled 1 2x Bilinear Multisampling 2 2x Quincunx Multisampling 3 FSAA disabled 4 4x Bilinear Multisampling 5 4x Gaussian Multisampling 6 2x Bilinear Multisampling by 4x Supersampling 7 FSAA disabled
GeForce FX, GeForce 6xxx, GeForce 7xxx, Quadro FX 0 FSAA disabled 1 2x Bilinear Multisampling 2 2x Quincunx Multisampling 3 FSAA disabled 4 4x Bilinear Multisampling 5 4x Gaussian Multisampling 6 2x Bilinear Multisampling by 4x Supersampling 7 4x Bilinear Multisampling by 4x Supersampling 8 4x Bilinear Multisampling by 2x Supersampling (available on GeForce FX and later GPUs; not available on Quadro GPUs)
so which value is invalid (apart from "-1")?
and as cyrille has pointed out, this has changed with newer driver versions.
people who don't want to think about it, should use a settings-tool that comes with their hardware (e.g. nvidia-settings)
mfg.asdr IOhannes
IOhannes m zmölnig a écrit :
people who don't want to think about it, should use a settings-tool that comes with their hardware (e.g. nvidia-settings)
yep, i made a modification month ago to allow this : if you don't send any FSAA message, then fsaa will not be set and you can then use driver default. (or change it in your nvidia-setting application) c
Hans-Christoph Steiner a écrit :
Then what about [FSAA 3(, [FSAA -1(, etc. I think a warning wouldn't hurt, you could add in 5 as a supported value.
i don't think there is any value that is not suported somewhere.
Another option is to have a [antialiasing 0/1( message with just gives a default AA level for people who don't want to think about it.
arggggghhhhhhhh. this is ugly. there is no default. everything is in balance from performance to quality.
c
.hc
On Mar 24, 2009, at 3:47 AM, cyrille henry wrote:
hello,
after few month without antialiasing, i realized that this as changed on new nvidia driver. FSAA 4 is no more valid. you should use FSAA 5 for the same effect (really can't understand nvidia logic).
So, no check / no warning is the best. C
Hans-Christoph Steiner a écrit :
I am documenting the FSAA message to the [gemwin]. It seems to me that it only takes the values 0,2,4,8. There is a check in the C++ code, but its currently a little bit useless: void gemwin :: fsaaMess(int value) { if (value == 2 || value == 4 || value == 8){ GemMan :: fsaa = value; } else{ GemMan :: fsaa = value; } } It seems to me that it should issue a warning if 'value' is not a useful value. Also, [FSAA 0( would be an intuitive way to turn off FSAA, it seems to work. .hc
I have the audacity to believe that peoples everywhere can have three meals a day for their bodies, education and culture for their minds, and dignity, equality and freedom for their spirits. - Martin Luther King, Jr. _______________________________________________ GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
As we enjoy great advantages from inventions of others, we should be glad of an opportunity to serve others by any invention of ours; and this we should do freely and generously. - Benjamin Franklin
On Mar 24, 2009, at 3:49 PM, cyrille henry wrote:
Hans-Christoph Steiner a écrit :
Then what about [FSAA 3(, [FSAA -1(, etc. I think a warning wouldn't hurt, you could add in 5 as a supported value.
i don't think there is any value that is not suported somewhere.
Even negative values? Is there anyway to detect which are currently supported?
Also, is this the only want to do anti-aliasing?
.hc
Another option is to have a [antialiasing 0/1( message with just gives a default AA level for people who don't want to think about it.
arggggghhhhhhhh. this is ugly. there is no default. everything is in balance from performance to quality.
c
.hc On Mar 24, 2009, at 3:47 AM, cyrille henry wrote:
hello,
after few month without antialiasing, i realized that this as changed on new nvidia driver. FSAA 4 is no more valid. you should use FSAA 5 for the same effect (really can't understand nvidia logic).
So, no check / no warning is the best. C
Hans-Christoph Steiner a écrit :
I am documenting the FSAA message to the [gemwin]. It seems to me that it only takes the values 0,2,4,8. There is a check in the C+
- code, but its currently a little bit useless:
void gemwin :: fsaaMess(int value) { if (value == 2 || value == 4 || value == 8){ GemMan :: fsaa = value; } else{ GemMan :: fsaa = value; } } It seems to me that it should issue a warning if 'value' is not a useful value. Also, [FSAA 0( would be an intuitive way to turn off FSAA, it seems to work. .hc ---------------------------------------------------------------------------- I have the audacity to believe that peoples everywhere can have three meals a day for their bodies, education and culture for their minds, and dignity, equality and freedom for their spirits. - Martin Luther King, Jr. _______________________________________________ GEM-dev mailing list GEM-dev@iem.at http://lists.puredata.info/listinfo/gem-dev
---------------------------------------------------------------------------- As we enjoy great advantages from inventions of others, we should be glad of an opportunity to serve others by any invention of ours; and this we should do freely and generously. - Benjamin Franklin
----------------------------------------------------------------------------
¡El pueblo unido jamás será vencido!
Hans-Christoph Steiner wrote:
I am documenting the FSAA message to the [gemwin]. It seems to me that it only takes the values 0,2,4,8. There is a check in the C++ code, but its currently a little bit useless:
void gemwin :: fsaaMess(int value) { if (value == 2 || value == 4 || value == 8){ GemMan :: fsaa = value; } else{ GemMan :: fsaa = value; } }
of course this is a bit redundant. and of course it also means, that any value is allowed.
It seems to me that it should issue a warning if 'value' is not a useful value. Also, [FSAA 0( would be an intuitive way to turn off FSAA, it seems to work.
"useful" is something depending on the hardware/driver (Gem just passes the parameter to the underlying hardware; at least on linux; and yes; fsaa on linux currently only does something with nVidia)
i cannot think of a way to document this (unlike you want to maintain a database of hardware/driver combinations)
mfgadsr IOhannes
-
cyrille henry -
Hans-Christoph Steiner -
IOhannes m zmölnig -
patrick