[Flightgear-devel] Re: compiling under Linux RedHat 7.2 and Accelerated 3D Summit openGL

2002-04-01 Thread stefan


Thank you for answer. I have submited to XIG this information. Indeed 
under Mesa's gl.h all the above definitons are included.

I have to see what XIG's answer is. Previous they made clear that Mesa 
does not stand 100% OpenGL ARB.


stefan


___
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel



[Flightgear-devel] Re: compiling under Linux RedHat 7.2 and Accelerated 3D Summit openGL

2002-04-01 Thread Andy Ross

Stefan wrote:
 > Thank you for answer. I have submited to XIG this information. Indeed
 > under Mesa's gl.h all the above definitons are included.
 >
 > I have to see what XIG's answer is. Previous they made clear that Mesa
 > does not stand 100% OpenGL ARB.

This is actually not an issue of conformance.  Mesa supports all of
the core OpenGL functionality, but not all of the defined extensions.
That's probably what they meant, and they are correct.

But some applications want to be able to use an extension if it is
available, yet fall back to a standard implementation if it is not.
For these applications, the extension definitions (GL_xxx definitions,
etc...) must be available at compile time, even if the implementation
on the current system does not support them.

In this case, XiG has a missing #define in their header, presumably
for a feature which they don't support.  This is wrong -- they need to
include the symbol definitions, and allow applications to fail over at
runtime via glXQueryExtension/glXGetProcAddress.

All that being said, I'd be curious as you how your experience is with
the XiG drivers.  These provide the only useful Radeon support for
Linux, right now, and are rather reasonably priced.

Andy

-- 
Andrew J. RossNextBus Information Systems
Senior Software Engineer  Emeryville, CA
[EMAIL PROTECTED]  http://www.nextbus.com
"Men go crazy in conflagrations.  They only get better one by one."
  - Sting (misquoted)


___
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel



[Flightgear-devel] Re: compiling under Linux RedHat 7.2 and Accelerated 3D Summit openGL

2002-04-01 Thread stefan


Thanks everybody for answers. In short:

There is no XFRee86, DRI support for Radeon 8500 3D. Radeon 8500 is 
different 
that standard Radeon therefore I think 3D engine has been changed. I hope 
4.3.0 XFRee will have support for that.

Yes gl.h from XIG has no definitions for GL constants found under 
xglUtils.c (some which ended with an error)- Now if this is a not a matter 
of conformance about OpenGL ARB 
whhat exactly are those GL constants ? I tried last night to search under 
OpenGL Reference book but with no success. They are not defined in there. 
But arent't GL constants part of OpenGL standard, like telling in version 1.2.1 what 
constants should 
we find under gl.h ? What GL constants then are added to gl.h in version 
1.3 and so on ?

I have to see what XIG's answer is. 

The drivers works really cool. I did not have any problems at all. the 
performance is really different from a starndard Radeon and the drivers 
are stable. Right now they have out 2.1.4 -

I had previously standard Radeon 64DDRAM, and now Radeon 8500 64 DDR.
I have no experience with nVidia cards so much. Is GeForce4Ti 4600 better 
than Radeon 8500 ? Do you guys have any experience with the new cards 
from ASUS, Creative or LeadTek ?

stefan


___
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel



Re: [Flightgear-devel] Re: compiling under Linux RedHat 7.2 and Accelerated 3D Summit openGL

2002-04-01 Thread Simon Fowler

On Mon, Apr 01, 2002 at 01:21:15PM -0800, Andy Ross wrote:
> All that being said, I'd be curious as you how your experience is with
> the XiG drivers.  These provide the only useful Radeon support for
> Linux, right now, and are rather reasonably priced.
> 
The DRI Radeon drivers aren't useful? They work nicely for me . . . 

Simon

-- 
PGP public key Id 0x144A991C, or ftp://bg77.anu.edu.au/pub/himi/himi.asc
(crappy) Homepage: http://bg77.anu.edu.au
doe #237 (see http://www.lemuria.org/DeCSS) 
My DeCSS mirror: ftp://bg77.anu.edu.au/pub/mirrors/css/ 



msg04773/pgp0.pgp
Description: PGP signature


Re: [Flightgear-devel] Re: compiling under Linux RedHat 7.2 and Accelerated 3D Summit openGL

2002-04-01 Thread Andy Ross

Simon Fowler wrote:
 > Andy Ross wrote:
 > > All that being said, I'd be curious as you how your experience is with
 > > the XiG drivers.  These provide the only useful Radeon support for
 > > Linux, right now, and are rather reasonably priced.
 >
 > The DRI Radeon drivers aren't useful? They work nicely for me . . .

Correct me if I'm wrong, but the last time I checked the DRI driver
lacked support for the (now almost a year old) 8500 cards, had no
hardware geometry acceleration, nor multitexture support.  These might
"work", in the sense of producing correct results, but they are only
semi-complete at best.  Performance-wise, ATIs windows drivers stomp
all over them.  Whether that consitutes "useful" or not depends on
your perspective, I guess.  I should have picked a better term.

The Xi Graphics folks at least claim to support the full feature set
of the cards.  I was curious as to how well it worked.

Andy

-- 
Andrew J. RossNextBus Information Systems
Senior Software Engineer  Emeryville, CA
[EMAIL PROTECTED]  http://www.nextbus.com
"Men go crazy in conflagrations.  They only get better one by one."
  - Sting (misquoted)


___
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel



Re: [Flightgear-devel] Re: compiling under Linux RedHat 7.2 and Accelerated 3D Summit openGL

2002-04-01 Thread Simon Fowler

On Mon, Apr 01, 2002 at 02:43:00PM -0800, Andy Ross wrote:
> Simon Fowler wrote:
> > Andy Ross wrote:
> > > All that being said, I'd be curious as you how your experience is with
> > > the XiG drivers.  These provide the only useful Radeon support for
> > > Linux, right now, and are rather reasonably priced.
> >
> > The DRI Radeon drivers aren't useful? They work nicely for me . . .
> 
> Correct me if I'm wrong, but the last time I checked the DRI driver
> lacked support for the (now almost a year old) 8500 cards, had no
> hardware geometry acceleration, nor multitexture support.  These might
> "work", in the sense of producing correct results, but they are only
> semi-complete at best.  Performance-wise, ATIs windows drivers stomp
> all over them.  Whether that consitutes "useful" or not depends on
> your perspective, I guess.  I should have picked a better term.
> 
The 8500 support is currently non-existant, yes, but tcl support is
available for the original Radeon chips, and I believe multitexture
support is likewise available, at least in the DRI cvs tree (which
is quite easy to install and use - certainly no harder than dealing
with FlightGear/SimGear/plib cvs trees). There are also binary
snapshots of the latest development code available.

Support for the 8500 /is/ planned - the docs are available, and the
reason there's no support yet is because the developers are working
on getting tcl support for the original Radeon working properly.
This includes developing the framework for hardware tcl support, so
that supporting the full feature set of the 8500 (and in fact /any/
chip that supports hardware tcl and similar) should be much easier.


As for a comparison with ATI's windows drivers . . . . . How many
man-hours has gone into them? Over how many years? Compared to about
three people working on the DRI code intermittently over the last
couple of years . . . At least they have full documentation now.

You might as well say that FlightGear isn't useful, because it
doesn't have everything that Fly! does . . .


Yes, the DRI drivers are a work in progress, but at present they
/do/ work, and they certainly work well enough that I'd choose to
use them rather than fork out cash for binary drivers.

> The Xi Graphics folks at least claim to support the full feature set
> of the cards.  I was curious as to how well it worked.
> 
They also ship with a broken gl.h, apparently . . . What that
suggests about their quality control is left as an exercise for the
reader.

Simon

-- 
PGP public key Id 0x144A991C, or ftp://bg77.anu.edu.au/pub/himi/himi.asc
(crappy) Homepage: http://bg77.anu.edu.au
doe #237 (see http://www.lemuria.org/DeCSS) 
My DeCSS mirror: ftp://bg77.anu.edu.au/pub/mirrors/css/ 



msg04777/pgp0.pgp
Description: PGP signature