• Welcome to OS2World OLD-STATIC-BACKUP Forum.
 

News:

This is an old OS2World backup forum for reference only. IT IS READ ONLY!!!

If you need help with OS/2 - eComStation visit http://www.os2world.com/forum

Main Menu

OpenGL ddk

Started by tj81, 2009.11.07, 18:17:06

Previous topic - Next topic

Which is more important to you for OpenGL?

High precision rendering
4 (19%)
Sacrificing precision for render speed
6 (28.6%)
A balance between the two (IF possible)
11 (52.4%)

Total Members Voted: 0

cytan

One suggestion is to use the svn on netlabs for your work. Maybe others might be interested to help you out.

demetrioussharpe

Quote from: cytan on 2010.10.24, 21:12:08
One suggestion is to use the svn on netlabs for your work. Maybe others might be interested to help you out.

Sounds reasonable enough. After I have something worth putting up, I'll look into that. If that's a bust, I'll probably end up at sourceforge.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

demetrioussharpe

Funny thing happened while I was looking through the OpenGL spec trying to separate commands into groups of interfaces between the components of the new GL/2. I realized that some of the commands that'll be linked into pgl actually go to the raster, which means that pgl will have to talk directly to both the pipeline & the raster. I guess the deeper you go, the more you learn about the flaws in your initial line of thinking. Lesson learned.

Also, I decided to go ahead & use version 4.0 of the OpenGL spec. If this goes correctly (which we all hope it does), I would imagine that it would be used for a long time & what better way to help future-proof it than to use the newest spec available. Besides, by the time this is fully functional, I expect more platforms to be using OGL 4.0.

It looks like version 2.0 is about the time that GLSL was entering the picture & becoming part of the standard, so it looks like I have no choice but to try to find a way to fit the shading language into the mix. Hopefully, when the time comes, there will be someone in the OS/2 community who has some ideas of how to create the shading units & help integrate them into this codebase.

In an earlier post, someone refurred to OpenGL on OS/2 as GL/2. For a lack of a better name, I think I'll just stick with that name.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

cyber

There is no better name than GL/2.  :D

djcaetano

Quote from: demetrioussharpe on 2010.10.25, 04:36:38
In an earlier post, someone refurred to OpenGL on OS/2 as GL/2. For a lack of a better name, I think I'll just stick with that name.

  Great name, indeed. :)
  Man, I believe most members on OS/2 community are now crossing their fingers, hoping for any development on this arena. There were many developments on important areas such as ACPI, QT, UniAud, Panorama, Samba, Flash and even Java; these are great to keep us alive, up to date... and adding real OpenGL stuff to this set of improvements will be awesome because we've been waiting for this support for almost 15 years!

   My best wishes and my kindest regards!

demetrioussharpe

I've been such a fool. I hope no one will be pissed at me, especially since the actual GL library is loadable by pgl, but I'm going to use TinyGL as my GL implementation. It's been sitting here staring me in the face the whole time. I'm also going to build upon earlier work that I was doing in regards to adding hardware acceleration to it. This is the fastest course of action. It also removes us from all of the *nix cruft code that are in other *nix centric implementation.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

cyber

Quote from: djcaetano on 2010.10.25, 16:19:50
  Man, I believe most members on OS/2 community are now crossing their fingers, hoping for any development

I do not want to be rude, but I do not believe that Demetrious can do that. Anyway, if hi is "the One", will this bring any improvement to existing applications, or we must wait to (no)one develop new apps which will be able to use this... and after all that... there is bunch of other problems/2 that wait for years to be fixed. Being /2 user so many years just lead to hmmmm... distrust ? that there is future for anything relative to /2 and new hardware in same sentence.   :'(

Demetrious, do not think that this text is offensive in any way, I'm really sorry that I can not help in any way, because my knowledge of programing is just at same level as my knowing knitting and crocheting.  :-\

demetrioussharpe

Quote from: cyber on 2010.10.25, 16:52:48
I do not want to be rude, but I do not believe that Demetrious can do that. Anyway, if hi is "the One", will this bring any improvement to existing applications, or we must wait to (no)one develop new apps which will be able to use this... and after all that... there is bunch of other problems/2 that wait for years to be fixed. Being /2 user so many years just lead to hmmmm... distrust ? that there is future for anything relative to /2 and new hardware in same sentence.   :'(

Demetrious, do not think that this text is offensive in any way, I'm really sorry that I can not help in any way, because my knowledge of programing is just at same level as my knowing knitting and crocheting.  :-\

No worries, man. I'll be the first to admit that this is a long shot. However, all through the years, everyone (including me) just kind of sat on our hands while OS/2 was being neglected. Other OS's got advancements & yet, nothing for OS/2. Is this deserving behavior towards an OS that started off as one of the most advanced OSes available to the general public? No, I don't think so. And yet, when these things started happening to other OSes, the users took matters in their own hands & made it their personal responsibility to ensure that their platform wouldn't die without a serious fight. But that wasn't necessarily the case for OS/2. There was the initial begging & demanding for OS/2 to be released as open source. When that failed, everyone just accepted it with a tear in their eyes. Sure, a few people released drivers for this video card or that sound card. But far & few in between were the guys that decided to tackle the system components knowing that no one else would do it for them. And when another company stepped forth, to fill the void, everyone cheered; never pausing to consider what would happen if this new company stopped producing it's OS/2 derivative. As long as we resign ourselves to be at the mercy of come company, we'll always live in fear of losing OS/2. Time is moving forward & it's carrying technology with it. We're going to have to take the responsibility of making sure that OS/2 is able to keep up. But, work like this is always an uphill battle & it can't be fought by just one man. Hopefully, the simple fact that someone is attempting to do something, will inspire others to do the same. If not, then it means that we didn't really deserve OS/2 & it was fate that took it away from us. Now, the bar for success is pretty low right now. To be honest, all I really have to do is prove that it can be done & that, in itself, will be success. But I think that we can both agree that our community deserves more than that. The *nix guys shouldn't be the only ones that are capable of rolling their own 3d graphics stack. Throughout the past couple of decades, many individuals have proved that it's possible to do so. We have the numbers & we have the ability, so lets take control of our futures. So, here's the challenge to everyone: If this thing succeeds, I challenge you to do something to contribute to the cause. It doesn't have to be coding, it could be something as simple as spreading the word. It could be as simple as writing documentation. If you have friends who're developers, convince them to consider developing for OS/2. Whatever it takes, whatever form it takes, I challenge you to do something, anything to contribute. Does that sound fair?

Sorry for the long post, I really needed to get that out.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

ivan

QuoteWhatever it takes, whatever form it takes, I challenge you to do something, anything to contribute. Does that sound fair?

More than fair.

When you manage to get working code I'll write the documentation for it - I've been writing tech docs for years - plus, I have several units here that have a collection of ATI video cards, stand alone and integrated,  if you want test beds.

ivan

demetrioussharpe

Quote from: ivan on 2010.10.25, 22:26:51
More than fair.

When you manage to get working code I'll write the documentation for it - I've been writing tech docs for years - plus, I have several units here that have a collection of ATI video cards, stand alone and integrated,  if you want test beds.

ivan

That sounds wonderful! In the mean time, I'll work on getting the code to at least compile & execute. When that's up & running, I'll be looking for feedback on functionality.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

cyber

This is to far optimistic. How can I trying to convince people to use /2 ?
First, they need to rob museum to got hardware that will working.
Second, they need to raise from average_idiot_home_user level just to install damn' thing.
Third, they need to pay for not just OS it self (even old one W4), they need to pay even for OOo which is on all other platform free program.
Then, there will me more problems, when buy new hdd larger than 500 Mb, when try to use printer wich is not from museum, when try to copy data from ordinary flash stick, to watch videos from Youtube, when try to backup data from smartphone...

Even I am a fanatic OS/2 user, I do not want to slipped other people to. There is alternatives, free of charge, wich isn't beautiful like /2, but work with any hardwer, and work with much less problems and faster.  :'(

demetrioussharpe

#41
Quote from: cyber on 2010.10.26, 09:14:34
This is to far optimistic. How can I trying to convince people to use /2 ?
First, they need to rob museum to got hardware that will working.
Second, they need to raise from average_idiot_home_user level just to install damn' thing.
Third, they need to pay for not just OS it self (even old one W4), they need to pay even for OOo which is on all other platform free program.
Then, there will me more problems, when buy new hdd larger than 500 Mb, when try to use printer wich is not from museum, when try to copy data from ordinary flash stick, to watch videos from Youtube, when try to backup data from smartphone...

Even I am a fanatic OS/2 user, I do not want to slipped other people to. There is alternatives, free of charge, wich isn't beautiful like /2, but work with any hardwer, and work with much less problems and faster.  :'(

You're right, I agree with you 100%. Which is why it's more important than ever for us to take control & reimplement OS/2 on our own. How many other communities must we sit back & watch while they set the example before we catch the hint & learn from their example? Do you think it was easy reimplementing OpenGL, Unix, Windows, BeOS, or DOS? It couldn't have been easy at all & yet we have Mesa3d, Linux, *BSD, ReactOS, Haiku, & FreeDOS. Reimplementing established systems is never easy, but it must be done. If you're not a developer, that's ok. But there's no need to discourage others from heeding the call. Others will take care of providing the technology, you just make sure you're there to use it & spread the word. Now, surely, that's fair.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

djcaetano

Quote from: cyber on 2010.10.26, 09:14:34
Even I am a fanatic OS/2 user, I do not want to slipped other people to. There is alternatives, free of charge, wich isn't beautiful like /2, but work with any hardwer, and work with much less problems and faster.  :'(

  Cyber, I believe every single OS/2 user is concerned about the future and how
to keep things going. Of course there are many obstacles to be beaten, but the
only way to get over them is one by one.
  While basic hardware support is gold, Serenity and other individuals have been
working on it for years... and more advancements are in the pipeline.
   That doesn't mean we do not need support for not-so-essential-technologies.
OGL support seems cosmetic, only for bells and whistles... but it's not. When
saying this will be just one more piece of not used library, keep in mind that SDL
should have support for OpenGL; it has not due to limitations on the available
alternatives at this moment. With a working and up-to-date SDL GL support,
several games would be available, and that would be even better if it is hardware
accelerated.

    If it is difficult to bring new users to eCS-OS/2, it can be even more difficult to
keep them using it when we do not even have the ability to run 3D applications or
even present him with a nice-looking operating system. Most advanced users
(not the lemming type) believe OS/2 is very advanced in the way it works and
in the way we use it, but they also think it looks like outdated and the support for
new technologies is non-existent. Something has to be done in this arena.

    Besides all I said previously, there is a new generation of monitors capable
of 3D which would never be reliably supported without a real 3D rendering stack.

    This is the reason I believe this effort is worth the community support. I do not
think it is a simple project but... even if the project fail, the effort shows at least
that our community will not fall without a fight. :)

demetrioussharpe

djcaetano: well spoken.
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!

demetrioussharpe

Here's an update:

Gotten TinyGL to compile into GLPIPELINE.dll.

Abstracted buffer code so that any renderer implementation will be able to provide it's own buffer support or let GLPIPELINE.dll default to one of the 2 internal buffer systems (either DIVE or system memory).

Implemented glGetError. It was already lacking from TinyGL.

Added function pointer table to GLContext to hold renderer's hooks.

Added void pointer to GLContext to point to renderer's private context data.


Next step:

Use TinyGL's rendering system to create internal version of GLRASTER.

Target internal renderer to use DIVE surfaces for buffers.

Fall back to rendering to buffers backed by system memory if DIVE is unavailable.

Implement code to load GLRASTER.dll, if available. Fall back to internal renderer if GLRASTER.dll is unavailable.


Next, next step:

Begin work on OPENGL.dll (PGL).
The difference between what COULD be achieved & what IS achieved
is directly relational to what you COULD be doing & what you ARE doing!