Return to Castle Wolfenstein and Enemy Territory

@samo79

You are right.The 3D driver must be much better in the future

@samo79

The bullshit he is talking about is probably your first statement about pizza. From the grammar point of view (lack of future tense) it looks like the port is already available... (still, for that it also lacks past tense) I figured out what you meant (though, I didn't get the nationality part joke) but reading it straight makes some confusion...

@tfrieden

Quote:

Nothing special... except it does not feel really integrated into the AOS API and I do not feel confident using it for writing handlers. But I didn't dig into it more than that, may be I'm mistaking ?

EDIT: not to say that currently this is not really threads but processes that we would get with our pthreads implementation.

@abalaban

Quote:

I'm using intensively a program who use pthreads intensively and pthreads work verry fine... For informations.

@Mrodfr

I'm not questioning the fact pthreads is working, but rather wondering on a *developer level* how it will interact with the rest of AOS APIs.

@All

A port to MorphOS is already been worked at...according to AmigaWorld.net.

@all

As you might have noticed, I have changed my focus a bit. I used to think that the way to advance the platform is games, because everyone likes to play. And while games and entertainment are an important part of it, I think the productivity and applications are more important.

I recently had a talk with someone, and I asked him "which of these headlines on a news site do you think would convince someone to switch to AmigaOS more?"

- Enemy Territory ported to AmigaOS
- Firefox ported to AmigaOS
- OpenOffice ported to AmigaOS
- Java JDK running on AmigaOS

At least for me, the answer is clear. Well not clear, but the above has three headlines that I would deem more important.

While I do consider games, and games infrastructure in the OS, quite an important part, I think that more consideration is needed for application development and/or porting. Timberwolf/Firefox was one of the things I wanted to do. OpenOffice and Java are certainly candidates for ports as well.

Anyway, my humble opinion. Feel free to disagree.

@abalaban

Quote:

I am not quite sure what you mean. The pthreads API is obviously not a native AmigaOS threads API, because we don't have one (yet). That is bound to change, and if you use pthreads now the pthreads library can be easily updated to work with a native AmigaOS thread API. If you roll your own, you'll have to do the work by yourself.

Right now it uses tasks, but you can rest assured it does things the way it is supposed to be done, after all, it's by the Kernel Guy (tm).

And we're using it in Timberwolf; Timberwolf has shown a few issues that have subsequently been corrected.

Just some points to consider.

@Rogue

No matter how much we want these games, you are of course right, Rogue. User applications and anything productivity are the most important things. To give
a platform a more serious focus than just games...

But of course, such games will breath new life in the Amiga platform, but first we need the missing tools that will bring these games to a new dimension on the Amiga platform..

@Rogue

I agree ,what else :)
On the other Side: I love Games ...
Maybe when Timberwolf finished, you can port some nice Game,
and yes i would like to buy GORKY17 in the Future, maybe :)

again , sorry bad english, i understand all but writing or speaking is another
Thing

@Rogue
You are 100% right here. Games it games, apps it apps. Let's other users/programmers worry about porting a games/demos to aos4, while developers who works on AOS itself, spend their time to the OS, and important application.

All in all maybe morphos users, or aros users will release the sources of their ports of that "return to castle and enemy territory". and then it can be easyly ported to aos4 as well. These games are not so "os specific" today (ogl / sdl /etc), so its not so hard to port, while working on the OO, Java, Firefox and all other nasty stuff like OpenGl/Gallium/etc are hard work, and more important.

Not the latest think, that while we not have fast OpenGL realisation, porting of such kind of games, will only show weeknes of the aos4 opengl realisation (like q3 already do), so better spend aos4's developers time to the OS/drivers/ and so on. Games will be done by the other coders soon or later :)

@Rogue

Quote:

Yes all this pthreads thing started because I was giving threads as an example of suffs I'd like to see in AOS APIs and mrodfr answered that pthreads.library already exists.[/quote]

@abalaban

Well the good news is that the scheduler will need a complete overhaul anyway once we move to dual core support, which should be directly after we released AmigaOS for the X1000. Threads will be on the list of things to do as well; it might not be in at once, but we're certainly going to keep things open enough.

The current scheduler is a 1-to-1 "map" of the original exec scheduler, and still mostly in assembler. As such, it's pretty fragile and difficult to maintain.

I think it must be one of the oldest parts of the system.

@netrot

Don't get me wrong, I love games and I think they are an important part of the platform. However, in absence of sufficient development resources (complicated way to say "coders"), you need to focus on those things that advance the platform.

It's what we have been doing in the last several months by providing better OS support for certain features in order to make application porting easier.

I know a lot of people frown upon ports (the same people that cheer when someone ports Quake 3 ;)) but right now, there is no way around porting software. I don't think anybody would want to start their own word processor from scratch.

@Rogue

well there is one : Claus Desler, the man behind cinnamon writer http://desler.be/

@Rogue

What you tell about the plans for X1000-support right after the initial release of X1000 with AmigaOS 4.1.2, are very interesting.

Can i ask if there will be a 64-bit support as well as the Dual-core support? I understand it that the mystical CPU
is a true 64-bit POWER CPU. The AmigaOS will benefit a lot
from the full 64-bit support, as well as being 32-bit compatible with games and software..

@Helgis

From my perspective I would imagine the OS team will be focusing more on multi-core support first then worry about 64-bit system wide support.

If you think about it its less than 1yr since Microshite and Apple released Win7 and Snow Leopard back in October of last year.

Uptake on the Mac in particular by developers to produce 64-bit versions of their software titles has been quite slow and for most intents and purposes 32-bit software is perfectly fine. Not a clue about Windows I could only surmise that uptake has been a little quicker.

Its more important to have faster running processors rather than software titles running in 64-bit mode that are more or less identical to their 32-bit counterparts.

Multi-core support must arrive on the X1000 from the onset (or very soon afterwards) whereas 64-bit system wide support won't be missed anytime soon.

@djrikki

Memory protection is more important i believe.
Average joe won't know if the os/programs are using both cores. He also has no clue about 64bit sw/cpu....
But i bet he notice when the crashed sw takes down the whole system.

@djrikki

You are right. The Dual-core will do greatly even in 32-bit mode, and should give the OS4 team plenty of time to work on great 64-bit support when that time comes..

I'm so excited!

@Antique

You both are right. Although, i believe 64-bit support in AmigaOS would be far more successful than seen in Windows, it might not be much necessary at the moment.

Still, i love the thought of both full 64-bit support and multi-core support..