Re:Finally, ABI stabilization. Now about optimizat
on
GCC 3.2 Released
·
· Score: 2, Informative
It makes it much harder to certify this though when the same behavior isn't present under gcc2.95x or other compilers, such as MSVC, MingW, or the *BSD compilers, etc.
Even if the aliasing case is indeed a bug in the code somewhere, the fact that -march=i686 works perfectly, while -march=athlon can cause X to segfault, the program segfault, or the program not show all the graphics leads me to believe it still needs help.
Don't get me wrong, I was happy to see the new Athlon option, but it doesn't do me much good when very minimalistic code gets generated incorrectly...
Re:Finally, ABI stabilization. Now about optimizat
on
GCC 3.2 Released
·
· Score: 1
I might believe it was the code if every version of gcc didn't change the behavior drastically, plus the fact this code is compiled under MingW, MSVC, etc. so it's much harder for me to believe that...
Some redhat versions of gcc 3.0x for example didn't have this issue, but every normal release from gcc3 people directly I've seen this issue with. So, *shrug*. I dunno.
Compilers aren't my forte unfortunately.
Re:Finally, ABI stabilization. Now about optimizat
on
GCC 3.2 Released
·
· Score: 1
Yes, but this switch wasn't necessary with 2.95, but is with gcc3. Technically it could be a problem in the code, but someone I know that is extremely familiar with assembly langauge looked over the output and compared the ASM output of gcc3 to gcc2's output and discovered what he considered to be errors in gcc3's output. Plus I know for certain that it generates int->float float->int "wrong" in certain circumstances, well maybe not wrong, but certainly less desireably than gcc2's output.
Finally, ABI stabilization. Now about optimization
on
GCC 3.2 Released
·
· Score: 5, Informative
I've been waiting for this. Building glibc in the past w/ gcc3 was PAINFUL beyond measure. There are still many optimization options that I have to use with programs otherwise gcc3's optimizers optimize away too much.
For example, if I compile the modified Quake engine project I work on without -fno-strict-aliasing bizarre graphical errors occur. (Or used, need to check 3.2 now:)
Or if I compile with -march=athlon I get fairly mixed results, code that sometimes segfaults for no apparent reason, etc.
I don't think I agree. The ole' Video toaster system that I used to use had color sprinkled everywhere, yet it is considered by many to be a fairly professional video system.
Now admittedly, I haven't seen what the Grass Valley system looks like, but...
The $100,000 Mac setup that someone I knew had in their front room with Avid desktop studio, etc. had plenty of splashes of color too.
Again, you may not like it, but some people it just doesn't bother. As they mentioned, the app is skinnable, so if you don't like it, change it, it's not like you don't have the source *cough*...
I won't claim to be a video expert, and I can certainly understand why the colors might bother you. But, *shrug*. They don't really bother me, even photoshop is fairly colorful these days...
Actually there WILL be Linux binaries *soon*. The game developers have promised us this much, and I believe the executive producer has pushed for this.
I hope blender's future is bright, but I think people should notice this nasty little surprise:
http://www.blender3d.com/deal.html
Sure, it'll be "GPL", but with a nasty little restriction apparently: To quote one of their pages:
"A. Membership services
After subscribing (gratis), you get access to the closed Membership area, which includes all user/artists services, all executable versions, all source codes. The license for the executables and codes will be the 'copylefted GNU GPL' license, also known as 'GPL' for short. This allows Members to freely use and redistribute the code,
but restricts building new derivations with Blender codes to other GPLed software projects."
(emphasis added by me)
Whoa! Hold on there. I just thought they said it was GPL. That isn't GPL if I can't use it in any other GPL project! Anyone else notice this?
My apologies, I could have sworn the original press release included 'Mac' as one of the current ones, it looks like it won't ship until this Fall:/ MacSoft has picked up the rights to publish the Mac version.
Indeed! Descent was the first true 6 degrees of freedom game afaik, and Descent 3 had force feeback support so I would have to question the title of "first".
Speaking of which, the source code to Descent I and is freely available at the site where I'm co-evil-overlord:
Actually, not quite. While they're releasing the MacOS and Win32 versions to the shelf.
Linux will see the online release of the Neverwinter Nights server at launch and the client shortly afterwards. However, Linux gamers will still need the Windows version of the game to register at the Neverwinter Nights community site and to import game resources into their Linux server and game. (The editor is currently not planned for Linux)
I stand corrected on the support thing, as of recent they have changed to the following info for 'required' and 'recommended':
What Linux distribution can I use WineX on?
WineX supports a wide range of systems, including RPM and Debian based systems. Currently it has been tested extensively on Mandrake Linux 8.1 Gaming Edition, and Debian Unstable.
Core library requirements are: o Linux Kernel 2.2 or higher o XFree86 4.0 or higher (4.10 recommended) o glibc 2.1.3 or higher o A working hardware accellerated OpenGL implementation
What hardware should I use with WineX?
Recommended minimum hardware is o 500 MHz or faster Pentium or Athlon CPU o 128 MB memory o NVidia GeForce graphics card o 1-2 GB free hard disk space for games
Note: o There are known performance and other problems with DRI based drivers for video cards such as the ATI Rage 128 and the 3DFX Voodoo series. TransGaming is investigating these problems, but it may be several weeks before performance with these cards is adequate. o Games that do not require 3D hardware acceleration, including The Sims, should run well with most video cards supported by XFree86 4.0 and higher.
However, I should note that Matrox cards aren't even fully DX 8.1 compliant under Windows, how could we expect them to be under Linux?;)
At last check they don't officially support DRI cards yet. You have a DRI driver, and sorry but to be quite honest, the state of the DRI drivers is pretty pathetic. Only NVidia cards as last check were officially supported. To me it's obvious why, NVidia is the only vendor that I know of that has released drivers for their card under Linux that has OpenGL 1.3 support, and is the fastest performing that i'm aware of.
I understand that you're upset, but you know what, this game didn't even run under windows for all of the people who bought it, so obviously things are going to be even narrower under Linux.
I agree that their support solutions need some work, but phone support is hardly something that any company could afford with measly $5 a month subscriptions. They have hinted that in the near future when the infrastructure is in place that they will have expanded support options which will understandably cost more.
Every question I have posed through the transgaming forum which if you flag as a support message goes directly to them, is esentially the same as direct email support, except that other users can benefit from the things that they tell you which helps keep their support times down. That way people can find the same answers that others were given before asking for support again (ideally).
Re:Nothing "Wrong" what Transgaming are doing here
on
Debian And WineX
·
· Score: 1
The funniest thing about all of this is that no-one is pointing back at the Lindows people. Where are the contributions from them, or their promises to give anything back at all?
Transgaming at least did these things! Anyone else realize this, or am I the only one?
No, of course not it's all evil Transgaming. Why? Because people a) don't ralize that Lindows is undoubtedly based off the old wine tree in some way b) Transgaming is in the spotlight while Lindows isn't...
"If they aren't going to contribute back to the base that their miniscule contributions are based upon, they don't deserve any of our financial support."
A lot of users do not consider their contributions to be miniscule. I for one am certain that they have contributed far more than your gross underestimation.
Re:Transgaming isn't bad.
on
Debian And WineX
·
· Score: 2, Insightful
"First off they don't run perfectly. There are tons of problems. They are not 100% directx 8.x compliant. And they are not acting like an open source company: they deserve no slack from the community since they get paid for their work. Now they need to support those products, period."
There are only a few products they officially support. And Diablo II does run 'perfectly' for me, just as well as it does under Windoze.
"Wine has been running games like starcraft for ages. It just takes a lot of effort to configure everything properly. Eventually the wine project will break through the proprietary SafeDisc stuff. You don't need a license to reverse engineer."
But with horrible 2d performance and unreliable battle.net, which they have FIXED. And actually, the DMCA and other laws may prevent SafeDisc stuff from being implemented in the main wine tree, it's one of the reasons I think (beyond the proprietary aspect) that they can't put it in the public AFPL'd source.
"I would guess that, too, since Loki released about 20 games, KDE includes a nice collection, as does GNOME. And wine supported a few. Winex gave me the ability to run Homeworld. And for that I feel my $60 contribution was worth it. But I still can't get diablo II to work and I was under the impression my money would go to fund sourcecode that eventually would fall under a GNU license. Was I wrong? Did I throw my money away? Yes. But I won't do it again."
It sounds like you have a personal vendetta. There are *MANY* people who are running Diablo II perfectly. You have to remember that they can't control all the factors, a lot of video cards under Linux have some pretty crappy drivers. Also, they have stated that their code would go into the wine tree when they got enough subscribers. I don't forsee that changing, especially since they haven't stated so.
In this particular case, it's not a matter of profit. It's a matter of existence, if you take their last stated number of subscribes (~3,000) x $5 US or so a month, that's hardly enough month to support several full time developers, and pay for rent on a space, etc.
It's sad that people don't understand this. They just want everything free. Somehow, someway Transgaming has to be able to recoup the costs of development.
By subscribing you not only get to vote on the direction of development, but you also ensure that they still exist tomorrow. And I got news for people, if they go out of business, all of their AFPL'd changes stay with them. The only changes they made that the public will get are all the 2d/dib/audio changes they contributed to the main wine tree before it went LGPL. (It was still X11).
Sadly, I wonder if they'll ever hit 20,000 at last check they only hit about 3,000. Personally, I wonder how much longer they can stay around. The saddest thing about this is stupid flamewars like this only end up hurting users.
Re:About time to dump Transgaming
on
Debian And WineX
·
· Score: 1
Hi there, BTW. They do contribute to free software. Or haven't you noticed the code merges that went into the MAIN wine tree in the past from them? Yeesh. Research a little, k?
Thanks for ignoring my statement that: "Really trying not to flame here, but if you're running Linux. The Radeon 8500 does *not* qualify as a "Real HTPC" *yet*."
Note that "if you're running Linux." part. Also, I feel as if I'm being turned into flamebait simply because I don't consider a windows black-box based solution a solution due to it's inflexibility and dependance on the vendor. Ok, fine I'm horrible at business wording everything but I am human and far from perfect. Give a guy a break. I even prefaced what I was saying with a common anchronism for "opinion." It would have been nice to have not bee moderated at all!
I am far from an RMS fanatic, and I despise being branded as one. Those who moderate posts to low levels simply because they don't like the peron's opinion infuriate me. I *have never* done that. It's a shame that the "do unto other as you would have them do unto you" rule doesn't work here.
Well, if you want put it that way, my friend has a Pentium 4 system with a hardware DVD decoder card and a DVD drive, and it often has problems with audio not syncing properly with the video, after some massive hardware tweaking the problem went away. But, it just goes to show that a set top box is often much easier / more reliable.
That's rather Trollish to say, note I said that this was my opinion, and "not working properly" is a completely flamebait phrase. There are solutions out there under Linux that "work properly" but they are far from polished.
Slashdot Mirror
It makes it much harder to certify this though when the same behavior isn't present under gcc2.95x or other compilers, such as MSVC, MingW, or the *BSD compilers, etc.
Even if the aliasing case is indeed a bug in the code somewhere, the fact that -march=i686 works perfectly, while -march=athlon can cause X to segfault, the program segfault, or the program not show all the graphics leads me to believe it still needs help.
Don't get me wrong, I was happy to see the new Athlon option, but it doesn't do me much good when very minimalistic code gets generated incorrectly...
I might believe it was the code if every version of gcc didn't change the behavior drastically, plus the fact this code is compiled under MingW, MSVC, etc. so it's much harder for me to believe that...
Some redhat versions of gcc 3.0x for example didn't have this issue, but every normal release from gcc3 people directly I've seen this issue with. So, *shrug*. I dunno.
Compilers aren't my forte unfortunately.
Yes, but this switch wasn't necessary with 2.95, but is with gcc3. Technically it could be a problem in the code, but someone I know that is extremely familiar with assembly langauge looked over the output and compared the ASM output of gcc3 to gcc2's output and discovered what he considered to be errors in gcc3's output. Plus I know for certain that it generates int->float float->int "wrong" in certain circumstances, well maybe not wrong, but certainly less desireably than gcc2's output.
I've been waiting for this. Building glibc in the past w/ gcc3 was PAINFUL beyond measure. There are still many optimization options that I have to use with programs otherwise gcc3's optimizers optimize away too much.
:)
For example, if I compile the modified Quake engine project I work on without -fno-strict-aliasing bizarre graphical errors occur. (Or used, need to check 3.2 now
Or if I compile with -march=athlon I get fairly mixed results, code that sometimes segfaults for no apparent reason, etc.
Anyway, congrats to the gcc3.2 team...
[opinion]
I don't think I agree. The ole' Video toaster system that I used to use had color sprinkled everywhere, yet it is considered by many to be a fairly professional video system.
Now admittedly, I haven't seen what the Grass Valley system looks like, but...
The $100,000 Mac setup that someone I knew had in their front room with Avid desktop studio, etc. had plenty of splashes of color too.
Again, you may not like it, but some people it just doesn't bother. As they mentioned, the app is skinnable, so if you don't like it, change it, it's not like you don't have the source *cough*...
I won't claim to be a video expert, and I can certainly understand why the colors might bother you. But, *shrug*. They don't really bother me, even photoshop is fairly colorful these days...
[/opinion]
Actually there WILL be Linux binaries *soon*. The game developers have promised us this much, and I believe the executive producer has pushed for this.
I AM A MORON. Seriously. I completely and totally misread that. Mod my comment down before some stupid freak actually believes what I said
http://www.blender3d.com/deal.html
Sure, it'll be "GPL", but with a nasty little restriction apparently:
To quote one of their pages:
(emphasis added by me)
Whoa! Hold on there. I just thought they said it was GPL. That isn't GPL if I can't use it in any other GPL project! Anyone else notice this?
My apologies, I could have sworn the original press release included 'Mac' as one of the current ones, it looks like it won't ship until this Fall :/ MacSoft has picked up the rights to publish the Mac version.
Indeed! Descent was the first true 6 degrees of
freedom game afaik, and Descent 3 had force feeback support so I would have to question the title of "first".
Speaking of which, the source code to Descent I and is freely available at the site where I'm co-evil-overlord:
http://d1x.warpcore.org/
It's sad to hear that you were disappointed with Morrowind. I'm still throughly enjoying, my only complaint about it is that it's too friggin' long
Actually, not quite. While they're releasing the MacOS and Win32 versions to the shelf.
Linux will see the online release of the Neverwinter Nights server at launch and the client shortly afterwards. However, Linux gamers will still need the Windows version of the game to register at the Neverwinter Nights community site and to import game resources into their Linux server and game. (The editor is currently not planned for Linux)
I stand corrected on the support thing, as of recent they have changed to the following info for 'required' and 'recommended':
;)
What Linux distribution can I use WineX on?
WineX supports a wide range of systems, including RPM and Debian based systems. Currently it has been tested extensively on Mandrake Linux 8.1 Gaming Edition, and Debian Unstable.
Core library requirements are:
o Linux Kernel 2.2 or higher
o XFree86 4.0 or higher (4.10 recommended)
o glibc 2.1.3 or higher
o A working hardware accellerated OpenGL implementation
What hardware should I use with WineX?
Recommended minimum hardware is
o 500 MHz or faster Pentium or Athlon CPU
o 128 MB memory
o NVidia GeForce graphics card
o 1-2 GB free hard disk space for games
Note:
o There are known performance and other problems with DRI based drivers for video cards such as the ATI Rage 128 and the 3DFX Voodoo series. TransGaming is investigating these problems, but it may be several weeks before performance with these cards is adequate.
o Games that do not require 3D hardware acceleration, including The Sims, should run well with most video cards supported by XFree86 4.0 and higher.
However, I should note that Matrox cards aren't even fully DX 8.1 compliant under Windows, how could we expect them to be under Linux?
At last check they don't officially support DRI cards yet. You have a DRI driver, and sorry but to be quite honest, the state of the DRI drivers is pretty pathetic. Only NVidia cards as last check were officially supported. To me it's obvious why, NVidia is the only vendor that I know of that has released drivers for their card under Linux that has OpenGL 1.3 support, and is the fastest performing that i'm aware of.
I understand that you're upset, but you know what, this game didn't even run under windows for all of the people who bought it, so obviously things are going to be even narrower under Linux.
I agree that their support solutions need some work, but phone support is hardly something that any company could afford with measly $5 a month subscriptions. They have hinted that in the near future when the infrastructure is in place that they will have expanded support options which will understandably cost more.
Every question I have posed through the transgaming forum which if you flag as a support message goes directly to them, is esentially the same as direct email support, except that other users can benefit from the things that they tell you which helps keep their support times down. That way people can find the same answers that others were given before asking for support again (ideally).
The funniest thing about all of this is that no-one is pointing back at the Lindows people. Where are the contributions from them, or their promises to give anything back at all?
Transgaming at least did these things! Anyone else realize this, or am I the only one?
No, of course not it's all evil Transgaming. Why? Because people a) don't ralize that Lindows is undoubtedly based off the old wine tree in some way b) Transgaming is in the spotlight while Lindows isn't...
Odd how that works...
"If they aren't going to contribute back to the base that their miniscule contributions are based upon, they don't deserve any of our financial support."
A lot of users do not consider their contributions to be miniscule. I for one am certain that they have contributed far more than your gross underestimation.
"First off they don't run perfectly. There are tons of problems. They are not 100% directx 8.x compliant. And they are not acting like an open source company: they deserve no slack from the community since they get paid for their work. Now they need to support those products, period."
There are only a few products they officially support. And Diablo II does run 'perfectly' for me, just as well as it does under Windoze.
"Wine has been running games like starcraft for ages. It just takes a lot of effort to configure everything properly. Eventually the wine project will break through the proprietary SafeDisc stuff. You don't need a license to reverse engineer."
But with horrible 2d performance and unreliable battle.net, which they have FIXED. And actually, the DMCA and other laws may prevent SafeDisc stuff from being implemented in the main wine tree, it's one of the reasons I think (beyond the proprietary aspect) that they can't put it in the public AFPL'd source.
"I would guess that, too, since Loki released about 20 games, KDE includes a nice collection, as does GNOME. And wine supported a few. Winex gave me the ability to run Homeworld. And for that I feel my $60 contribution was worth it. But I still can't get diablo II to work and I was under the impression my money would go to fund sourcecode that eventually would fall under a GNU license. Was I wrong? Did I throw my money away? Yes. But I won't do it again."
It sounds like you have a personal vendetta. There are *MANY* people who are running Diablo II perfectly. You have to remember that they can't control all the factors, a lot of video cards under Linux have some pretty crappy drivers. Also, they have stated that their code would go into the wine tree when they got enough subscribers. I don't forsee that changing, especially since they haven't stated so.
In this particular case, it's not a matter of profit. It's a matter of existence, if you take their last stated number of subscribes (~3,000) x $5 US or so a month, that's hardly enough month to support several full time developers, and pay for rent on a space, etc.
It's sad that people don't understand this. They just want everything free. Somehow, someway Transgaming has to be able to recoup the costs of development.
By subscribing you not only get to vote on the direction of development, but you also ensure that they still exist tomorrow. And I got news for people, if they go out of business, all of their AFPL'd changes stay with them. The only changes they made that the public will get are all the 2d/dib/audio changes they contributed to the main wine tree before it went LGPL. (It was still X11).
Sadly, I wonder if they'll ever hit 20,000 at last check they only hit about 3,000. Personally, I wonder how much longer they can stay around. The saddest thing about this is stupid flamewars like this only end up hurting users.
Hi there, BTW. They do contribute to free software. Or haven't you noticed the code merges that went into the MAIN wine tree in the past from them? Yeesh. Research a little, k?
Thanks for ignoring my statement that: "Really trying not to flame here, but if you're running Linux. The Radeon 8500 does *not* qualify as a "Real HTPC" *yet*."
Note that "if you're running Linux." part. Also, I feel as if I'm being turned into flamebait simply because I don't consider a windows black-box based solution a solution due to it's inflexibility and dependance on the vendor. Ok, fine I'm horrible at business wording everything but I am human and far from perfect. Give a guy a break. I even prefaced what I was saying with a common anchronism for "opinion." It would have been nice to have not bee moderated at all!
I am far from an RMS fanatic, and I despise being branded as one. Those who moderate posts to low levels simply because they don't like the peron's opinion infuriate me. I *have never* done that. It's a shame that the "do unto other as you would have them do unto you" rule doesn't work here.
Well, if you want put it that way, my friend has a Pentium 4 system with a hardware DVD decoder card and a DVD drive, and it often has problems with audio not syncing properly with the video, after some massive hardware tweaking the problem went away. But, it just goes to show that a set top box is often much easier / more reliable.
Oooohhhh....candy bar ;) When I have the time/cash flow I will certainly look into this :)
Thanks!
Tut, tut... Many people's beloved European Union is moving towards DMCA style legislation too, so speak carefully.
"Fully Legal" IMHO, means legal in all countries. Yes, there is such a thing for a lot of software.
I'll take your last comment as pure opinion, which you certainly have the right to voice.
That's rather Trollish to say, note I said that this was my opinion, and "not working properly" is a completely flamebait phrase. There are solutions out there under Linux that "work properly" but they are far from polished.