The flaw is neither company will release their specs to the open source community, but they will release it to anyone who pays
Microsoft doesn't pay ATI or NVIDIA for hardware specifications; the vendors write their own drivers. And then vendors pay Microsoft to have their drivers tested.
All they have to do is tell us how the damn parts work and we'll write the driver.
Will you personally take the time to write a driver? The vendors probably make more money from working closed-source drivers than they would from assisting third party volunteers create open-source drivers. No one is going to buy a brand new expensive GPU for use in Linux if they have to wait for the community to create a stable driver for it (and potentially wait for their distributor to upgrade packages that include the new driver...)
ummm.... not quite. In Vista the graphics card driver interfaces changed to support Vista Display Driver Model Architecture.. They effectively embedded a DirectX graphics subsystem into the kernel (Dxgkrnl.sys) which includes a display port driver, video memory manager, and CPU scheduler.
A new graphics driver interface was added to enable Aero, but the old "XPDM" interface still works. This is why non-Aero compatible GPUs like the GMA 900 continue to work with Windows Vista and Windows 7.
Why do you think NVidia had such struggles getting a stable driver at Vista release time?
Intel certainly didn't have as many struggles compared to ATI or NVIDIA. In August 2006 (four months before Vista's retail release), Intel WDDM Aero-capable drivers had better performance than ATI's drivers at the time, even though all of the ATI GPUs were faster...
I don't think this is correct. While Windows Vista and 7 support Win2K drivers for many devices unchanged, video driver model in particular was completely rehashed in Vista. I'm not aware of any video drivers for XP, much less 2K, working in Vista or 7.
Every GPU that doesn't have a WDDM driver uses a 2000/XP driver. This includes chipsets like the GMA 900 (famous for not having Aero support).
In "real world" you can just install an old graphics driver in Windows 7. You can even use the XP drivers for GPUs that have native Windows 7 drivers; of course you are then limited to the features of the old interface.
Then explain why my printer doesn't work on Windows 7, when it works in XP, VIsta, OSX and Ubuntu? Out of the box in two of those four(I'll let you guess which).
I said graphics, not printer.
I would also like you to explain why my 4 year old Dell laptop doesn't let me pick the appropraite resolution in Windows 7, when it will in XP, Vista and Ubuntu.
It will matter in 5 years when you want to use Linux on a computer and you discover that the proprietary driver hasn't been updated in two years. So then what do you do? Use a 2 year old version of Linux?
This is a flaw in Linux/X.org.
Windows 7 manages to support the same graphics driver interface as did Windows 2000, which is nearly ten years old.
So it's no shock that these kids, of which very little was ever demanded or expected of them, should suddenly find themselves failing college once the gloves come off.
I like to think the cause is much simpler: colleges have increased enrollment considerably, and in order to do so, reduced their admissions standards.
But the code that calls it needs to be able to execute natively - because WINE Is Not an Emulator. In theory you could run Windows NT Alpha/PowerPC code with the corresponding version of WINE...
x86 Wine is/was supported using qemu's user mode emulation.
Sorry, but Wine and its derivatives can only ever work on x86 hardware (or hardware with x86 compatibility) as I understand it. You can move binaries between OS's by emulating.intercepting.translating system calls, but not between architectures.
It (is/was) possible to run x86 Wine on qemu's user mode emulation. This was the original motivation to port Wine to Darwin/Mac OS X (before Apple switched to Intel).
Wine's Win32 implementation can also be used to compile native executables using the Win32 API. I think some commercial Windows applications were actually ported to non-Windows operating systems this way. It appears to support PowerPC.
Re:Cheating
on
PS3 Hacked?
·
· Score: 4, Insightful
Microsoft has been banning Xbox 360s for years.
The process for "fixing" a banned Xbox 360 console involves cloning the NVRAM from another Xbox 360. That's hardly working around a ban.
Found this interesting, granted this A9 board is not running at 2ghz but is already very competitive with Atom.
Just watch the video; the Atom system is faster every time. The ARM isn't so slow as to be useless for web browsing, but it is clearly slower.
The Atom tested here is the single core N270, one and a half years old. Its replacement (N450) is already on the market. There are dual-core Atom CPUs as well, but only the single-core is sold for use in netbooks.
Polycom's HDX systems (since 2007) separate the codec (rackmount or tabletop unit with camera/mic inputs and display/speaker outputs and the network connection) from the cameras and microphones. It should be possible to create a system suitable for any room size, as long as there's a good location for the camera.
Polycom's own camera has optical zoom, rotation, etc. and I think they even support 1080p30 and 720p60. The codecs support input via DVI, VGA, component, etc.
Doesn't make an iota of difference what decodes your video, if you're not legally allowed to encode it in your jurisdiction using a patent-encumbered codec in the first place.
You called x264 a codec. It is not a codec; it is an encoder. Furthermore, it is completely legal to encode video with a patent-encumbered encoder if you have a license to do so.
It says "clean room implementation from the specs". This is not the same as "reverse engineering" as you claimed twice. Furthermore, stackoverflow is hardly a reliable source for this kind of thing.
The specs for FAT32 are available, too. Didn't seem to do TomTom much good, did it?
I didn't make any claim about the specs being freely available meaning anything with regards to patent licensing. Since I was questioning your erroneous claim of x264 being "reverse engineered", I provided the link to the freely downloadable H.264 specification to demonstrate that reverse engineering would be unnecessary.
The point being that at some point dual processor ARM based systems will be available. Do you want to pay the premium for an Intel Atom (locked in) processor now?
Just because someone uses an Atom today doesn't mean they can't use some other CPU in the future. Those using open source and portable software aren't locked in at all.
ARM isn't some open-source design managed by a benevolent nonprofit. There are plenty of companies who use ARM CPU designs, but they all pay licensing fees to ARM.
People generally buy computers because they want to use them *now* to do something useful. Only a small minority of a minority of computer users will wait for a non-Intel CPU.
There are a lot (probably 100+ million people) who work with desktop systems running Pentium (IV) systems -- which are not likely to get changed anytime soon.
Relevance?
The newest Pentium 4 systems are probably 2-3 years old; corporations with three-year replacement schedules will be completely rid of the Pentium 4 in the near future.
the better question for/. is "What is the most efficient CPU/OS combinatin for users to use.
How many people use energy efficiency as their only criteria when selecting software and hardware?
ARM announced a "2GHz Capable Cortex-A9 Dual Core Processor" in Sept. of 2009.
Have there been any systems with this CPU? Even samples/demo units?
then you ought to be taking a serious look at whether your supplier is locked into "Intel/Microsoft"
All of them? I find this highly unlikely.
The ARM based systems *should* be available, even with multiple cores (which for a "normal" user means one core does nothing (I defy you to come up with a scenario where 2 cores are kept busy 24/7 unless you are running SETI@Home (or equivalents) on your laptop/netbook.
One need not actually use the full CPU capacity at all times.
At any rate, I mentioned dual core because the system in the article does have a dual core CPU.
At which point you have to begin to ask serious questions as to whether or not the CPU (power/CO2 emissions) use is justified in the possible expansion of the human knowledge base.
Justified? It doesn't matter if it's justified; we do not have an ideological-driven centrally planned economy. What matters is that people are willing to pay for the energy and any environmental costs involved in generating it.
Anyway, modern CPUs have plenty of power saving features.
Even if you don't do 3d - 2d window drawing and other operations seem to cause X to use a lot of CPU for some reason. This only happens with intel.
This is not a problem with the GMA 950; it is an X/driver problem. Some driver/kernel/X server/library combinations resulted in acceleration not working; I don't remember the details. This is partly the fault of the distributors who shipped these combinations without adequately testing them on what is one of the most popular GPUs. (Problems like this have led me to avoid OpenSUSE, Fedora, and non-LTS Ubuntu, and keep me from even considering using Linux on the desktop.)
Intel graphics work fine in Mac OS X or Windows, even on the GMA 950 and even with desktop compositing.
If you are going to run a non-Windows (free software, open source) system you should be looking at ARM based systems.
Please let me know where I can find an ARM system with as much performance as a dual-core Atom.
I'm sure they're coming real soon..
Part of being an informed consumer is recognizing monopolies (both software and hardware) and making purchasing decisions that do not promote said monopolies.
Intel has no monopoly on the low power x86 market. VIA and AMD both have competing CPUs for low-power desktops; Dell sells a low-power system with an AMD CPU and GPU for under $300. Intel's main advantage is in battery powered netbooks; but this system is a PC, and uses a dual-core Atom instead of the single-core model used in netbooks.
Most respectfully, you are living in a dream world. I regularly get 20 or 30 PowerPoint presentations for a meeting, with nearly as many formats, wrappers, containers, and resolutions as you could want. To top it off, often made with different versions of PowerPoint.
They simply will not all work correctly.
I am talking about one container and its codecs: WMV.
So what if the file is big? A small file that doesn't play correctly is a big failure.
It can take a long time to send large files over a WAN or Internet connection.
And do you find it difficult to find software to create avi's? I can make them directly from all my Video software.
I don't have a problem with creating AVIs. But users have software that will create WMV files. There's even such a program in Windows XP.
Your saying that the OS supports WMV does not mean that PowerPoint on another machine will play it correctly.
Indeed; that's why I also said that WMV is supported by PowerPoint.
You didn't read my post, did you? Windows XP supports the WMV container and codecs out of the box.
PowerPoint - for one example - really really likes AVI's, despite Microsoft "not supporting the format.
Plays WMVs just fine.
What will likely happen is that when it comes time to play the video, you will be kicked out of PowerPoint, taken into Windows Media Player, then the screen will resize to the video's dimension, and the projector will go blank, or act weird in some other way as it adjusts to the new image.
No. This isn't how playback in PowerPoint works - and the output mode should only change if someone has changed their video output settings to theater mode/etc.
Thing is, the presentation machine and the playback machine better have the same codec.
Everyone's machine really does have WMP and therefore supports WMV as long as they have Windows XP or newer. And they do, since this is 2010 and not 2002.
So you can either make it AVI with Cinepak, or something that won't work.
And then you can choose having a small poor-quality file or a huge file. And that's after you get software to create an AVI file.
Good luck with telling them about the great new codec you know about that is sooo much better than AVI.
WMV is not a "great new codec"; the container and decoders have been supported in Windows for almost (if not over) a decade. AVI is not a codec.
The only way to ensure that the video will play on everyone's computer is to make it AVI, and make it Cinepak compression. I don't like it either, but that's how it is.
Every Windows XP or newer system supports WMV. Furthermore, this doesn't change that MPEG-2 containers are still more popular than any other format.
I always thought the main point of VLC is that it does not rely on the often broken collection of installed codecs. It "just works" even if the rest of the system is messed up in all kinds of ways.
MPC-HC has built-in decoders which it uses by default, but it will use Windows DirectShow filters for anything unsupported or disabled.
Slashdot Mirror
Microsoft doesn't pay ATI or NVIDIA for hardware specifications; the vendors write their own drivers. And then vendors pay Microsoft to have their drivers tested.
Will you personally take the time to write a driver?
The vendors probably make more money from working closed-source drivers than they would from assisting third party volunteers create open-source drivers. No one is going to buy a brand new expensive GPU for use in Linux if they have to wait for the community to create a stable driver for it (and potentially wait for their distributor to upgrade packages that include the new driver...)
Yes, Windows Vista/7 do support 2000/XP (XPDM) drivers.
It's unfortunate that you can't get your hardware working in Windows 7, but that's no reason to insult anonymous people on the Internet.
I am not your tech support.
It sucks that you can't get your hardware working, but your anecdote does not change that Windows 7 supports drivers using the 2000/XP interface.
Of course it will work - this is how non-Aero capable GPUs work on WIndows 7.
A new graphics driver interface was added to enable Aero, but the old "XPDM" interface still works. This is why non-Aero compatible GPUs like the GMA 900 continue to work with Windows Vista and Windows 7.
Intel certainly didn't have as many struggles compared to ATI or NVIDIA. In August 2006 (four months before Vista's retail release), Intel WDDM Aero-capable drivers had better performance than ATI's drivers at the time, even though all of the ATI GPUs were faster...
Every GPU that doesn't have a WDDM driver uses a 2000/XP driver. This includes chipsets like the GMA 900 (famous for not having Aero support).
In "real world" you can just install an old graphics driver in Windows 7. You can even use the XP drivers for GPUs that have native Windows 7 drivers; of course you are then limited to the features of the old interface.
I said graphics, not printer.
I'm not your tech support.
This is a flaw in Linux/X.org.
Windows 7 manages to support the same graphics driver interface as did Windows 2000, which is nearly ten years old.
I like to think the cause is much simpler: colleges have increased enrollment considerably, and in order to do so, reduced their admissions standards.
x86 Wine is/was supported using qemu's user mode emulation.
It (is/was) possible to run x86 Wine on qemu's user mode emulation. This was the original motivation to port Wine to Darwin/Mac OS X (before Apple switched to Intel).
Wine's Win32 implementation can also be used to compile native executables using the Win32 API. I think some commercial Windows applications were actually ported to non-Windows operating systems this way. It appears to support PowerPC.
Microsoft has been banning Xbox 360s for years.
The process for "fixing" a banned Xbox 360 console involves cloning the NVRAM from another Xbox 360. That's hardly working around a ban.
Just watch the video; the Atom system is faster every time. The ARM isn't so slow as to be useless for web browsing, but it is clearly slower.
The Atom tested here is the single core N270, one and a half years old. Its replacement (N450) is already on the market. There are dual-core Atom CPUs as well, but only the single-core is sold for use in netbooks.
Polycom's HDX systems (since 2007) separate the codec (rackmount or tabletop unit with camera/mic inputs and display/speaker outputs and the network connection) from the cameras and microphones. It should be possible to create a system suitable for any room size, as long as there's a good location for the camera.
Polycom's own camera has optical zoom, rotation, etc. and I think they even support 1080p30 and 720p60. The codecs support input via DVI, VGA, component, etc.
You called x264 a codec. It is not a codec; it is an encoder.
Furthermore, it is completely legal to encode video with a patent-encumbered encoder if you have a license to do so.
It says "clean room implementation from the specs". This is not the same as "reverse engineering" as you claimed twice. Furthermore, stackoverflow is hardly a reliable source for this kind of thing.
I didn't make any claim about the specs being freely available meaning anything with regards to patent licensing.
Since I was questioning your erroneous claim of x264 being "reverse engineered", I provided the link to the freely downloadable H.264 specification to demonstrate that reverse engineering would be unnecessary.
x264 is only an encoder for the H.264/MPEG-4 AVC standard. It does not handle decoding.
Citation definitely needed. The project calls itself a H.264/AVC encoder.
Being implemented through reverse-engineering doesn't matter. Anyway, it seems the specification is freely downloadable.
Just because someone uses an Atom today doesn't mean they can't use some other CPU in the future. Those using open source and portable software aren't locked in at all.
ARM isn't some open-source design managed by a benevolent nonprofit. There are plenty of companies who use ARM CPU designs, but they all pay licensing fees to ARM.
People generally buy computers because they want to use them *now* to do something useful. Only a small minority of a minority of computer users will wait for a non-Intel CPU.
Relevance?
The newest Pentium 4 systems are probably 2-3 years old; corporations with three-year replacement schedules will be completely rid of the Pentium 4 in the near future.
How many people use energy efficiency as their only criteria when selecting software and hardware?
Have there been any systems with this CPU? Even samples/demo units?
All of them? I find this highly unlikely.
One need not actually use the full CPU capacity at all times.
At any rate, I mentioned dual core because the system in the article does have a dual core CPU.
Justified? It doesn't matter if it's justified; we do not have an ideological-driven centrally planned economy. What matters is that people are willing to pay for the energy and any environmental costs involved in generating it.
Anyway, modern CPUs have plenty of power saving features.
This is not a problem with the GMA 950; it is an X/driver problem. Some driver/kernel/X server/library combinations resulted in acceleration not working; I don't remember the details. This is partly the fault of the distributors who shipped these combinations without adequately testing them on what is one of the most popular GPUs. (Problems like this have led me to avoid OpenSUSE, Fedora, and non-LTS Ubuntu, and keep me from even considering using Linux on the desktop.)
Intel graphics work fine in Mac OS X or Windows, even on the GMA 950 and even with desktop compositing.
Please let me know where I can find an ARM system with as much performance as a dual-core Atom.
I'm sure they're coming real soon..
Intel has no monopoly on the low power x86 market. VIA and AMD both have competing CPUs for low-power desktops; Dell sells a low-power system with an AMD CPU and GPU for under $300. Intel's main advantage is in battery powered netbooks; but this system is a PC, and uses a dual-core Atom instead of the single-core model used in netbooks.
I am talking about one container and its codecs: WMV.
It can take a long time to send large files over a WAN or Internet connection.
I don't have a problem with creating AVIs. But users have software that will create WMV files. There's even such a program in Windows XP.
Indeed; that's why I also said that WMV is supported by PowerPoint.
You didn't read my post, did you? Windows XP supports the WMV container and codecs out of the box.
Plays WMVs just fine.
No. This isn't how playback in PowerPoint works - and the output mode should only change if someone has changed their video output settings to theater mode/etc.
Everyone's machine really does have WMP and therefore supports WMV as long as they have Windows XP or newer. And they do, since this is 2010 and not 2002.
And then you can choose having a small poor-quality file or a huge file. And that's after you get software to create an AVI file.
WMV is not a "great new codec"; the container and decoders have been supported in Windows for almost (if not over) a decade. AVI is not a codec.
Every Windows XP or newer system supports WMV. Furthermore, this doesn't change that MPEG-2 containers are still more popular than any other format.
MPC-HC has built-in decoders which it uses by default, but it will use Windows DirectShow filters for anything unsupported or disabled.