Let's look at the reason Windows has better driver support than Linux. How many video drivers has Microsoft written? How many video drivers has the XFree86 project written?
The problem is that all the video drivers for Linux are being written by a third party. Nowadays, if you buy a new video card and use it with Windows, most of the time *IT WILL NOT WORK IN ANYTHING BUT 640x480x4bpp*
The reason? Windows has no driver support.
"Say what?"
That's right kids. Windows doesn't have driver support; the manufacturers do. In fact, if you were to compare Windows to Linux (XFree86 in particular) in terms of video card support, Linux is miles ahead. The reason devices are supported in Windows is that manufacturers write drivers and give them to you with the video card. THAT is what makes the card work.
Now let's analyze why manufacturers can't release drivers for Linux. Support for video has been traditionally put into the X server (with the exception of fbcon, of course). Adding support means that you would either have to make your own X server, hand over the specs, or force you to sign an NDA for the specs.
Now how can we prevent this?
I'm no expert, but as far as I'm concerned, writing a set of drivers for Windows is relatively easy for the designers of the chipset who know all the features inside out. But writing a driver for XF86 means releasing the source. And Windows video drivers are more of a "plug in and go" thing. So what we need is an X server which supports binary-only video drivers which can be loaded when it is starting. That way, a manufacturer can write a driver, compile it, release the binary, and the user can specify it in the configuration and go.
Poof. No NDA, no harrasing, less trouble. We don't get the benefit of Open Source, but we do get more hardware support. If the drivers are buggy, well, we can't do much about it. But hey, it can't be THAT much worse than Windows, can it?
It certainly won't play 128k mp3s. The PC110 is based on a 486SX-33, so you will be using Linux's FPU emulation code to play. If it were DX-based, you would be able to play lower bitrate MP3s (64k?), but that would be pushing the CPU pretty hard.
" Disable the onboard stuff, slap your new whizbang video card in, and be happy. Or else just slap your new whizbang video card in, and have _two_ heads on your machine, or _two_ soundcards, or _two_ whatevers... "
Actually, I have a friend with a Houston mobo with everything integrated. The sound is a CM8330 or something like that, which is at least half decent. The video is an SiS 6326 - I've had bad experiences with SiS in the past, but...
The thing is, the SiS is an AGP chipset, but there's no AGP slot on the board. So if you wanted to replace the video card, you had to go PCI, which is slowly fading away...
Slashdot Mirror
Let's look at the reason Windows has better driver support than Linux. How many video drivers has Microsoft written? How many video drivers has the XFree86 project written?
The problem is that all the video drivers for Linux are being written by a third party. Nowadays, if you buy a new video card and use it with Windows, most of the time *IT WILL NOT WORK IN ANYTHING BUT 640x480x4bpp*
The reason? Windows has no driver support.
"Say what?"
That's right kids. Windows doesn't have driver support; the manufacturers do. In fact, if you were to compare Windows to Linux (XFree86 in particular) in terms of video card support, Linux is miles ahead. The reason devices are supported in Windows is that manufacturers write drivers and give them to you with the video card. THAT is what makes the card work.
Now let's analyze why manufacturers can't release drivers for Linux. Support for video has been traditionally put into the X server (with the exception of fbcon, of course). Adding support means that you would either have to make your own X server, hand over the specs, or force you to sign an NDA for the specs.
Now how can we prevent this?
I'm no expert, but as far as I'm concerned, writing a set of drivers for Windows is relatively easy for the designers of the chipset who know all the features inside out. But writing a driver for XF86 means releasing the source. And Windows video drivers are more of a "plug in and go" thing. So what we need is an X server which supports binary-only video drivers which can be loaded when it is starting. That way, a manufacturer can write a driver, compile it, release the binary, and the user can specify it in the configuration and go.
Poof. No NDA, no harrasing, less trouble. We don't get the benefit of Open Source, but we do get more hardware support. If the drivers are buggy, well, we can't do much about it. But hey, it can't be THAT much worse than Windows, can it?
- ed
It certainly won't play 128k mp3s. The PC110 is based on a 486SX-33, so you will be using Linux's FPU emulation code to play. If it were DX-based, you would be able to play lower bitrate MP3s (64k?), but that would be pushing the CPU pretty hard.
"
Disable the onboard stuff, slap your new whizbang video card in, and be happy. Or else just slap your new whizbang video card in, and have _two_ heads on your machine, or _two_ soundcards, or _two_ whatevers...
"
Actually, I have a friend with a Houston mobo with everything integrated. The sound is a CM8330 or something like that, which is at least half decent. The video is an SiS 6326 - I've had bad experiences with SiS in the past, but...
The thing is, the SiS is an AGP chipset, but there's no AGP slot on the board. So if you wanted to replace the video card, you had to go PCI, which is slowly fading away...
I've read about an adapter that would let the SNES play old NES games by emulating it in hardware...
Can anybody say "Microsoft employee forum poster"?
If we want Toshiba Japan to make their decision, why are we filling in the form on Toshiba of America's site?