Posts: 14
Joined: May 2007
Reputation:
0
Hi yuvalt,
I checked my mesa version. It is 6.5.2. Although it is new enough I reinstalled it. Guess what. Compilation now works. Thank you for pointing me into the right direction.
Regards,
riker
Posts: 51
Joined: Jan 2005
Reputation:
0
If xbmc gets released for a certain hardware-box it should at least be one with hdcp support, no matter what the connector interface is. (DVI or HDMI)
Posts: 952
Joined: Feb 2005
Reputation:
0
gav1577, dude... a port does not mean that the old xbox is abandoned. it still has it purpose like you describe. the port is for content which you have no use for anyways on your tv, so stay with the xbox.
the only problem when writing new source is that developer in some cases has to do special things for either xbox, linux or pc for it work on all supported platforms
Posts: 26,215
Joined: Oct 2003
Reputation:
187
Does glxgears work? What about some SDL apps (eg just download some simple SDL demos)
Posts: 14
Joined: May 2007
Reputation:
0
I'm a long time Linux/Windows/Xbox-xbmc user/programmer, and I have owned many shuttles and have tried to use them as HTPCs. I have switched over to Apple as my primary computing platform this year. My personal preference would be mac mini support. It has many benefits over most off the shelf shuttle type boxes. Built in IR, Bluetooth, WLAN, it's whisper quiet, small and likely to be a pretty static platform. I appreciate the fact AppleTv is cheaper but it's a somewhat closed platform, one of my main frustrations with the current Xbox as a platform is that it is a hack. It really limits the audience and the ability for your average users to get themselves out of a mess.
Even though I think the ideal platform would be mac mini. I do think that the Linux port should be the primary first effort. I think that a port from a well written linux port to os x would not be that difficult.
Another reason Linux is ideal is virtualization. I notice there is a lot of talk about what hardware this should run on, video cards, etc. I see the benefit of having a limited target platform. I think that one of the target platforms, if not the target platform should be VMWare or Parallels. This would not have been an option 6 months ago, but now both VMWare and Parallels are promising DirectX and OpenGL support (with Parallels OpenGL support already in Beta). If we fast forward to 12 or 18 months now and we have a flawless xmbc that uses a virtual target platform - both Parallels and VMWare will have pass through 3d support, off the shelf computing power even when taking the hit of virtualization will be much more powerful than the original Xbox's hardware. The end result is we get the advantage of a commercial enterprise doing the heavy lifting of hardware compatibility support, and now we have an XBMC that will run on Linux, Windows, and OSX.
Lastly, I think even if we did end up preferring a native version of xbmc (be it on linux or osx), there is an advantage to developing in a VM. Beyond the debugging, snapshotting and rollback advantages, it would allow us to all work on one hardware set on day one.