Wine 1.1.33 Gains More Direct3D 10 Functions

[phoronix]

Phoronix: Wine 1.1.33 Gains More Direct3D 10 Functions

It was a year ago that Wine developers began working on Direct3D 10.0 support and then in March developers at CodeWeavers began working on the Direct3D 10 support too. Recent Wine releases in particular have brought better Direct3D 10.x coverage and this work has continued with the just-released Wine 1.1.33. Wine 1.1.33 implements more Direct3D 10 functions, the Gecko browser engine is now installed at wineprefix creation time, there is better support for certificates in crypt32, improved sound support in mciwave, and many cleanups for issues spotted by the Valgrind memory debugging utility...

http://www.phoronix.com/vr.php?view=NzcwMw

[Henri]

It was a year ago that Wine developers began working on Direct3D 10.0 support and then in March developers at CodeWeavers began working on the Direct3D 10 support too.

It's a bit misleading to put it like that, the "Wine developers" that started working on d3d10 support last year (me) work for CodeWeavers.

[lucian]

I've always wondered whether it would be easier to implement Direct3D over Gallium3D, instead as a layer over OpenGL. This would be a bit problematic at first since many cards don't have Gallium3D pipeliles, but it may pay off in the long run.

[Craig73]

I've always wondered whether it would be easier to implement Direct3D over Gallium3D, instead as a layer over OpenGL. This would be a bit problematic at first since many cards don't have Gallium3D pipeliles, but it may pay off in the long run.

Gallium3d isn't on OS X, one of the various platforms WINE is targeted for. And, iirc, the developers don't feel a re-write would yield any substantial benefits at this time (especially considering other development work that needs to be completed)

Search for this on phoronix, it's come up before.

[Craig73]

What I'm curious about is the status of the roadmap... how are we doing moving towards a 1.2 version? What do they envision 1.2 including at this point (ie - are there still plans for a GDI engine or can anyone agree on what a proper implementation of that even looks like)

[Henri]

I've always wondered whether it would be easier to implement Direct3D over Gallium3D, instead as a layer over OpenGL. This would be a bit problematic at first since many cards don't have Gallium3D pipeliles, but it may pay off in the long run.

The infrastructure for something like that is mostly there, but we don't think it would be worth the effort (both writing it in the first place and the added maintenance) at this point. Also note that Gallium3D doesn't have a stable/public interface, so something like this would probably take the form of a D3D specific state tracker in Mesa.

[Henri]

What I'm curious about is the status of the roadmap... how are we doing moving towards a 1.2 version? What do they envision 1.2 including at this point (ie - are there still plans for a GDI engine or can anyone agree on what a proper implementation of that even looks like)

The current plan is that the main new Wine 1.2 feature will be 64-bit infrastructure. I.e. getting the architecture right, not necessarily supporting a lot of 64-bit programs. 1.2 probably won't have a DIB engine.