DirectFB 1.6 Release Is Imminent With New Features

[phoronix]

Phoronix: DirectFB 1.6 Release Is Imminent With New Features

DirectFB 1.6 is about to be released this month and it will bring new features to the Direct Frame-Buffer project...

http://www.phoronix.com/vr.php?view=MTA0MzI

[Chewi]

I used to be quite interested in DirectFB for low end systems but I feel that Wayland has perhaps made it obsolete. I didn't know webOS uses it though. Maybe Wayland is still relatively heavy but it also has a much bigger scope than DirectFB. Hardware has moved along too so Wayland should be viable in most cases. I've not tried it yet but I'm looking forward to it.

[timofonic]

FB? Developers still acting retarded, involved on legacy stuff...

What about DirectKMS? KMS is quite more interesting and it's a lot undervalued, someone needs to make it's potential shine and more embedded drivers appear.

[Chewi]

FB? Developers still acting retarded, involved on legacy stuff...

What about DirectKMS? KMS is quite more interesting and it's a lot undervalued, someone needs to make it's potential shine and more embedded drivers appear.

Unless I'm mistaken, KMS is just another framebuffer driver. DirectFB and KMS actually compliment each other.

[RealNC]

FB? Developers still acting retarded, involved on legacy stuff...

What about DirectKMS? KMS is quite more interesting and it's a lot undervalued, someone needs to make it's potential shine and more embedded drivers appear.

Uhm, the FB is provided by KMS.

[VinzC]

I've only read a little about DirectFB and I always wondered how it plays with nVidia cards. I suppose the binary from nVidia is not necessary with DirectFB right? If so hardware acceleration would be quite limited compared with the maximum performance the BLOB offers, am I right?

I'm asking because I vaguely planned to build up my own media server for my Giada N20 that comes with an NVidia card. I wondered if it was best to play with Xorg and the proprietary driver than explore the framebuffer way.

Can anyone shed some lights on this? Thanks in advance.

[timofonic]

Uhm, the FB is provided by KMS.

It uses an incomplete famebuffer emulation layer, mostly done for at least able to run fbconsole.

There were intentions to develop kmsconsole and promote apps to use it directly, but not work has been done for that.

That was said to me over IRC by some project developers, not sure if Radeon or Intel or Xorg. Some of you can ask about it, to check the fact and the rest of people learn something about the critical KMS situation (something similar happens to Gallium3D and such in my opinion).

Old framebuffer must die ASAP, the same as non-Gallium3D drivers in mid term. The new KMS infraestructure can provide proper and efficient GPU acceleration to Xorg-less world if done properly, making most part of the Xorg project obsolete in it's current form (so being a lot more lightweight and integrated with the Linux/BSD/*NIX graphics arquitecture) and new embedded/lightweight/advanced alternatives to get more support too.

FOSS must evolve, this must not be just a play and catch to propietary software but show the real potential of the Open Source developer's community.

A real Open Source Revolution must happen, or the Copyright lobbyists will kill it. First the SOPA try, then patents and more.

And despite Android being a nice smartphone competitor, it doesn't benefit so much compared to a stronger Open Source ecosystem with full hardware support just using Open Source drivers. Imagine able to lobby hardware companies to release the specs or even write the drivers, when Linux could be everywhere and most important hardware already has 100% fully featured support.

Support Copyleft!