AMD Has Open-Source Driver For HD 8000 Series

[madbiologist]

AMD announced the availability of the desktop "Richland" A-Series APUs on 5th June 2013. Model numbers are A10-6800K, A10-6700, A8-6600K and A8-6500. They contain Radeon HD 8670D, Radeon HD 8670D, Radeon HD 8570D and Radeon HD 8570D graphics respectively. Further specs and info about performance increases is available at http://www.amd.com/us/press-releases...013june05.aspx

[madbiologist]

The "Richland" PCI IDs have now been added to Mesa - http://cgit.freedesktop.org/mesa/mes...ee0141b433887a

[madbiologist]

AMD announced the first availability of the "Richland" A-Series APUs on 12th March 2013. Model numbers are A10-5750M, A8-5550M, A6-5350M and A4-5150M. They contain Radeon HD 8650G, Radeon HD 8550G, Radeon HD 8450G and Radeon HD 8350G graphics respectively. Further specs and info about performance increases and battery life is available at http://www.amd.com/us/press-releases...2013mar12.aspx

[madbiologist]

It's official. AMD officials revealed to journalists during a conference call on February 15, 2013 that AMD will be releasing new Radeon desktop GPUs "from top to bottom" in 2013, though you won't be seeing the next-gen Radeon 8000 series graphics cards until late in the year. Instead, the company will be firmly focused on the Radeon 7000 series cards currently available on the market for the first half of 2013. In the meantime they plan to focus a lot of attention on aggressively developing drivers (I assume that means the Catalyst driver) and courting gaming industry partners.

See http://www.pcworld.com/article/20284...s-in-2013.html

[przemoli]

I have asked for clarification re: what we have officially said about Richland -- can't comment otherwise until I hear back.

We are trying to have launch-time code for all of our hardware now. The downside is that we need to understand and coordinate with our launch plans, which is a new challenge for the engineering team

You poor things

ok ok. end of joking. I need to grab my fedora iso... its time for checking out systemd, gnome3 and r600g (unless fedora welcome me with llvmpipe )

[bridgman]

I have asked for clarification re: what we have officially said about Richland -- can't comment otherwise until I hear back.

We are trying to have launch-time code for all of our hardware now. The downside is that we need to understand and coordinate with our launch plans, which is a new challenge for the engineering team

[madbiologist]

Thanks bridgman.

Do you have any information about Richland? I realise that you can't confirm or deny the release date I quoted if it hasn't been announced by AMD marketing, but can at least re-share what they _have_ said so far? And maybe also tell us if AMD's open-source driver team are working internally on as yet unreleased driver code for Richland?

[bridgman]

I was surprised to read that AMD postpones the release of the new GCN2 ASICs to late 2013...

We were surprised to read that too. It's funny how when rumour A is replaced by rumour B people automatically assume that *we* changed something

[curaga]

MrCooper, agd5f, thanks for the clarification.

[entropy]

Just noticed that the article links the just-released driver support for Oland to an HD 8800-series part, presumably based on September 2012 rumours which described Oland as a 3.4 billion transistor upper-midrange part.

Oland is actually a 6 CU part (ie same ALU count as the GPU on Trinity, but GCN rather than VLIW4). See Anandtech description of Oland/Mars :

Oland : http://www.anandtech.com/show/6570/a...eries-for-oems

Mars : http://www.anandtech.com/show/6571/a...n-8000m-series

The "Oland" support we pushed recently was for the HD 85xx/86xx OEM and mobile parts announced in January, not for "HD 8800".

Thanks for clarification!

I was surprised to read that AMD postpones the release of the new GCN2 ASICs
to late 2013 while the first bits of driver support hit the repositories.