AMD Developers Begin Making Open-Source FreeSync/AdaptiveSync Plans

[starshipeleven]

Well. The HAL was the main reason, why the code didn't get merged.

There is no mistake, they published the code to get feedback on what to do, they didn't think the first source release was going to be merged in its current form.

It's a bit hard to ask kernel devs about what to change in their code if they didn't show it.

Then the reaction was exaggerated by media for the sake of clickbait.

[leipero]

Yet another reason why GNOME 3 developers should start working on mutter wayland support for refresh rates (now dynamic), nothing urgent tho, since I'm sure mutter on X11 is ready for dynamic refresh rates (it handles changes of refresh rates well and "on fly"), but something to think about implementing.

[darkbasic]

Does FreeSync even have a future? Nobody seems willing to answer this question, when AMD already announced its successor will be PROPRIETARY.

[Guest]

Does FreeSync even have a future? Nobody seems willing to answer this question, when AMD already announced its successor will be PROPRIETARY.

Yes, it has. FreeSync is more like a protocol specification. Any monitor can support it, but it doesn't guarantee proper adaptive sync on all FPS.
FreeSync 2 defines minimum quality requirements (for example, minimal ratio between minimum and maximum refresh rate). It guarantees that low frame rate compensation will work, and other stuff... Therefore, certification process is required.

[leipero]

darkbasic Both FreeSync and G-Sync were done by wrong people, it is technology that display manufacturers should have introduced and GPU vendors followed. Unless AMD, Intel and nvidia come up and follow one standard, none of those technologies have future.

At this point there's really no technical reason why almost every display on planet could not support dynamic refresh rates technology, we are lagging at least 5 years behind potential in this sphere IMHO.

[haagch]

When enabled, this will request FreeSync via the hybrid amdgpu DDX's
AMDGPU X11 protocol extension.

Are there patches around so you can try it out until it gets properly implemented?

[donkeykong]

I'm glad to see this happening: I invested in an AMD Gpu and a Freesync monitor some time ago, believing in their promises. I'm glad AMD continues supporting the Linux eco-system despite being only a fraction of their benefits. Great !

[zanny]

darkbasic Both FreeSync and G-Sync were done by wrong people, it is technology that display manufacturers should have introduced and GPU vendors followed. Unless AMD, Intel and nvidia come up and follow one standard, none of those technologies have future.

At this point there's really no technical reason why almost every display on planet could not support dynamic refresh rates technology, we are lagging at least 5 years behind potential in this sphere IMHO.

VESA standardized adaptive sync in 2014. The problem was they didn't make it mandatory on Displayport 1.3+ which let Nvidia get out of actually supporting it.

And I have no idea what Intel is doing. They should have had support for this years ago.

[Brisse]

Does FreeSync even have a future? Nobody seems willing to answer this question, when AMD already announced its successor will be PROPRIETARY.

This statement seems a bit misinformed. Freesync 2 is Freesync, only with stricter requirements on the monitor manufacturers, because they were too lax before which resulted in a lot of cheap monitors with sub-par Freesync implementations which I think is in nobody's interest.

And what exactly is proprietary in this case? The certification process for licensing the trademark, or the technology that drives it, because it's still VESA Adaptive Sync you know...

[microcode]

Another benefit of Adaptive Sync is the ability to play video content at its native framerate (24, 24000/1001, 30, 30000/1001, 25, 50, 60000/1001) without jitter in the presentation time. Another is the correctly-timed playback of gifs (which are inherently variable-frame-rate, and never synced to today's displays on account of having inter-frame delays specified in whole number milliseconds).