running openGL application remotely using ssh -X and cygwin/x ,extension "NV-GLX" missing on display "localhost:10.0

[Jon TURNEY]

On 28/04/2014 22:02, Linda Walsh wrote:
> Jon TURNEY wrote:
>> Yes, this should work.
> ----
> *But*, I'm pretty sure it doesn't anymore since the "Xgl" extension
> that was used to transport the openGL commands between client/server
> was removed from xorg's Xserver.

You seem to be confusing Xgl (an X server implementation) and GLX (an X 
protocol extension).

While they do contain the same letters in a different order, they are 
very different things.

> AIGLX doesn't work with client's native openGL drives when the
> DISPLAY isn't local. Instead, it sends full-frame-buffer updates to
> simulate what would be happening -- something that "appears" to work
> correctly for small OpenGL windows. But is entirely 'faked' (not
> really remote openGL that used the Server's acceleration Hardware.
>
>      Which would give you unaccelerated frame-buffer updates to simulate
> the effect.  Not quite what used to be available.

This is also totally wrong.

You are (more or less) describing how mesa's direct software rendering 
works (which is usually the default path for remote displays) which is 
completely different to AIGLX (where GL commands are sent via the GLX 
protocol to the X server and rendered using acceleration there)
	
So, again, please stop spreading misinformation.

I'm sure there are bugs in and limitations with OpenGL and the XWin 
server, but if you have a problem, please don't hijack someone else's 
thread, but report it in sufficient detail for me to try to reproduce it.

-- 
Jon TURNEY
Volunteer Cygwin/X X Server maintainer

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://x.cygwin.com/docs/
FAQ:                   http://x.cygwin.com/docs/faq/