you care about the functionality (which will be there at some point one way or another) or you just want network transparency for your own reasons???Quote:
Originally Posted by zoomblab
if its the second you plainly won't have it with wayland.
Printable View
you care about the functionality (which will be there at some point one way or another) or you just want network transparency for your own reasons???Quote:
Originally Posted by zoomblab
if its the second you plainly won't have it with wayland.
Aside from other correct answers to your question, it should also be noted that between audio and visual, there is a MASSIVE difference in the amount of data that has to be shuffled through the stack. Video requires orders of magnitude greater bandwidth than audio. The bandwidth requirements for audio are negligible by today's standard. Video still takes everything you've got.Quote:
Originally Posted by zoomblab
Φίλε μου, μαζί μιλάμε χώρια καταλαβαινόμαστε… :)Quote:
Originally Posted by 89c51
I don't care. In fact I am all in favor of the Wayland architecture. My post was an observation meant for discussion.
Again:
- Wayland: Drops network transparency.
- PulseAudio: Introduces network transparency.
Point taken, audio < video. However Pulse is known to add latency, and besides that also adds complexity.Quote:
Originally Posted by droidhacker
Well, I've studied ancient greek at school, but this is quite different. The beginning sounds like "My dear, learn...", but after this I'm lost.Quote:
Originally Posted by zoomblab
Back on topic, I think today audio network transparency is considered more useful than video one (media streamers etc), and it has been realized that putting network transparency at the protocol level was not such a great idea for video (I've never seen a modern desktop running through the network without becoming inusuable, except maybe on an empty gigabit lan).
Short answer they didn't move to opposite directions. Wayland will implement the same functionality as X in a different way (for simplicity reasons according to the wayland FAQ) and Pulseaudio added functionality that ALSA didn't have (afaik). Where you will introduce a function is just a design decision and could be done a million different ways with pros and cons to each design.Quote:
Originally Posted by zoomblab
Καλύτερα τώρα?? :rolleyes:
Yes, because Ethernet speeds are at a standstill and are not increasing. :rolleyes:Quote:
Originally Posted by droidhacker
Some of us have been running X sessions over IP since Thicknet and Token Ring dominated the scene.
I hope Wayland comes with some kind op C-style networked PostScript-SVG hybrid that can be accelerated with Gallium (SVG) and also a non-networked stripped Xlib solution for backwards compatibility.
We can dream...
Quote:
Originally Posted by locovaca
Latency is the problem, not bandwidth.
Some other points:
* There are many other remote desktop technologies being used today that are without a doubt superior solutions to X for what people need. It's been around for years, but has largely been superceeded by other technologies which are now commonplace in businesses for terminal and remote desktop for users and administrators.
* You don't need to let your X Server drive your video card in order to use X Windows networking.
* X Windows networking works just fine on Microsoft Windows and on OS X, yet nobody on those platforms is clamoring for developers to produce X Windows applications for them.
* Wayland will have it's own X Server which will work with compositing.
* No amount in increase of bandwidth is going to make X Windows better.
Put all that together and then try to figure out what that means.
Quote:
Originally Posted by 89c51
Wayland and Pulseaudio are both servers/managers. One server manages your display, while the other manages your audio.
XFree X Server, is also a display manager server.
How much latency does that userland server add to your button clicks?