X@FOSDEM 2009: RandR 1.3, GEM, Gallium3D, Etc

[phoronix]

Phoronix: X@FOSDEM 2009: RandR 1.3, GEM, Gallium3D, Etc

All of the recordings from the X.Org meetings that took place during FOSDEM 2009 are now available on Phoronix. There were nine topics in total from RandR 1.3 to shader compiler optimization strategies. Novell's Matthias Hopf had explained RandR 1.3 with all of the work involved in this update to the Resize and Rotate extension for the X Server that now has panning support and other new-found capabilities...

http://www.phoronix.com/vr.php?view=NzA2Ng

[curaga]

Thank you for posting a link to the vids!

Finally Phoronix acknowledged that not everyone likes flash.

It still required a view of the source at blip.tv to get the mpg version, as it has a different filename. But that is their problem, not Phoronix's.

[DragonsTear]

Just out of curiosity :

Was there a topic at the FOSDEM about the future organization and development of Xorg?

I think it's great to here that so many things happen actually in the OpenSource development, even with support ( more or less ) from the big organizations, but except for features, pretty less is done in my humble opinion...

Everybody is working hard on the features, but there are so less people that the rest of the Xorg part is pretty much inactive

Are there any comments to this topic at the FOSDEM or am I totally wrong?

[drag]

Ya I may have complained about flash in the past, but I do appreciate Phoronix working to make these recordings easily available.

[Tomasu]

So do they all have understandable audio now? The ones posted earlier all tend to be way too quiet, or have super annoying audio glitches.

[bridgman]

Was there a topic at the FOSDEM about the future organization and development of Xorg?

Don't think so; that kind of discussion normally happens more at the XDC/XDS conferences which are organized by x.org.

I think it's great to here that so many things happen actually in the OpenSource development, even with support ( more or less ) from the big organizations, but except for features, pretty less is done in my humble opinion... Everybody is working hard on the features, but there are so less people that the rest of the Xorg part is pretty much inactive

Right now I think the feeling is that the features being worked on right now are so important for the future of "the Linux Desktop" that they really *should* be the priority for a few more months, ie it's not just developers being distracted by shiny things. Once kernel modesetting, RandR1.3, DRI2, GEM (with or without TTM underpinnings) and RDR are all in place, I expect that the longer term discussion and planning will ramp up again.

There does seem to be some important work happening on the X input side as well, although I'm not really up on the details.

[miles]

The good news, however, is that the audio can be easily cleaned up through Audacity.

You can't. I tried before with similar files, either with Audacity or with professional programs (not any better actually). Denoise filters and the like can't magically clean the files, contrary to the kind of knowledge that comes from watching too much spy movies. You can, with hours of manual efforts, clean up a small part of a recording, but that's a painful process that would take ages on files as long as those. Noise removal filters mess up the audio at least as much as they "remove" the noise, which means for a noise as bad as the one on the videos, you'll create too much mess for the audio to get better.

Bottom line is, the only way to get decent audio is to get a decent audio at recording time - plugging directly to the sound system is great when the speakers have microphones, or at least putting a micro close to the speaker (or if there's no other way, something like an "mp3" recorder, provided you choose a good position).

I'd love to be able to listen to these talks though.

[bridgman]

Bottom line seems to be that Michael was getting decent audio when he checked it, but the adapter was crapping out randomly during the sessions and the problems weren't noticed until after the sessions had finished.

You can listen to the talks (I've listened to about half so far), but (a) you need a fast hand on the volume control, and (b) you need to use your imagination to fill in the gaps where the noise overwhelms the speaker's voice ;)

[miles]

Yes, but non-native speakers have a harder time ;)

As for Michael, it's ok, first times are always like that, you need to learn the hard way ;)

Best setup :
* a sound recorder with line-in, a sound cable with different Jack sizes for when you can convince the sono to plug it in;
* a mic, a preamp and a sound recorder (with plenty of spare batteries) for the other cases. Preferably a sound recorder with internal mic too for the times the connection with the mike is bad.

Never trust cameras for sound recording...

[bridgman]

I think this was a netbook with a microphone

[Michael]

The setup was a microphone with TRS connector connected to a netbook via a quarter-inch to mini jack adapter (what seems to have caused much of the problem). A Logitech web camera then recorded the video separately.

[Tomasu]

Yes, but non-native speakers have a harder time ;)
What made it bad for me is that many of the speakers had very heavy accents (as compared to mine anyhow ;D), they can be tough to understand normally, let alone with noise and buzzing going on.

Also not being able to see the slides or the speakers made the video pretty much useless.

[miles]

The setup was a microphone with TRS connector connected to a netbook via a quarter-inch to mini jack adapter (what seems to have caused much of the problem). A Logitech web camera then recorded the video separately.

I'm not sure how the adapter (usually just a piece of conductive metal) could have caused the problem. It doesn't sound like normal interferance noise, more like the noise you get with bad microphones or lack of a preamp.

Since you used a quarter-inch to mini jack adapter, does it mean your microphone was a HIFI/Sono microphone, in which case the mic in of any computer would provide insufficient gain (i.e. you can get loud enough sound, but you'll have to increase the gain so much you'll get the noise cranked up too)? For those you'd need a preamp between the mike and your computer mic in.

There's mostly two types of microphones:
- the ones for computer, that can be plugged on a computer mic in as is, but don't provide great sound unless the speaker takes extra care (impossible to use in a setting like these talks - you'd get same result you had);
- the ones for HIFI, that provide good audio, but need a preamp before you plug them on a computer (since they're not made for computer use - without a preamp you'd also get the same result you had).

There's also the mikes that comes with a preamp built in, which are easy to spot since you'll usually have to provide batteries.

[miles]

What made it bad for me is that many of the speakers had very heavy accents (as compared to mine anyhow ;D), they can be tough to understand normally

That, and the fact you don't see the speakers on most videos (seeing the speaker can be a great help). I'd rather see the speakers than have a video of the slides ;) (for the slides, a link to the .pdf is better than a youtube video).

[fart_flower]

Let's just blame Pulse Audio for the sound problems. It has certainly caused all of my Linux audio issues for the past year. In fact, Pulse Audio is so awful I'm sure some researcher will soon discover it even causes cancer, AIDS, and global warming.

Okay, maybe not cancer and global warming, but after attempting to watch the aforementioned videos, it nearly did give me AIDS. (Well, hearing aids at any rate...)

[llama]

Phoronix: X@FOSDEM 2009: RandR 1.3, GEM, Gallium3D, Etc

http://www.phoronix.com/vr.php?view=NzA2Ng

Thanks for incorporating our suggestions about audio quality warnings, and links to easier-to-download vids, that were posted in the forum thread about the previous article.
:)