Nightwulf|work: hi all
corvvs: I'm having an issue with the radeonhd driver... that's probably not *directly* related to it, but I was hoping you guys could help
corvvs: I've been googling one more time because I wanted to make sure that I couldn't find something easily first :)
corvvs: but udev isn't creating my /dev/dri/card0 the way it should
Nightwulf|work: corvvs: are you very sure, it isn't the driver in combination with your card? if it's a r6xx and you don't use version 1.2.4 of the driver, dri *can't* work
corvvs: ahhh, that's probably the case - I'm using 1.2.3
corvvs: thank you :)
Nightwulf|work: corvvs: but don't get me wrong...1.2.4 does support DRI but no hw acceleration whatsoever on r6xx/r7xx
corvvs: none at all?
corvvs: oh well
corvvs: anything in the git tree? :)
Nightwulf|work: not even video acceleration since beginning from r6xx you need nearly full 3D support to do 2D stuff
Nightwulf|work: well, 1.2.4 is what's in the git tree :P
corvvs: that sucks - heh, I got this motherboard because I remembered hearing ati was supporting free software drivers for their stuff now
corvvs: I might be picking up a 20 dollar old school radeon (or geforce) in the interim :)
corvvs: I got burned by them saying that back in 2000 too, IIRC - maybe I should stop trusting them :P
Nightwulf|work: corvvs: well, they actually DO
Nightwulf|work: corvvs: but you can't write a driver for a modern card within some days
Nightwulf|work: the documentation for r5xx does consist of >600 pages!
corvvs: ha, gotcha :) I didn't really think about how recent this chip may be (RS780) -- I haven't been paying a great deal of attention to the video card wars since about 2004, so I just got the ati one when I ordered a mobo for my new box
corvvs: out of curiosity, are the docs available online, or is there an NDA or something?
yangman: it'll happen soon
yangman: both DRI code and accompanying doc are still under IP review
yangman: the first round of release is supposedly ready. just waiting for OK
corvvs: ah, ok - they have to make sure they're not treading on anybody's patents or something
yangman: avoid releasing anything that can lead to reverse engineering of the DRM schemas, mostly
corvvs: thanks for the info :)
Nightwulf|work: corvvs: just to state that clearly...I'm not an dev on this project....just a user like you
Zajec: corvvs: it is not about how old is some chip
Zajec: corvvs: the driver is new
corvvs: Nightwulf: well thanks for the info - you've been following it longer than I have, I'm sure :)
Nightwulf|work: corvvs: yep...i had to...my hd3870x2 isn't supported even without acceleration by the official catalyst drivers (still not)
corvvs: Zajec: I've got it now - like I said, I haven't really been paying attention to video stuff for a few years, so I'm used to older cards and just working out of the box. I hadn't been following the project at all until about four hours ago :)
corvvs: well, when I say "working out of the box" I basically mean "not stopping working since 2003" which was the last time I bought any video hardware :)
corvvs: well, thanks everybody
corvvs: see you later
lupine_85: still no docs for r6xx, I take it? ;)
bridgman: so the first release of code & info will include both EXA and Xv for 6xx/7xx
bridgman: trying to get it out in the next week or so; not sure if we'll make Christmas but trying to get something out for the holidays
lupine_85: well, that's nice to hear
phercek: is rv6xx different from r6xx from the point of view of radeonhd implementation?
Mark__T: phercek, afaik r6xx == rv6xx != rs6xx
phercek: Mark__T: that is what I understood from some web info too, but not sure, thanks
maney: hi. following the path of least resistance
yangman: maney: you can just add a minimalistic config file yourself
Jin^eLD: maney: or run X -configure and it will create one for you, you could then have a closer look at it, that's how I did it
Jin^eLD: btw, do I understand this correctly, there is no xv support for rv6xx based cards yet? is there anything usable for watching movies fullscreen?
Zajec: no xv
Jin^eLD: anything else that could be used for fullscreen video playback?
Zajec: you can probably watch in FS using just CPU
Zajec: using "x11" CPU will be used for scaling video
Jin^eLD: does not really work out
Jin^eLD: slowish and not nice at all..
Zajec: well, i think then you will have to wait for xv
yangman: depends on how well the software scaler is implemented in the player
Jin^eLD: Zajec: btw I read that the problem was, getting those specs from ATI, and that this was the main reason why noone started with xv stuff yet
Jin^eLD: has the doc stuff been worked out yet?
adamk_: Jin^eLD, Shortly. AMD is planning a release of an updated r600/r700 DRM hopefully in the next few days, and an updated DDX to enable EXA and Xv.
adamk_: "planning" be the key word... They'd like to get it out before Christmas.
Jin^eLD: sounds nice
Jin^eLD: at least it seems I made the right choice going for ATI, I figured - with nv you simply have no chance to get an open source driver, with ATI it may be a long way, but there one can at least help testing and bug the devs in case things go wrong :)
Zajec: adamk_: DDX with EXA/Xv won't be released in "next few days"... someone quoted yesterday bridgman saying that
Zajec: adamk_: AFAIR in this quote bridgman said about docs this year, mesa with acceledated glxgears in January
adamk_: I just said that on #radeon in the last half hour and bridgman chimed in, concurring.
Jin^eLD: well, I'd already be happy to know that they *will* be release and "soon" is good enough for me :>
bridgman: docs probably won't be this year, we don't have enough review coverage, but still trying to get drm + ddx + demo out this year
bridgman: and the code has all the registers from docs in the header files, just no descriptive info about what the registers & fields do
adamk_: Zajec, That doesn't contradict what I said.
bridgman: but we're trying to provide working code for everything so
Jin^eLD: ...so hope for "xv" in a somewhat near future?
bridgman: btw I have to ask that the word "guess" remain associated with my comments ;)
bridgman: sometimes we get in trouble when our SWAG estimates gradually turn into hard commitments without us ;)
maney: returns, after several distractions... oh, so I shouldn't expect to see a change in glxgears? 'cause it's RADEONHD all over the logs but ol' gears is reporting just the same as under RADEON
Zajec: bridgman: thanks for comment :)
bridgman: I'm not sure of Xv status, will ask agd5f; last I heard was that he had Xv running except for colour space conversion; on 6xx and up we need to do that in a shader rather than having the texture engine do the conversion; we have a first-cut shader written and compiled but not sure if it actually *works* or anything ;)
bridgman: oh well, back to reviewing ;(
Jin^eLD: hm, still sounds promising :>
firewing1: When I use RadeonHD, X won't start unless a monitor is connected... Is there an option I can specify in xorg.conf to get around that?
yangman: firewing1: laptop? or are you actually trying to start X on a headless system?
firewing1: yangman: Yeah, I'd like to remote-control my desktop downstairs via VNC
firewing1: It has a Radeon HD 3850 with no monitors attached at the moment
yangman: I don't know if X is supposed to be able to start without an output
dremdr: can you help me. i want to buy radeon hd 4850. but...
dremdr: i dont know whitch manufaturer to chose, and...
dremdr: the starndard ati desing in not very well, so i want to get some with better cooling ( 85 C is really too much)
dremdr: i have seen Asus have better cooling
dremdr: Saphire Toxic
dremdr: MSI ( it takes 2 slots)