Free software drivers for the Intel 965GM Express Chipset
| From: | Keith Packard <keith.packard-AT-intel.com> | |
| To: | lwn-AT-lwn.net | |
| Subject: | Announcing free software drivers for the new =?ISO-8859-1?Q?Intel=AE?= 965GM Express Chipset | |
| Date: | Wed, 09 May 2007 19:05:18 -0700 | |
| Cc: | keith.packard-AT-intel.com |
The Intel® 965GM Express Chipset represents the first mobile product that implements fourth generation Intel graphics architecture. Designed to support advanced rendering features in modern graphics APIs, this chipset includes support for programmable vertex, geometry, and fragment shaders. Extending Intel's commitment to work with the X.org and Mesa communities to continuously improve and enhance the drivers, support for this new chipset is provided through the X.org 2.0 Intel driver and the Mesa 6.5.3 releases. These drivers represent significant work by both Intel and the broader open source community. In addition to Intel® 965GM chipset support, the X.org 2.0 driver adds native video mode programming support for all chipsets from i830 forward. The driver supports automatic video mode detection and selection, monitor hot plug, dynamic extended and merged desktops and per-monitor screen rotation. These Intel-developed features are built in to the X.org 1.3 X server release and will eventually be supported across most of the open source X.org video drivers. Additional information available on the web: http://intellinuxgraphics.org "Intel's committment to providing high-quality drivers that meet the needs of the mobile Linux community is second to none." Matthew Garrett, Ubuntu Mobile Linux Engineer -- [email protected]
Posted May 11, 2007 0:59 UTC (Fri)
by dhess (guest, #7827)
[Link]
Posted May 11, 2007 1:12 UTC (Fri)
by drag (guest, #31333)
[Link] (20 responses)
But it's not like this is a hell of a lot better then any other graphics company! So Intel won't get my thanks(well I am a bit greatfull. :) ), but they will get my money. (just guess which one they will appreciate the most!)
So this is the chipset that will be used in the "Santa Rosa" Centrino-pro laptops.
Other features of the platform are.. what?
EFI instead of BIOS.
How well is that other stuff supported in Linux?
I see that Intel has not openned up a sourceforge website for their wireless yet.
They have this: http://intellinuxwireless.org/?p=iwlwifi
Iwlwifi, the mac80211 based intel wireless support. Is the 4965AGN support going to come out of this?
Now I know this isn't Intel's problem, but a FS issue.. But is there a way to take advantage of the onboard flash so that I can leave my drive asleep for longer periods of time or get better performance shutting down or starting up?
Maybe stick Ext3 full journalling on this flash device? Or maybe turn up swappiness and put the swap partition on the flash?
I am just curious because I realy would like to get one of these laptops and it would be nice if I could finally get some cutting edge PC hardware support for Linux for once.
Posted May 11, 2007 4:43 UTC (Fri)
by josh (subscriber, #17465)
[Link] (11 responses)
Excuse me? I can think of five graphics card vendors that remain even remotely relevant, and of those only Intel has done anything to support Free Software 3D drivers for their cards. Intel has actually written Free Software drivers, hired some of the top X developers to do it and do it right, pushed X infrastructure changes into X even though that helps other vendors, and released Free Software drivers along with the Windows drivers, before the hardware even goes on sale. Compare that to the other four: Intel gets my hearty thanks, and my money on my next laptop purchase. Thank you, Intel, and thank you, Keith!
Posted May 11, 2007 5:31 UTC (Fri)
by khim (subscriber, #9252)
[Link] (10 responses)
I think you are missing the point. Intel is still far from being good free software citizen: wireless, etc. So no thanks for them. Intel is the best of the bunch and rightfully will get the money (it's far from being good free software citizen, but it's way better then other graphics vendors). Just shows the sorry state we are in drivers-wise: major components include binary blobs, support only limited functionality and/or use reverse-engineered drivers (and in a lot of cases it's all three simultaneously). It's true for products from all vendors (including Intel). BTW it is possible to get Intel graphics without WiFi, for example: my US-sold IBM T43p has Intel's WiFi, but my friends Europe-sold IBM T43p includes something else (I don't remember if worse or better Linux-wise). Unfortunately it's typical story: laptop vendors freely change components without saying anything in documentation or on site. If it's not major (HDD, CPU, Video) - you can get one deal and you friend in the same store - different one for the same model. Rare, but it happens. If we're talking about different countries - it's rule rather then exception :-( At least with Intel you can be sure you'll see pictures on screen!
Posted May 11, 2007 6:24 UTC (Fri)
by drag (guest, #31333)
[Link] (3 responses)
They've made me switch from the faster AMD cpus to the Intel CPUs and Intel-based chipset (which is better then any other PC chipset I've ever used... which I like it a lot)
They got my money. If some other company (say ATI) does one better and actually releases specifications and/or allows developers to operate without having to sign restrictive NDAs then that company will be GODLIKE in my eyes. I will go out and by the newest things, I will post reviews on the internet and give details about how to get the most out of it. I will make sure that everybody knows it.
But there is this irritating trend that Linux developers do.. That is they tend to follow after big industry names and give them a lot of attention while smaller players tend to get ignored. It's dissapointing.
For instance wireless:
You have Intel, which is probably the best supported chipset aviable for Linux. They do code drops, they don't release specifications. For regulatory reasons or something like that.
But they get the attention for this.
And even broadcom cards, they get the attention.
But what about Ralink cards?
These are very common devices. You can go to most any consumer electronic store and pick up cards that use them. They are high quality, usefull, and very cheap.
They are supportive of Linux and other Free systems. They are supportive with hardware, documentation, and drivers.
Why isn't support for these devices in the kernel yet?
Why does Intel (which I am not upset about), and even worse.. BROADCOM have IN-KERNEL drivers and support while Linux users with RAlink cards are still struggling to get them working?!
For craps sake, it's not like these things haven't been around for many years. Why is the Linux kernel developers ignoring them? Or if they aren't ignoring them then why aren't they more supportive?
This is a top notch company, about as good as anybody can expect and yet because Intel has the big name they get the big attention. I WANT to tell people to buy these devices, but I can't because kernel support for them is shit, almost non-existant. I can't be certain that they work or anything... even though they some of the most common and most inexpensive devices you can find.
Drivers can be found at:
"""BTW it is possible to get Intel graphics without WiFi, for example: my US-sold IBM T43p has Intel's WiFi, but my friends Europe-sold IBM T43p includes something else (I don't remember if worse or better Linux-wise). Unfortunately it's typical story: laptop vendors freely change components without saying anything in documentation or on site. If it's not major (HDD, CPU, Video) - you can get one deal and you friend in the same store - different one for the same model. Rare, but it happens. If we're talking about different countries - it's rule rather then exception :-( At least with Intel you can be sure you'll see pictures on screen!"""
Ya..
Well with Intel they've gotten smart a long time ago and that is why they have the 'Centrino' trademark.
In order for a laptop to be labled 'Centrino' you have to have Intel cpu, Intel chipset, and Intel Wifi. So if you buy a laptop with Centrino stickers and you get a broadcom wifi then they are breaking the Intel trademark by selling you the laptop and advertising it as 'centrino'. It shouldn't be difficult to get your money back from something like that.
The only thing that Centrino does not garrentee is the video chipset and I suppose other doo-dads like webcam and such.
So the answer to 'What linux laptop should I get' is fairly easy:
Of course even there is enough differences to causeLinux users problems. Thank goodness for System76.
The 965MG will include the next generation graphics, the ones that come after GMA X3000, maybe GMA X3500 or GMA X4000? I don't know.
It's not a huge difference from the X3000, but they are suppose to support DirectX 10, which is a big deal for windows users.
These should meet the basic requirements for games up to Doom3 or Quake4 in terms of speed and capabilities. I'd probably make sure to get the fastest ram. Also I don't know how optimized they are and what all hardware features the drivers support.
For example with the GMA 950, which I use, the hardware has features for motion compinsation for mpeg2 playback acceleration. Almost a requirement for higher HD resolutions. aka XVMC support.
But the drivers don't support that.
The GMA X3000 should support features like shaders and hardware transformations and lighting, I don't know if the Intel drivers support that though... I don't know of any place I can go to find out this information.
Posted May 11, 2007 8:41 UTC (Fri)
by ca9mbu (guest, #11098)
[Link] (1 responses)
I went and bought myself a Ralink based PCI (rt61 chipset) card because of their supposed good driver support (I didn't want to have to use binary drivers, ndiswrapper, etc.). There are currently 3 different flavours of Ralink drivers (4 if you include Ralinks binary drivers):
1) rt61 - legacy driver, supports WEP encryption.
In the mean time though, confusion abounds as google is filled with out-of-date information, or inaccurate info based on the enormous confusion caused by the various rewrites of the driver and their various configuration quirks. The following maybe full of inaccuracies too but it's the best my failing memory can recall at the moment:
Ubuntu Edgy wanted to load rt61pci by default, which wouldn't scan for wireless networks. One had to blacklist that and get the legacy rt61 module to be loaded instead. Even then, that didn't work for me. Only with Feisty, where they reverted to the legacy rt61 driver can I finally connect to my wireless network, though I've had to drop down to WEP as it doesn't support WPA!
Hopefully Gutsy will come with the new wireless stack, compatible ralink driver and I'll have one less thing to concern myself as it enters the "Just Works" list.
Matt.
Posted May 11, 2007 9:58 UTC (Fri)
by drag (guest, #31333)
[Link]
Then you have to use the rutilT program. http://cbbk.free.fr/bonrom/
It's a very simple program, just scans the network then feeds the essid/channel/ap information to the internet and runs dhclient, but I think the author is able to take into account quirks and such that the normal network manager stuff doesn't.
You have to make sure that you have the device up before starting the program (I put ifconfig ra0 up).
Still though sometimes I have to take the interface down, unload the module, and reload the module for it to work.
And it's just to bad, this driver has been in constant development for about 3 or more years now.
I have owned 3 devices.. rt61 PCI card, rt2500 PCI card, and a Rt2570 USB card. With the Rt61 I was able to even perform some advanced things like packet injection attacks for making wep protected networks spew arp packets. (for research only!! Honest!!)
Also besides the ralink stuff, I've owned a broadcom bcm43xx (Apple Airport Extreme) (which sucked; but thank goodness for the good folks who reverse engineered documentation and the good folks that took that documentation and made it work.), a 'Prism54' PCI card, and a Prism54 PCMCIA card.. both were the original 'FullMac' devices
(totally depressing story about conextent destroying prism54 cards compatibility and then lied about it. Those prism54 drivers were one of the few early Linux 802.11g success stories until Conextant bought up the company that made them).
Posted May 11, 2007 10:21 UTC (Fri)
by smitty_one_each (subscriber, #28989)
[Link]
Posted May 11, 2007 6:28 UTC (Fri)
by elanthis (guest, #6227)
[Link] (5 responses)
Their latest cards allow all of the FCC required crap to be in the firmware, allowing 100% Free drivers with no binary kernel blob or daemon. I don't believe their older cards can do this, but that's a legal issue, not Intel's fault. They went way out of their way to extend hardware functionality just for Free Software users.
You can maybe argue that not giving away firmware source still makes them evil, but I've always found that silly. Those argument disappear if the firmware were hard-coded in the ASIC - the only purpose of firmware is to allow the hardware vendor to fix bugs in the firmware without forcing you to buy upgraded components. It's not something that users can really hack with any meaningful value, unless the firmware is just needlessly crippled - and the wireless firmware is crippled for legal reasons, which any company is definitely going to consider a strong need here in the real world.
"support only limited functionality"
In the case of limitations on functionality in Intel's latest batch of drivers, it's only because the developers just haven't gotten to it yet. The missing features in the X.org drivers are documented and scheduled to be finished this year.
Posted May 11, 2007 6:55 UTC (Fri)
by drag (guest, #31333)
[Link] (1 responses)
I wouldn't buy their devices if it required a propriatory blob, weither or not it existed as a kernel driver or daemon.
Firmware files, while I'd like to avoid, I can live with without complaint.
What we should be aiming for in vendors, though, is still specifications and/or avoid restrictive NDAs.
I can respect NDAs if the intention and purpose and effect is designed to protect things like manufacturing proccess, or future product releases.. Anything that is not nessicary to write drivers for the hardware.
Say if some Linux or X.org hacker wanted to add XVMC support to the Intel drivers...
Would Keith be allowed to help him with this? Would he be allowed to talk about it? Is that other developer going to be stuck pulling the old ATI and Nvidia game with reverse engineering the hardware and mucking around capturing traffic between the hardware and the Windows drivers?
Is that proper, so that your still not allowed to know how to operate the hardware, that developers can't talk about it, can't spread information or use their skills to help get support for other hardware?
So that's still distastefull for me. I care more then just about being able to redistribute source code, being able to communicate and share knowledge is very important, even potentially more important.
Although I'll take what I can get in terms of hardware support.
Posted May 13, 2007 4:29 UTC (Sun)
by keithp (subscriber, #5140)
[Link]
Of course I'd love for that to happen, our list of things we'd like to see the drivers do is fairly extensive. But, unless we can provide reasonable documentation, it's going to be really hard to break into a big new area of functionality like this. The current header files document what the driver uses, but not a lot beyond that.
What we are doing is trying to get some minimal XvMC support done this year so that the hardware functionality is reasonably well described in both code and comments. At that point, I'm hoping others will be able to help expand support for additional encodings, and other activities.
Any help we can get will be appreciated. Questions about specific chip operations that can be answered by consulting the specs are something I'd like to spend more time answering.
Posted May 11, 2007 18:09 UTC (Fri)
by bfields (subscriber, #19510)
[Link]
So however silly the centrino branding may otherwise be, for new laptops it looks like it will provide some reasonable assurance that the things I care about will Just Work on any sufficiently recent free linux distribution.
Posted May 13, 2007 4:41 UTC (Sun)
by keithp (subscriber, #5140)
[Link] (1 responses)
The big ticket items on the list for graphics are OpenGL 2.1 and hardware accelerated video playback. Other than that, the new drivers expose quite a bit more of the hardware than the 1.3 version, with native mode programming for everything we could get our hands on.
OpenGL 2.1 support is largely a matter of working with the Mesa project to add needed functionality to the library and DRI subsystem in the kernel. We're busy on both fronts there; Zou Nanhai is busy taking the new GLSL compiler and generating Gen4 instructions, and Eric Anholt is working on the TTM memory system to finally provide for FBOs and other necessary features.
For media playback, Xiang Haihao is working on XvMC support. Adding iDCT support and H.264 decode support is also on the list, but those are limited by the lack of any API capable of exposing them to applications. We're working with some of the media playback developers to figure out what might work here.
There's lots of work to do just to get things functional; beyond that, we've got plenty of low-hanging optimizations that should keep everyone busy.
Posted May 13, 2007 11:36 UTC (Sun)
by drag (guest, #31333)
[Link]
You guys working on this stuff definately deserve praise. It's going to be great when nobody can accuse 'Open Source Graphics Drivers' of being 'Second Class Graphics Drivers'
Every time I look at projects like OpenCroquet, Beryl, OpenRT, and other such things.. and relatively lack of stability of drivers (closed and open source) it just makes me think about how much all this ultra-proprietory attitude is realy holding back progress.
The technology X is now able to bring to the table is amazing. You have things like DMX for X sessions spreading across multiple computers. Chromium for distributed OpenGL acceleration.
Then support for multiple pointers in MPX. Then you have Multi-seat support in X, for true multi-user PC environments. The ability to migrate applications across displays with xmove.
Seriously, using Blender over a encrypted ssh with hardware acceleration thanks to AIGLX makes me giggle.
People talk about oh web2.0, but I don't think that the web's REST architecture is realy suitable for what people are trying to do with hosted applications and such.. but FreeNX allows me to use my desktop quite well over the internet. X on a modern 'broadband' internet, I think, has real potential.
And all sorts of stuff like that.. it's all very exiting and compelling. But it's all limited by the lack of very good drivers for most people.
Right now Intel 945g chipset is the only graphics that have stable 3D and 2D acceleration support out of box for Linux.
If Intel is able to get things stable and help/let you guys keep/get things open for the high-end graphics that people say are coming in a couple years then I think that we can start expecting very wonderfull things with X, Linux, OpenGL, and Intel.
Posted May 11, 2007 5:27 UTC (Fri)
by walken (subscriber, #7089)
[Link]
That said, I'm quite happy in general with Intel's work on linux drivers for their platforms. This was actually one of the factors I considered when choosing a motherboard. At this point I think Intel's doing much better than their competition in that regard.
Posted May 11, 2007 9:37 UTC (Fri)
by Zenith (guest, #24899)
[Link] (1 responses)
Taken from this Ars Technica news article.
Posted May 12, 2007 9:21 UTC (Sat)
by njs (subscriber, #40338)
[Link]
Posted May 11, 2007 12:48 UTC (Fri)
by brugolsky (guest, #28)
[Link] (4 responses)
Intel opened the specs to AHCI, and that is doing for SATA what ATAPI did for CD-ROMS. [ Remember when nearly every CD-ROM interface had its own driver? ] We now have good AHCI support.
The efforts to open up the WiFi drivers, and to do it in concert with building a better core WiFi stack, is also the right move.
On the graphics front, Intel is playing catchup, and this is a good play for them. Their integrated graphics produces significant power savings over discrete graphics components.
Where they really fall down is in documenting the rest of their chipsets, because they push utter crap like ACPI, EFI, DRM etc., and refuse to cooperate with LinuxBIOS. I spend a disproportionate amount of time trying to work around BIOS deficiencies. The BIOS from major vendors like DELL is worse than useless. If Intel doesn't want workarounds like ACPI DSDT override in the kernel, then they should spend time and money getting vendors to fix the BIOS. Where's the "Intel-Certified" campaign?
AMD, on the other hand, has published great technical specs of the components that they produce: the AMD Opteron and Athlon64 processors, basic chipsets, etc. Unfortunately, their partners including Nvidia and recently acquired ATI have been less forthcoming, so this great advantage has been squandered.
Posted May 11, 2007 12:59 UTC (Fri)
by drag (guest, #31333)
[Link]
BTW with the Santa Rosa platform they claim they are going to use EFI instead of a BIOS. It'll be interesting to see how it affects you.
(well most of it, the mac80211 (which seems to be the Linux stack of the future) was originally from DeviceScape, a wireless hardware provider that used Linux in it's designs. Of course it had to be beaten into shape to make it work with kernel.org Linux properly. It's due for inclusion with 2.6.22. The Intel stack has grown into disfavor and on their latest devices intel has rewriten their drivers to use the mac80211)
I like AMD a bit more then Intel.. but Via seems to have given up the ghost on new chipsets, Nvidia are bastards when it comes to Linux support and ATI is worse.
I just realy hope that AMD rubs off on ATI much much more then ATI rubs off on AMD.
Posted May 11, 2007 14:20 UTC (Fri)
by arjan (subscriber, #36785)
[Link] (2 responses)
Intel is actually actively working on helping Dell and others get the BIOS "mess" right: http://www.linuxfirmwarekit.org is done EXACTLY for that reason, and Dell, HP, IBM and many others are now using this.... it doesn't solve existing systems but new ones have a standing chance now.
If you have a specific bios bug you run into and you think it'd be useful to test for... let us know.
Posted May 14, 2007 13:40 UTC (Mon)
by brugolsky (guest, #28)
[Link]
As I'm sure that you are aware, if you want to help Linux on Intel boxes, get Intel to participate in LinuxBIOS, and just get us to userland as quickly as possible. After putting the machine in the rack and cold booting, I have almost no use for the BIOS - I want to kexec into my next kernel, and I want the hardware to work properly. If SMI is needed for anything, it should be
fixing up my hardware state when I kexec -- other than that, it should
get out of the way.
I am currently grappling with trying to force-enable HPET on a number of motherboards -- between the HPET doc and the patches posted to LKML, I'll probably get it working - but what a waste of time, and why? Serial consoles on a large number of motherboards have inconsistent settings (some do 9600, some only do 57600, some on ttyS0, some on ttyS1), have broken terminal handling, and hang in various ways. A hint to BIOS vendors: terminal escape sequences and lack of flow control don't mix well; the whole business of exporting the console display is just braindead. I want a boot log, I want it stored somewhere, and I want to be able to dump it into /var/log/ once my machine is up. Why it can't have a real settable verbosity level. Even GRUB can do better than that. We have Intel-SE7320VP2 motherboards with Marvel and Intel E1000 NICs; of course the machine can't PXE boot from the E1000. Every BIOS seems to have a different PXE enable/disable behavior, and it presents so many headaches in a VLAN and security rich environment that we use USB sticks on our DMZ machines instead.
My brand-new Core2 Duo ICH7-based laptop is currently using ata_piix, for lack of a config option to switch to AHCI. :-(
I went to the EFI sessions at OLS a few years ago, and I was appalled. I apparently still can't do something like:
Perhaps the engineers should have a look at the "lxbios" tool.
Where are the config commands that would allow me to do a "setpci" to workaround a misconfigured bridge? Perhaps it is there now, but I wouldn't know, because I walked away in disgust.
EFI introduces yet another driver model. One would think that nobody at Intel is aware of the problems with MILO, SILO, GRUB, OpenBoot etc. EFI is so bloated, many BIOS flash parts now have enough room for a whole mini Linux distro. I guess I should count that as a blessing; someday LinuxBIOS may run on the handful of new Intel chipsets, and then we'll have plenty
of room to install a mini-distro in flash.
Posted May 17, 2007 16:29 UTC (Thu)
by zooko (guest, #2589)
[Link]
I wish that there was a central, organized, updated web site with all of this kind of information so that I could go to that web site before buying any new hardware. The FSF has a list [2], but somehow it doesn't seem to be complete, informative, organized, or fresh enough for me to trust. (It doesn't help that I submitted a couple of e-mails to them giving information about some hardware and they never wrote back nor, as far as I noticed, added my information to their list.)
[1] http://www.fsf.org/campaigns/free-bios.html
[2] http://www.fsf.org/resources/hw
Posted May 11, 2007 7:16 UTC (Fri)
by pheldens (guest, #19366)
[Link] (19 responses)
Thanks.
Posted May 11, 2007 8:18 UTC (Fri)
by drag (guest, #31333)
[Link] (18 responses)
Anybody here with a 965G care to talk about 3d performance?
glxgears -info ?
Tremulous, torcs, Nexiuz, and such should be aviable via apt-get if your using Debian. You should be able to get quake3 performance or enemy-territory performance results pretty easily.
Posted May 11, 2007 8:21 UTC (Fri)
by drag (guest, #31333)
[Link]
It looks very simple to do benchmarks with it...
Posted May 11, 2007 8:40 UTC (Fri)
by bobort (guest, #5019)
[Link] (16 responses)
I don't have any performance numbers handy, and we've never used them for games.
Both of them crash the machine far too often, but G965 is worse at present.
Posted May 11, 2007 9:04 UTC (Fri)
by nix (subscriber, #2304)
[Link] (15 responses)
Maybe you have some other hardware problem, or you're doing heftier things than I am...
Posted May 11, 2007 12:15 UTC (Fri)
by tialaramex (subscriber, #21167)
[Link] (4 responses)
So far as I can tell the big two vendors (I haven't tried Intel) design hardware for performance first and foremost, then cost, and reliability is definitely an afterthought. It would be possible to design these chipsets so that you had to do something fairly stupid to lock them up, but instead they're incredibly sensitive to everything from timing to data layout.
The vendors own drivers aren't able to make this stuff actually stable, as gamers on Windows soon learn. You can expect that any new game will cause a few crashes or lockups until the drivers get tweaked to make them less likely or you lose interest in playing. This isn't with "known bad" hardware either, a mid-range IBM Thinkpad laptop running World of Warcraft might crash to the Windows "Your video system died and I can't fix it" screen once a week or so, leaving the player to reboot their PC. Players just learn that this is "normal" and put up with it.
Posted May 11, 2007 12:48 UTC (Fri)
by drag (guest, #31333)
[Link] (3 responses)
If you think about it people have been saying that 'software rendering' is the future for a long time now. And although very slow software rendering is very stable and gives a very good visual image. (if you don't mind slideshows, of course)
Intel seems to want to go with a approach that leads to a massive number of very simple x86-like cores that able to perform massive amounts of parrallel graphic-related calculations very quickly.
Make x86-like to make it easy for normal folks to program. Have them all pipelined together so you can essentially program your own GPU to match what you need. Then since it's generic you can program it for lots of different sorts of stuff like media encoding or maybe OpenRT (realtime raytracing).
Then since they are x86-like it makes sense to me that with some code optimizations you could recompile Mesa to use these devices and get essentially hardware-acceleration speeds with these cores. Pretty much remove most of the distinction between GPU and CPUs.
They are aiming to release a discrete video card in 2009 that is designed to compete head to head with Nvidia. Supposedly up to 16 cores for the high-end model. Several times more powerfull then the current high-end cards.
Posted May 11, 2007 20:33 UTC (Fri)
by nix (subscriber, #2304)
[Link] (1 responses)
Posted May 11, 2007 22:56 UTC (Fri)
by drag (guest, #31333)
[Link]
I just hope that Intel does open up everything enough that just your normal average C hacker will be able to make stuff to configure and run on these cores themselves.
It would be a very cool to be able to leverage your video card's proccessing power to do special things like rendering raytracing, or do media encoding and such things. I think that we'd be a lot further along in terms of 3d graphics and virtual environments, also, if it wasn't for these things being so damn proprietary all the time.
Posted May 12, 2007 9:36 UTC (Sat)
by njs (subscriber, #40338)
[Link]
It'd be nice if it were possible to get at those massively parallel CPUs without running Windows, though.
Posted May 11, 2007 12:48 UTC (Fri)
by rankincj (guest, #4865)
[Link] (4 responses)
May I ask which versions of X and Mesa you are running, please?
Posted May 11, 2007 16:04 UTC (Fri)
by AJWM (guest, #15888)
[Link] (2 responses)
Posted May 11, 2007 20:35 UTC (Fri)
by nix (subscriber, #2304)
[Link] (1 responses)
Posted May 11, 2007 23:54 UTC (Fri)
by AJWM (guest, #15888)
[Link]
Posted May 11, 2007 20:35 UTC (Fri)
by nix (subscriber, #2304)
[Link]
Posted May 11, 2007 15:16 UTC (Fri)
by bobort (guest, #5019)
[Link] (4 responses)
The G965 machines crash 3-5 times/day on occasion when somebody is really using the 3D hard. I've seen several reports of this same crash from others as well (it emits a DRI error before hanging, so it's also pretty easy to spot). I have higher hopes for this to get fixed in the future, G965 is still pretty new. I doubt the r200 crash will ever be fixed at this point, it's probably a hardware bug that ATI isn't disclosing the workaround for.
Posted May 11, 2007 17:02 UTC (Fri)
by airlied (subscriber, #9104)
[Link] (2 responses)
Posted May 11, 2007 20:37 UTC (Fri)
by nix (subscriber, #2304)
[Link] (1 responses)
So it's not just that VIA sucks (although this VIA-based Athlon 4 system
Posted May 11, 2007 23:07 UTC (Fri)
by drag (guest, #31333)
[Link]
These things were originally designed for passive cooling and are just very massive. Suprisingly they aren't not heavy, you don't have to worry about the weight of the thing tweaking heatsink mounts or anything like that.
I have one on my 2.8ghz dual core Pentium-D proccessor. Pentium-Ds are infamous for being a very hot CPU, essentially being 2 Pentium4 cores.
I made a small duct for it out of stiff paper going back to the rear exhaust fan, which is a 800rpm 120mm fan. Then directly above the cpu I have the intake for the powersupply, which is another slow 120mm fan.
Those two fans are the only fans on my system and they keep my cpu ticking at barely above room tempurature. It's just amazing how efficient these heatpipe heatsinks can be.
The only serious problem is that you need to make sure that adiquate cooling is aviable for the big power conditioners/transistors typically found close to the cpu socket. If you overclock these will get very hot without air directly blowing down on them like you find a traditional cpu heatsink. Small aluminum heatsinks with thermal epoxy are more then adiquate to take care of that.
but I don't overclock so it's a non-issue.
that's another thing I like a lot about the onboard video: No need for a VGA card fan. This does wonders for making things quieter.
My old Dell 600mhz firewall hidden in the closet makes more noise through the door then my desktop currently does when running full-tilt.
Posted May 23, 2007 12:20 UTC (Wed)
by daenzer (subscriber, #7050)
[Link]
Those are the symptoms of a GPU lockup, which can be caused by a lot of different things.
Keith, if you're out there, can you post the output of Which OpenGL extensions?
glxinfo running on this driver?
To bad your not releasing specifications necissary to program with teh drivers also.Free software drivers for the Intel 965GM Express Chipset
NAND-based Flash memory to augment disk drive.
IDA, Intel Dynamic Acceleration.. WTF is that?
And then the new 4965AGN for 802.11a/b/g/n support.
possible WiMax?
I know this is about the graphics chip, but you can't get the graphics without this other stuff.. it's a bundled item.
Free software drivers for the Intel 965GM Express Chipset
But it's not like this is a hell of a lot better then any other graphics company! So Intel won't get my thanks...
Free software drivers for the Intel 965GM Express Chipset
What do you think Intel wants more? My gratitude or my money?Free software drivers for the Intel 965GM Express Chipset
http://rt2x00.serialmonkey.com/wiki/index.php/Downloads
http://rt2x00.serialmonkey.com/wiki/index.php?title=Rt2x0...
make sure to get a Centrino laptop with Intel onboard graphics.
"Drivers can be found at:Free software drivers for the Intel 965GM Express Chipset
http://rt2x00.serialmonkey.com/wiki/index.php/Downloads
http://rt2x00.serialmonkey.com/wiki/index.php?title=Rt2x0..."
2) rt61pci - rt2x00, purportedly supports WPA encryption but I've never got this driver to be able to scan my wireless network. Even if it did, NetworkManager wouldn't be able to deal with it because it doesn't use WirelessExt to provide this support. From what I've read, I think this means one has to download binary firmware blobs and edit a config file to get WPA support, but can't confirm that.
3) The new git head code - this is based on the mac80211 stack. Now that things have started settling down in the upstream kernel in terms of the favoured/de-facto standard wireless stack I'm pretty confident that this version of the driver will find its way upstream too, and all of us with Ralink hardware will get out-of-the-box support for it. You can bet your bottom dollar on there being plenty of announcements when this (and other out-of-tree drivers that are in a similar position) get merged upstream!
For best performance I found you'll have to completely get rid of network manager, plus any thing like dbhcdbd that cache dhcp information.Free software drivers for the Intel 965GM Express Chipset
drag:Free software drivers for the Intel 965GM Express Chipset
You got it.
While abstractions are important, the metric that alters corporate behavior is cash.
I bought a Dell D620 from emperor linux and chose intel across the board, and was rewarded by getting better video performance as a result of this driver.
Now, you could ask yourself, what mysterious reason makes video and wireless drivers lag so much in the open source realm, but why screw up a lovely Friday with caustic speculation?
Best,
Chris
"Intel is still far from being good free software citizen: wireless, etc."Free software drivers for the Intel 965GM Express Chipset
Ya.. moving the regulartory crap to the firmware was a very good move for Intel.Free software drivers for the Intel 965GM Express Chipset
> Say if some Linux or X.org hacker wanted to add XVMC support to the Intel drivers...Free software drivers for the Intel 965GM Express Chipset
Yeah, free software to run on the actual wireless card itself would be way cool, but having at least fully free (and, hopefully soon, mainline) kernel drivers is a big step in the right direction.Free software drivers for the Intel 965GM Express Chipset
> "support only limited functionality"Free software drivers for the Intel 965GM Express Chipset
Wonderfull. Free software drivers for the Intel 965GM Express Chipset
If we're going to turn this into a laundry list, I wish there were lm-sensors drivers for my two Intel boards (DQ965GF and DG965RY). This is the only thing I have not got to work on my systems - actually I did not even think about checking it before buying because I never even thought it might not work. Oh well.Free software drivers for the Intel 965GM Express Chipset
Free software drivers for the Intel 965GM Express Chipset
IDA, Intel Dynamic Acceleration.. WTF is that?
The thing you're talking about is real, but I believe IDA is something different. It's hard to find good technical details (or I'm not very good at finding them, at least), but it looks like IDA has to do giving dual-core chips more flexibility when dealing with serial or asymmetric loads -- supposedly by ramping up the clockspeed on the chip that's actually working, while the other one sleeps in a low-power state, thus making more effective use of the same number of watts.Free software drivers for the Intel 965GM Express Chipset
Much of Intel's open source strategy seems to focus around the strategic value of its chipsets and the Centrino brand. While AMD processors were handily besting them on performance, Intel was busy standardizing its platform components and ensuring good support for them.Intel's uneven embrace of Free Software
I absolutely agree with you. Intel's uneven embrace of Free Software
(disclaimer: I work for Intel)BIOS compat issues
Thanks, I took note of your original announcement, and then it passed off my radar. This and PowerTOP look useful, and I certainly appreciate the efforts of you and others at Intel to improve the situation. So take the following as the rantings of someone driven mad by machine room noise. :-P
The problems is not compatibility, it is a broken model.
Is that the "sham open source bios" project that the FSF linuxbios people decry? [1]BIOS compat issues
I'd like to see some real world quake3/doom3 performance please, no glxgears.Free software drivers for the Intel 965GM Express Chipset
Does it outperform Ati X850 on r300_dri.so? or even 9250 on r200_dri.so?
hear hearFree software drivers for the Intel 965GM Express Chipset
Vdrift is aviable at least through Debian unstable apt-get..Free software drivers for the Intel 965GM Express Chipset
http://www.michaellarabel.com/index.php?k=blog&i=133
We use both r200 and G965 for molecular visualization type stuff. Comparing a 2x2.4GHz opteron w/ r200 vs. 2x2.9GHz core2 w/ G965, the latter is noticeably faster/smoother, but not tremendously so. It's hard to say whether the cpu or gpu is more responsible for this, and since r200 is AGP-only and the Intel boards are PCIE-only a proper comparison is impossible.Free software drivers for the Intel 965GM Express Chipset
Interesting. I wonder what's different between your system and mine? My r200 (9250) has not once caused a crash in X, no matter what I did with it.Free software drivers for the Intel 965GM Express Chipset
I've owned a variety of hardware with R100 and R200 family chipsets. On some of the hardware 3D it crashes pretty easily (e.g. one machine wouldn't last more than a few minutes running Quake 3), on other hardware it's quite stable, but it's still definitely the least reliable part of the system.Free software drivers for the Intel 965GM Express Chipset
Hopefully improvements in technology will render most of the traditional 'discrete hardware acceleration' mostly obsolete.Free software drivers for the Intel 965GM Express Chipset
Sounds more like a Cell's SPUs than anything else to me, only a SPU with Free software drivers for the Intel 965GM Express Chipset
high-speed access to the video controller.
Well you know, in terms of cpu technology IBM has always lead the pack.Free software drivers for the Intel 965GM Express Chipset
Modern graphics cards AFAICT basically *are* general purpose processors running a custom-built software renderer -- the hardware is a bit different from your average CPU (massively parallel array of little chips with weird instruction sets and cache structure), but whatever. It's a supercomputer, except it retails at ~$1k and it goes in a PCI slot. (I know at least one person who is literally using a high-end NVidia to replace an honest-to-goodness cluster).Free software drivers for the Intel 965GM Express Chipset
Interesting - I own a Radeon 9200 and it has definitely been known to crash X. (celestia, WoW etc.) In fact, the other week it brought my entire PC down with it! However, I have been assuming that this is because I am on Mesa's bleeding 7.0 edge.Free software drivers for the Intel 965GM Express Chipset
Just as a datapoint, I have a 9250 and haven't had any problems - celestia, flightgear flightsim, etc. are just fine. Mesa 6.2.1 and the card's in an AMD64 box. Using the open source driver, of course. Mostly though I just do 2D, I only picked up the card for flightsim.Free software drivers for the Intel 965GM Express Chipset
Mesa 6.2.1 is a bit old, isn't it? Do you mean 6.5.1?Free software drivers for the Intel 965GM Express Chipset
No, 6.2.1 is correct. It's a relatively old install (Suse 10.0) and I haven't got around to updating it yet.Free software drivers for the Intel 965GM Express Chipset
No trouble with xserver 1.1.1 or 1.2.0 and xf86-video-ati 6.6.3 combined Free software drivers for the Intel 965GM Express Chipset
with Mesa 6.5.0 or 6.5.2. (All compiled from upstream sources.)
We have 2 G965 machines and 7 r200 machines, and they all crash. The r200 bug has been well known for 2-3 years and many others have reported it (it involves the X server spinning the CPU with a particular set of ioctls, it's pretty easy to recognize). It happens much more often when there are 2 DRI apps running simultaneously. I have an r200 at home that I've never seen crash, but I hardly ever use DRI there.Free software drivers for the Intel 965GM Express Chipset
A lot of the r200 crashes were AGP bugs that the open source world never found about it.. so r200 in Intel AGP might be fine, stick it in via or amd and it sucks..Free software drivers for the Intel 965GM Express Chipset
00:01.0 PCI bridge: VIA Technologies, Inc. VT8366/A/7 [Apollo KT266/A/333 Free software drivers for the Intel 965GM Express Chipset
AGP]
has been persistently prone to spontaneous lockups and is terribly
overheating-sensitive: CPU temp above about 50C means a lockup, so it
needs a hugely overspecced fan. Perhaps I should underclock it.)
If you have the space for it I suggest taking a look at the massive scythe ninja heatsink.Free software drivers for the Intel 965GM Express Chipset
http://www.silentpcreview.com/article251-page1.html
> The r200 bug has been well known for 2-3 years and many others haveFree software drivers for the Intel 965GM Express Chipset
> reported it (it involves the X server spinning the CPU with a particular
> set of ioctls, it's pretty easy to recognize).
Copyright © 2007, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds