Mark Shuttleworth, the self-made millionaire and leader of the Ubuntu project, has been very vocal about the adoption of free software and that "the free software approach is a better device driver development model." But what does Mark think about AMD's announcements this week with the 8.41 display driver and the just-announced program where AMD will be handing out specifications under NDA and helping out the open-source community? Mark Shuttleworth has provided Phoronix some of his initial thoughts on ATI/AMD's new Linux push.
Rumors and speculations have been flying around for months about ATI/AMD opening up the source-code to their Linux display driver or providing their GPU specifications to community developers. This for the most part had started after Henri Richard's statement at the Red Hat Summit earlier this year. Well, those rumors can finally be put to rest. AMD will be providing NDA specifications, an open-source library, and there is a new open-source graphics driver as a result. AMD will continue producing a closed-source proprietary driver; however, they are opening the source-code to a critical library with accompanying GPU specifications for X.Org developers. To get the ball rolling, AMD is also funding the development of a new open-source R500/600 driver.
So far today at Phoronix we've published the AMD 8.41 Display Driver Preview, ATI R300/400 Linux Performance, ATI R500 Linux Performance, and ATI Radeon HD 2900XT Linux Performance -- yet, it's only ten minutes past midnight! However, looking at AMD's official statement on the Linux driver, there is a rather interesting sentence about good things to come...
Since Linux support for the Radeon X1000 "R500" series was introduced some 17 months ago, there has been a performance imbalance. The past fglrx drivers have worked, but they've had a difficult time competing with the competition presented by NVIDIA and their Linux driver when it comes to the frame-rate performance. Though as we have been sharing in all of our articles today, the new ATI/AMD Linux driver will be made available in the coming days and it offers some striking performance improvements. In our Mobility Radeon X300 testing, the new driver allowed Doom 3 and Quake 4 to run at playable frame-rates, which simply wasn't possible with the previous 8.40.4 fglrx driver a month ago! The Radeon X800XL performance was also improved significantly, which allowed it to outperform the NVIDIA competition. Now onto the Radeon X1000 family, as you can probably suspect, there are massive performance improvements to be found with these graphics cards as well. In this article we have tested the Radeon X1300PRO, X1800XL, X1800XT, and X1950PRO with the new ATI Linux driver and have found significant performance improvements.
To some extent, ATI's R300 and R400 series is more popular than the R500 (and now R600) series for Linux users. The R300/400 series has a reliable open-source driver and while the performance of the X.Org Radeon driver lags behind ATI's binary driver, it's currently the fastest for offering open-source 3D performance on dedicated graphics cards. With the cards being around much longer, they are also much cheaper and have become somewhat popular for Linux desktops that can power Compiz and Beryl "eye candy" effects. However, it's just not the R500 and R600 series that receive a nice performance boost from the new ATI/AMD Linux driver, but so does the R300/400 series. Using the fglrx 8.41 driver on this older graphics hardware will allow for a sizable performance improvement compared to the older binary drivers. Much of AMD's focus is on the R500 and R600 series, as is our focus, but we've completed some benchmarks comparing the new and old fglrx drivers using an ATI Mobility Radeon X300 64MB and ATI Radeon X800XL 256MB graphics card.
While many thought this day would never come or that it was some form of propaganda, it's coming and it's coming this month. The AMD fglrx 8.41 driver features a brand new underlying code-base that has been under development for well over a year. At Phoronix we have literally spent hundreds of hours using and testing this new driver, and to say the least it is truly a new experience on the ATI Linux front. The new driver delivers massive performance improvements, Radeon HD 2000 (R600) support, and a whole lot more. This article is the first of five articles that will be published today at Phoronix as we examine the new AMD Linux driver extensively on all fronts from seeing how it can handle the flagship ATI Radeon HD 2900XT graphics card to going back to past graphics card families to show the dramatic performance improvements.
On the Phoronix Forums we have been running a Q&A with the developers of the Nouveau project. For those out of the loop or new to Linux, the Nouveau project aims to provide an open-source 2D/3D graphics driver for NVIDIA hardware. After collecting a number of questions from our readers, KoalaBR and Marcheu have answered these questions. The questions range from whether there will be open-source SLI support to asking if NVIDIA has ever contacted the Nouveau developers.
Back in April of 2006, a proprietary driver from ATI that had supported the Radeon X1000 "R500" product family had finally greeted Linux users. This driver, v8.24.8, had supported the entire desktop and mobile Radeon X1000 lineup with 3D support and even Avivo video playback capabilities. For the six months prior, Linux users were stuck in the dark without any Linux support for the R500 series while the ATI Windows Catalyst customers had support that was continually improved. Of course, back in 2006 there was no open-source R500 driver either. Over the past 16 months with R500 support in the fglrx driver, the features have continually improved with an AMD Catalyst Control Center for Linux, support for the newer R500 graphics cards, and there are far less bugs in the driver now then there was in the past. However, these changes really have not affected the performance as it has stayed the same for the most part with minimal improvements. In this article today we are going back to test the original 8.24.8 display driver and compare it directly against the latest driver available as of today, the 8.40.4 fglrx release.
Scheduled for release tomorrow is X.Org 7.3. Among the new features for X.Org 7.3 include the Xorg server 1.4, an application for adjusting a display's backlight, updated display drivers, and support for font catalog directories. Version 1.4 of the X.Org server contains such features as RandR 1.2 support, input hot-plugging, KDrive enhancements, Solaris DTrace support, and EXA improvements. In this article today, we will briefly go over some of the changes found in X.Org 7.3 and we will follow up with some benchmarks in early September.
One of the new features to land with Ubuntu 7.10 Gutsy Gibbon Tribe 5 is the displayconfig-gtk. This new GTK control panel is similar to NVIDIA's nvidia-settings panel and ATI's Catalyst Control Center, which allow for GUI-based modifying of your xorg.conf to control such options as the display resolution, setting up multiple display heads in either an extended mode or mirroring, and changing your graphics driver. In this article, we are taking a brief tour of this early displayconfig-gtk application.
Through a joint effort between Phoronix and the Nouveau project, we will now be publishing the Nouveau Companion on Phoronix to update the progress made by these developers working on an open-source display driver for NVIDIA's complete selection of hardware from the NV0x chips to the latest models in the GeForce and Quadro series. This is the 26th edition of the Nouveau Companion which provides an updated status on the Nouveau driver with the progress being made in the areas of X-Video, video overlays for the different NVIDIA GPU series, and renouveau-parse for parsing XML dumps. These developers still need help especially with the NVIDIA GeForce 8 series hardware.
Thanks to recent commits in the RandR 1.2 branch for the open-source X.Org Radeon driver, it's now possible to use S-Video and composite TV output on your ATI graphics card without any patches. This is not limited to the R200 series but will also work with the R300 series and theoretically any graphics card supported by the xf86-video-ati driver. In this guide have outlined the instructions for enabling TV output support from the Radeon driver git code as well as some of the current limitations.
Last month marked the introduction of Fedora 7 support with the fglrx 8.39.4 ATI/AMD driver but not much more than that was to be found aside from an undocumented XML file. This driver did also present watermark problems that led to the driver being recalled and a corrected driver being pushed out. Today the fglrx 8.40.4 driver has been released and while many had speculated the 8.40 series would mark the introduction of the new Linux driver code-base, that's not the case for 8.40.4 driver. What changes does the 8.40.4 driver hold in store for Linux users? We'll tell you today but there isn't much.
Jerome Glisse and his posse of open-source developers have been making good progress with an open-source "Avivo" graphics driver for the ATI Radeon X1000 (R500) series hardware. However, one roadblock they have hit along the way is with TMDS setting issues on the Radeon X1200, X1300, X1400, and X1900 series. This roadblock had also led to postponing the Avivo 0.1.0 release. Now if you are an ATI R500 owner and have been wondering how you can help with the advancement of this open-source driver, there is a way even without prior development experience and that is by providing VBE mode-setting BIOS dumps. As you have probably never created such dumps before, we have written a brief tutorial on using vbespy/vbetest for the first time.
Since publishing our Avivo versus fglrx driver GtkPerf benchmarks that compared the GTK performance between the community open-source driver and ATI's official driver, we have received a number of requests for more of these 2D benchmarks with different graphics cards and different drivers. While this is not one of our formal articles, we have completed a few more GtkPerf tests with NVIDIA and ATI graphics cards to see how the GTK performance stacks up.
If you have been reading our ATI/AMD Linux display driver reviews for some time, you will know that there are periods where it doesn't look like the fglrx driver is actively being worked on, but in fact changes are being made "under the hood". We are going through one of these droughts right now where not many new features or bug fixes are being introduced, though a lot of work is going on internally. We have decided to dissect the last 19 months of driver releases from ATI/AMD to expose some interesting facts and what should be coming in the future.
Last month was an interesting time for AMD and their ATI Linux display driver. The Radeon HD 2400 and 2600 series were introduced, but as we have unfortunately come to expect, there was not a supported R600 driver that day or even that month. In fact, there still is no Linux driver to support any graphics card in the R600 series. Though making the month unique were two display drivers being released in the same week. The 8.38.6 driver had introduced Red Hat Enterprise Linux 5 support as well as some minor fixes, but nothing to spark interest in the minds of end-users. The second Linux driver released, 8.38.7, had come down the pipeline as a hot-fix release to correct a bug where aticonfig would crash and remove your xorg.conf. Moving forward to the 8.39.4 driver released today, we still have no AIGLX or R600 support, but we do have Fedora 7 support!
While the Avivo driver doesn't yet contain 3D functionality or support a number of features found in the official fglrx driver and the community Radeon driver, it is making steady progress despite its age. Most recently with the open-source R500 driver implementing shadow frame-buffer support, we have experienced a noticeable increase in performance. As we have begun to receive messages from those interested in this driver wondering about the performance capabilities, we have carried out a brief GtkPerf test comparing the Avivo git code to ATI's official binary "fglrx" display driver.
It was just a month ago that the open-source Avivo driver for the ATI Radeon X1000 (R500) series was introduced to the public, but in this time we've seen some great progress made. This open-source R500 driver now contains RandR 1.2 support, support for a variety of R500 graphics cards, and most recently support for Shadow Framebuffer was added. The Avivo driver still isn't comparable when it comes to the features found in the fglrx driver or even the open-source Radeon driver for the R200/300/400 series, but it's a work in progress. If you are running into problems with the fglrx driver, stuck using the VESA driver for one reason or another, or just want to get rid of the binary blob and experiment with this open-source driver, we have written a guide for setting up the Avivo driver from source on Ubuntu.
It was earlier this month that version 2.1.0 of the xf86-video-intel driver was released, which among other things had introduced open-source Linux graphics support for the G33, Q33, and Q35 chipsets as well as fixing a horde of bugs and adding PCI IDs for the 945GME, 965GME, and 965GLE chips. As our last Intel graphics performance article was looking at the Q965 back in May, in this article we have enclosed some benchmarks from Intel's GMA 950 IGP using the new xf86-video-intel 2.1.0 driver.
It was on July 20, 2006 that I had issued The State of ATI Linux while ending off the ATI Redblog, which was a fifty-day experiment for using the ATI fglrx driver under Linux exclusively to see how well the driver really could compare to that of NVIDIA's binary competition. It's going on a year later and it's now time for this year's address as far as what I have seen from the driver in the past year and where I hope and believe the driver is going in the near future.
The last time we had written an article on Nouveau, the community project working on developing an open-source 3D display driver for NVIDIA hardware, was this past January after being enlightened by glxgears finally being able to run on NVIDIA's NV4x GPUs. Since then many developments have occurred with this open-source NVIDIA driver as we have covered in several news postings. In this article today we will share with you where the Nouveau project stands today from a user perspective as well as testing out the driver on a few more NVIDIA systems.
The train has gone off the tracks. In The Truth About ATI/AMD & Linux, AMD's Matthew Tippett had shared with us that the AMD driver release cycle is like a train and that "...we are on the train, and to add a new carriage or update the carriage, we have to do it while the train is running, without stopping the train, or letting anything fall off." However, this morning the train (AMD's release schedule) has went off the tracks for the penguin riders. AMD has issued its first-ever hot-fix release for the Linux fglrx driver. This release corrects an aticonfig crash that removes the xorg.conf if you are using the vesa driver.
David Stevens, a Linux user and Phoronix reader, had written a letter to ATI/AMD expressing concern over their lack of providing an official free software driver. David had asked this letter be made openly available and shared with our readers. You may share your thoughts on the fglrx driver and ATI/AMD's lack of an open-source driver in the Phoronix Forums. Like this letter, Phoronix accepts quality editorials and pieces written by the community.
Hitting the web this afternoon was the AMD 8.38.6 display driver. Introduced in this driver was official Red Hat Enterprise Linux 5 packaging scripts. Fixed in this driver was moving a video window between two monitors in a Big Desktop configuration no longer crashing the X.Org server and playing multiple videos when TexturedVideo is enabled no longer results in corruption.
At Phoronix we are constantly exploring the different display drivers under Linux, and while we have reviewed Sun's Check Tool and test motherboards with Solaris in addition to covering a few other areas, we have yet to perform a graphics driver comparison between Linux and Solaris. That is until today. With interest in Solaris on the rise thanks to Project Indiana, we have decided to finally offer our first quantitative graphics comparison between Linux and Solaris with the NVIDIA proprietary drivers.
It was just two weeks ago that NVIDIA had introduced the 100.14.09 display driver, but today we have another new Linux, Solaris, and FreeBSD driver out from the green binary camp. The NVIDIA 100.14.11 display driver adds support for their new GeForce 7 integrated graphics, fixes console restoration problems, and improves some other areas of their Linux (as well as FreeBSD and Solaris) driver.
Last week the first open-source ATI R500 (Radeon X1000 series) driver had entered the world. This new driver (named the xf86-video-avivo) is very early into development, but a small set of developers have been working on reverse engineering this GPU class for the past couple of months. This driver does not yet contain any 3D functionality or support for features that most end-users expect. At this point, the driver just contains very basic initialization and set video mode support for a portion of the Radeon X1000 family. Even with this very basic R500 driver, we couldn't help but to explore the Avivo driver for the past few days.
We have previously looked at the NVIDIA 100.14.03 and 100.14.06 display drivers, but this afternoon we finally have our hands on the stable NVIDIA 100.14.09 driver. New in version 100.14.09 is added GeForce 8 and Quadro product support, improved notebook GPU support, improved RenderAccel support for sub-pixel anti-aliased fonts, added Xv brightness and contrast controls, improved interaction with newer kernels, and fixing an issue with nvidia-settings. The new yet-to-be-released GeForce 8 names have also tipped up in the release notes. We've seen many of these features introduced in the previous 100.14.xx builds, but they are now officially supported.
Last week we had published The Truth About ATI/AMD & Linux, and to no real surprise, the feedback ranged from beliefs that it was propaganda to others being grateful that AMD finally shared some additional information with their Linux customers about the fglrx development cycle. While the article was far from being propaganda, what had outraged a number of open-source developers were AMD's comments on the R200 support or there the lack of. In this article, we have a few additional comments to share along with what some open-source developers had to say about AMD's information.
795 display drivers articles published on Phoronix.