Riding the 64-bit Chariot: Choosing Socket 939 and Linux
by Kristopher Kubicki on July 19, 2004 12:05 AM EST- Posted in
- Linux
Subjective Testing
One of the most immediate and discouraging issues with SuSE/VIA/NVIDIA was an issue using the YAST updated NVIDIA driver. It seems as though our installation was extremely picky about the order of the installation, so we just installed the driver manually after several failed attempts with YAST.There were other various "gotchas" as far as Linux and these reference boards go, in general. Even VIA's core logic is still identified as K8T400, the previous code name to the chipset rather than K8T800.
Audio and integrated Ethernet were critical components in this analysis. Although we were not going to benchmark and debate whether NVIDIA or VIA has better audio quality, what we did want to see is which platform provided better support. NVIDIA tends to incorporate all of its components into single chips, but VIA spreads their chipsets out a little farther. It turns out that both methods have drawbacks in Linux.
| Chip | Comments |
| VIA VT8237 (Southbridge) | SATA RAID support seems incomplete on x86_64 |
| VIA VT6306 (Firewire) | Fully Supported |
| VIA K8T800 (Northbridge) | - |
| VIA VT6122 (Gigabit PHY) | - |
| VIA VT1720 (Envy Audio) | Only partially supported? |
| VIA VT6421 (SATA RAID) | - |
| VIA VT1616 (AC'97) | ALSA supported |
| NVIDIA nForce3 250GB | Gigabit Ethernet not supported |
| Realtek ALC850 | Fully supported |
Our via82xx blanket driver detects some sound device, but SuSE never recognizes a device. The ALSA project, which is integrated into SuSE 9.1, provides some support for the chip, but it seems as though there are a lot of missing features for Envy. Parts of the source are released for those brave enough.
We also run into issues with AGPGART on the VIA board (both 64-bit and 32-bit).
agpgart: Unsupported VIA chipset (device id: 0282)
We would like to see a Hyperion driver for Linux, particularly since sorting through all the little nitpicks becomes particularly discouraging for Linux. Our drivers are supposed to be installed in the kernel, but clearly some things are not.
NVIDIA is not without problems either. Getting either the VIA or NVIDIA Gigabit Ethernet working out of the box became more trouble than it was worth. The second VIA reference board (built by MSI) uses a Realtek 8110S, which is supported right in the kernel. Even after installing the nForce platform drivers for Linux on AMD64, we were unable to get the integrated GB Ethernet working. The secondary strong point of nForce GB, NVIDIA's firewall and configuration software, seems moot compared to the robustness and completeness of Linux's in-kernel packet filtering.
VIA has traditionally been behind the curve (compared to NVIDIA anyway), but they definitely deserve credit for the amount of work that they have done. A unified Linux driver would put them in a much better position for further adoption.
Facebook
Twitter
RSS
24 Comments
View All Comments
MNKyDeth - Monday, July 19, 2004 - link
10 - Posted on Jul 19, 2004 at 1:23 PM by tfranzese"hardcore linux gamer", I'm sad for you. ;)
Yeah, well, some of us would rather play the games linux has and not have to worry about a blue screen while we play. I won't use wine or wineX either, unless it upports 90% of all windows games and works 100% all the time.
RyanVM - Monday, July 19, 2004 - link
It seems to me that optimizing the binaries would certainly be useful for seeing just how good performance can be. The vast majority of Linux users will be compiling their own code anyway...sprockkets - Monday, July 19, 2004 - link
Hmmmm, although you say that you didn't optimize using flags during compile of Lame or anything else compiled, doesn't the configure script do this for you automatically, or at least, detect your processor type and compile accordingly?Comments on chipset support: It is nice to be able to install SuSE without a floppy, unlike EVERY version of windows out there. Did the SATA work though on the NF3? Sound would work since it's just the usual Intel 8x0 driver, right?
VIA audio to some extent works on my motherboard with the 8237 SB, and while the codec seems fully supported, which is a SoundMAX, sound only came out once.
I wouldn't call the fact that the kernel calls the chipset the K8T400 a necessary gotcha. One of the best things I like about linux is the fact that it refers to hardware by either it's real name or codename, unlike stupid windows which refers you your hardware from the marketed name of the product, making you think it really is something unique. For instance, SuSE 9.1 refers to the 964L SB from SIS as having a 963 ATA controller. On the outside it says 964L, but really, it probably uses the same IDE controller from before.
KristopherKubicki - Monday, July 19, 2004 - link
Doom3 get a linux port but probably not right away. You all know we will have benchmarks first :)Kristopher
tfranzese - Monday, July 19, 2004 - link
"hardcore linux gamer", I'm sad for you. ;)Anyway, there's no need to included UT2K3 when they've got UT2K4 to benchmark. Including Quake3 is old and tired. Is Doom3 getting a Linux port?
MNKyDeth - Monday, July 19, 2004 - link
I am very glad to see linux benches from a reputable site for a change. It's just nice having some hard numbers/facts, that I can use to compare my hardware choices with. I am a hardcore linux gamer and would like to make a suggestion if possible. RTCW:ET and UT2k4 are great benches, but there are many other games you can bench with on linux like Savage and Quake3, UT2k3 and maybe even Medal of Honor:AA. It's just a suggestion to include more games if you do a vid card roundup on linux later on or even if it is just a small comparison.RZaakir - Monday, July 19, 2004 - link
Javescript links are wacky no matter how you look at it. There are myriad ways to design links to work when Javascript is disabled.Anyhow, I echo #5. I am very encouraged by the performance gains that we are starting to see. Around 30% on UT2004 is amazing. Hopefully Microsoft will get their act together so that we'll start seeing similar performance on Win32.
Jeff7181 - Monday, July 19, 2004 - link
It's just too bad a crappy FX5600 was used for the gaming tests. Couldn't at least have dug up a FX5900?Oh by the way... I'm writing this using Firefox as well... maybe someone needs to learn how to use Firefox's features that make it a "web browser of choice." :)
srMatanza - Monday, July 19, 2004 - link
Another awesome article in an awesome series. I think it's great that quality linux benchmarking articles are finally starting to show up in reputable forums.I can't wait for the linux video comparison. I think a good 64-bit distro review would also serve a purpose, especially if it focused on usability and maturity.
Keep up the good work.
AlexWade - Monday, July 19, 2004 - link
WOW! Finally, some REAL benchmarks between 32-bit and 64-bit.And I must say, if gains were saw across the board with non-64-bit optimized code, imagine what the jump will be with 64-bit optimized code!