Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
Prices: CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ Search Prices:

LOSTCIRCUITS	SHORTCUTS: Distributed Computing Crossfire - SLI RX200 Chipset Overview At One Glance What You Get Layout VRM On-Board Peripherals BIOS Test Configuration / Debugging Memory, Audio, USB, RAID DOOM3 3DMark'05 Canyon Flight FarCry and 5.12 Drivers Quake4 F.E.A.R. Call Of Duty2 Final Words Comment on this review on the LC Forums
Sapphire Pure Crossfire A9RD480 There was something we meant to say .. but we forgot
(Review by MS, Dec 22 2005)

We got the Crossfire platform, we set it up, we cursed and swore for about a week until we figured out some of the stability issues and Crossfire woes and then we still had issues that - in the final instance - were never resolved. Unfortunately, this experience does not appear to be isolated, rather, there are a few common threads running through the experiences of everybody we talked to, even if the concerns were not always vocalized too openly.

Every new platform and new technology has initial problems. Some of the problems, however, especially those concerning the SouthBridge have been known for years now and yet, it does not appear as if anybody would actually take the initiative and solve the problems with, for example, the USB performance. It is certainly no small task to design a working SouthBridge, that's for sure, even with buying components but that's not where the efforts can stop. Interestingly, the driver signatures suggest that the IDE controller may be licensed from VIA Technologies - but we have no hard proof of that. What we are getting at, though is that the entire chip appears a hodgepodge of different components, all of which are great by themselves but they don't appear to get along with each other too well.

Maybe, however, this is just an issue with drivers. For years, ATI has struggled with their drivers and even though recently there has been some real progress, there are still issues. One of the more anecdotal problems relates to the addition of all kinds of applets to the drivers. A case in point is the SiliconImage 3212 SATA II controller driver, which is a total of 286 kB when downloaded from SiliconImage. The same driver package is some 15 MB download from Sapphire which is explainable since it contains an install shield and a configuration JAVA applet - which, as a side effect features a number of links to JAVA-Sun, encouraging the user to download this and that. All of that is still excusable, even though in a similar approach, Phoenix Award almost took a nose-dive with their embedded shopping BIOS a few years ago (for those old enough to remember). Interestingly, however, the installation of the JAVA runtime interferes with enabling Crossfire mode. In order to solve this problem, the SATA drivers with the embedded JAVA console need to be uninstalled first. In combination with the DOTNet mandatory log-in, this one almost qualifies for a Darwin Award. There are ways around all of this but…. point made.

After three weeks of troubleshooting and twisting and turning the PC-A9RD480 to look at it from every possible angle, we got some positive results, overall the board has some great hardware and design features and gets more stable with every BIOS and driver revision but by the end of the day, the entire platform is still just shedding the Lanugo fur. There is definitely some potential there and the setup has a certain odd charm but in terms of being a competition for SLI, Crossfire still has some ways to go. Admittedly, some of the issues we were running into may be isolated incidences confined to the test sample, some may be weird compatibility issues that will be resolved over time and time will tell whether they can be resolved in time before yet another chipset emerges. One short term possibility is - along these lines - the ULI M1575 but, what would happen to the audio in this case?