Microsoft has been under fire with the differentiation between Vista Capable and Vista Ready, resulting in a class-action lawsuit. Could graphics drivers be to blame for end-user woes? Or is this just more smoke and mirrors in an effort to salvage Vista's reputation?
Earlier this month, I told you about the class-action lawsuit that had been given the nod by Seattle's Judge Marsha Pechman. As a result of that go-ahead, some 300 documents, including 158 pages of e-mail, were made public. At that time, the discussion was focused on whether Microsoft had downgraded its Vista logo program to differentiate computers that would only be capable of running Vista Home Basic and computers that will run any flavor of Vista.
The issue was that Intel had a large inventory of motherboards on shelves with the 915 chipset. The 915 chipset does not have the graphics power necessary to power the Aero display. Since Vista Home Basic doesn't include the Aero display, PCs with the 915 chipset were thought to be "Vista Capable," or at least they were thought to be by Microsoft executives trying to facilitate the sale of existing stock.
While the 915 chipset is indeed a valid issue, it seems that there were some other things that may have played a part in the end-user experience with Vista, regardless of the version — like graphics card drivers.
In that package of documents is a list of driver crashes in Vista that was compiled by Microsoft. The time frame is ambiguous — it is only identified as "Period 2007."
The list documents "Crashes by Organization" for all file type drivers. At the top of the list is Nvidia with roughly 480,000 crashes, or 28.81 percent of the total, followed by Microsoft (17.97 percent), Unknown (17.07 percent), AMD-ATI (9.30 percent) and Intel (8.83 percent). Webroot Software (3.99 percent), Realtek Semiconductor (3.34) and Creative Labs (1.09) are the only other organizations with more than 1 percent of driver crashes on the list.
It came as little surprise to sources that drivers associated with discrete graphics from Nvidia and AMD-ATI topped the list when you eliminate the ambiguous "Microsoft" and "Unknown" categories.
"Graphics drivers in particular are the most complex in the whole stack. The complexity in those drivers is in 3D animation effects," said an AMD spokesperson.
Nvidia had a rocky start with Vista, but according to spokesman Derek Perez, it "fixed the bugs in the order that they came in and the ones with the most complaints got fixed first."
"When any OS is launched, there's a number of bugs coming out of the gate. That's pretty typical. We had driver bugs and the community let us know. One thing I was proud about was, given the response from the community, we took an outside-in approach to fixing bugs," Perez said.
Adrian Kingsley-Hughes of ZDNet suggests that the numbers may be misleading in a recent blog. While the number may indeed be 30%, that doesn't mean that the system experienced a BSOD or a hang.
From Adrian's blog on ZDNet:
The important thing to note here is that an nVIDIA or ATi driver crash doesn't automatically mean a BSOD, hung system or a full-blown crash under Vista. This is because of a mechanism called TDR - Timeout Detection and Recovery. Here's how Microsoft describes TDR:
Windows Vista attempts to detect these problematic hang situations and recover a responsive desktop dynamically. In this process, the Microsoft Windows Display Driver Model (WDDM) driver is reinitialized and the GPU is reset. No reboot is necessary, which greatly enhances the user experience. The only visible artifact from the hang detection to the recovery is a screen flicker, which results from resetting some portions of the graphics stack, causing a screen redraw. Some older Microsoft DirectX applications may render to a black screen at the end of this recovery. The end user would have to restart these applications.
The thing about these TDR events is that if your system is set to report crashes, information on each one of these events is sent to Microsoft. While each one certainly classified as an undesirable event, they're not crashes in the XP sense of the word.
The question in my mind is what Microsoft could possibly have done differently to avoid the problem. InformationWeek's Dave Methvin noted that a longer beta probably would not have solved the problem, as vendors weren't going to get serious about making their stuff work with Vista until they were sure that it was really going to ship. That, coupled with the new software interface called "Windows Display Driver Model" or WDDM, is likely to be the foundation of the challenge.
If Microsoft wanted to improve Vista's quality, the best thing they could possibly have done would be to provide more support for driver developers. In particular, they should have carpet-bombed the major display hardware makers with assistance, treated them to "driver ed" classes, and even paired Microsoft experts with the hardware developers to make sure the drivers would be excellent. Or, as Steve Ballmer would say: "Developers, developers, developers, developers."
The question to be answered by the court is one of willful negligence. It seems to me that the whole discussion of display drivers, while intriguing, is merely an avoidance of the real issue. What are your thoughts?
The Vista Capable fiasco: to hell with system recs! (Ars Technica)