Even arstechnica doens't like them:
"You need only look up "VIA 686b" with any search engine to get a picture of the scope of the problems with one aspect of recent VIA-based motherboards; look up "VIA and Sound Blaster Live" and you'll read more woes. Call the tech support department at M-AUDIO.com about USB audio glitches and, hey, it's a known problem with VIA chipsets. They say VIA has even been conferring with them about it.
Look up George's PCI latency patch to see just how deep the problem goes, affecting everything from sound cards and digital video cameras to system freezes and lockup issues when transferring large amounts of data over the IDE bus. In fact, there are glitches galore, most of which can be resolved by diligently updating to the latest patches -- many of the most serious problems involve a component known as a "4 in 1" driver.
USB Glitches
The most commonly reported glitch affects USB peripherals. We've seen cases where a Microsoft SideWinder 2 joystick seems to work -- the Game Controllers control panel in Windows finds the device and reports its status as "OK," but the device just doesn't work right. Substituting a different USB controller (available as a PCI add-on card for about $20) solves the problem -- and points a finger squarely at VIA's USB implementation. It's worth mentioning that this problem is not a result of the USB device drawing too much power -- the Microsoft joystick has its own AC power supply.
Indeed, we've seen several USB-based peripherals that now specifically warn of compatibility problems with VIA chipsets. One such devices is the Roland UA-30. the package lists a number of restrictions on its use with certain types of PCs. For example, the package says you may expect good results with systems based on PCI-to-USB controllers from Intel and ALi. However, it warns that USB host controllers from VIA Tech and SiS may not function reliably. It also notes that if your PC is fitted with a non-Intel USB host controller, audio reproduction may be interrupted when using games or other software. You may be unable to play music CDs on your PC through the UA-30. Your PC, warns Roland, may not supply sufficient power to the USB port to reliably power the device. In some cases (specifically, with the SiS7001 PCI to USB Open Host Controller), even the use of a USB mouse can cause audio interruptions.
PCI glitches, most often causing audio stuttering on Athlon-based systems, are also commonplace.
Graphics Glitches
The whole fiasco reached something of a climax in Dec. 2001, when nVidia released an updated (23.11) version of its Detonator drivers that seemed to cause no end of troubles for Windows XP users of motherboards based on VIA chipsets. VIA responded to allegations of compatibility and stability problems in places such as the Via Arena forum by saying it was an innocent victim, and that the graphics drivers themselves were the source of the problems. However, this didn't stop the company from releasing a beta patch incorporating a fix expected to be part of the company's next version of the 4in1 drivers -- along with an apology for using end users as "guinea pig testers."
Technically, the VIA patch closes the RX55 memory register in BIOS, fixing a problem that Microsoft's KnowledgeBase weakly suggests may alternately be addressed by disabling graphics acceleration entirely. Says VIA, "in a nutshell, it?s a memory timing problem that only happens when the RX55 register is opened. Some motherboard manufacturers have already released new BIOS that have the register closed. In other instances, this patch is needed." At any rate, this wasn't the first such incident we've seen, and it probably won't be the last. The Inquirer has details.
Another issue apparently limited exclusively to systems based on VIA chipsets is detailed on our Windows XP troubleshooting page. Users upgrading to Windows XP who see an "UNMOUNTABLE BOOT VOLUME" error message may find themselves in a situation where they can't boot Windows XP, nor can they easily uninstall Windows XP and revert to an older operating system (e.g., Windows 98 or Me, etc.).
VIA-based system users may face other problems when attempting to add hardware to their systems. Consider, for example, the warning on the website of Steinberg, a leading music software and hardware developer: "We're afraid that we cannot recommend to use [sic] a mainboard with VIA chipset together with the Project Card."
As well, motherboards based on VIA chips typically offer lower performance than do Intel-based boards, especially in ATA/133 and SCSI performance.
And, even after you download sometimes further trouble-inducing BIOS updates and firmware patches to allow AGP to work as expected on a VIA-based system, the performance isn't all that great. As Tom Pabst demonstrates with a series of benchmarks, the popular VIA Pro 133 chipset's primary weakness is in AGP performance. (The Rev. A of the Pro 133 supports AGP4x; early revisions do not.) There are many documented examples of users finding VIA chipset-based computers less capable than Intel-based ones when attempting to use today's newer AGP4x video cards. Even the ageing Intel 440BX chipset outperforms the Pro 133 in 3D graphics tests, particularly when a lot of texturing and 3D operations are performed, as you might when running games.
Here are a few reviews that discuss key VIA weaknesses:
www.tomshardware.com: A preview of the VIA Pro 133
www.anandtech.com: review of the Tyan Trinity 400
www.gamersdepot.com: review of the Tyan Trinity 400
viahardware.com: Issues with VIA 686B and Sound Cards
viaarena.com: Issues with VIA chipsets and digital video
KT266A
VIA 266A-based motherboards seem prone to glitches, too, as a search of Google's newsgroups amply demonstrates. At Ars-Technica.com, Hannibal in Feb. 2002 said he's "given up on VIA," noting that the KT266A-based Soyo K7V Dragon+ IDE RAID + WinXP combination he tried to get working just didn't work properly. (Yes, he claims he tried all the latest drivers and patches.) He eventually gave up on his WinXP-on-an-IDE-RAID-1-partition plans, saying "in my old age, I've gotten to where I like things that work."
A Solution?
On Oct. 24, 2002, VIA released a new PCI latency patch for its trouble-plagued chipsets. It is, the company says, a Windows driver that will patch motherboards that have VIA chipsets. It is intended to solve many problems, including these:
-Files copied from one IDE hard disk drive to another are corrupted
-The computer stops ("freezes", "hangs", "locks up") when playing sounds, especially to a Creative SoundBlaster LIVE! card
-The computer stops when copying to or from an IDE DVD, CDROM, or CDRW drive Sounds played by a Creative Labs sound card are distorted
NOTE: It does not apply any fixes to AMD chipsets, such as the 750 or 761.
Download it from Fileconnect.
Conclusion
Can these problems usually be resolved? Yes. Is a VIA-based system a viable alternative to one based on an Intel chipset? Yes. Are VIA-based systems less trouble? No. Do we recommend VIA-based motherboards. Emphatically, NO."
WHY exactly should he not purchase what has been shown to be a stable product and have the best numbers?
Facebook
Twitter