The fix is a single file called VIAPFD.SYS. There is no "readme" included, and the installer is very simple. You just click next twice, and it asks you if you want to reset. No INF is installed. Only a single registry entry, VIAPFD, is added to the driver section of the registry and provides Windows the path to the .sys file for execution. The .sys file version number is 5.0.2195.100, which is, consequently, a Windows 2000 build number. I have no information at this time to say if it will work on NT4 or any of the 9x variants of the OS. Given that it is a single .sys file, it's within reason for this file to be universal across the Windows platforms. The installer, which adds the registry key, would simply need to be aware of the differences among the Windows databases.
Windows will execute this driver during boot, and it will make the appropriate register changes to the NorthBridge/SouthBridge. At this time, I don't know what register settings it changes, but if you don't need this fix, then I would avoid installing this driver. Performance loss is a possiblity. I am also unsure if this driver will help 686A users.
Removal is an easy process. Windows' Find Files and Folders will allow you to delete copies of the viapfd.sys file and a regedit will locate the single relevant key by searching for viapfd.sys. If you delete the driver, though, the registry key will be ignored. So this isn't too dangerous to test. Just make sure you are in a position to reach a DOS prompt in order to delete the file in case of a boot failure. NT4 and Windows 2000 NTFS users, you should be leery unless you have Sysinternals NTFSDOS Professional Edition. Good luck.
Update: George Breese, the creator of many VIA patches, including one for this very issue, has provided some info to VIA Hardware about the changes the viapfd.sys file makes:
The patch is very specific. You must have a KX133 or KT133x, 686B, and a Creative sound card before the patch will allow itself to work.
The result, observed in WPCREDIT, is that register 76, bits 4&5, are set to 01 in the main chip.
Well, I guess that leaves you 686A users hanging out to dry. It also will not help any Apollo Pro users who are experiencing issues. I would like to be able to convey to you what exactly is being changed at bits 4 and 5 of register 76, but I am afraid that VIA is not as forthcoming as Intel with register sheets for their chipsets.
George also hinted that this fix alone may not be enougha BIOS fix may complement the viapfd.sys file. Without knowing exactly what those register changes are doing, I can't say one way or another.
|1. BIF - $340||2. Ryu Connor - $250||3. mbutrovich - $250|
|4. YetAnotherGeek2 - $200||5. End User - $150||6. Captain Ned - $100|
|7. Anonymous Gerbil - $100||8. Bill Door - $100||9. ericfulmer - $100|
|10. dkanter - $100|
|AMD drops prices on the Radeon RX 460 and RX 470||40|
|Reports: Radeon RX 470D is a budget Polaris card for China||9|
|Examining reports of slow write speeds on the 32GB iPhone 7||27|
|Cellular Insights dissects iPhone 7 Plus modem performance||11|
|Deals of the week: scads of high-performance storage and more||9|
|Tobii's Eye Tracker 4C knows where your head is||4|
|GeForce driver 375.57 is prepared for Titanfall 2||7|
|Phanteks Eclipse P400 gets a tempered glass option||0|
|Radeon 16.10.2 drivers add support for October's big games||10|
|A real "console monitor" would be 720p @ 30 Hz ;P||+62|