This is an edit, but I can't find an edit option.
After more reading:
* RootKitRevealer installs itself as a Windows service and is meant to work within the running OS. I haven't found any “outside the box” scanners currently available, despite all the articles that said such scans were the most reliable [many articles said only] way to make sure a rootkit was not present. I also haven’t found a description of how such a scan is done. If rootkit code – even unpublished/unknown rootkit code – were readily and reliably identifiable when the Master Boot Record and other files were read or parsed or whatever from outside the OS then I’d think bootable rootkit scanners would be readily available and they’re not.
* From reading articles I mostly don't understand, there currently seem to be two basic types of rootkit scanners available. One type looks for the signatures of rootkits whose code is publicly available and I'm guessing of rootkits found by the antimalware people. I'm assuming Trend Micro, avast, avg, bitdefender, etc., scan for signatures.
*The second type of scanner, which I think is called a cross-view scanner, does two types of scans from within the running OS - a "high level" scan that uses the win api to list files and a "low level" scan that obtains its file list by reading from the ntfs directly. The results of the two scans are compared looking for files listed in results from the "raw content" scan that are not listed [therefore possibly intentionally hidden] in the high level scan. Cross-view scans have the capability of finding rootkits w/out known signatures but their results apparently require considerable expertise to interpret.
* Rootkitrevealer and Blacklight [no longer a free download] are cross-view scanners. The technet rootkitrevealer article dated 2006 said it was possible, but very unlikely, that a rootkit could fool rootkitrevealer [http://technet.microsoft.com/en-us/sysinternals/bb897445.aspx]
I don't know if that still holds true in 2010.
* RootRepeal scans for signatures. Since it looks for “typical symptoms” it may also use heuristic methods. Since it IDs hidden files it sounds like it also does cross-view scans comparison – unless it crashes and you change the disk access level. I’m assuming [again] that “disk access level” refers to gaining kernel-level access to read the MFT & registry hives directly – what Rootkitrevealer calls “raw content’- to generate its "low level" file list.
* Most actual-information articles that aren't completely technical greek that I've found about rootkits and non-signature antirootkit detection methods are from 2005-2006. I hope that's because current AVs w/ signature-type antirootkit protection are effective for home computers.
* I would think/guess the average home computer would be much more likely to encounter rootkits w/ a published code than be targeted by someone capable of writing their own rootkit code. I haven't, however, found any good rootkit articles that provide info/recommendations based on risk probability/threat type for home computers. Maybe no one has that data.
* Having done a LOT more reading about rootkits I now know a lot more about what I don't know.