I think this guy has a pretty good explanation of the limits of 32 bit systems.
http://www.dansdata.com/askdan00015.htm On Mon, Mar 24, 2008 at 8:02 PM, Ben Scott <[EMAIL PROTECTED]> wrote: > On Mon, Mar 24, 2008 at 9:17 PM, Ken Schaefer <[EMAIL PROTECTED]> > wrote: > >> beyond that, which are not clear or well documented. The poorly > >> documented/unclear part is that XP apparently sometimes recognizes > >> much less than 4 GiB of RAM -- far more so than that which can be > >> explained solely by other hardware using up physical address space. > >> That isn't about PAE. > > > > Do we have any properly documented examples of this happening? > > I don't have any references to hand, no. I had seen one really good > blog post from an MS employee that seemed to make sense of it all, but > now I can't find it again. :-( And I should qualify this by saying > that it's entirely possible, even likely, that it's not Win XP alone > that's causing these issues, but some combination of Win XP on certain > motherboards, or maybe certain drivers, or some such thing. > Unfortunately, there's so much murk and misinformation surrounding > these issues that good information is hard to find. Trying to do a > web search just finds cargo cult voodoo, repeatedly demonstrating that > a little knowledge can be a dangerous think. That's part of the > reason why I think it's important to be clear when stating what limits > cause what -- virtual address size, address bus, PAE, motherboard, > chipset, OS imposed, etc. > > > I'm also interested in where you may have found "poor documentation" > around this issue. > > Well, if it was well-documented, I'd have lots of references. :) > > -- Ben > > ~ Upgrade to Next Generation Antispam/Antivirus with Ninja! ~ > ~ <http://www.sunbelt-software.com/SunbeltMessagingNinja.cfm> ~ > -- Mike Sullivan [EMAIL PROTECTED] ~ Upgrade to Next Generation Antispam/Antivirus with Ninja! ~ ~ <http://www.sunbelt-software.com/SunbeltMessagingNinja.cfm> ~
