Originally posted by: taltamir
You think I don't know that? It has nothing to do with what I said...
Actually, the point is that what
you said had
nothing to do with anything said by the person to whom you were responding (Zap).
He recommended Vista 64-bit over XP 64-bit for driver support reasons, and you disagreed with "that conclusion". Now you are saying that you were actually recommending Vista 64-bit over XP 64-bit for driver support reasons, when you were disagreeing with Zap's recommendation of Vista 64-bit over XP 64-bit for driver support reasons?
Are you f-cking high?
"Lemme paraphrase..."
Are you f-cking on drugs?
I have seen many drivers that support XP 64bit (2003 kernel) and vista 64bit...
Of course, for the reasons I stated. They install and run for the same reason that Windows 2000 drivers can be installed on Windows XP. Hell, even many Windows 98 WDM drivers can be installed and will function on Windows XP. That doesn't mean Windows 98 WDM drivers "support" Windows XP. It is the opposite, Windows XP can support those drivers.
I have yet to see a SINGLE XP64bit driver that does not support vista 64bit. (I saw one very VERY early on, but it immidiately added vista 64bit support).
Then you have extremely limited experience with or exposure to the spectrum of mainstream hardware. Among the numerous devices that maintain different function, class, or filter driver binaries and/or implementations for XP 64-bit and Vista 64-bit include (but certainly not limited to):
Intel Wired LAN
Intel Wireless LAN
NVIDIA chipsets*
ATI chipsets*
Realtek Audio
Realtek LAN
* = with some exceptions such as ACPI BIOS and SMBus devices
Need I go on? Common driver binaries and/or implementations covering both XP 64-bit and Vista 64-bit are in fact the exception, not the rule. Storage drivers are particularly unlikely, since most developers have begun using the Storport model in Vista but continue using the older SCSIport model for their XP 64-bit drivers. I excluded graphics hardware for the obvious reason (WDDM is unique to Vista).
Also... all drivers have to be written to work properly with PAE... if microsoft disabled the so called "artificial limitation" they would end up invalidating a lot of existing drivers, requiring them to be remade. That is NOT acceptable regardless of how much the OS costs.
So its perfectly acceptable for drivers to break a $5000 server system that might cost a company tens of thousands of dollars in downtime or data loss, but completely unacceptable for drivers to break a $500 home PC that might prevent someone from playing Mine Sweeper? Ooo-Kay!
If this dreaded problem were so, umm, dreaded, Microsoft would have "protected" those with the most to lose from encountering it - enterprise segments. Yet it did the exact opposite. The bottom line is, Microsoft is perfectly willing to expose you to this awful compatibility risk - for a hefty premium.
Especially considering that any of microsofts inexpensive 64bit operating systems can address "17.2 billion gigabytes, 16.8 million terabytes, or 16 exabytes of RAM". So, no, it isn't to sell more copies of their server operating systems which could address a mere 64GB of ram in 32bit.
Compatibility and hardware upgrade costs determine why someone would select 32-bit server OS with PAE over native 64-bit, not memory limits.
BTW, the most that any current Microsoft 64-bit Server SKU will support is 2TB. Microsoft only supports configurations it can actually test.
Next I suppose you're going to tell us that memory support for these SKUs are artificially capped for honest-to-goodness technical reasons rather than marketing ones?
- 32-bit Windows XP Starter Edition = 512MB
- 32-bit Server 2003 and Server 2008 Standard Edition = 4GB
- 64-bit Server 2008 Standard Edition = 32GB
We're all stocked up on bullsh-it here, thank you. Try selling that crap down the hall, last door on the left.