Re: [OM] question for Doctor Monitor, if he exists

[Chuck Norcutt]

No more complicated than 20 bit segmented addressing on an 80286.  :-)

Chuck Norcutt


Mike Lazzari wrote:
>> But with PAE the 32 bit addressing limit is circumvented by bank switching.
> This is hardware dependent. A workaround implemented by Intel unique to 
> intel cpu's as per the article you linked. Also software and OS 
> dependent. Seems to be a hardware/OS cobbleled together way to page out 
> to a virtual RAMdisk by allowing the processor to address 36bits. 
> WindowsXP limits page size to 4kb and physical address space to 4gb. 
> Many other limits. This one may to apply to PS users.
>> Although this allows the obvious benefits of virtual memory, the 
>> downside is the performance impact on applications that have one or 
>> more of the following characteristics:
>>
>> •    
>>
>> Use a large amount of physical memory for their data sets
>>
>> •    
>>
>> Do a great deal of I/O
>>
>> •    
>>
>> Have large executable working sets
>>
> read the article:
> <http://www.microsoft.com/whdc/system/platform/server/PAE/pae_os.mspx>
> 
> I'm still skeptical that a RAMdisk has any substantial benefit without 
> an apples>apples test. Maybe I should give it a try. But I think Win7 
> and a small (80gb) SSD will be a better over all option before the end 
> of the year.
> 
> Mike
> 
-- 
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/