Actually, most Windows programs are multithreaded, otherwise the UI would freeze up anytime you told it to do something, like check mail or render a web page.
Why are you talking about servers in response to a quoted question that says non-servers?
I'm talking about a development desktop.
Assuming that the application you're running will generate sufficient cache hits. For some applications this performance gain is minimal.
Now you're talking the minority.
I'm just saying that in many applications, a system with with a slightly faster dual core will beat out a quad core or two quad cores.
I'm sure that is true. A single-threaded application that crunches small chunks of data in a 2K window would probably fit that. To make it true you have to keep as much as possible on the CPU, getting most of your data and instructions from the cache. Thus more cores, more cache and a faster bus wouldn't matter much.
But I certainly don't think it's at all reasonable to say they compete in the 'bang for the buck' region in purely terms of hardware.
Horses for courses as always. Blanket statements in computers usually don't work well. I do simple video editing, barely amateur, and video conversion in backing up the DVDs before the kids can destroy them. I wish I had that 8-core monster, as my little $23 shareware video converter can use them. It would also be nice to run Windows in a VM and give it two cores while leaving two for the host OS.
You know, I do a compile of a huge application and it's not the compile that takes most of the time, it's the build I/O.
Having a lot of fast RAM should help a bit due to the operating system's disk cache. It would be cool to set up a system to compile off a RAM disk. When converting video the hard drive just blips every few seconds.