Windows Server 2008 is better than Vista, but why?

IT infrastructure veteran Mark Wilson asks:

It seems that, wherever you look, Windows Server 2008 is almostuniversally acclaimed. And rightly so - I believe that it is afantastic operating system release (let's face it, Windows Server 2003and R2 were very good, too) and is packed full of features that havethe potential to add significant value to solutions.
So, tell me, why are the same journalists who think Windows Server2008 is great, still berating Windows Vista - the client version of thesame operating system codebase?

The short answer is that Server 2008 delivers new features thatcustomers wanted, whereas Vista delivers new features that Microsoftthought its customers should want. However, it seems there may be moreto it than that. Maybe Server 2008 really does perform better thanVista.

According to this post from Windows performance project exo.performance.network,Server 2008 is 11 - 17 per cent faster than Vista SP1, running a coupleof benchmarks that test typical client applications. Backoffice serverand systems admistrator Christian Mohn concurs:

Windows Server 2008 performs better, even with the Aero featuresenabled, than Vista ever did on the same hardware. To me, this a bitstrange, even if a lot of services are still disabled, as the codebaseis pretty much the same as Vista.

Mohn's example is less scientific, though: he never ran Vista SP1, and also moved from 32-bit to 64-bit.
Server 2008 has a "Desktop Experience" feature, which installsthings like Windows Media Player, Aero GUI effects, and other fluffthat doesn't belong on a server. My assumption had been that once youinstalled this, Server 2008 would perform in a similar manner to Vista.Apparently this is not the case.
It seems to me there are a few possibilities. One is that Microsoftisn't being straight with us about this "same codebase" stuff. It wouldbe interesting to analyze the core DLLs and work out which are thesame, and which are different.
The second possibility is that there's stuff in Vista that is notpart of the core, nor part of the Desktop Experience, but that slugsperformance. If so, it would be great to identify it and turn it off.
The third explanation is that the testers are wrong, and thatperformance is actually similar. For example, maybe Vista was running abackground update or backup during tests. Background processes make ithard to conduct truly rigorous performance comparisons.
I'd like to see Microsoft platform and services division fellow Mark Russinovich get his teeth into this. I'm also tempted to try the Server 2008 desktop experiment myself.
This article originally appeared in ITWriting.