Generational Windows Multicore Performance Tests

snydeq writes "Windows XP, Windows Vista, and (soon) Windows 7 all support SMP out of the box, but as InfoWorld's Randall Kennedy notes, 'experience has shown that multiprocessing across discrete CPUs is not the same thing as multiprocessing across integrated cores within the same CPU.' As such, Kennedy set out to stress the multiprocessing capabilities of Windows XP, Windows Vista, and Windows 7 in dual-core and quad-core performance tests. The comprehensive, multiprocess workload tests were undertaken to document scalability, execution efficiency, and raw performance of workloads. 'What I found may surprise you,' Kennedy writes. 'Not only does Microsoft have a firm grasp of multicore tuning, but its scalability story promises to keep getting better with time. In other words, Windows Vista and Windows 7 are poised to reap ever greater performance benefits as Intel and AMD extend the number of cores in future editions of their processors.'"

Comments

So what?

[fpophoto]

Are we supposed to be surprised that the leading OS vendor, who's had deep, intertwined relationships for decades with hardware makers is exploiting that hardware properly?

Honest question: where's the news here?

Re:So what?

[Anonymous Coward]

The news is that nobody can say anything non-critical of Microsoft without being accused of advertising or astroturfing.

Oh wait, this is Slashdot. Nevermind.

118% slower

[TheMeuge]

Roughly 118 percent slower than XP on dual-core

Some great mathematics in this review... it also appears as if Vista isn't just not solving the problems presented to it, but also adds new ones to increase its own workload.

Fascinating...

The Money Quote

[d3ac0n]

Basically, this article states the obvious: Windows XP 64 is just plain faster than Vista 64 or Win7 64. By a factor of 20-40%. But to understand why, you need to read the MONEY quote. Here it is:

In the end, it all comes down to the complexity of the execution path. With its simpler legacy kernel devoid of DRM hooks and other performance-sapping baggage, Windows XP provides a cleaner code path for the workloads to navigate as they execute. This, in turn, translates into better overall performance with lower consumption of CPU cycles.

It's the DRM baby. You strip that out of the Kernel, and Vista and Win7 will EASILY outpace XP with their more advanced and flexible SMP capability. Until Microsoft understands that people DO NOT WANT DRM and removes it from their newer OSes, these new OSes will continue to suffer from performance problems, and thusly, acceptance and sales problems.

Come on Microsoft. Apple has figured it out, DRM is a sales loser. Do you really want to keep wasting time on a loser technology in the midst of a global recession? You blew it with Vista, but you still have a chance with Win7. Offer people a DRM-Free kernel and Win7 will FLY off the shelves.

 

Re:The Money Quote

[quo_vadis]

I have never been a "windows fanboi"( In fact this is being posted from a linux computer) and I am no defender of Microsoft's business practices. However without doing code analysis, it is impossible to say that this slowdown is because of DRM. Nowhere in the article does it suggest that they were able to do a profile analysis of the kernel codes and compare what modules on the path were causing the delays. So while it is theoretically possible(and likely) that the source of the delay was DRM related, one cannot be sure. If you possess knowledge otherwise, please feel free to cite it and correct me.

Re:The Money Quote

[rossjudson]

This is yet another one of those times when I wish Slashdot wasn't so ridiculously hostile to Microsoft. What we need here is some informed, possibly even official commentary from someone in the know at MS. Exactly why is a workload slower on Vista? Where's that time going? Right now something like 60-70% of corporate workloads still run on Windows OS, so gaining an understanding of exactly why is important.

When the difference is on the order of 20-40% (if the article is to be believed), we're looking at some level of system-call "tax" under Vista, or we're looking at a different _capability set_. If the workload on Vista is in a secured environment, and the same workload runs faster on XP in an unsecured environment, we're talking apples and oranges.

It could be the case the even for workloads running as root equivalents in Vista execution times are worse...but we don't really know from what's quoted in this article, and there isn't any response from MS.

I think Vista is a pretty important upgrade for most users. Even if its security mechanisms are intrusive, at least they're _there_, and that's a step in the right direction.

Re:Interesting

[blankinthefill]

Try running Win 98 on Vista's minimum hardware. Hell, lets go whole hog here. Try running Win 3.1 on Vista's minimum hardware. (Okay, you might have to do a lot of work to get it running, but I'm just trying to make a point). I guarantee you that both 98 and 3.1 will run faster than XP on that hardware. By a lot. If it surprises you that vista and xp run slower on the same hardware than xp, then either you're not thinking things through, or you're not very bright. As stated, they are far newer. This means they have a much higher assumed baseline of technology that they can run against, which means that they have far more assumed resources to play with. So yeah, on the same system, Vista runs slower than XP. No surprise (at all, as far as I'm concerned). Honestly, all this speed stuff is pretty pointless. The question with OSes is never really about the fastest, or we would all still be using DOS. The question with OSes is are the fast ENOUGH. This is very subjective, but it basically boils down to: will they run what we want them to run in an acceptably small amount of time. On its original release, Vista did not. However, right now Vista is certainly running fast enough for me, and I expect Win 7 will to. But you're ALWAYS going to take a performance hit moving to an OS that utilizes new technology, and I don't care what OS you use.