Windows out-WoWs OS X
p2p news / p2pnet: With Apple’s Boot Camp out-booted by two indie apps, one from Blanka and Narf, which came well before the Apple effort, and the other a triple boot application which adds Linux to the line-up, we now learn that stacked against Windows, the Mac OS comes out as #2, at least when WoW is involved.
"Boot Camp is a pretty slick program," says Gabe on Penny-Arcade. "You go through a standard Windows install and then you throw your driver disk in and boom it just works. Once I had XP on it was time to test it out and that meant installing WOW."
And here’s what he was using:
MacBook Pro
2 GHZ Intel Core Duo
2 GB Memory
Radeon X1600
Gabe says he played World of Warcraft at 1440 x 900, "with all my graphic settings jacked all the way up" and OS X I hovered between 15 and 20 FPS and was, "just barley playable". For "raids and stuff," he’d, "drop down to a lower resolution and take off a few of the fancy effects".
Under Windows and with the settings, he averaged between 35 and 40 FPS and, " was even able to maintain that frame rate in major cities and in a raid out in SS".
BIG difference, he states, adding:
"It’s a hell of a laptop. On the other hand I’m disappointed that I can’t get that kind of performance under OS X. I honestly prefer using OS X at this point and it was sort of nice to have a game I could play that didn’t require me to go upstairs and start up my Windows box. But it runs twice as fast under Windows ………."
Also See:
Blanka and Narf - Windows on Macs, April 6, 2006
triple boot - Boot an Intel Mac with Linux, April 15, 2006
Penny-Arcade - Bootcamp, April 19, 2006





p2pnet - rss feed: 
April 21st, 2006 at 12:11 pm
This large difference in preformance on the exact same hardware is interesting in light of the fact that video card manufactorers refuse to tell anyone how to use the hardware they produce.
Video card manufacturers are not trustworthy, they have been caught cheating time and again. Producing incorrect images in order to get better preformance stats, detecting the software that’s running in order to do the former, disabling unstable features only during WHQL certification, etc…
Here’s a smatering of supportting links from a google search:
Detecting the exe name and other cheating, how primitive.
http://techreport.com/etc/2003q2/3dmurk03/
A microsoft employee says they not only cheat benchmarks, but WHQL bug testing as well.
http://blogs.msdn.com/oldnewthing/archive/2004/03/05/84469.aspx
cheating on 3dmark (couldn’t linkify, cut and paste)
http://www.extremetech.com/article2/0,3973,1105259,00.asp
I predict that if anyone leaked the source to the drivers of either of the two major players’ offerings, we’d be seing revelation after revelation of dubious application specific optimizations (cheating).