I’m working on a (rather large) project which uses OpenGL across three different platforms (OSX, Windows, and Linux). I recently ported the newest version (which I had been working on on OSX) over to Linux and Windows. It was a relatively seamless port, as people on this project have been doing it for a couple of years.
When I ported the program to Linux, I noticed a rather large difference in rendering times, favoring Linux (3D rendering, real-time rotation, zoom, translation… you know the deal). I figured it was just Linux, and I was rather impressed with it. However, I then ported it over to Windows, and the same thing happened… a large jump in performance, in the favor of Windows. This is when I decided it was a little suspicious.
I’ve checked throught optimization settings on the compiler. I’ve even tried using a Unix makefile (almost exactly the same as the one used on the Linux machine) on the Mac, as opposed to XCode, which I had been previously using (and will continue to use). I’ve looked through Apple’s Developer website, and I found it all but un-navigable (or I’m just looking in the wrong places… that’s possible). The Apple I’m working on is a G5… the Linux machine is a stock Dell (not new), as is the Windows machine (I think… not too sure on that one).
So, is there any reason you can think of that there should be a large gap in performance, especially between a G5, and two inferior machines? Any ideas would be helpful. Thank you for your time.
P.S. - Note that I’m not exactly skilled in the ways of XCode… I don’t know much about it at all.