PC / Mac comparative benchmark?

[ripkord]

Is there such a beast?

Just want to compare my Wintel to my Mac....

[ipodandimac]

Originally posted by ripkord
Is there such a beast?

Just want to compare my Wintel to my Mac....
i think the best way to "benchmark" them is to just time some real world tasks. if you do video, time renders and things like that. the other thing you have to consider is stability. no matter how much faster a wintel is (if it is at all), at least the mac won't crash while you're testing it. :)

[ripkord]

heh, it seems to be quite bloody fast thats all :)

[Karl Kuehn]

Originally posted by ripkord
Is there such a beast?

Just want to compare my Wintel to my Mac....

One word answer: no.

Longer explanation: you have to define what you are comparing, and by the time you properly do that, you probably don't want the complex answer you are going to get.

[iPoster]

Originally posted by ripkord
Is there such a beast?

Just want to compare my Wintel to my Mac....

CineBench 2003 (http://www.maxon.net/index_e.html)

Bare Feats uses it to benchmark Mac Vs PC.

or one of the shared games, like Unreal Tournament or Quake3..

[Kickaha]

All of which only tests particular algorithms and whether or not they have been well optimized for the various platforms.

No benchmark matters other than how long it takes you to get your tasks of interest done.

[a_greer]

In this day and age, benchmarks are pretty much worthless, desktop rendering of video or images or 3d is likely within seconds of each other on equal Macs and PCs, the differances that you need to weigh are:

1: Stability and security - the Mac wins here.

2: what software packeges do you like: nearly every pro level tool is availible on both, Macromedia Studio MX, Avid, Adobe *cs, MS Office (sans Access), and so on, but some are platform specific, Like FCP(Mac) or Autodesk Autocad (Windows)(there are other cad applications for Mac)

3:Your comfort, do you like where you are at? do you want to change? do you need to change?

4:Your budget, Software replacement for platform change can be cheap (Adobe will trade one for the other IIRC) or expencive - like replaceing premere pro with FCP.

[curiousuburb]

You could test similar algorithms optimized (to a degree) for each platform. Though admittedly, these would not be not ideal comparisons, as some tasks are better optimized for some platforms and/or threading.

PC, Mac, and 'Nix versions of the following might offer some clue based on time per work unit:


SETI@Home (http://setiathome.ssl.berkeley.edu/)
Folding@Home (http://folding.stanford.edu/)
Distributed.net RC5 cracking (http://distributed.net/)

Even comparing the top performers from the data charts of these sites wouldn't necessarily provide accurate benchmarks. Different platform versions of the code might be at different optimization levels and revs.

I think there will always be a measure of apples-to-oranges dichotomy in most benchmarks.
How efficiently and productively you can work may be more important than how fast the machine can work.