Can we corrolate specific machines to a performance level??

Question

I was soap-boxing in another post about buying the machine you need now as opposed to trying to future proof yourself. The unpopularity of that idea aside, it does help one keep from loosing money by purchasing from Apple's stable of 'First Adopter/Performance Junkie' machines.
I'd be really curious to come up with a way to chart the speed of a specific set of tasks over a line of computers, to show how adequate a machine might (or might not) be for any given photographer.
I have some ideas as to the task list, and I was also thinking along the lines of charting out the performance of not only the machine, but also the machine using different megapixel cameras. In this way you could see that Machine X with camera Y would be acceptable, but Machine X+ would be better.
It would also help 'future-proof' in a more realistic way than 'buy the best you can', because it would be based on the actual performance specs of actual machines–hard data, in other words.

Answer 1

Hi Matthew,
I carried out about 24 tests between a Quad G5 and a MBP Core2Duo recently ... Canon 20D 8.2Mp RAW's
http://discussions.apple.com/message.jspa?messageID=3445975#3445975
Maybe a useful idea as far as it goes. Several of these tests could be taken out into your suggested matrix, and others added.
However, per the other thread you note, your premise fundamentaly tells you about today, but offers no headlights into performance under Leopard, or Aperture v2, etc. You cannot extrapolate. But to your point, it does quantify the value currently for someone considering a purchase and the cost/benefit they would realize short term.
BTW, I have NO desire to assist Apple's burgeoning profits either, but nor do I wish to buy something like my Quad that - despite performance improvements in Aperture 1.1, 1.1.x, 1.5 and 1.5.1 still does not perform adequately for me. Yes, I will be buying an X1900 card, which speaks to the value of an upgrade path.
G.