Your Apple can do no wrong attitude is the main cause for the dregradation of discussion around here.
People bash the nMP because it's an expensive non-upgradable or serviceable computer that is less performing than the machine it replaced.
myself.. i think the dregadation happens because a lot of you place far too much emphasis on benchmark results as signifier of performance.
here's why benchmarks are stupid:
they tell you which component has less crappy speed.. note- they don't tell you which component is better.. they tell you which one is not as crappy.
benchmarks measure bottlenecks.. and they tell you how much waiting to expect when clogging up an avenue.
what they don't tell you is this-> "yes, you will wait.. or-- no, you will not wait"..
no matter what, you're waiting on something and waiting sucks.
say i move 20GB data in 4minutes.. then a geekbench player pipes in "pfft.. that's soo slow.. my thing does it in 2 minutes!"
the glitch in the bench player's thinking here is that they're now thinking their system is fast..
when, it's actually super slow.. it's not fast.. you're waiting 2minutes to move 20GB data.. it's freaking slow.. it's a tad less crappy than my score but still crappy nonetheless.
if a benchmark is a true indicator of your machine's performance/usability then your computing experience sounds miserable.. all you're doing is sitting around waiting on a computer to do something.. click something.. wait.. click something else.. wait.. do this.. wait.. do that.. wait.
and really, my experience has nothing like the above happening.. most (like- very high percentage most), of the things i do on a computer are instant or perceptibly instant.. what exactly are you all doing with your computers which sets you in this click&wait cycle? i honestly have a hard time believing anybody's workflow reflects the type of numbers/tests being shown by benchmarks.. they're irrelevant and and should only be used at a very basic level of judging a machine or assessing how a computer is going to react in real world scenarios..
another problem with benchmarks is they completely eliminate a user's ability/ideas/skill/knowledge/taste etcetc from the equation and puts judgement on a computer as if it's the computer creating the content.
very similar to tasting a new dish and wondering "how long did it have to sit in the oven for?".. and thinking something that cooked for 12 minutes is inherently better than something that cooked for 20minutes.. you've eliminated the chef entirely.. you've eliminated the hours of work required by the chef to do their thing as if none of their work is important and are only thinking the time in the oven.. the time when the chef is no longer working.. as the key factor in this new dish.
i'm sorry but i really do feel if one's main argument regarding computer performance is a benchmark score then it only shows how very little actual work you do on a computer.. because if you did actually work on the things then you'd know there are only minimal enhancements to be gained by a computer producing higher benchmark score..
until that particular benchmark gives an infinity score (or a score of zero when the smaller graph means faster) then you're waiting on whatever that particular benchmark is testing.. regardless of how 'good' your computer is, you're waiting..
an instant result is a game changer.. that's when the particular operation has now become good instead of crappy.. but once the technology behind a particular operation leads to instant results, it's no longer tested by the benchmark.. it's now a good thing but since benchmarks are only testing for crapness, it's not included in the test..
