Yes, I was going to reference Passmark results too, but in general I read these synthetic tests are ~ par. I feel benchmarks help categorizes CPU so one can then proceed to applications specific tests to find out who truly is better, once a full PC system is assembled using those CPUs; you'll find that other substems: e.g. mobos and memory, play a part on overall performance for the final most vital test: a user's experience.
Which brings us to the concept of obsolescence. Since the user experience is primary, a simple rule was established by Miller in 1968, and subsequently reinforced by other researchers as a way to optimize system response time. It guides all phases of design, both hardware and software. It won't matter whether the technology is fresh or decades old, from desktops through the network latency, speed is all that matters, as well as reliability, and total operational costs.
http://www.nngroup.com/articles/response-times-3-important-limits/If the task you need completed: wordprocessing, spreadsheet crunching, video editing etc., whether local or virtual, can be completed in 1 second or less, your machine and network is optimal. If not, one can consider upgrading. Currently, my only issue with local processing using my Q8200 is high end video editing which fails the 1 second rule. The CPUs may also fail certain gaming apps, but since I don't game, it this context is doesn't apply to my system.
My quad core q8200 CPU has actually appreciated to $160+ since I bought it
and yet it is slower than $80 Pentium G870. You cant really base comparison of old technology on price. People are irrational and will pay more for something that they feel is better value, not something that categorically is better value.
Actually...
http://www.cpubenchmark.net/cpu.php?cpu=Intel+Pentium+G870+%40+3.10GHz
http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core2+Quad+Q8200+%40+2.33GHz
OC the Q8200 to 3.1 and it'll be even further ahead.
They're both rather far behind even the 4+ -year-old Nehalems though.