stalinx20
ManBearPig
stalinx20
No, I said there was no decent "consumer based" app/program that can utilize "more than 4 cores" sufficiently. Do not twist my words.
Not twisting your words at all, you yourself said it.
stalinx20
ManBearPig
On topic: give me 22 cores 
with ~20 of them sitting idling. 
You said that if I had a 22 core CPU that 20 of them would be sitting idle.
Exactly. You all are missing my point out of this whole entire thread; Why are they only stacking more cores on top and not allowing cores to take on more than 2 commands at once? Why not have Hyperthreading which does 4 commands at once for each core? Why can't they make a single core stronger than what it currently is? Why can't they make the processors run faster passed 5.0ghz on standard cooling? Why why why. No, I mean for real, that's no joke. Yes, most of those cores will sit idling for the things you do MBP. But by all means, we can continue arguing about it if you'd like, I know for a fact you're not into heavy CAD or even Video Editing, don't even deny it. You'd waste your money, and a lot of it. But that's ok right? YOu get to tell everybody you have a 22-core proc. Good for you. I still rest my case.
each core doing 4 commands at once - then it wouldn't be 4 core or the instruction set would be kinda complex
what you want are instruction pipelines to each core and that adds chip complexity and heat as well - so slower chips (but better utilized if the OS is made to handle it)
currently intel uses a 14 stage pipeline for each core
http://www.gamedev.net/page/resources/_/technical/general-programming/a-journey-through-the-cpu-pipeline-r3115 if you want better performance look at M$ for software design - bad OS design leads to poor utilization of the current chips we have now
M$ 6,7 or 10 is not a symetrical OS
and again i will remind you this
Heat and leakage is the problem to most chips so there is speed barrier we are hitting
the closer you get the traces/circuits together in the chip - the traces act like capacitors and you have leakage - power jumping circuits - causing heat
let alone the imperfections in the etching/fab process that makes the traces rough/imperfect-(closer together) in some spots leading to further leaking or arching
and pushing more power to circuits(to have higher clocks) only lends to more leakage and arching and more heat
then add in circuit and chip communication latency - and more transistors complexity doesn't mean more speed
try reading VSLI design
https://books.google.ca/books?id=gdRStcYgf2oC&pg=PA91&lpg=PA91&dq=chip+circuit+latency&source=bl&ots=MJDOazZdXM&sig=FBRNJ9OwrXbeyhFdXjZQj8qbWv0&hl=en&sa=X&ved=0ahUKEwjM_NKeyLzQAhVpq1QKHcvYBTc4ChDoAQhKMAc#v=onepage&q=chip%20circuit%20latency&f=false and we keep pushing the circuits closer together - hence we have less room to push the voltage - and less overclocking room
shrinking doesn't necessarily improve speed
what we need is better bandwidth(memory - on chip and onboard)
and better core utilization - and that means optimization at the OS level you can have all the speed in the world but if the software is not designed for multiple cores it will use fewer cores
we see that with some games only using 2 core some only 1 core and some using as many as they can
and your are still at mercy of the OS which doesn't assign cores well either specially M$ - linux is better
more cores are better in my opinion - AV software is getting big and cpu intensive
you want to protect you system 24/7 even while gaming that takes cpu cores and you still need cpu core for the OS and background tasks at the same time - AND you want to game too??
more is better
like compresssing a video file in DivX and runing AV software and gaming simultaneouesly - that be nice