Intel® Core™2 Duo processors

duderoo

duderoo

Expert Member
Joined
Jun 7, 2005
Messages
3,421
Reaction score
250
Location
Earth
Howcome the processor speeds are lower namely 1.86Ghz to 2.66 Ghz surely it should be higher than say 3Ghz as I have a P4 3 Ghz CPU?
 
Because they finally realized AMD was right all along, you dont need high clock speeds for a good CPU. It must still be settling in with all the Intel people who thought their chips were better because they were clocked higher than AMD's.

No, there is a real technical reason, im sure someone will be able to tell you, I however can only spread AMD propaganda.
 
Because it's two physical cores? You've basically got 2 X 2.66Ghz CPUs on one die.
 
It's instructions per clock, more than the clock speed.
say you have 2 cpu's. CPU X and Y.
X = 3 intstructions per clock, Y = 6 instructions per clock.
If you need to do 600 instructions.
CPU X needs 200 cycles to complete the instructions, but CPU Y only needs 100 cycles.
See pentium4 is like CPU-X but Core2Duo is CPU-Y.
2GHz Core2Duo = 4GHz Pentium4 :)

It's not exactly like this and is far, far more involved, but you should get the general picture :)
 
It's not totally accurate, but this is an analogy to use.

Take your car engine. You can of course go out and grab yourself one of several options. You want power, acceleration etc, but also fuel economy. You could buy a 4.0 litre model, which has plenty power and speed, but requires the economy of a small African country to run, or you could get a 2.0 litre model, with turbo. Similar but different type of performance. With the fuel saving. Originally Intel went for grunt, hoping that they'd be able to play down the power consumption, or improve as time went by. AMD went with the turbo option. Core 2 Duo is Intel deciding that pumping out 6.8 litre fuel suckers isn't really viable, as you can't get much bigger, and they generate a lot of heat. And heat means bigger cooling systems. And nobody wants a bulky techno gadget. So they decided to go the turbo route as well. But with Core 2 Duo, they've gone from grunt, to whish, with some serious turbo tech. They've realised that it's easier to get big performance gains by using the turbo with low power consumption, than the increase in clock speed. They've also given themselves headroom to increase clock speed on these models, thereby again increasing performance. Obviously AMD also have this capacity. Now it's a race of who can come up with the best turbo technology for their procs. Hope that helps
 
ShockG actually puts it better and more acurately than I do. I'm suffering from severe lack of sleep, and disappointed I couldn't think of such a simple explanation.
 
So its not really the fact that its two cores is it?

They are simply going the way of AMD, with more instructions per clock? I wasnt really sure, but this is what I thought.
 
Thanks KingRob

kingrob said:
graviti, you need to get some sleep... seriously. :)
Click to expand...

Was it that far off... Next time I'll sleep before I act. But in reply to the question. Dual core has plenty to do with it. And so does instructions per clock cycle, and amount of cache per core, and many other factors. One of the biggest deciders for the slower clock speed, apart from being able to release them at a later stage, was that higher clock speeds require more power which equals more heat, which equals more cooling, which equals more space required, which equals big, which equals ugly, which means people don't buy, if you get what I mean. Less power also means lower electricity bill. i.e if you live in CT, sell that power hungry P4, and get a Core 2 Duo. Sell that AMD and get a Core 2 Duo. However this only is NB if you live in CT, the land of darkness. Either way, Core 2 Duo accomplishes more than the higher speed processors, coz it does more in one cycle, because it has more to do it with. My favourite example, again not always accurate. Pile of rocks to move from one place to another. Ferrari with boot space for one rock at a time, and fuel consumption to boot, or Toyota hilux bakkie, with space to carry the whole lot in two loads. Ferrari is faster, but Toyota is more efficient. The general consumer wants efficiency, so that's what they are finally getting.
 
I guess all thoughts should be directed to ShockG.

btw, Shock, do you have a copy of crysis running yet? I guess theres not muhc use for all your fancy hardware without anything to push it.
 
graviti said:
It's not totally accurate, but this is an analogy to use.

Take your car engine. You can of course go out and grab yourself one of several options. You want power, acceleration etc, but also fuel economy. You could buy a 4.0 litre model, which has plenty power and speed, but requires the economy of a small African country to run, or you could get a 2.0 litre model, with turbo. Similar but different type of performance. With the fuel saving. Originally Intel went for grunt, hoping that they'd be able to play down the power consumption, or improve as time went by. AMD went with the turbo option. Core 2 Duo is Intel deciding that pumping out 6.8 litre fuel suckers isn't really viable, as you can't get much bigger, and they generate a lot of heat. And heat means bigger cooling systems. And nobody wants a bulky techno gadget. So they decided to go the turbo route as well. But with Core 2 Duo, they've gone from grunt, to whish, with some serious turbo tech. They've realised that it's easier to get big performance gains by using the turbo with low power consumption, than the increase in clock speed. They've also given themselves headroom to increase clock speed on these models, thereby again increasing performance. Obviously AMD also have this capacity. Now it's a race of who can come up with the best turbo technology for their procs. Hope that helps
Click to expand...

graviti said:
Thanks KingRob



Was it that far off... Next time I'll sleep before I act. But in reply to the question. Dual core has plenty to do with it. And so does instructions per clock cycle, and amount of cache per core, and many other factors. One of the biggest deciders for the slower clock speed, apart from being able to release them at a later stage, was that higher clock speeds require more power which equals more heat, which equals more cooling, which equals more space required, which equals big, which equals ugly, which means people don't buy, if you get what I mean. Less power also means lower electricity bill. i.e if you live in CT, sell that power hungry P4, and get a Core 2 Duo. Sell that AMD and get a Core 2 Duo. However this only is NB if you live in CT, the land of darkness. Either way, Core 2 Duo accomplishes more than the higher speed processors, coz it does more in one cycle, because it has more to do it with. My favourite example, again not always accurate. Pile of rocks to move from one place to another. Ferrari with boot space for one rock at a time, and fuel consumption to boot, or Toyota hilux bakkie, with space to carry the whole lot in two loads. Ferrari is faster, but Toyota is more efficient. The general consumer wants efficiency, so that's what they are finally getting.
Click to expand...

I still like ShockG's explanation better... :p Maybe use more spacing graviti? ;)

ShockG said:
It's instructions per clock, more than the clock speed.
say you have 2 cpu's. CPU X and Y.
X = 3 intstructions per clock, Y = 6 instructions per clock.
If you need to do 600 instructions.
CPU X needs 200 cycles to complete the instructions, but CPU Y only needs 100 cycles.
See pentium4 is like CPU-X but Core2Duo is CPU-Y.
2GHz Core2Duo = 4GHz Pentium4

It's not exactly like this and is far, far more involved, but you should get the general picture
Click to expand...
You're the man!! :D
I can definitely see speed improvements from my 3.2GHz P4 to Core2Duo 2.33GHz.
Any pointers for a newbie wanting to do some benchmark tests to confirm how sweet his ride is? :p It's for a notebook.
 
The Intel Core architecture is clearly the heir and descendant of the hugely successful P6 architecture. However, it has state of the art technology on board such as micro-op/macro-op fusion, memory disambiguation and massive SIMD/FP power.

Compared to the excellent AMD K8/Hammer architecture, the Core CPU is simply a wider, more efficient and more out of order CPU. When I suggested to Jack Doweck that the massive execution resources may not be fully used until SMT is applied, he disagreed completely. Memory disambiguation should push the current limits of ILP in integer loads a lot higher, and the massive bandwidth that the L1 and L2 can deliver should help Core to come close to the execution utilization percentages of the current P-M. 33% more execution potential could thus come very close to 33% more performance, clock-for-clock.
Click to expand...

Very interesting article below comparing the K8 vs Core

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2748
 
hmm old against new. :/
We will have to re-evaluate when Barcelona arrives early next year. Then Barcelona vs Conroe will be a more fair comparison. :)
Having said that, Core2 is really nice :D
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X