Clock Speed

[haripriyaaithal]

What is the meaning of Speed of the processor is 3Ghz,?

Does it mean that my processor can perform 3 billion instructions per seconds???

 

Please anyone help me to know this!  

[Godlygamer23]

It means the processor is able to achieve 3 billion cycles per second. Cycles are not the same as instructions. Clock speed is comparable to RPMs in a car. It only tells you how many rotations the crankshaft is performing - it tells you nothing about the actual performance of the engine.

[Glenwing]

 

What is the meaning of Speed of the processor is 3Ghz,?

Does it mean that my processor can perform 3 billion instructions per seconds???

 

Please anyone help me to know this!  

 

 

It doesn't mean anything too specific. It means the processor performs 3 billion "cycles" per second. What exactly a "cycle" entails depends on the processor design and type of task you're doing. Some operations are performed in a single cycle, other instructions take multiple cycles to finish, so it depends on what the program is doing and what instructions/operations are being used.

 

The important thing to take away is that it means different things for different processors. It isn't an absolute measure of "speed", so it's not very useful to compare the "GHz" number between different processors unless they're exactly the same in every other way.

[haripriyaaithal]

Okay, then what exactly one "cycle" mean?

[Godlygamer23]

Okay, then what exactly one "cycle" mean?

It is when the processor completes one operation - what entails an "operation" is dependent on the architecture.

[Glenwing]

Okay, then what exactly one "cycle" mean?

 

This question has no answer, as "cycle" does not have any exact meaning in performance terms. The CPU's internal clock (3 billion ticks a second, or 3GHz) is just how every part of the CPU coordinates with all the other parts. Every time the clock ticks, every part of the CPU takes one step forward. Some instructions may take 1 cycle to complete, other instructions may take 5 cycles to complete. Every CPU design and every different type of instruction will be different in this respect.

[haripriyaaithal]

Thank You friend

[D3MON]

Okay, then what exactly one "cycle" mean?

you're going to have to do allot of googling on this but the simplest way to understand it is a light switch. (the engine rpm is a good one aswell, especially if you've ever had an engine receive a signal from a cam sensor every time the crank rotates).

if you turn it on then some part of an instruction is done, then its turned off and then on again and the next part is executed.

the faster you do it the faster the cpu can perform a task, IPC is how many tasks can be done per cycle.

more of either increases performance.

id also recommend checking out pipelines and superscaler architectures, simple but cool stuff.

[Comessy]

It is when the processor completes one operation - what entails an "operation" is dependent on the architecture.

This is how many each arcitecture does per cycle, according to :

http://stackoverflow.com/questions/15655835/flops-per-cycle-for-sandy-bridge-and-haswell-sse2-avx-avx2

 

 

 

 

Here are FLOPs counts for a number of recent processor microarchitectures and explanation how to achieve them:

Intel Core 2 and Nehalem:

• 4 DP FLOPs/cycle: 2-wide SSE2 addition + 2-wide SSE2 multiplication
• 8 SP FLOPs/cycle: 4-wide SSE addition + 4-wide SSE multiplication

Intel Sandy Bridge/Ivy Bridge:

• 8 DP FLOPs/cycle: 4-wide AVX addition + 4-wide AVX multiplication
• 16 SP FLOPs/cycle: 8-wide AVX addition + 8-wide AVX multiplication

Intel Haswell/Broadwell:

• 16 DP FLOPs/cycle: two 4-wide FMA (fused multiply-add) instructions
• 32 SP FLOPs/cycle: two 8-wide FMA (fused multiply-add) instructions

AMD K10:

• 4 DP FLOPs/cycle: 2-wide SSE2 addition + 2-wide SSE2 multiplication
• 8 SP FLOPs/cycle: 4-wide SSE addition + 4-wide SSE multiplication

AMD Bulldozer/Piledriver/Steamroller, per module (two cores):

• 8 DP FLOPs/cycle: 4-wide FMA
• 16 SP FLOPs/cycle: 8-wide FMA

Intel Atom (Bonnell/45nm, Saltwell/32nm, Silvermont/22nm):

• 1.5 DP FLOPs/cycle: scalar SSE2 addition + scalar SSE2 multiplication every other cycle
• 6 SP FLOPs/cycle: 4-wide SSE addition + 4-wide SSE multiplication every other cycle

AMD Bobcat:

• 1.5 DP FLOPs/cycle: scalar SSE2 addition + scalar SSE2 multiplication every other cycle
• 4 SP FLOPs/cycle: 4-wide SSE addition every other cycle + 4-wide SSE multiplication every other cycle

AMD Jaguar:

• 3 DP FLOPs/cycle: 4-wide AVX addition every other cycle + 4-wide AVX multiplication in four cycles
• 8 SP FLOPs/cycle: 8-wide AVX addition every other cycle + 8-wide AVX multiplication every other cycle

ARM Cortex-A9:

• 1.5 DP FLOPs/cycle: scalar addition + scalar multiplication every other cycle
• 4 SP FLOPs/cycle: 4-wide NEON addition every other cycle + 4-wide NEON multiplication every other cycle

ARM Cortex-A15:

• 2 DP FLOPs/cycle: scalar FMA or scalar multiply-add
• 8 SP FLOPs/cycle: 4-wide NEONv2 FMA or 4-wide NEON multiply-add

Qualcomm Krait:

• 2 DP FLOPs/cycle: scalar FMA or scalar multiply-add
• 8 SP FLOPs/cycle: 4-wide NEONv2 FMA or 4-wide NEON multiply-add

IBM PowerPC A2 (Blue Gene Q), per core (supports 4 hyperthreads):

• 8 DP FLOPs/cycle: 4-wide QPX FMA every cycle
• SP elements are extended to DP and processed on the same units

IBM PowerPC A2 (Blue Gene Q), per thread:

• 4 DP FLOPs/cycle: 4-wide QPX FMA every other cycle
• SP elements are extended to DP and processed on the same units

Intel MIC (Xeon Phi), per core (supports 4 hyperthreads):

• 16 DP FLOPs/cycle: 8-wide FMA every cycle
• 32 SP FLOPs/cycle: 16-wide FMA every cycle

Intel MIC (Xeon Phi), per thread:

• 8 DP FLOPs/cycle: 8-wide FMA every other cycle
• 16 SP FLOPs/cycle: 16-wide FMA every other cycle

[BurgerBum]

you're going to have to do allot of googling on this but the simplest way to understand it is a light switch. (the engine rpm is a good one aswell, especially if you've ever had an engine receive a signal from a cam sensor every time the crank rotates).

if you turn it on then some part of an instruction is done, then its turned off and then on again and the next part is executed.

the faster you do it the faster the cpu can perform a task, IPC is how many tasks can be done per cycle.

more of either increases performance.

id also recommend checking out pipelines and superscaler architectures, simple but cool stuff.

essentially, your proccessor is the light switch.

how it preforms is based on how many light bulbs are wired up to it and how fast the todler is messing with it. 

 

 

the only way you can calculate how your proccesor preforms is by knowing both IPC and clock rate. although simply googling bench marks form a reliable source is just easier.

[Faceman]

Thank You friend

You shouldn't weight the strength of a processor based on its speed or number of cores.  The architecture of a processor has much more to do with its strength than cores or speed, especially for gaming performance.