[SOLVED] Base clock vs boost clock

[nuwandhammika]

Imagine there are two CPUs . ( 6 core/ 6 thread CPUs)

1. 2.8 GHz base clock and 3.9GHz turbo boost speed .
2. 3.4 GHz base clock and 3.8 GHz turbo boost speed.

Rest of other cpu specifications are equal( equal number of cores , equal cache memory )
If cpu cooling and motherboard power delivery are good enough , which cpu may perform better in gaming ?

*does CPUs come with relatively lower base clock combined with somewhat around 4 GHz turbo boost clock are enough for Gaming ? Or higher base clock make a different in gaming ?
How base clock and boost clock are affect in gaming ? (Consider cooling and power delivery are good enough in all scenarios.)

 

[kanewolf]

Imagine there are two CPUs . ( 6 core/ 6 thread CPUs)

1. 2.8 GHz base clock and 3.9GHz turbo boost speed .
2. 3.4 GHz base clock and 3.8 GHz turbo boost speed.

Rest of other cpu specifications are equal( equal number of cores , equal cache memory )
If cpu cooling and motherboard power delivery are good enough , which cpu may perform better in gaming ?

*does CPUs come with relatively lower base clock combined with somewhat around 4 GHz turbo boost clock are enough for Gaming ? Or higher base clock make a different in gaming ?
How base clock and boost clock are affect in gaming ? (Consider cooling and power delivery are good enough in all scenarios.)

Unknown. Clock speed is not the only determing factor. Even with clock speed "boost clock" is not a uniformly defined thing. CPU #1 might hit 3.9 on one core and CPU #2 might hit 3.8 on all cores. And then there is the question of software. There is no "paper" answer. That is why benchmarking is so important.

 

[NightHawkRMX]

You cant look at it that way since you dont know what frequency they will actually run at in any particular workload.

You just have to look at the performance vs competition rather than clockspeed.

 

[nuwandhammika]

Unknown. Clock speed is not the only determing factor. Even with clock speed "boost clock" is not a uniformly defined thing. CPU #1 might hit 3.9 on one core and CPU #2 might hit 3.8 on all cores. And then there is the question of software. There is no "paper" answer. That is why benchmarking is so important.

What if cpu 1 also hit 3.9 on all cores ?

 

[NightHawkRMX]

Assuming the CPUs are otherwise identical, in an all core workload the one that runs at 3.9ghz all core would beat the one that run 3.8ghz all core.

 

[kanewolf]

What if cpu 1 also hit 3.9 on all cores ?

It is still UNKNOWN. CPUs have different "efficiencies". That can be instructions per clock, cache design even memory interface. A CPU with quad channel memory will generally have better performance than a dual channel design with the same RAM timings.
If you have a specific pair you would like help evaluating, then we might be able to help. These "hypothetical" CPUs aren't worth my time to continue to say "UNKNOWN" ...

 

[nuwandhammika]

Assuming the CPUs are otherwise identical, in an all core workload the one that runs at 3.9ghz all core would beat the one that run 3.8ghz all core.

how long those CPUs stay in turbo clock (full cpu load) if cooling and power delivery are enough ?

 

[NightHawkRMX]

how long those CPUs stay in turbo clock (full cpu load) if cooling and power delivery are enough ?

It depends on the boosting algorithms.