ONE dual core CPU or TWO single core CPU's

[BlahBlahYouToo]

lets pretend the architecture is the same except for what's mentioned in the thread title.

will the performance be equal or close to equal?

(the 2 single cores are on the same motherboard, 2 sockets)

 

[OVerLoRDI]

I think 1 physical cpu with 2 cores would be faster in every way keeping GHz, memory amount per core, and cache per core constant. I base this on shared L3 cache and decreased latency between the 2 cores.

In reality it would depend a lot on the application and the difference would likely be small.

 

[jones377]

Look up old reviews around the Athlon 64 X2 and PentiumD days, I believe some compared to dual socket systems.

 

[classy]

If I remember correctly when dual cores first came out they proved to slightly faster than a dual cpu system.

 

[aigomorla]

this is a rediculously difficult question to answer.

Because ive seen platforms which were 2 dualcores beating a quadcore any day of the week because of the more ram which is required on a dual cpu board.

I would go with the single dual core cpu, because its less things that can break on the board, vs a dual cpu board.

However dual cpu boards are more durable then single cpu boards.

 

[Scali]

In the case of Pentium D and Athlon X2, the difference is minor. The cores are completely isolated and communicating via a bus, very similar (in the case of Pentium D even identical) to how two sockets communicate.

In theory the Core2 Duo is faster than two single-core Core2's, because of the shared cache between the cores.

Ofcourse if we get into NUMA configurations (only possible if each CPU has memory controllers), then it can go either way. In some cases, it's better to dedicate memory to each CPU, and in that case, two CPUs would be better. But if you just share memory (and thus a memory controller), then the single CPU will probably do slightly better.

 

[tweakboy]

The dual core CPU is goign to be newer technology thus faster. Using one core vs a single core cpu. the dual core using its one core will win, because its newer generation new technology and at lower Mhz then the single core also.

those single core cpu's were amazingly weak , good ole dayz,, x800xt pe nightmares, LOL

 

[lambchops511]

depends-- if your platform supports NUMA.. going dual sockets might give u the memory bandwidth advantage depending on the app

 

[mv2devnull]

@tweakboy: OP did not ask for apples to oranges comparison.


Yet, the answer is dependent on era. There was a time, when cache was not in the CPU. There was the time, when memory controller was not in the CPU. There was a time, when dual-core was implemented by dumping two single-cores into same die.

The question is, where is the bottleneck? That does not depend only on the hardware, but also on the resource usage pattern of the algorithm. It is easy to write a (useless) loop that fits into the registers within core, just like it is easy to make a mainly network-bound routine.

As already said, forcing the dual-core(s) to share something (like L3 cache) that the separate CPU's don't, might be either good or bad. Thus, while it might be difficult to answer the "what is faster", one could answer what components are shared or not.

Alas, we are back in the "which era". Current generation of CPU's would say that memory controller is common/separate (ie differs), but in earlier generations it was on the motherboard, so there was no difference.

But in the end it is difficult to say whether the hardware differences of this hypothetical comparison actually cause substantial performance difference. Perhaps.

 

[Anarchist420]

I think the Sega Saturn would've been been faster if it had a single dual core SH2.

 

[Lorne]

Era has the biggest influence, The C2 did not use IMC and therefore required FB dim in 2 CPU systems, FB memory is slow and each core shared memory bandwidth and only single channel if I remember correctly.
So a single C2D/Q would be a better choice.
The C10/I series uses a IMC and can use dual or tri channel none FB dims, A 2 socket C10 has more meory bandwidth then a single socket C10 duo or quad and makes more use of the QPI.

Toms has and article (intel xeon 5600 series) showing the differences between the I7 980, dual 5500 and 5600, It gives a good example on this subject.

As for AMD the AMD 64 was also a step up with its IMC, Improved later with HT, They were just ahead of the game in that area till the C10/I came out.