Comparing Pentium 4 to Single Core of Dual Core Processor

[smstx]

I'd like to compare the speed of a Pentium 4 processor to a single core of a Dual Core Processor.

Based on this chart, would it be accurate to say that a single core in the dual core processor is almost four times faster than the Pentium 4 processor? (that seems high to me). I'm comparing the Intel Core 2 Duo E4500 @ 2.2 Ghz to the Pentium 4 2.53Ghz. The Core 2 Duo had a CPU mark of 1178, while the Pentium 4 had a mark of 350. Is the Core 2 Duo's score gaining an advantage from having two cores, or is the score a reflection of a single core speed? Are the scores then an accurate indicator of the ratio of calculations per clock cycle between the two?

To say it another way, I'd like to compare 2.2 Ghz in a single core to the 2.53 Pentium IV directly, in case I have software that can't take advantage of two cores to maximize performance (like older games and such).

Thanks for your help.

[David (PassMark)]

The Core 2 Duo has a much higher score as a result of,
1) It having 2 cores (which are fully used by the benchmark that made the graph)
2) The Core 2 being a more modern CPU higher performance architecture than the P4, even with the same number of cores and at the same clock speed.
3) The Core 2 almost definitely being used on a newer motherboard with faster RAM and BUS.
4) More of the Core 2 CPUs being used with Windows 64bit rather than 32bit.

The short answer is that the Core 2 chips will beat the Pentium chips with pretty much any software. In some cases the advantage might be small, in other cases the advantage will be huge (particularly with multi-threaded software).

[smstx]

Thanks, that's very helpful. I am curious, though, how one might compare a Pentium 4 to a single core of a dual core processor. Is there any way to do this? I'm pretty content knowing that the dual core will beat the Pentium in just about every case. I'm just wondering exactly how much better it actually is in a core-to-core comparison, as the Ghz don't seem to communicate much of anything any more, between different chip types.

[David (PassMark)]

The difference will depend on the application used, and if all other components are the same (e.g. MB and RAM) and all software is the same (like the O/S and device drivers).

[smstx]

Great, thanks for the helpful feedback.

[folst02]

I would like to compare a single processor system to a dual processor system.

Single processors:
Intel Xeon E3-1220 V2 @ 3.10GHz - 6,517
Intel Xeon E3-1240 @ 3.30GHz - 8,147

Dual Processor:
[Dual CPU] Intel Xeon E5504 @ 2.00GHz - 4,987 Version 8 of benchmark
[Dual CPU] Intel Xeon E5504 @ 2.00GHz - 6,254 Version 7 of benchmark

I am assuming the benchmarks can be compared between dual and single processors systems as long as they use the same version of the benchmark. So according to the information extracted from the benchmark tables, the single processor systems should be comparable or faster then the dual processor systems. Is this correct?

[David (PassMark)]

Yes, one of the new Xeons will beat two of the old Xeons in most tasks.

The Xeon E5504 is about 5 years old now, and only ever had a slow 2Ghz clock speed.