Speed comparisons

[Dave2002]

I have several machines, and in particular an Apple MBP and a Mac Mini. I thought these two had similar specs, but today I checked, and this is what I found:

Both are i7s with 8 Gbytes of memory.

MM: 2.6 GHz
Processors 1
Cores 4
L2 256kb
L3 8 Gybte
Intel Graphics HD 4000 chipset (same as MBP) - VRAM 768 Mbytes

MBP
3 GHz
Processors 1
Cores 2
L2 256 kb
L3 4 Gbyte
Intel Graphics HD 4000 chipset - also 768 Mbytes VRAM


So which should be faster? I don't know. At first glance the MM should be slower, with the lower clock rate, but it's got more L3 cache, and runs at 2.6 GHz rather than 3 GHz. OTOH the MM has 4 cores, so depending on the software and how it uses the processor, that MM might actually be faster.

[Anastasius]

I'm sure you'll find your machines and the answer to your question here. https://browser.primatelabs.com/mac-benchmarks

Or failing that, roll your own http://9to5mac.com/2015/05/14/how-to...ree-downloads/

[Dave2002]

I'm sure you'll find your machines and the answer to your question here. https://browser.primatelabs.com/mac-benchmarks

Or failing that, roll your own http://9to5mac.com/2015/05/14/how-to...ree-downloads/

Difficult to be sure. Looking down the list it seems to me that the closest models of MBP and Mac Mini are similar, but with the MBP slightly ahead. OTOH 4 cores plus more cache of the mini might still win on some more complex tasks, even with the slightly lower clock speed, and I suppose it also depends on how software uses the machines.

I may try some tests, but without giving the machines something hefty to chew over, and with good code, I'd expect the MBP to do slightly better. Optimised software could make a big difference and reverse the order, possibly by quite a big margin. However, finding good quality code to exploit the machine usefully might not be so easy. No real point for regular type processing, but some applications, e.g. video rendering or conversion might show benefits - the sort of tasks which run for hours on nearly 100% cpu.