I have written some code which runs faster on a system with 2x16 core intel chip than a system with a 4x16 core AMD cpu. The AMD cpu is of a slightly slower clockspeed, and I have ensured my Threading.@threads loop has more than 64 items. I have found that the total runtime using the intel chip is around 60% that of the AMD chip. Does anyone have any explanation for this, or the behavior of threading on multiple CPUs?
It’s very hard to say what could be the cause of this without knowing what the code does, but this could be explained by better caching behavior on the intel chip.