Is there a way to get one task to run at a time and use all the CPUs available, I presume the task would finish faster??

Is there a way to get one task to run at a time and use all the CPUs available, I presume the task would finish faster??

Only for those projects specifically coded to use more than one CPU.

Milkyway: nbody

gpugrid: quantum chemistry (if they have any work units)

Amicable Numbers

there may be others.

I would add that some projects such as CPDN will never support multi-core tasks because the nature of the work is that each calculation depends on the result before it. CPDN gets at least two or three messages on their boards each year asking why they don't support it. Some projects I guess it may be just that no one wants to do the work of programming for multi-core tasks. (I have no idea how much extra work is involved.)

It is really down to the particular project(s) you run.
Some projects have already developed multi-core applications, many have not for a variety of reasons, not the least of which being time to develop and test - the project needs to be certain that each and every application it distributes returns correct results (and that includes errors when a data-set has poor quality data that falls outside the valid range of values).

LHC and Cosmology have multi-threaded executables. They run under VirtualBox only. Unfortunately my experience is that the VMs are badly behaved with more than one heavy thread, with or without hyperthreading. I had to limit tasks to one thread (LHC) or run them on an unattended box where user responsiveness doesn't matter.

---