I don't believe that there is any hard-coded limitation from the Qlik side else it will depend from the executed task to how many cores/processors it could be divided, for example a single bar-chart on a sheet will probably not much benefit from a 32 core-machine but if there are 10 charts which needs to be calculated at the same time it will increase the overall performance.
Beside the capabilities from Qlik to distribute the task to n cores it depends from the used processor-type and their caching-approach how well it worked because the processor needs to handle all the threads in a certain order and the cores need to communicate with each other. This means there will be always some amount of overhead and it could even delay the processing itself. AFAIK the old AMD Opteron were often faster by using only 2 or 4 cores and disabling the other n cores because the queueing of all threads needed more time as the real calculation. I think the best deduction what might be working will you get from the various postings within the Qlik-Scalability.
I assume my answer won't really satisfy you but I think there is no easy statement that n processors/cores for a certain kind of task are the min/max or especially suitable for your requirements.
- Marcus
