How come CPU tasks take so much longer to complete than GPU tasks.

Is it because GPU’s are much faster/optimized for these tasks, or are the CPU tasks put together, like CPU tasks equals 10, 20, or even 50 GPU tasks, and that’s why it takes so much longer to complete?

Just wanna know if my CPU time is better spent on another project, just like GPU’s running Folding@Home generate many times more points per day than CPU’s do, why you should run BOINC projects for your CPU instead.

I would say, that is because your gpu has CUDA cores & or tensor cores. (I forget what the AMD side of the house calls them.) They are blazing fast. Like my CPU has 192 cores, it takes about an hour to one packet, where as my 4090 can do the same workload in under 2 & 1/2 minutes.

The CPU application is outdated and doesn't fully leverage SIMD capabilities, whereas the algorithm is more GPU-friendly. For math-intensive and parallelizable workloads, GPUs deliver substantially better performance, as they are specifically designed for this type of computation.
I recommend allocating your CPUs to projects that either lack a GPU version or where the GPU version isn't as efficient as the CPU counterpart...

The CPU application is outdated and doesn't fully leverage SIMD capabilities, whereas the algorithm is more GPU-friendly. For math-intensive and parallelizable workloads, GPUs deliver substantially better performance, as they are specifically designed for this type of computation.
I recommend allocating your CPUs to projects that either lack a GPU version or where the GPU version isn't as efficient as the CPU counterpart...

For Intel Macs, the Einstein CPU applications are much more efficient than the GPU applications. For Intel Macs, the Einstein CPU applications are much more efficient than the Rosetta CPU applications. The latter is a bit of a moot point, because Rosetta is once again largely out of work. Happy crunching!

---