Installed the license for Zorro S to be able to use all 8 cores for WFO training. I can run multiple Zorro instances and I was able to use all cores for WFO for a couple of days. Unfortunately this stopped working and I am not able to use more than one core for training. I can still run multiple Zorro instances but I can only use one core per instance.
I had Zorro S 1.60 and I have now the same problem with Zorro 1.66.

Anyone having similar issues ? Thanks in advance.

So I open Workshop6.c and set NumWFOCycles = 10. It spawns 6 processes and it works fine. I then set NumWFOCycles = -10 (negative number this time) and run Train again. This time it uses only one processor. No other changes required to reproduce the issue.

Maybe you're confusing WFO with multiple cores? A negative number just means that only 1 cycle is trained.

The documentation states that you can opt for Anchored Walk Forward Optimization by choosing a negative number (NumWFOCycles < 0). And it implies that you can choose more than one cycle to train.

Am I reading the documentation wrong and only the Rolling WFO can be trained on multiple cores?

EDIT: Here is what the documentation sais:

Number of cycles in a Walk Forward Optimization / Analysis (default = 0 = no Walk Forward Optimization). If NumWFOCycles is set to a positive number, rolling walk forward optimization is enabled with the given number of cycles; if it is set to a negative number, anchored walk forward optimization is enabled.

Could it be that anchored WFO recycles testing data from the first run?

It recycles training data from the first run, but testing data is not recycled. And @Orbiter, you're right - now I understand what you mean. A negative number anchores WFO to the first cycle. Multiple cores are used in normal WFO, not in anchored WFO.

Thanks guys. It makes sense now why the anchored training would use only one core as the second core would have to wait for the first one to finish the training, the third one for the first two ... and so on.

A note added to the documentation would not hurt plus some heads up on the fact that multi core training does not kick in if the source file is not in the Strategy folder. I also tried to "organize" the scripts and I created another folder. The confusion was even bigger but everything seems to be clear now.

Thanks.