I just bought my new Mac Studio M1 and trying to run my old workflows on KNIME is becoming a nightmare. R packages like edgeR cannot be downloaded for M1 chips, only for Intel chips. Consequently, R-snipped for Gene-expression-analysis does not work on my new Mac Studio.
Also, most of the processes that are supposed to run on the GPU are using the CPU.
If you know a solution for the R problems or for GPU usage I will be very happy.
Thanks in advance.
I’m afraid on KNIME side there isn’t much we can do once we hand off to R. It will take some time for the developers to port their packages to M1. However, on R forums, I would imagine you may already find some workarounds and possible configurations to mitigate performance cuts and compatibility issues when emulating the x86 builds.
@marvin.kickuth Yeah I kind of figured it out about R, but it is not only limited to R but also Python since KNIME only leverages GPU acceleration from INVIDIA cards only. This will dramatically reduce the number of Mac users utilizing KNIME for serious ANN work.
Regarding R, I did find a workaround. You need to install R for Intel and it works. Of course, it is not optimal because it impacts the performance. It is like using a Ferrari in Florida where you cannot exceed 100 km/h.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.