Feedback & Feature Requests
As far as I can tell, it is only possible to transpose Input matrices. This would normally be done in the following way:
MW = MI.1.1.T
I am not able to do the same with working matrices, like so:
MW = MW.T
I find this very annoying as it means that if I want to make any changes to an input matrix before transposing it, I need to do these operations in separate MATRIX programs.
Ideally, I would like to see the transpose procedure implemented in the same way as the built-in MATRIX functions ( e.g. Transpose(MW) ).
Your suggestion is a really nice idea. We will contact our developers to look into this. Hopefully, this feature will be included in our next release of Cube.
For now, you need to have a separate MATRIX program to transpose the working matrices.
Thank you for sharing your idea!
Yes, I agree.
It would also be nice to say someting like:
MW = MW+2*MW+MW.T
I just want to throw another 'yes, please!' to this... especially since I'm about to code this manually AND because I have an orange box on the screen that says "REVIVE THIS TOPIC? It's over 340 days old".
This improvement and the improvement I emailed in a few months ago (tear-off tabs that can be dragged to another screen) would make the next Cube version quite nice.
I am not sure how transposing a working matrix will work. The matrix program will do zone by zone processing and as such, at any stage, the working matrix will only be partially complete up to that zone, which is currently being processed. Transposing this data will give incomplete transpose matrix. For example, the figure below shows how a working matrix transpose will evolve at each zone processing. Could you give us some use cases of such a transpose for our reference? And also there could be run time performance implications when doing transposes for each zone processing.