Hi there, after a quick glance at the screenshots, I can see one issue: you haven't selected a population to display (instead the menu is on "Preview Populations"; this displays all cells including dead cells that autofluoresce in many channels). As a result I can't clearly see a positive and a negative population for each channel. It's possible this is making it harder for you to observe the compensation (but I can also see that your plots are not really compensated- so this can't be the only issue).

Why don't you try to manually compensate in flowjo using the compensation workflow? This will ask you to select a subset of cells on which to compensate. The resulting matrix may look very different to what you have acquired, and can also be edited further. To do this, display the channels as FL1 vs everything, FL2 vs everything and so on- as in your screenshot, then copy the matrix and tweak the individual cell values if needed. The fluorochrome currently being displayed (selected in "Preview Sample") corresponds to the row labelled with the same name. The columns are the amount of spillover from all other channels that you will be removing from the fluorochrome you are currently displaying. Aim to align the MFIs of the positive and negative population (i.e. most plots will look "square").

If your settings on the machine were correct, you will not have to edit the matrix much, or at all.

However, be careful with this- compensating by eye can be misleading and create false populations. It is easier if using very obvious yes/no markers on your single-stained controls (like CD3, CD4 or beads with a positive and negative population).