Multi-Sub Optimizer Tutorial (page 15)
Earlier in the tutorial, in order to compare the results before and after optimization, you saved the project as tutorial_1.msop before optimizing, and obtained the graph of Figure 31. After optimization, you saved the project as tutorial_2.msop and obtained the graph of Figure 30. This kind of comparison can be done within a single project using the Configurations feature.
The Configurations feature allows for optimization using different types of filters, different filter parameters and so on. Save the current project and open up the tutorial_1.msop. Save it as tutorial_3.msop so you can still go back to the original if you make a mistake. Select the name of the configuration, Config 1, and press F2. Rename the configuration to Original. Right-click on Original and choose Clone Configuration. This will cause the Clone Configuration dialog to be launched. In this dialog, enter Optimized in the Choose a name for the new configuration box. Make sure the checkbox labeled Clone associated graphs is checked. This will create an identical copy of the Optimized configuration. In addition, all graphs of the Original configuration will be copied, and the traces of the copy will be changed to refer to the new Optimized configuration. This means all your graphs will be set up for the new configuration automatically. After collapsing the tree structure of both Original and Optimized, the result should look as below.
The Original configuration should be kept as-is to enable performance comparison before and after optimization. To protect it, you can lock all the parameters of all the filters in the configuration. Select the topmost node of the Original configuration as shown above in Figure 33. Right-click and choose Lock All Filter Parameters of this Configuration from the context menu. This will set the Optimization allowed property of all parameters of all filters in Original to False. Verify that this is the case by expanding the Original node. The icons for all the filters in the Original configuration will now be a dark gray color, indicating that the optimizer will not adjust any of their parameters.
Press the Start optimization button on the toolbar. A dialog box pops up, asking you to specify which configuration you want to optimize. This dialog only pops up when there is more than one configuration present. Select Original and click OK. You should get the message The configuration named “Original” has no optimizable parameters. This is expectred, since you just locked all its filter parameters. Since this project originated with the tutorial_1.msop that was saved prior to running the optimization, you'll need to run the optimization again to optimize the filter parameters for the Optimized configuration.
Activate the Data View by clicking on the Data View tab at the lower left of the main window. Rename the Combined Responses graph by clicking on the Combined Responses text under Graphs and pressing F2. Rename it to Combined Responses (Original). You'll notice that in the process of cloning the original configuration, a new graph named Optimized: Copy of Combined Responses has been created. Select this name, and using F2, rename it to Combined Responses (Optimized) as shown below.
Notice that the trace names in the newly-created graph are the same as the trace names of the original. This was done automatically in the Clone Configuration operation. To avoid possible confusion caused by multiple open graphs, close the Combined Responses (Original) graph window if it is open, so that only the Combined Responses (Optimized) graph is shown.
Now is a good time to save your work, so press Ctrl+S to save or choose File, Save from the main menu to save the project.
Press the Start optimization button on the toolbar again, but this time choose Optimized as the configuration. In the dialog, you can either click OK or double-click the configuration name to start the optimization. This is a complex optimization, so let it run for the full 30 minutes to ensure the best results.
Since the optimizer uses a random number generator in its optimization algorithm, the results vary somewhat from one equally long optimization run to another. Because of this, open the tutorial_3_prefab.msop in the Sample Projects sub-folder of the tutorial to ensure your results match the tutorial. Then, use File, Save As to save it as tutorial_3.msop, overwriting the tutorial_3.msop you created earlier.
As an exercise on your own, try out what you learned earlier in Getting Information About the DSP Filters to create a graph named Optimized Filter Responses showing the frequency responses of the Sub 1, Sub 2, Sub 3 and Sub 4 channels of the Optimized configuration. Choose manual scaling for the y axis and name the traces Sub 1, Sub 2, Sub 3 and Sub 4. You can also try the step of doing a filter report for the Optimized configuration. That exercise has been worked out and can be found in the tutorial_4_prefab.msop in the Sample Projects sub-folder of the tutorial.