Creating an experiment

Experiment view creating an experiment

When you create an experiment, you take a "snapshot" of the system’s current parameters and (if required) results. Therefore, you simply set the parameters as required for the experiment, and optionally, perform a simulation.

Once you are ready to create an experiment, you have three options:

Icon	Experiment type	Description
	Create a linked experiment	The experiment is saved in a user-specified folder (not in the circuit's .ame file). When saving the experiment, make sure you do not use unicode characters in the experiment filename as the experiment might fail after being applied to a model.
	Create an embedded experiment	The experiment is embedded within the circuit's .ame file.
	Create a parameter-only experiment	The experiment only contains the circuit's parameters (not any results that were generated).

Note

These three options are also accessible through the contextual menu.

Warning

If you create a linked experiment, and then move your circuit to a different location (without access to the folder where you stored your linked experiment), the link will of course be broken:

Note

You can also create an experiment directly from the Run Monitor. See The Run monitor

When you select one of these options, a new entry appears in the Experiment view:

Figure 131: A saved experiment

Experiment view editing

The entry lists the status, name, description, and creation date for the experiment. You can edit the name and description through a right-click:

If you now change some parameters in your system and perform a new simulation, you can click the Save to experiment button again, to create a new entry in the list of experiments:

Figure 132: The list of experiments

The default name and description are incremented. Obviously, this does not give you any useful information on the different parameters used in each experiment. This is why it is useful to be able to edit the description and name each time you create an experiment.

Note

You cannot modify an existing experiment’s parameters. An experiment is simply a snapshot of the system at a particular moment. If an experiment is ‘wrong’, or unneeded, simply delete it and create a new experiment as required.

If you save an experiment without first performing a simulation run, you can only save the parameters (either as a linked or embedded experiment); you will see that the Create embedded experiment icon is greyed-out: