C++ C Prolog
Switch branches/tags
Nothing to show
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
core
imgs
include
interface
lib
module
plugin
.gitignore
COPYING
COPYING.LESSER
Doxyfile
INSTALL
README
gecko.pro
manual.doxy

README

This is a README document for the Generic Experiment Control Kit and Online
analysis framework GECKO.


INSTALLATION
------------

Find installation notes in INSTALL


USAGE
-----

After installation, type 

./gecko

on the command line.


GET STARTED
-----------

* Interfaces:

In order to communicate with the hardware, at least one interface has to be added
to the list of interfaces on the right.

Click with the right mouse button on the "Interfaces" item in the list and a
dialog window pops up, where the new interface can be configured.

Choose a name for the interface, then choose the interface type and (for VME
interfaces) assign a base address.

The first interface is the so called "main interface", which has some special
privileges. See below for more information on Main Interface.

When an interface is selected, you can change its settings in the right part
of the window or access the hardware manually. For VME interfaces it is
possible to send single shot VME calls.


* Modules:

Each piece of hardware is modeled in a software module, which can be added to the
list in the same manner as the interfaces.

A unique name has to be given to each module and then a type has to be chosen.
Also it is necessary to assign a default interface to each module.
The software module will use this interface to communicate with the hardware.
For VME modules the base (hardware) address has to be assigned.

When a module is selected from the list, it is possible to change its settings
in the right part of the window. This is the most important part of the setup,
so make sure that you know what you are doing. Please refer to the respective
user manual of each module, if you are unsure.


* Channels and triggers:

Once one or more modules have been added to the list, a number of channels and
triggers will appear in the list boxes on the "Run setup" page.
Here you can choose, which channels are important for the current run setup,
and which module will serve as the main trigger for the data acquisition.


* Plugins:

To make use of the data from the modules, the data has to be processed by plugins
after readout. Special demultiplexing plugins (demux) take care of unpacking the
raw data from each module, so that for each acquisition cycle the new data can
be accessed from the respective channel.

Plugins connect to the channels and collect the incoming data, process it and
provide access to the processed data.

Adding a plugin to the list is similar to adding a module, but the plugins are
also sorted into groups. So first choose a plugin group, then the plugin.
Give it a name and adjust free parameters, if necessary in the bottom list box.

Once the plugin appears in the list, you can select it and edit its preferences
in the right part of the window. Most importantly you have to connect the
plugin to a module channel or an output of another plugin.


* Plugin connections:

The upper part of each plugin settings area contains two list boxes, the left
one showing the input connectors and the right one showing the output connectors.
Using the right mouse button you can now select an opposite side for each
connector to connect to. For instance one input of a histogram plugin can be
connected to the third channel of a QDC module.


* Run setup:

Once all modules and plugins have been set up, the important channels have to be
selected in the "Run Setup" tab. Unselected channels will be read out, if they
happen to contain data, but an acquisition cycle is valid, if only the selected
channels carry data. So the notion is to have "important" channels which MUST
have data and "unimportant" channels which MAY have data.

In the list of triggers, one module must be selected to serve as the trigger
source (if a module can indicate a trigger in several ways, several different
trigger entries are visible for the same module).


* Saving settings:

Use the "File -> Save..." menu entry to choose a name for the current configuration.
All settings of the current session will be saved to the file. Attention: There
is NO autosave. So please remember to save your settings again, if you've made
important changes.


* Loading settings:

With the "File -> Load..." menu entry a *.ini file can be selected from which the
configuration will be loaded.


* Run control:

In the "Run Control" area you first have to select a directory, where all data for
the current "run" will be saved. This includes all data, the settings file,
special info files containing information on the run itself and also manually saved
data, like histogram snapshots or similar.

The next action should be to describe the current run in the "Run Info" text area.
Here the basic principle "more is more" is generally true.

Then double check that all settings are fine and save the settings again (Ctrl+S).

Make sure, that the trigger contition is met and hit "Start Run" to, well, start
the run.

Run statistics will be displayed in the upper half of this window.

Note that it is not possible to modify the module settings or the plugin connections
during a run. This would lead to unexpected results and most definitely to a crash.

To stop the run, simply press the "Stop Run" button.


REMOTE CONTROL
--------------

Gecko can discover other running instances of the software on the same network
via UDP broadcasts. This can be done on the "Remote Control" page.

Once a peer is discovered after pressing the "Discover" button, a connection 
can be established by pressing the "Connect" button. Of course, if the IP address
of the peer is known, it can also be typed into the address field. The status of
the peer is then updated on a regular basis. If no continuous update is necessary,
a single shot update can be requested via the "Update" button.

Using the "Start Remote Run" button, the remote instance can be controlled, i.e.
started and stopped. This is only possible, if the remote instance is not already
remotely controlled by a third party. If that is the case a notification showing
the controlling IP address pops up.