Usage: Logging

Logging is an option for those users who need to know more about a simulation than what is shown in the simulation window and the population graph. It also provides a sometimes-surprising glimpse into the dynamics of the evolution of one or more species.

When logging is enabled the user is prompted for a new filename (an extension of .LOG is assumed). After correctly entering and confirming a filename the file is created and logging begins if file did not previously exist, otherwise logging information is appended to the existing file. In either case simulation statistics are logged to the tab delimited file until is exceeds the maximum size specified by “M File Size” (above). If a simulation file and logging file are specified from the command line but logging is disabled in the *.SIM file the fact that logging was specified in the command line will override the simulation file and logging will be enabled to the specified log file.

The log data is stored in a standard CSV format that can be easily imported into your favorite spreadsheet or database program. To do so for MS-Excel for example, first import the *.LOG file as a space delimited file. Once in a standard spreadsheet format you would arrange all columns desired to be graphed so they were adjacent. Then select the appropriate column headers and all the fields in the columns you wish to graph. Then click on graph button, draw a rectangle with the crosshairs that the mouse cursor becomes and answer all questions appropriately. For more detailed instructions consult the user manual for your spreadsheet or database software.

You will probably want to make separate graphs for the species population and the species Standard Deviation. These two usually give an interesting perspective of the simulation

If file previously existed then a header is first appended showing when logging started. Then the column header is added as shown in the example below. The file is then updated with all applicable fields at the rate determined by the “Log Period” until logging is stopped, the program is halted, or the size of the log file reaches the maximum file size as specified by “M File Size”.

All fields are delimited by one or more spaces and are defined as follows:

Column

Description

D

# of days simulation has run

Time

Hours:Minutes:Seconds simulation has run

C

# of cycles elapsed

Pop

# of bugs currently on screen

BactS

# of bacteria on Screen (not including garden)

BacG

# of bacteria in Garden

P0..5

Population of each of the 5 possible species

SD0..5

Average Standard Deviation for each of the 5 possible species

MStr

Average Maximum Strength of all bugs

Mage

Average Maximum Age of all bugs

Repr

Average Reproduction Strength of all bugs

MRSt

Average Age of Maturity of all bugs

MS

Mutation Severity (G) – as defined in the menu

MB

Mutation Severity (B) – as defined in the menu

GH

Garden Height

GW

Garden Width

SL

# of bacteria placed on screen per turn

GL

# of bacteria placed in garden per turn

P

Whether garden protection is enabled (0 or 1)

As an example of how log file analysis can provide a more detailed look into the simulation dynamics, here are two graphs of a basic simulation using default parameters and no Garden of Eden.
The first graph (below) shows the relative population of the Bacteria vs. the Bugs for the first 89 cycles, prior to the development of any distinct “species”. For clarity, the population values of the bugs and the bacteria are shown relative to the maximum population of each so that they are presented on a common scale, similar to that of the the simulation graph. As no species exist prior to cycle 89, there is no standard deviation value to graph.

In the first graph (above) you can see that the Bug population oscillates in a somewhat erratic fashion but is generally out of sync with the bacteria population sine when more bugs are present they are consuming bacteria at a faster rate which causes the bacteria population to fall. When the bacteria population falls too far it is insufficient to support the bug population and the vulnerable bugs begin to die off, allowing the bacteria population to increase only to start the cycle over again.

The second graph (below) shows the relative population of the Bacteria vs. the Bugs starting at the point where a population of forward moving bugs first evolves (cycle 90) and is detectable via the standard deviation threshold of the program. As in the first graph, the population values of the bugs and the bacteria are shown relative to the maximum population of each.

In the second graph (above) you can see that with the evolution of a distinct forward moving species, the Bug population oscillates in a much more distinct fashion and is clearly 180 degrees out of phase with the Bacteria population. During the period graphed the bugs have stumbled on the optimal movement pattern (forward movement) to take advantage of the simulation environment and any other changes in the genes are due primarily to random mutation rather than environmental pressures.
VN:F [1.7.9_1023]
Rating: 0.0/10 (0 votes cast)
VN:F [1.7.9_1023]
Rating: 0 (from 0 votes)
Share and Enjoy:
  • Print
  • Digg
  • Sphinn
  • del.icio.us
  • Facebook
  • Mixx
  • Google Bookmarks
  • Blogplay

Leave a Reply

 

 

 

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>