|
Why is it free?
e-den is free because it started off as a hobby, with no commercial intent.
Furthermore, the e-den project needs to find people prepared to donate some simulation time
to the e-den master grid. Users will get some free entertainment, hopefully, and some insight into
evolution or neural nets. The e-den project, on the other hand, gets the benefit of distributed
computing, which seems like a fair trade.
Why aren't my organisms doing anything?
Lazy organisms may survive and reproduce in e-den if conditons are too easy. If the 'climate' parameter is set very high, grass will grow next to the organisms fast enough that they don't have to move. Furthermore, by staying still, they avoid the risk of a collision. Get rid of the current crop of lazy organisms with the 'reset' button and then try again with the climate set to a low value (20 or so).
Why don't my organisms reproduce despite having enough metabolic fuel to do so?
e-den doesn't force sensible behaviour on its organisms.
They can only breed when they have a 'metab' value greater than (metab-div + metab-child) but,
even then, they must decide to breed. The neuron responsible for this decision is number #008
and you can make an organism more likely to breed by setting the 'drift' parameter for neuron #008
to a positive value. This is best achieved in the 'Bug Lab' window but check that it has worked
by inspecting the organisms brain in the 'Net' window. Another problem might be that the organism fails to grow to adult length. The neuron responsible for 'extending' is number #006 and this, too, should be given a positive value for drift. (A more sophisticated solution is to make an excitatory link between the 'metab' neuron and the 'extend' neuron so the organism extends when well-fed). NOTE: the default herbivore shipped with some beta versions of e-den has a negative value for drift in these neurons... For this species to be viable, this needs to be fixed and then it is worth saving the genome as C:\biology\protectd\org1.ini, overwriting the defective species. You will find that progress is very slow until you have at least one species that is truly viable, able to reach generation 100 over a few hours. You might do better trying this sample genome ...
Why doesn't my organism have the parameter I just set in the 'Bug Lab' window?
The e-den genetic syntax allows polygenic specification of all continuous variables.
Included in this category are all metabolic parameters and neural parameters such as 'drift',
'flux' and 'threshold'. During traversal of the genome during embryological development,
the first time a value is specified for one of these parameters, it is set to that value.
The next time the old value and the new are averaged, and thereafter the new value and the
current value are averaged in a rolling process. (Neural connections, by contrast, are considered
discrete and new specifications overwrite earlier ones). If you add a gene in the 'Bug Lab' it
won't overwrite previous values but will contribute to the rolling average. If you add the same
gene 3 or 4 times, the rolling average will move closer and closer to the desired result. The moral is:
always check that the organism has the value you think it should have.
Why haven't my organisms evolved much after several hours of continuous computer time?
Not even Darwinian evolution can escape the fact that it takes time to accumulate mutations.
If a 20 digit sequence is needed to specify some great new neural reflex, then this would take
20 generations to appear even if every reproductive event were accompanied by a mutation in the
relevant gene and each mutation were made to order by a guiding deity. If only 1% of reproductive
events were accompanied by a mutation in that gene, then even with guided mutations it would take
2000 generations. Parallelism and Darwinian selection can improve the efficiency with which the
design space is searched but they cannot produce a suite of mutations in the relevant gene all at
once. So please, be patient... e-den will only take off when it runs in its distributed mode and,
even then, will only achieve its true potential when gigaherz chips arrive.
Do I have to learn the genetic syntax to create new organisms?
Learning the e-den syntax is a little like learning a new programming language...
Furthermore, designing an e-den organism requires programming skills, to get the organism
structure you want and then, on top of that, an intuition for the fuzzy complexities of
interconnected neurons... It's not for everyone. But you can still get value out of e-den
by joining the master grid and providing an extended digital world for other people's
creations, or by trying selective breeding. A set of default organisms with varying behaviours
will soon be provided and these may be used as a starting point. Cloning an organism is
as simple as right-clicking it, to copy its genome, then left clicking wherever you
want a new organism. Don't forget, carbon-based biology was studied long before
Watson and Crick discovered the double helix. You could study e-den biology in complete
ignorance of e-den genetics.
Do I have to give up my e-mail address to join the master grid?
The author uses a hotmail address for mail but his service provider's mail account for e-den.
This is probably the best solution. Otherwise, a manual organism exchange is recommended
before you read your e-mail. If you forget to do this, you'll get some biological spam and kill
a few organisms but that's no big problem. An iconic email-flusher will ultimately be provided
with e-den to make this easier. e-den reads your non-eden e-mail but leaves it with your POP
server so you don't have to worry about losing your regular e-mail.
How do I join the e-den master grid?
The e-den master grid is not up and running yet because not enough people have the software.
In fact, it is probably at least a few months away but the author would be delighted if you
registered your interest now. If just a tiny fraction of the tamagotchi enthusiasts ran e-den on
their PC, e-den would have a very good chance of achieving worthwhile results.
How do I know if I have the latest version?
Check the e-den download page. Your version number is shown if you click the "?" button. The latest version is 0.54.
|