|
General Forum |
 |
|

Ettina
 
|
12/24/2011 | |
You know when it says '# of genetics errors corrected automatically'? It's actually changing the genome without your permission! For example it got rid of a mutation that causes a pigment gene to code for 'unknown pigment 19' instead of blue. (This could potentially alter a norn's coloration, since mutations over time can cause a high gen's coloration to be coded for by very few genes.)
By the way, I don't care if anyone steals my ideas for their own work, as long as you don't try to stop me from making my own stuff. Many ideas I mention are things I don't have the time or skill to actually do. |
 Lodestar
Doringo
   

|
12/24/2011 | |
I mean technically that actually was an error that didn't really do anything ingame? I haven't really ever had this become an issue when gengineering, which is probably what this thing is more designed for. |
 Peppery One
Papriko
    
|
12/24/2011 | |
One of the reasons why I use Genetic Lab and not Genetic Kit.
Lets play plants! Photosynthesis... Photosynthesis... Photosynthesis... |
 Lollipop Lord
C-Rex
    

|
12/24/2011 | |
I never knew that :0. Maybe that's the reason I had trouble making the genomes for my Salamander Norns... |

N1


|
1/4/2012 | |
Well to be fair a gene for pigment 19 is clearly invalid, and the Creatures engine most probably just ignores it.
The Genetics Kit is only able to display a somewhat valid subset of all possible genomes (the same is true for LiveGMS by the way), and it cannot technically display a pigment gene for a color other than red, green and blue. I personally prefer it to make the genome consistent with what it shows you at least, but I agree that it's not a perfect solution.
It gets even more complicated: For a pigment gene it might make more sense to just delete the gene if it's for an invalid pigment. Because as you say, resetting it to some default pigment will alter how the creature looks like, and that might be undesired. But what should it do with a Brain gene that has an invalid SV rule? It cannot just delete it because that would do even more harm. So I guess setting that rule to something like STOP is a reasonable thing to do. As long as that interpretation of an invalid rule is consistent with what the Creatures engine does when encountering such a gene that is.
Semantically invalid (or let's say undefined) genomes seem to be a tricky thing to handle properly...
Edit: When I try to load a genome with an invalid pigment gene into Genetics Lab, it crashes by the way:
"Unhandled error": "Invalid property value in module GENELAB: LoadGenome()". Seems to have a minor glitch there.
But it seems to be able to preserve all kinds of invalid SV rules for example, and will show them as "Unknown #something". Makes sense I guess.
(genome file I used for testing: http://danielmewes.dnsalias.net/~daniel/invalid%20pigment.gen) |
 Peppery One
Papriko
    
|
1/4/2012 | |
Invalid genes can be a pain. Once I made the mistake and tried a mutating genus gene. They started as norns, became grendels, morphed into geats and finally did my game crash telling me that it is unable to create a creature of type 7.
Lets play plants! Photosynthesis... Photosynthesis... Photosynthesis... |
 Patient Pirate
ylukyun
     Manager
|
1/4/2012 | |
IMO it would be much better if they were validated. Even the brain can't mutate, because if you set it to mutable it goes to engine-crashing values. This doesn't seem to have been as much of a problem in the other games, so I wonder why they decided to change it.
Now if the engine could decide to do something with pigment nineteen, that would be a different story. Or if norns had 100 pigment genes, a few mutating might not matter so much. But they just end up uncolored whenever the mutation rate is too high. |