Andy we have talked about this some years ago. But I still keep bringing it up because I think it would take G-Force to a whole new level.
Think about a central server that is running a Genetic Algorithm with the best configs. When a user is running G-Force, a number of Configurations are being downloaded. Then G-Force is going through those configs. When the user likes a config, he can press the + key, when a user doesn't like a config, he can press the - key (or multiple times). Then the ratings of the configurations are being transmited back to the server. The server is then using the combined ratings of all users to provide each configuration with a fitness. Then the best configurations survive and do cross over (with a slight mutation) to produce new configurations to be rated.
I think this would take this to a whole new level, pretty much weeding out bad looking stuff while improving the good looking stuff. This might even be platinum service product...
Years ago your answer to that was that tastes are different, that's true, but when there are different tastes, the Genetic Algorithm will improve each taste (local maximum) - that's what GA's are good for.
At the end what we have is one big evolving thing, constantly producing amazing grafics. Maybe even music speed could be included, which would result in different great looking configs depending on music speed.
How about a Genetic Algorithm for Config Selections (IDEA)
Moderators: BTT, andy55, b.dwall, juxtiphi
I do though see problems with this.
People out there don't understand what each config is. I mean are you talking delta-maps, colour maps, etc.
As G-Force constantly changes it is hard from one instance to say that is what a person likes. Do you store all the different config information? What about say the transistion of say the waveshapes?
I fear it could cause more problems than it solves.
People out there don't understand what each config is. I mean are you talking delta-maps, colour maps, etc.
As G-Force constantly changes it is hard from one instance to say that is what a person likes. Do you store all the different config information? What about say the transistion of say the waveshapes?
I fear it could cause more problems than it solves.
With a config(uration) I mean a set delta map, a set color map, a set particle etc... just everything needed to produce one display.
G-Force is constantly changing, meaning it changes to other color maps, other delta maps, other particles etc, but at any given point (other then when there is a transition) there is one color map, one delta map etc. present.... The idea would be when you see something you like all you do is press a plus or minus, or a number, or something where you tell the program when you like or dislike something. That's all the user does. So for the user other then providing a rating and having an internet connection, nothing is changing they don't need to know what a config is etc... So for the user it's really simple.
Regarding transitions, I would not rate them (or at least not for the beginning). If it is decided to rate transitions, then 2 configs need to be modelled in the Genetic Algorithm - which would leverage this idea even more.
The complicated part is defenitely not on the user side. The complicated part is that someone needs to set up a server, which costs bandwith, and someone needs to program the GA(s) - but hey maybe a platinum subscription could help to pay for this - which at the same time would provide a constant income to Andy - without him being forced to constantly improve G-Force and add feature to keep income coming.
A Genetic Algorithm would MAXIMIZE G-Force visualisations the same way evolution has maximized life on this planet (it's the same principle).
Think about the most awsome stuff you have ever seen on G-Force and imagine G-Force constantly producing this stuff - and then imagine this to constantly improve....
G-Force is constantly changing, meaning it changes to other color maps, other delta maps, other particles etc, but at any given point (other then when there is a transition) there is one color map, one delta map etc. present.... The idea would be when you see something you like all you do is press a plus or minus, or a number, or something where you tell the program when you like or dislike something. That's all the user does. So for the user other then providing a rating and having an internet connection, nothing is changing they don't need to know what a config is etc... So for the user it's really simple.
Regarding transitions, I would not rate them (or at least not for the beginning). If it is decided to rate transitions, then 2 configs need to be modelled in the Genetic Algorithm - which would leverage this idea even more.
The complicated part is defenitely not on the user side. The complicated part is that someone needs to set up a server, which costs bandwith, and someone needs to program the GA(s) - but hey maybe a platinum subscription could help to pay for this - which at the same time would provide a constant income to Andy - without him being forced to constantly improve G-Force and add feature to keep income coming.
A Genetic Algorithm would MAXIMIZE G-Force visualisations the same way evolution has maximized life on this planet (it's the same principle).
Think about the most awsome stuff you have ever seen on G-Force and imagine G-Force constantly producing this stuff - and then imagine this to constantly improve....