fromderinside
Mazzie Daius
- Joined
- Oct 6, 2008
- Messages
- 15,945
- Basic Beliefs
- optimist
Let's say the focus is our Sun. It has 1057 atoms. If you approximated it and used Machine Learning to help with the accuracy, you could create a plausible simulation of it with a fraction of the atoms... e.g. 1015 atoms.... plus in a simulation you could instantly change any aspect of the simulated Sun....
OK. Now can you validate whether at -15 simulation will produce usable (real) as would be the -57 real thing? How would you do so? I think all you really get is a nice 'what if' game. Testing SW as I recall is a very labor intensive and time viewing monster. For instance, I understand analysis of experiments using the Hadron Collider requires trillions of 'experiments' because one need at least probabilities in the fraction of billions to one likelihoods to produce a positive result.
Scaling up isn't all it's cracked up to be.
I understand we have a time and space problem being beings that last fewer than 100 years in a world that is billions of light years in extent. Still, are you comfortable with what models suggest? I was extended as a freshman to get good results from a cannon and ball experiment when I estimated to four decimals using an 80 inch equivalent Picket slide rule. Computer back then filled rooms to provide just a few thousand words of processing power.