Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
1 online now:
Newest Member: popoi
Post Volume: Total: 915,819 Year: 3,076/9,624 Month: 921/1,588 Week: 104/223 Day: 2/13 Hour: 0/1


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Cosmic evolution after the Big Bang: Can simulations 'test' theory ?
EZscience
Member (Idle past 5154 days)
Posts: 961
From: A wheatfield in Kansas
Joined: 04-14-2005


Message 1 of 7 (213539)
06-02-2005 1:08 PM


The following quotes are taken from a news piece by the BBC available in its entirety here.
"Astronomers have used supercomputers to re-create how the Universe evolved into the shape it is today."
"We have learned more about the Universe in the last 10 or 20 years than in the whole of human civilisation," said Professor Carlos Frenk, Ogden professor of fundamental physics at the University of Durham and co-author on the Nature report.
"We are now able, using the biggest, fastest supercomputers in the world, to recreate the whole of cosmic history," he told the BBC.
"English Astronomer Royal, Sir Martin Rees told the BBC: "Now we have the Millennium Run simulations, we have the predictions of the theory in enough detail that we can see if there is a meshing together of how the world looks on the larger scale and the way we expect it should look according to our theories. It's a way to check our theories."
So the topic I propose for discussion (Big Bang and Cosmology Forum) is as follows:
Can mathematical simulations of evolutionary theories (cosmic OR biological) serve as tests of these theories in any way?
That is, can we use the results of such simulations to justifiably increase our confidence in the underlying assumptions of the theory, and thus our confidence in the validity of the theory itself?
This message has been edited by EZscience, 06-02-2005 12:09 PM

Replies to this message:
 Message 3 by coffee_addict, posted 06-02-2005 1:44 PM EZscience has replied

  
AdminJar
Inactive Member


Message 2 of 7 (213543)
06-02-2005 1:16 PM


Thread moved here from the Proposed New Topics forum.

  
coffee_addict
Member (Idle past 477 days)
Posts: 3645
From: Indianapolis, IN
Joined: 03-29-2004


Message 3 of 7 (213560)
06-02-2005 1:44 PM
Reply to: Message 1 by EZscience
06-02-2005 1:08 PM


EZ writes:
Can mathematical simulations of evolutionary theories (cosmic OR biological) serve as tests of these theories in any way?
The problems with using such a test is that (1) we have no way of knowing with a reasonable certainty that our current mathematical model reflect the true nature of the universe and (2) we may never be able to confirm any finding on such a grand scale.
That is, can we use the results of such simulations to justifiably increase our confidence in the underlying assumptions of the theory, and thus our confidence in the validity of the theory itself?
If we are talking about confidence and nothing else, then absolutely. The model itself is built on many mathematical principles that were developed over millenia, some of which were developed independently of some others. We can gain more confidence in the theory because technically the theory is confirmed by many different sources.

This message is a reply to:
 Message 1 by EZscience, posted 06-02-2005 1:08 PM EZscience has replied

Replies to this message:
 Message 4 by Orlando Dibisikitt, posted 06-02-2005 1:51 PM coffee_addict has not replied
 Message 5 by EZscience, posted 06-02-2005 2:47 PM coffee_addict has not replied

  
Orlando Dibisikitt
Inactive Member


Message 4 of 7 (213564)
06-02-2005 1:51 PM
Reply to: Message 3 by coffee_addict
06-02-2005 1:44 PM


I guess that with the correct algorythm and adequate hardware, anything can be simulated.

This message is a reply to:
 Message 3 by coffee_addict, posted 06-02-2005 1:44 PM coffee_addict has not replied

Replies to this message:
 Message 6 by EZscience, posted 06-02-2005 2:52 PM Orlando Dibisikitt has replied

  
EZscience
Member (Idle past 5154 days)
Posts: 961
From: A wheatfield in Kansas
Joined: 04-14-2005


Message 5 of 7 (213575)
06-02-2005 2:47 PM
Reply to: Message 3 by coffee_addict
06-02-2005 1:44 PM


When models converge...
GAW writes:
The problems with using such a test is that (1) we have no way of knowing with a reasonable certainty that our current mathematical model reflect the true nature of the universe and (2) we may never be able to confirm any finding on such a grand scale.
Yes. Both quite true. In the case of (1), it is possible to produce a simulation that generates the expected results (consistent with theory) for the wrong reasons. For example, two erroneus modelling assumptions might sometimes cancel one another out. One could conceive of many other scenarios that might lead the modeller to assume that the model was correct when it wasn't.
GAW writes:
If we are talking about confidence and nothing else, then absolutely. The model itself is built on many mathematical principles that were developed over millenia, some of which were developed independently of some others.
I agree wholeheartedly. But all we can ever really strive for in science (despite the protestations of theists in search of 'absolute truth') is an increased level of confidence in our theory.
GAW writes:
We can gain more confidence in the theory because technically the theory is confirmed by many different sources.
Truly the crux of the matter. Modelling is a heuristic exercise. Models that produce credible inconsistencies with theory can cause the theory to be re-examined and 'tweaked'. Models that replicate theory perfectly should still be doubted, examined, and their assumptions tested. This can be done by direct experimentation (not always feasible, as you point out) but also by the construction of other models with different sets of assumptions. As these converge in agreement with a particular theortical framework, our confidence in the veracity of that framework can be GREATLY increased.
But I have little doubt we will see some 'scientific dissenters' show up on this thread.

This message is a reply to:
 Message 3 by coffee_addict, posted 06-02-2005 1:44 PM coffee_addict has not replied

  
EZscience
Member (Idle past 5154 days)
Posts: 961
From: A wheatfield in Kansas
Joined: 04-14-2005


Message 6 of 7 (213576)
06-02-2005 2:52 PM
Reply to: Message 4 by Orlando Dibisikitt
06-02-2005 1:51 PM


SImulate anything?
I guess in theory - yes.
But you would still have to specify all relevant components of the problem and then make a long list of all the implicit assumptions before building the algorithm. That is still going to be a daunting task for more complex simulations such as this one. You have got to respect the patience and skill these guys had to have.

This message is a reply to:
 Message 4 by Orlando Dibisikitt, posted 06-02-2005 1:51 PM Orlando Dibisikitt has replied

Replies to this message:
 Message 7 by Orlando Dibisikitt, posted 06-02-2005 3:50 PM EZscience has not replied

  
Orlando Dibisikitt
Inactive Member


Message 7 of 7 (213583)
06-02-2005 3:50 PM
Reply to: Message 6 by EZscience
06-02-2005 2:52 PM


Re: SImulate anything?
oh yes.. total respect, didn't mean to belittle it. I also know that getting the right algorithm is the hard bit but it would be good wouldn't it

This message is a reply to:
 Message 6 by EZscience, posted 06-02-2005 2:52 PM EZscience has not replied

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024