From: nicolaas.vroom@pandora.be   
      
   On Saturday, 20 October 2018 17:02:35 UTC+2, Phillip Helbig wrote:   
   > In article <37f31736-60df-46b0-bead-39a9c54dc0e6@googlegroups.com>,   
   > Nicolaas Vroom  writes:   
   >   
   > > What I want to know is why my simulations of clock in a linear   
   > > accelerator or centrifuge are wrong.   
   >   
   > This is the question which should be addressed.   
   >   
   > In general, while a simulation might visualize something which is   
   > otherwise difficult to grasp, in terms of physics one gets out only what   
   > one puts in.   
      
   That is 100% correct.   
      
   On Wednesday, 24 October 2018 01:20:18 UTC+2, Tom Roberts  wrote:   
   > On 10/20/18 5:38 AM, Nicolaas Vroom wrote:   
   > > The clocks used in the book SpaceTime Physics are not pointlike   
   > > and the reason that tick is also because they are not pointlike.   
   > > The reason that they behave differently is because the mirrors   
   > > can be parallel or pendicular to the direction of motion.   
   >   
   > I'll bet that in the book they are all moving inertially. Because if a   
   > light clock with parallel mirrors is accelerated along a direction not   
   > parallel to its light path, it will cease to operate, because the   
   > mirrors will be accelerated away from the bouncing light ray.   
      
   That is correct.   
   My simulation shows the same behaviour.   
   However my simulations shows more:   
   The whole object of the simulation is to simulate the behaviour of a clock   
   and to test if it is in agreement with the equation:   
       t' = \integral sqrt(1-v^2/c^2) dt   
   The simulation shows that that is true when the light signal is perpendicular   
   to the direction of movement and the mirrors are parallel.   
   The simulation also shows that equation is wrong in the opposite case i.e.   
   lightsignal parallel and and mirror perpendicular.   
      
   Does that mean that this simulation is wrong?   
      
   [[Mod. note -- The key question is whether the *limit* of the   
   simulated clock time approaches   
       t' = \integral sqrt(1-v^2/c^2) dt   
   as the clock gets smaller and smaller.  If the simulation doesn't   
   show that, then it's wrong.   
   -- jt]]   
      
   > 	(That does not happen for any inertial motion.)   
   >   
   > > My simulation assumes the same.   
   >   
   > Then it cannot accurately model acceleration of a light clock (except   
   > along the direction of its internal bouncing light ray).   
      
   Why not?   
   When the clock moves horizontal its bouncing light ray can be vertical.   
   In that case the mirrors are also horizontal.   
   This case is discussed in the book SpaceTIme Physics.   
   The bouncing light ray can also be horizontal.   
   In that case the mirrors are vertical.   
   Both cases can be used to simulate acceleration in horizontal   
   direction.   
      
   The importance of the simulations are that when you use such a clock   
   in a real experiment you have to take special precaution to use   
   the results.   
      
   > > How can I understand a thought experiment when in reality a clock   
   > > has 3 dimensions and in the thought experiment not. (only 1)   
   >   
   > You need to learn about approximations, and how to determine when they   
   > are appropriate (or inappropriate) to a given physical situation.   
      
   I do think the issue is much more about accuracy.   
   As such the question is if the following sentence is 'correct':   
   "General speaking assuming different physical situations the   
   speed of light can vary and is not constant. However a good   
   approximation for many circumstances is to treat it as a constant"   
      
   > 	BTW that is an essential aspect of an education in physics.   
   > 	Except for certain gedankens, physicists are always making   
   > 	approximations....   
   >   
   > This seems to me to be the root cause of your confusions.   
      
   If physicists make approximations they should clearly explain   
   that with and without the approximations the results are the same,   
   because for example how longer the experiment takes and how larger   
   the distances involved approximations can influence the outcome.   
      
   In the book SpaceTime Physics at page 38 we can read something about   
   approximations (The subject is clock synchronization):   
   "Moreover, the error can be made as small as desired by carrying the   
   portable clock around sufficiently slowly"   
   This is not true for a clock which light ray bounces in the direction   
   of movement.   
   When you move a clock from A to B and back to A the extra time for the   
   light signal to travel the same distance is not taken into account   
   (is not counted).   
   This does not depend if you move the clock fast or slow.   
      
   > > What I want to know is why my simulations of clock in a linear   
   > > accelerator or centrifuge are wrong.   
   >   
   > That depends on myriad details....   
      
   I like to know these details.   
      
   > > A more basic question is: why do we consider for example the surface   
   > > of the earth at rest.   
   >   
   > This depends on whether such an APPROXIMATION is appropriate to the   
   > problem at hand. For playing sports it is both appropriate and very   
   > accurate. For mapping stars and planets it is inappropriate and   
   > hopelessly inaccurate.   
      
   Here we both agree. This is an accuracy issue.   
   The behaviour of clocks is also an accuracy problem.   
   That is why IMO you should treat a clock in a simulation   
   first and for all in 3D.   
      
   > Tom Roberts   
      
   Nicolaas Vroom   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)