Concept of time in Croquet world

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Concept of time in Croquet world

askoh
Administrator
What is the unit of time is a world? The world showing the planets must have time in term of years while the world with fish is in seconds. How is that set? How do I retrieve the absolute time of the world in the appropriate units?

Aik-Siong Koh
Reply | Threaded
Open this post in threaded view
|

Re: Concept of time in Croquet world

Joshua Gargus-2
"Island time" in Croquet is in seconds.  This is independent of the time
of any simulation that might be occurring in a Croquet world.  For
example, you might have a space where there are two floating 3D
simulations, one of a solar system and another of Brownian motion in a
protein molecule.  Each of these simulations has it's own distinct
time-scale, both of which are unrelated to island time.  You can even
have a rewind button on one of the simulations... time moves backward in
the simulation, but island-time keeps marching steadily forward.

Is this clear?

Josh


askoh wrote:
> What is the unit of time is a world? The world showing the planets must have
> time in term of years while the world with fish is in seconds. How is that
> set? How do I retrieve the absolute time of the world in the appropriate
> units?
>
> Aik-Siong Koh
>  

Reply | Threaded
Open this post in threaded view
|

Re: Concept of time in Croquet world

Joshua Gargus-2
One small error (that I realized just as I hit "send")... the time is in
milliseconds, not seconds.

Josh


Josh Gargus wrote:

> "Island time" in Croquet is in seconds.  This is independent of the
> time of any simulation that might be occurring in a Croquet world.  
> For example, you might have a space where there are two floating 3D
> simulations, one of a solar system and another of Brownian motion in a
> protein molecule.  Each of these simulations has it's own distinct
> time-scale, both of which are unrelated to island time.  You can even
> have a rewind button on one of the simulations... time moves backward
> in the simulation, but island-time keeps marching steadily forward.
>
> Is this clear?
>
> Josh
>
>
> askoh wrote:
>> What is the unit of time is a world? The world showing the planets
>> must have
>> time in term of years while the world with fish is in seconds. How is
>> that
>> set? How do I retrieve the absolute time of the world in the appropriate
>> units?
>>
>> Aik-Siong Koh
>>  
>

Reply | Threaded
Open this post in threaded view
|

Re: Concept of time in Croquet world

askoh
Administrator
In reply to this post by Joshua Gargus-2
I think I need to know the following Croquet definitions:
Island, World, Space, Simulations.
What are the relationships between them?

Thanks,
Aik-Siong Koh


Josh Gargus wrote
"Island time" in Croquet is in seconds.  This is independent of the time
of any simulation that might be occurring in a Croquet world.  For
example, you might have a space where there are two floating 3D
simulations, one of a solar system and another of Brownian motion in a
protein molecule.  Each of these simulations has it's own distinct
time-scale, both of which are unrelated to island time.  You can even
have a rewind button on one of the simulations... time moves backward in
the simulation, but island-time keeps marching steadily forward.

Is this clear?

Josh


askoh wrote:
> What is the unit of time is a world? The world showing the planets must have
> time in term of years while the world with fish is in seconds. How is that
> set? How do I retrieve the absolute time of the world in the appropriate
> units?
>
> Aik-Siong Koh
>  
Reply | Threaded
Open this post in threaded view
|

Re: Concept of time in Croquet world

Joshua Gargus-2
Start with: http://opencroquet.org/index.php/Croquet_SDK

There is no Croquet concept called World.

A Space is basically the root of a 3D scene-graph.  Spaces aren't part
of the core Croquet model (you could use Croquet to create a replicated
2D application), but in practice everyone seems to use 3D spaces
(probably since they're already available).

Simulation is not a Croquet concept.  By simulation, I just mean any
program that you call a simulation if you wrote it to run on a single
computer.  You would probably call the solar-system a simulation, but
you wouldn't call a word-processor a simulation.  Same thing when you
write these as replicated Croquet applications.

To carry this a bit further, a word-processor needs no explicit notion
of time; when you press a key or click the mouse, it responds
immediately.  However, a model of the solar system needs to take time
into account, because it keeps moving according to deterministic laws,
even when there is no user input.  In a non-replicated application, the
rate of time in the simulation is some multiple of time in the real
world.  For example, the simulation of one earth year might take one
minute in the real world.

You can basically consider island time to be the same as real time (due
to changes in network latency etc., island time does not proceed at
exactly the same rate as real-time...).  So, when you re-write a
single-computer simulation to run as a replicated Croquet application,
you would replace any code that uses a system call to get the real time
with a message send to get the island time.  Croquet itself provides no
framework for constructing simulations with different rates of
time-flow; you have to write that code yourself.

The link at the top of the email probably explains this better.

Is that more clear?

Josh


askoh wrote:

> I think I need to know the following Croquet definitions:
> Island, World, Space, Simulations.
> What are the relationships between them?
>
> Thanks,
> Aik-Siong Koh
>
>
>
> Josh Gargus wrote:
>  
>> "Island time" in Croquet is in seconds.  This is independent of the time
>> of any simulation that might be occurring in a Croquet world.  For
>> example, you might have a space where there are two floating 3D
>> simulations, one of a solar system and another of Brownian motion in a
>> protein molecule.  Each of these simulations has it's own distinct
>> time-scale, both of which are unrelated to island time.  You can even
>> have a rewind button on one of the simulations... time moves backward in
>> the simulation, but island-time keeps marching steadily forward.
>>
>> Is this clear?
>>
>> Josh
>>
>>
>> askoh wrote:
>>    
>>> What is the unit of time is a world? The world showing the planets must
>>> have
>>> time in term of years while the world with fish is in seconds. How is
>>> that
>>> set? How do I retrieve the absolute time of the world in the appropriate
>>> units?
>>>
>>> Aik-Siong Koh
>>>  
>>>      
>>
>>    
>
>