On this page 

A particle viewer pane is a Viewer pane with its Network menu set to Particles (POP). The viewer shows you the particle simulation you are editing.
Note
The particle viewer simulates how a POP SOP would convert the particle system you are editing into geometry. The display options available in the particle viewer closely resemble the parameters of the POP SOP.
POP Viewer State Parameters  Standard Page
Note
All of these parameters are actually the parameters of the POPNET (inside of which the POPs live, just like SOPs live inside Objects) being viewed. Thus these parameters can also be changed by going to the parameter dialog of the POPNET being viewed.
Start Time /timestart
The start time is the time at which the pop simulation starts cooking. Before this, the particle system will be empty. The default is Tstart(frame one).
Preroll Time /timepreroll
How many seconds of the pop simulation to bypass, after the reset time is reached. For example, if you put the number 33 into this field (and reset is at Tstart), frame one will show the simulation that was at a time of 33 seconds. In other words, the first thirtytwo seconds have been bypassed, and the time at thirtythree seconds is shifted to frame one. The first thirtytwo seconds must still be calculated in order to compute the status of the points, so you will notice some delay upon reset.
Initial State
Instead of starting from an empty particle system, you can use a geometry file to specify the initial state for the simulation. Create a geometry file by using a Geometry Output OP, and specifying a POP sop as its output.
Oversampling
Oversampling is how many times in between frames to cook the simulation. For example, a value of 1 means to cook once per frame. A value of 2 means to cook at frame 1, frame 1.5, frame 2, etc.
Access Oversampling in the POPnet > Parameters, or in the POP sop.
Note
The Sampling Rate in the POP Editor is not associated in any way with the oversampling in the SOP. The SOP’s oversampling is the only one used when rendering, or for display of the SOP’s data. The POP Viewport oversampling is only for the POP editor (there may be multiple SOPs referencing the same POP network with different oversampling rates).
Max # of Particles
Specifies the maximum number of particles allowed to exist in the system.
Remove Unused Points
Removes all unused points from the input geometry. When points are removed, memory is conserved. On the other hand, reusing points saves the time needed to purge the points from memory.
POPs Viewer State Parameters  RBD Page
The RBD solver requires a number of new parameters to specify how to cook the network. These parameters can be found on the Rigid Body Dynamics page of the Particle CHOP, the POP SOP, and the POP Viewer parameter dialog. The following sections describe each of the new parameters.
Rest Threshold
This parameter is used to specify the relative velocity between objects that is considered as resting contact by the RBD solver. Resting contact is treated differently from collision contacts, which is why this division is necessary. Usually the default value should be used here unless your objects are very large.
Contact Tolerance
This is the distance between two objects at which the RBD solver considers them to be touching. If this value is made large, objects will collide and bounce in obviously unrealistic ways (i.e. before they actually contact). If this value is made small, you should increase the Max Time Splits value which will allow more accurate simulation of collisions but will also slow the simulation.
Max Time Splits
This parameter specifies how finely the RBD solver is allowed to split up a time step. The solver splits time steps to find the precise moment at which two object collide. The larger the number of time splits you allow the more accurately collision can be simulated, but also the longer the simulation may take. This parameter should take into account the maximum velocity at which particles will be moving in the simulation, the Contact Tolerance value, and the size of each time step (which is one frame time divided by the Oversampling parameter on the Standard page).
Constraint Time
This parameter specifies the approximate time (in seconds) that the RBD solver can take to satisfy an unsatisfied constraint (see the Constraint POP for information about setting up constraints). The smaller this value is, the more rigidly constraints behave, but the larger the forces are that get applied to the objects. The larger this value is, the less likely you are to produce an unstable situation in your simulation.
Solver Type
Choose from: Euler (fastest, but least accurate), Midpoint, and RungeKutta (slowest, but most accurate).
POP Viewer State Parameters  Input Geo Page
Object n / SOP n
Specify the Input geometry objects here by specifying their Object and SOP from the popup menus provided. These let you set the context geometry to use for the POP network.
Context Geometry
All POP networks can have 'Context Geometry'. These context geometries are set in the parameters of the POPNET, the Particle CHOP, or the POP OP. In the POP OP they can also be set by wiring OPs into the inputs of the POP OP. These context geometries can be referenced by the following POPs: Source POP, Collision POP, SoftBody POP, Attractor POP, Creep POP (All of which have the parameter  Geometry Source > Use Context Geometry). So, instead of hardcoding a specific piece of source geometry or collision geometry, a single POP network can be used by several POP OPs to generate several different outputs, depending on the geometry input into the POP OP.
Context Geometry  Example
The main point of 'Context Geometry' is that a POP network can be used as a macro instead of a dedicated effect. For example, suppose you want to have a piece of geometry emit particles from it’s points, and then have those particles fly over to the points of some other piece of geometry. You create a Source POP that references an object, an Attractor POP that references some other object, then you put a POP SOP into an object somewhere and you're done. You don’t even provide any inputs into the POP SOP. Now suppose you have another pair of objects where you want to have this same effect. Without Context Geometry you would have to create a whole new POP network that was identical to the first, except it referenced different objects in the source and attractor POPs. Suppose you have 50 of these pairs of objects where you want to use this effect  that’s 50 POP networks.
Enter Context Geometry. Change the Source POP to use Context Geometry 1, and the Attractor POP to use Context Geometry 2. In your POP SOP, connect the source geometry to input 1 and the attractor geometry to input 2. But now you want to do this same thing with 50 pairs of objects. You still need 50 POP SOPs. Each one takes as input the pair of geometries you want to work with. But you only need one POP network. This is a tremendous advantage if you ever want to make a change to the POP Network.
In the POP Viewport, you go to the 'Operation Parameters' dialog ( P), and to the Input Geo page. On this page, you specify the Object and SOP to use for each of your four Context Geometries. So, for this example, when you're designing your POP network, you would set the 'Object 1' and 'SOP 1' to point to the source geometry. You would set the 'Object 2' and 'SOP 2' to point to the destination Attractor geometry. Now in the POP Viewport, the Source POP will behave exactly as if you had entered these Object and SOP values into its parameters. Even templating geometry works.
POP Viewer State Parameters  Viewer Page
Transform Object
When you create a pop network, you must specify a geometry source from where the particles are birthed. The pop network will automatically assume the coordinate space of the geometry in that object, just as if the source geometry contained an Object merge sop. You select which object  whose space you would like to use  from this menu (this is the same menu in the POP Viewport > Select state > subicons Transform Object).
Cache Size
Specifies the size of the cache  which will hold the result of the pop simulation for faster subsequent playback. A zero (0) turns off caching, 1 sets infinite caching, and any other value sets the maximum number of frames to cache.
Note that to get an uptodate and accurate particle simulation, you should reset the simulation with the yellow Recook simulation from Start button at the top of the POP Viewport.