Jump to content

  • Log In with Google      Sign In   
  • Create Account


view frustrum culling


Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.

  • You cannot reply to this topic
4 replies to this topic

#1 Anddos   Members   -  Reputation: 471

Like
0Likes
Like

Posted 07 September 2013 - 03:57 PM

Would using view frustrum culling benifit players fps performance in a multiplayer game?

Has anyone tried it with say 100 players and only render the players that are within a set radius from you ? but for all the other players you would still need to know about there x,y,z positions right so the server always sends there positions to you that make the client decide if they should be rendered or not.

 

Could you do this for terrain and geomtry as well or that that be expensive on the server?


Edited by Anddos, 07 September 2013 - 04:02 PM.

:)

Sponsor:

#2 cozzie   Members   -  Reputation: 1408

Like
0Likes
Like

Posted 07 September 2013 - 05:18 PM

Without knowing more, I'd say yes.
Everything (outside the frustum) you can reject early and don't render is potentially increasing performance. Keeping track of players positions etc, also outside the frustum is another process/ on the cpu I assume

#3 Norman Barrows   Crossbones+   -  Reputation: 1835

Like
0Likes
Like

Posted 07 September 2013 - 07:59 PM

i couldn't imagine NOT doing it! i mean you're already massively multi-player, with network lag and overhead slowing you down, so you start out with two strikes against you vs a single player title.  AND YOU DON"T CULL THE SCENE?  obviously, you'll still need all entity data, but physics and graphics should be unrelated, except for the fact that they both use the entity's location and rotation to do their job.


Norm Barrows

Rockland Software Productions

"Building PC games since 1988"

 

rocklandsoftware.net

 


#4 GaldorPunk   Members   -  Reputation: 922

Like
0Likes
Like

Posted 07 September 2013 - 08:30 PM

I'd just add that frustum culling is purely a graphical task, and should be done client side and not on the server. Culling shouldn't have any effect on the amount of data you're sending, since you're presumably going to be sending the positions of each player either way.

 

Also, frustum culling terrain and static objects can be expensive if you've got large and detailed maps, so ideally you should have some kind of quad tree structure to limit the number of checks you need to do, but in the end any kind of culling is going to be much less expensive than attempting to draw all the off-screen assets.



#5 Lodeman   Members   -  Reputation: 581

Like
0Likes
Like

Posted 08 September 2013 - 02:50 AM

For most scenarios a quadtree structure + frustum culling will get you very far. Unless you're working extensively in 3 dimensions, in that case you could consider an octree.

It will definitely be worth implementing based off your description, 100 player models, and I'm assuming an averagely detailed terrain with scenery on it.






Old topic!
Guest, the last post of this topic is over 60 days old and at this point you may not reply in this topic. If you wish to continue this conversation start a new topic.



PARTNERS