Managing Latency

Discussion in 'Multiplayer Meeting Room' started by wrenchinator, Mar 9, 2016.

  1. wrenchinator

    wrenchinator Commander

    Joined:
    Jan 6, 2016
    Messages:
    79
    Likes Received:
    86
    Reading through the Red vs Blue thread piddlefoot had mentioned the prealpha conditions and lack of smoothing code. I noticed a real problem in Slams footage with ships bouncing around a lot. I wondered where are they going with this? I can imagine that it would be hard to predict the projected path of our ships. So when they do smooth things out are they managing the latency and predicting positions? We have a ship that goes from 0-111.3 m/s in 1.45 seconds, it stops on a dime and is just as powerful sideways, it would be hard to determine where it will be in .5 seconds if we started to input any other variables.
     
    #1
  2. =TCR= Big_Red

    =TCR= Big_Red Rear Admiral

    Joined:
    Nov 16, 2015
    Messages:
    684
    Likes Received:
    634
    Game can cheat. It knows your X-Y position all the time witch it can pass over to the enemy. Why do you think the aliens shoot at you through the walls and hill sides. Lol
     
    #2
  3. wrenchinator

    wrenchinator Commander

    Joined:
    Jan 6, 2016
    Messages:
    79
    Likes Received:
    86
    I was more on about MP and such, in order to present the right stuff on each front end smoothly there has to be some predicting of position and some account of varying pings. I was just wondering how that was being approached.
     
    #3
  4. Brokenshakles

    Brokenshakles Rear Admiral

    Joined:
    Feb 14, 2015
    Messages:
    478
    Likes Received:
    301
    This is all a rather advanced network programming topic. So to condense it a bit, yes, there is currently network prediction working in the code right now, and no, it's not very good yet. Especially in cases of high latency, like our Ontario server. ESPECIALLY in the case of high latency + high acceleration of player viewpoint. You may not realize this, but 35m/s is about one tenth of the speed of sound yet we can go from 0 to that in .75 seconds, and we are supposed to believe that pilots can survive this obviously 15g+ acceleration? The acceleration of just about everything in the game is way, way to high, and the top speeds are too low. This also has the negative effect of effectively shrinking the game space as it is actually experienced. It also seriously messes with network prediction and tracking.
     
    #4
    Last edited: Mar 9, 2016
  5. wrenchinator

    wrenchinator Commander

    Joined:
    Jan 6, 2016
    Messages:
    79
    Likes Received:
    86
    Exactly, any idea how they are approaching it? I also wondered about the G forces. All of what you mention could help to reduce the predictive errors I was alluding too. I used to play a game that was very good at managing these issues but it was easier to predict the future positions of the aircraft involved, they had less ability to change direction quickly.
     
    #5
  6. Brokenshakles

    Brokenshakles Rear Admiral

    Joined:
    Feb 14, 2015
    Messages:
    478
    Likes Received:
    301
    No clue, I am not privy to Eleons internal codebase, the farthest access I have is a submission account on the bugtracker, which only shows bugtesting discussions. I only know the general theory behind it and what I can figure out from the behavior I have seen in my extensive number of testing hours.
     
    #6
  7. piddlefoot

    piddlefoot Rear Admiral

    Joined:
    Mar 4, 2015
    Messages:
    1,841
    Likes Received:
    1,612
    Yea I agree, the game needs some sort of limit for acceleration so that its not only believable to look at ingame, but also to assist the game engine to overcome some of the latency issues.

    And we are starting to see uber nerfed ships being blueprinted in that defy logic, gravity , and every Newtonion law out there !
     
    #7

Share This Page