User Tag List

Page 2 of 2 FirstFirst 12
Results 11 to 20 of 20
  1. #11
    Moderator ][X][~FLuKE~][X]['s Avatar
    Join Date
    Feb 2012
    Posts
    1,367
    Country:
    its basically like the platform is removed/destroyed but the collision is still there.

  2. #12
    Rampage Feralidragon's Avatar
    Join Date
    Jan 2011
    Posts
    374
    Country:
    Just checked the code, the problem is obvious.
    The platform spawns an hidden collision (because as platform is a pawn, thus it cannot serve as platform as you would keep jumping on it instead of staying), then it uses a simulated non-native event to destroy it (-> unreliable way to destroy actors on client).

    Plus, given that the ICH is bNetTemporary=False, it cannot be destroyed in the client, and when you forcefully try (like it is happening now), once the server destroys it, odd things may happen client-side.

    Use the native destroyed event to destroy the ICH together with the platform, take out the reference from the ICH to the platform (only the reference platform->ICH is needed) and do it server-side only, and problem solved.

  3. #13
    Whicked Sick |uK|Grimreaper's Avatar
    Join Date
    Feb 2011
    Posts
    1,315
    Country:
    whatever he said ^


  4. #14
    Administrator SAM's Avatar
    Join Date
    Jan 2011
    Posts
    8,296
    Country:
    Grim can you stop posting useless spammy posts in constructive discussions like this please?

  5. #15
    Whicked Sick |uK|Grimreaper's Avatar
    Join Date
    Feb 2011
    Posts
    1,315
    Country:
    huh what i was one of the few people that asked about this bug


  6. #16
    Whicked Sick Higor's Avatar
    Join Date
    Apr 2012
    Location
    Full sail ahead!
    Posts
    3,676
    Country:
    Yeah, the platform should be a normal, replicated actor.
    That means extra bandwidth but maybe there's another way.

    Feralidragon, there's one thing I've been wondering for a long time and never decided to try it out:
    On clientside replicated actors, do their PostBeginPlay(), PostNetBeginPlay(), and Destroyed() get called every time they enter relevancy or does that happen once?

    If it happens everytime, we could make the platform collision appear in a totally clientside fashion and serve the same purpose the non-replicated collision serverside, then destroy it just like the graphic effects.
    As a matter of fact, if the events don't happen, we can just make an extra graphic effect that works as collision so we don't have to rewrite a lot of code.
    DELETED TEXT: Stupid statement.

    PD: Was the forum domain name hijacked for the last days or two? The site was showing something completely different for me a few hours ago.


    EDIT:
    I noticed that due to this extra collision, the platform can't take non-explosive damage from above.
    I suggest using a workaround so that the platform itself doesn't receive damage, while the collision (which should be bigger both above and below than the actual platform) forwards TakeDamage calls to the platform.
    Last edited by Higor; 05-18-2012 at 02:02 PM.

  7. #17
    Administrator SAM's Avatar
    Join Date
    Jan 2011
    Posts
    8,296
    Country:
    Quote Originally Posted by Higor View Post
    PD: Was the forum domain name hijacked for the last days or two? The site was showing something completely different for me a few hours ago.
    No it wasn't. I pointed the NameServers to a wrong IP /amateur mistake.

  8. #18
    Rampage Feralidragon's Avatar
    Join Date
    Jan 2011
    Posts
    374
    Country:
    Quote Originally Posted by Higor View Post
    Feralidragon, there's one thing I've been wondering for a long time and never decided to try it out:
    On clientside replicated actors, do their PostBeginPlay(), PostNetBeginPlay(), and Destroyed() get called every time they enter relevancy or does that happen once?
    It only happens once.

    Quote Originally Posted by Higor View Post
    If it happens everytime, we could make the platform collision appear in a totally clientside fashion and serve the same purpose the non-replicated collision serverside, then destroy it just like the graphic effects.
    As a matter of fact, if the events don't happen, we can just make an extra graphic effect that works as collision so we don't have to rewrite a lot of code.
    Yes, as long the building runs simulated functions/events, it could just make a Level.NetMode != NM_DedicatedServer and spawn the effects.
    Personally I did that a lot, the only reason to not do it is when the spawn context of an effect happens server-side (well, but even for that we can define replicated functions, but most of the times just spawning the effect in the server isn't problematic as long as the actual processing and variables doesn't get replicated as well, which can occupy a lot more bandwidth over time than the spawn alone).

    Quote Originally Posted by Higor View Post
    I noticed that due to this extra collision, the platform can't take non-explosive damage from above.
    I suggest using a workaround so that the platform itself doesn't receive damage, while the collision (which should be bigger both above and below than the actual platform) forwards TakeDamage calls to the platform.
    Yes, that would be the best approach (actually that's pretty much the "always-correct" kind of approach for anything using an external collision hull).

  9. #19
    Moderator .seVered.]['s Avatar
    Join Date
    Jun 2011
    Location
    Near a River and Under a Bridge
    Posts
    2,125
    Country:
    Couldn't we make the regular supplier supply everyone standing in it?

  10. #20
    The Best There Ever Will Be! |uK|B|aZe//.'s Avatar
    Join Date
    Jan 2011
    Location
    London, United Kingdom
    Posts
    6,860
    what would be the point of having a super sup then

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Code review: Supplier ammo-giving optimization.
    By Higor in forum #siegepug Discussion
    Replies: 19
    Last Post: 06-07-2012, 07:34 AM
  2. Code review: sgHUD
    By Higor in forum Code Reviews
    Replies: 6
    Last Post: 06-01-2012, 02:45 PM
  3. Replies: 4
    Last Post: 05-12-2012, 11:41 PM
  4. Code review: Supplier (50%) and WeaponFiding optimization.
    By Higor in forum #siegepug Discussion
    Replies: 4
    Last Post: 05-12-2012, 11:41 PM
  5. Supplier Ammo Increase
    By SilverWing in forum zp| * * * -=[SIEGE]=- |uK| One Night Stand -=[SIEGE]=- Server * * *
    Replies: 9
    Last Post: 02-27-2012, 10:11 AM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •