[ENet-discuss] Some ENet issues

Philip Bennefall philip at u7142039.fsdata.se
Wed Jul 14 01:30:52 PDT 2010


Hi Blair,

Oh yes, the unreliable version gets more and more accurate over time but it takes a while. It was to decrease the time that this takes that I thought of decreasing the ping interval, in order to have more reliable messages to correct it with - so to speak.

Kind regards,

Philip Bennefall
  ----- Original Message ----- 
  From: Blair Holloway 
  To: enet-discuss at cubik.org 
  Sent: Wednesday, July 14, 2010 10:08 AM
  Subject: Re: [ENet-discuss] Some ENet issues


  Does the unreliable version of your code converge after a longer period, perhaps? Try running it for a minute or two and compare the results.


  - Blair


  On Wed, Jul 14, 2010 at 4:30 PM, Philip Bennefall <philip at u7142039.fsdata.se> wrote:

    Hi Blair,

    What I'm doing as a test is to set up both a server and a client using ENet, connected through localhost. The server echos back anything it receives, and the client prints out statistics after 10 seconds.

    Kind regards,

    Philip Bennefall
    ----- Original Message ----- 
      From: Blair Holloway 
      To: enet-discuss at cubik.org 
      Sent: Wednesday, July 14, 2010 8:23 AM
      Subject: Re: [ENet-discuss] Some ENet issues


      Generally, no. 500ms should be adequate; pinging more frequently is just going to take up more bandwidth from ping responses (reliable acks). 


      What are you using for comparison as a "valid" ping? The output of your platform's ping command?


      - Blair


      On Wed, Jul 14, 2010 at 3:25 PM, Philip Bennefall <philip at u7142039.fsdata.se> wrote:

        Hi Blair,

        Do you think it'd be a good idea to decrease the ping interval? Maybe to 200 milliseconds?

          Kind regards,

          Philip Bennefall
          ----- Original Message ----- 
          From: Blair Holloway 
          To: enet-discuss at cubik.org 
          Sent: Wednesday, July 14, 2010 7:06 AM
          Subject: Re: [ENet-discuss] Some ENet issues


          By default, Enet sends (reliable) ping packets every 500ms, if no other reliable traffic was sent in that interval. If you're sending reliable packets 30 times per second instead of 2 times, it's possible Enet is deriving a more accurate average round trip time. 


          - Blair


          On Wed, Jul 14, 2010 at 2:45 PM, Philip Bennefall <philip at u7142039.fsdata.se> wrote:

            Hi Lee and others,

            I am having some minor issues with ENet.

            First, I'm trying to get the average up and downstream for each peer by using the appropriate data fields in the peer structure but it always returns 0 for some reason. The same seems to be true with the host structure as well.

            Second, when I look at the average round trip time for a peer, this value is only correct if I send out a few reliable packets. on localhost, for instance, I ran a test where I sent 30 unreliable packets every second. I poll the network every 5 milliseconds, but got an average round trip of 44 milliseconds. When I changed it to reliable packets, however, I got an average of 12 which seems much more reasonable. Is this intended behavior?

            Thanks in advance for any help.

            Kind regards,

            Philip Bennefall

            _______________________________________________
            ENet-discuss mailing list
            ENet-discuss at cubik.org
            http://lists.cubik.org/mailman/listinfo/enet-discuss






----------------------------------------------------------------------



          Internal Virus Database is out of date.
          Checked by AVG - www.avg.com 
          Version: 8.5.432 / Virus Database: 271.1.1/2662 - Release Date: 02/01/10 12:37:00



        _______________________________________________
        ENet-discuss mailing list
        ENet-discuss at cubik.org
        http://lists.cubik.org/mailman/listinfo/enet-discuss






--------------------------------------------------------------------------



      Internal Virus Database is out of date.
      Checked by AVG - www.avg.com 
      Version: 8.5.432 / Virus Database: 271.1.1/2662 - Release Date: 02/01/10 12:37:00



    _______________________________________________
    ENet-discuss mailing list
    ENet-discuss at cubik.org
    http://lists.cubik.org/mailman/listinfo/enet-discuss






------------------------------------------------------------------------------



  Internal Virus Database is out of date.
  Checked by AVG - www.avg.com 
  Version: 8.5.432 / Virus Database: 271.1.1/2662 - Release Date: 02/01/10 12:37:00
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cubik.org/pipermail/enet-discuss/attachments/20100714/12b5f04e/attachment-0001.html>


More information about the ENet-discuss mailing list