Have you been affected by "lag" during large group fights?

edited December 2015 in North of Thera

There's been some discussions recently about people experiencing "lag" during fights, usually those involving larger groups but that's not always the case. I've experienced it myself, and receiving from the game seemed to be fine, but my commands were very delayed in going through. The idea of this poll is to try to identify how wide ranging this issue is and to identify if there's any correlation with client and curing support systems (Svo and Wundersys specifically, since they're the two most widely known/used). If you pick either of the "I use another client" options, please identify which client you're using in a post.

Note: "use" means "have installed", even if you have most of the features disabled.

«1

Comments

  • Someone suggested it wasn't a curing system issue, but certain offense systems. If it's related to queueing/serverside aliases (and I have no idea why it would be, but it seems as plausible as anything else), certain offensive systems or coding habits might lead to too much queuing/setaliasing at once, or something.


  • I've had it a lot recently. Twice raiding Targ and last night defending the Isle from the Village People spring to mind.

    (Party): Mezghar says, "Stop."
  • I'm at the point that I'm considering just changing my attack setup again, so that I can toggle between using queue and just holding down the enter key.  I love queue, but it really does feel like this is a queuing issue. With that said, I have been getting very heavy and strange spikes during group fights (I can see people doing stuff at regular pace, but my stuff isn't), and I have had a few instances of hunting causing some issues, particularly when leading over 4.
  • edited December 2015
    It's definitely a room-localized issue, though, because when I move out of a laggy room to snipe or whatever, the lag is magically fixed. Of course, that'd imply it'd be client-side, but nothing else seems to point at that.
  • Kiet said:
    Someone suggested it wasn't a curing system issue, but certain offense systems. If it's related to queueing/serverside aliases (and I have no idea why it would be, but it seems as plausible as anything else), certain offensive systems or coding habits might lead to too much queuing/setaliasing at once, or something.


    Tecton says there's been no server-side lag on processing commands, so the issue seems to be client-side (primarily getting the commands sent to the game, it seems). If the issue was the volume of commands being sent I'd expect the same issue to be seen in 1v1 and other smaller fights, since I imagine if people are spamming in group fights they're probably also spamming in 1v1 fights too (I know that I definitely don't take a different approach to how I enter commands between the two).

    I know Dunn said he had the same issue in Twins, but I'm not aware of any other reports of it happening for people in small fights. The volume of input (i.e. text and GMCP data coming from the game) seems to be at least a factor.

  • Kiet said:
    Someone suggested it wasn't a curing system issue, but certain offense systems. If it's related to queueing/serverside aliases (and I have no idea why it would be, but it seems as plausible as anything else), certain offensive systems or coding habits might lead to too much queuing/setaliasing at once, or something.


    I wasn't queuing last night still happened during that thing. You know the one I'm talking about. Even manual inputs were lagging. 

  • This used to happen to me occasionally (years ago, in multiple clients and in different states with different ISPs), and it's still common when I try playing from my phone. I would receive output from the game normally with no lag (or no more lag than expected, on a slower connection), but sent commands would be delayed a lot, sometimes there'd be 30+ seconds between sending a command and it actually going through, while seeing incoming text from the game normally the entire time. It seemed to mostly be linked to spammy situations, receiving a lot of messages in a short time (on my phone it could sometimes be triggered just by following someone while they move at 2 rooms per second for too long).
  • edited December 2015
    But why would it start happening now, when computers are much more powerful, if it's a client-side issue? And why does it seem to affect people on crappy computers as much as powerful ones? Is there some situation where the incoming text becomes far too much to process with systems for mudlet? That seems like it'd have to be a lot of text these days. Like even in the old huge ctf's when 300+ players online were common, I never experienced quite this issue. Maybe it's just the way mudlet works, though, I don't know.
  • Kiet said:
    But why would it start happening now, when computers are much more powerful, if it's a client-side issue? And why does it seem to affect people on crappy computers as much as powerful ones? Is there some situation where the incoming text becomes far too much to process with systems for mudlet? That seems like it'd have to be a lot of text these days. Like even in the old huge ctf's when 300+ players online were common, I never experienced quite this issue. Maybe it's just the way mudlet works, though, I don't know.
    I was sitting here last night seriously going what the fuck... I'm running a 3.9 ghz unlocked quad core apu with 16 gb ram. I haven't had lag like that since I was on dial up using a HP Pavilion from 1999.... 

  • edited December 2015
    Last CTF I walked into a room where fight is going on. I lagged and mudlet went irresponsive. I had to end task. 

    This time tumblerace I can't do QL when we all stand at the bridge with Tecton, or it will be a 10s lag before the room desc shows up. 

    Client is definitely the issue, between SVOF, mapper, ndb and offensive scripts. 
  • The Mudlet mapper and especially the various ndb/highlighting scripts are notorious for bogging down Mudlet in a variety of situations.
  • JonathinJonathin Retired in a hole.
    edited December 2015
    I haven't been in any 'large' group fights lately, 6 people at most. No lag there though. I only have 13 hops though and an average ping of 19ms over the last 50 pings.

    I did have a single spike a couple days ago during a meeting, but that's about it.

    On another note, I have been using Svof 4 in Shaman and Svof 7 on Runewarden. Will update and test again.
    I am retired and log into the forums maybe once every 2 months. It was a good 20 years, live your best lives, friends.
  • Would the mapper really bog while standing perfectly still, though? I purposefully don't use ndb/highlighters other than simple triggers for key people, and I still get the same thing.
  • edited December 2015
    Right, by keeping a light profile my comp doesn't even slow down from the massive spam drops, so I find it weird that somehow it's choking in a 10 person fight in such a bizarre way.
  • edited December 2015
    Kiet said:
    Right, by keeping a light profile my comp doesn't even slow down from the massive spam drops, so I find it weird that somehow it's choking in a 10 person fight in such a bizarre way.
    I hadn't even been having that issue until last night. I just raided Hashan not too long ago with zero issues and they rushed us... (this was uhh 3 rl days ago maybe... what ever night I got the tell from  @Fendrel telling me I should perhaps ease off) 

    oh and my normal ping is .039ish... with a system lag of .071ish 

  • edited December 2015
    I get the kind of lag @Daeir describes during group combat, I don't remember a time I haven't though. I wonder if I get it worse due to being quite far from the servers. My regular ping is a slightly rough 0.238 and that's about the best I can get in my country. There's a reason I like sniping in group combat, playing in room I can barely respond before I'm dead.

  • I use my own system on mudlet and it happens to me. 

    I get input from Achaea just fine and my echos are firing so mudlet is obviously not locked up.  Is there some priority in Mudlet that prioritizes receiving input over sending output and the sending just doesn't catch up?

    I guess there's no way to know exactly when an input is sent to the server, only when the 'send' function has been called and when you get the response back from the server seeing it process in game?

    image
  • I was thinking maybe it has something to do with GMCP input being too much for Mudlet, and then like Jhui said maybe it gets handled before sending output? Since all of the GMCP expansions are more new, so that would maybe explain why it's become such a big deal more recently.

    It doesn't seem to be queueing related because I stopped using queueing for some fights and still had the same lag.

  • This has been very frustrating. In group fights, nothing will go through, including trying to just walk from the room, then suddenly everything will at once. It's been between 10 to a good 20 seconds lag before I get spammed with all the backed up commands. I haven't changed or added to anything I have recently, and it all worked just fine up until the last few weeks. 
  • TectonTecton The Garden of the Gods
    I just tweaked a couple of things and adding in some more datalogging to see if I can track down the cause if it's anything on our end. The CTF tonight should be a good test.
  • On a side note, this is an issue that happened to a small number of people back in the zMUD days. Never found the exact cause, but it seemed to stop after I cleaned up my highlights, echoes, and gags.

  • Dochitha said:
    Last CTF I walked into a room where fight is going on. I lagged and mudlet went irresponsive. I had to end task. 

    This time tumblerace I can't do QL when we all stand at the bridge with Tecton, or it will be a 10s lag before the room desc shows up. 

    Client is definitely the issue, between SVOF, mapper, ndb and offensive scripts. 
    I found out that too much use of regex when not needed can cause a bit of this. I had the same problem, walking into big rooms would be 10-20s of lag, mostly ending up in me being dead before mudlet started responding again.

    What I did is any trigger I had that didn't use a matches character I changed it to substring instead of regex. For whatever reason this actually fixed that particular issue.




    Penwize has cowardly forfeited the challenge to mortal combat issued by Atalkez.
  • When Mhaldor raided the cliffs in Targ a few days ago during one of the rushed my lag felt pretty bad so I tried to leave the room. I spammed S like 50 times to try and leave. The fight went on around me with no response for my movement spam. Then about 45 seconds later I finally moved and all the spa commands went through. I have mudlet lag occasionally this wasn't it. 
  • HalosHalos The Reaches
    edited December 2015
    The exact same thing happened to me during that fight. I got the command spam and the iconic postmortem "Duanathar" sent through after I was dead though, of course. 

    I felt like "Noo the irl mhaldorian ddos attacks again" but it was just lag. 

    A frenzied cleric screams, "Like more than one halo!"
  • I don't use queues for my attacks and it definitely happens to me.
  • edited December 2015
    hate it when there is rustle lag
  • I've only had freezes when someone dies and drops a large number of items.
    retired
  • Recently switched to the HTML5 client now that I have a computer that can handle it, but I haven't gotten a chance to try it out in any big combats. Previously I used Atlantis with no offence system and server side curing. Big fights lagged me out very badly, and in many cases I would get immediately disconnected as soon as anything started happening.
  • Cooper said:
    On a side note, this is an issue that happened to a small number of people back in the zMUD days. Never found the exact cause, but it seemed to stop after I cleaned up my highlights, echoes, and gags.
    It happened to me back then, but it happened in zMUD, Mushclient, and the old Java client (before old Nexus), so it didn't seem to be client/script-related (my guess was some kind of network issue, something like incoming packets working fine but outgoing packets being lost/delayed for some reason, but I don't know enough to guess beyond that).
  • Seemed to be way better during the CTF.
Sign In or Register to comment.