Page 3 of 10 FirstFirst 12345678910 LastLast
Results 41 to 60 of 191

Thread: Player warning: Crowd Limiting System coming up

  1. #41

    Research the problem a bit more

    Quoting: "To call this "lag" is actually a wrong definition of the problem. Most of the time this is not network lag, or the server running slowly, but what we call the "Resource Manager" loading new meshes and textures from the hard-disk of your computer or the loading of textures and meshes into the 3D card - to be able to display it all gloriously on your monitor "

    This is not true. When the ping reaches 10K+, my harddisk is hardly working, memory bandwidth (RAM and 3D card memory) is not strained at all and the network monitor shows very little activity the problem is not MY computer, it's the Internet and YOUR cluster of servers.

    I am probably connected to one of the best connections available. (100Mbit fully switched fiberoptic net directly connected to Uninets backbone). When I recieve very little information from your servers, the problem is either the Atlantic cable or your system. To test the Atlantic cable I can connect to a N-American server. Upon recieving 500Kilobyte+ per second, I can't see any other problem than your servers.
    Corianin - TL6 NT - eqp

  2. #42
    Three things:

    First, since people would not be allowed to zone into an area, they could end up trapped in a shop (the Mort area comes to mind). That seems to me to be very troublesome.

    Second, what about rez. If I die and have to rez into a crowded place, what happens to me?

    Third, I don't see how we can plan attacks and defense if "attackers will actually be moved out. (And vice versa.)" Who will be moved out? What will be the computer's criteria?

    Scorus

  3. #43
    Talk on the test server prior to 14.6 was that they were revamping the HSR (Hidden Surface Removal) routines for 14.6 and 14.7

    I have a sneaking hunch that 100 Atrox all in full MK and Albrecht wearing beams would lag you out almost as much as 100 totally mixed up breeds/gear.

    They seem to have issues with both with streamlining object models (objects as in code objects) and properly exposing only what is needed. Scares me to death to think that your client looks at an Atrox in a Graft armor helmet and the server tells your client that he's wearing a Graft armor helmet rather than just Mesh 98563 on model 481. I doubt it's that bad, but it's an example where your client would have to go look up a Graft Armor helmet in the local database and then determine the appropriate model and mesh to render AND determine if the mesh has already been loaded into the memory of your vid card.

    100 identical Atrox enforcers should load lightening fast. First thing I thought of was how MP3's work. If you've already uploaded a similar bit sequence, you don't upload it again. You just point to where it was used before. One big burp of data and then 99 more saying 'hey it's just like that one'.

    Then there's the whole thing regarding whether we really give a crap about what some other guy is wearing during a PvP battle. As far as Tower Wars are concerned, they could all be blue boxes, grey spheres, and pink triangles. Asynchronously broadcasting 'eye candy' data during lower traffic periods and allowing our client to render based on this information when it has some free cycles is another way to go. I really don't care if you have a beard or are wearing a bikini.

    Mostly I like this idea as it has the potential to promote some good battles with even numbers of combattants. As much as those who wish to bring 300 people to a battle and enjoy the 'thrill' of winning a non-contest may speak out against it, I think even they will feel this is all for the best once they experience a fair fight for once.

    As I said, you are going to have to put more thought into strategy and generalship. Think of an american football game where you are only allowed 11 men on the field at once to battle the other 11 men. Who you put out there and how you handle your substitutions could be crucial. The large bloated orgs still have the advantage of a never ending supply of subs.

    Imagine an american football game where one team could put 80 men on the field while the other could only put 11. Sound sporting? Assume those 80 men were all fat smokers. They would still probably beat the other 11 no matter what shape the 11 were in. Rough analogy but, it's kinda why I like the proposed idea.

    A note should be made that the 'liittle guy' still has to field a full team. If he's got to get 40 people roused, that's no small feat for a 'small' guild. Also, once one of his 40 dies and is sitting at reclaim, there may very well be no sub to fight in his stead while he's in rez shock.

    As I said, this could very well be the best thing for PvP in AO or just another bungled disaster.

    A side note: SW: G is handling PvP like this somewhat when it generates 'battlefields'. A battlefield is autogenerated and a call to arms goes out. You can then sign up and fight for your faction. The battlefield closes when it is full. More than a few similarities.

  4. #44
    Originally posted by wrecklass


    So you think it would be a POSITIVE thing if they reduce the number of players in the game to make it more enjoyable to a few? That's where this ends up going after awhile. If they lose customers, which they are likely to if this is implemented, then they won't be able to afford their staff and AO disappears.

    I don't think that is as positive a thing as you make it out to be.

    I'm not being negative, I'm being a consumer. If they implement this and it means I can't go to some playfields, or enter some city or dungeon, then I have to look for another game to play. I won't keep paying for my account if that happens too often. Simple economics.

    I just don't think they have thought of other options very well. They didn't ask us all to comment with our praise of their brilliance. If that is all they wanted, they shouldn't have started this discussion. They wanted our opinions, and they are getting them.
    I'm just saying there is such thing as constructive crticizm, that's all. Not get rid of your programmers, or your programmer suck blah blah. It doesn't help anything... Not many people on these boards could do much better than their programmers. And if they can, go work for funcom and fix the problems.

    Yes they wanted your opinions, but not your opinions on how competent their programmers are.

    If you don't like this solution then comment on that, don't bash their programmers, that's all... I didn't go back and re-read your posts, so if you didn't bash their programmers that was not intended for you and i appologize.

    AO doesn't know how EQ works therefore comparing the 2 and expecting funcom to figure out something based on another companys proprietary code is... I don't have a word for it...

    I do not think they will limit entire *playfields* or zones. There are not enough playfields to handle any kind of limits like that. 80% of the people would have no zone to go to... Maybe I misread, but i don't remember him saying that. My guess is that they will limit based on the smaller areas, like 40m areas, or 80m areas, like the general area around the towers.

    Or in dungeons, where if the max is reached they can spawn a new dungeon for you to go to.

    I don't think their goal is to make you not able to play or go anywhere in the game. If that is there goal, and I am wrong, then fine. I'm a total idiot.

    But to answer your question, i'd rather NOT goto an area if it's so congested it will lock me up or knock me off AO. Simple. Therefore if me not going, makes the experience of those other 50+ people that made it in better, then so be it. It's not like that 1 place, is the only place to be on Rubika... Go fight somewhere else. Each guild can have 5 main towers not just 1...

  5. #45
    This is unacceptable though....preventing people from taking part in things in efforts to reduce exploitation via lag and lag itself??

    They just need to stop messing around with content patches and work strictly on making the game engine solid. Whether or not they have the programming power to do that is another story....
    .: Naraya :.

  6. #46
    The way I read it (no one else did it seems), the 50/50 limiter will only kick in when You reach the max number of people of the zone. If the zone can handle 100 people then there's no problem with 80 attackers and 20 defender. But when defender No. 21 enters one of the attackers will be kicked. I'm guessing it will be the lowest level char that get kicked. 2 reasons:

    1) Seems most fair. Otherwise You might end up with only highlvls on one side and only lowlvls on the other.
    2) It prevents people from logging in lowlvl alt's of the opposite side to hog the spots.
    Ghump - a member of Ragnarok

    My hovercraft is full of eels

  7. #47
    Originally posted by Ghump
    The way I read it (no one else did it seems), the 50/50 limiter will only kick in when You reach the max number of people of the zone. If the zone can handle 100 people then there's no problem with 80 attackers and 20 defender. But when defender No. 21 enters one of the attackers will be kicked.
    Yes, that is exactly how I understood it, and I think most of us undestood that as well. But I can guarantee, whatever the number is, it will be reached. If you have a one hundred player limit on the PF, then their will be 100 people there. There are guilds who can certainly make sure that they fill up every available slot, and will do so if it gives them any advantage at all.

    Originally posted by Ghump
    I'm guessing it will be the lowest level char that get kicked. 2 reasons:

    1) Seems most fair. Otherwise You might end up with only highlvls on one side and only lowlvls on the other.
    2) It prevents people from logging in lowlvl alt's of the opposite side to hog the spots.
    The problem is that a lot of us are guessing because they didn't say much in the article. Basically they warned us that they were gonna do this, and they appologized (numerous times) for having no better ideas. I doubt the system will be this straight forward. More likely it will kick the last player that came into your team when a new player shows up on the other side. That way the first people to show up would have priority for staying in.

    But that is just a guess as well. I suspect that keeping track of who entered at what time would be too complicated, and it will just randomly toss out some folks.

  8. #48

    Hmm.. easilly exploitable

    All it would take to unbalance this system is for one side to warp in sit-in characters of the opposite side for the purpose of limiting the number of people who could come in.

    Example:

    Let's say it's camelot, and the limit is 100 people, 50 clan, 50 omni.

    Let's say I'm Omni. A dirty trick would be to warp in 25 clans and park them. Then Clan could never have a substiantial enough force to take box. Or, vice versa. With the number of people with multiple accounts/multiple computers, it becomes very easy to exploit... and it's not "really" an exploit... I mean.. there's no rule saying that the player on a separate account can't be there, and not doing anything... right?

    Another thing....

    Would this mean that if Omni left 50 clan parked in Camelot, that Omni could *always* own that GS (and vice versa)? That seems silly.


    Proposed solution:

    Leave it the way it is, but make beacon warp fail when limit is reached. Indoors PFs, the current GS mechanism works great, except for warping.
    Last edited by Talys; Nov 7th, 2002 at 20:52:19.

  9. #49
    I believe some limit on player density will have to be implemented, but it could be a lot higher than is possible now with engine improvements.



    Originally posted by Bionitrous

    Then there's the whole thing regarding whether we really give a crap about what some other guy is wearing during a PvP battle. As far as Tower Wars are concerned, they could all be blue boxes, grey spheres, and pink triangles. Asynchronously broadcasting 'eye candy' data during lower traffic periods and allowing our client to render based on this information when it has some free cycles is another way to go. I really don't care if you have a beard or are wearing a bikini.
    I agree here 100%. While I have made no attempt to examine the game engine internals, it seems all textures must be loaded before a player can be drawn. In cases where either bandwidth or client limitations (texture uploading) would result in some form of lag, I would like to see generic models used. I think three models, one for each side, would suffice.

    Once these generic models are drawn, the full models should updated only when conditions allow, prioritizing on movement, damage, and effects with bandwidth limitations and frame-rate with client side limitations. Some small percentage of resources though, such as 10 to 20 percent, should be reserved to update models, even in these resource limited situations.

  10. #50
    Im reading this the same way as the rest of you and it seems an impossible thing to police and do right.

    Plus I think some of you are being overly generous with the numbers. You get over say 40-50 people in a tight area and the lag starts even on a fairly powerful machine.

    100? 50 per side? right.

    I was on a raid once with about 60 or so people and it was nearly unbearable.

    The alt issue completely destroys this entire idea. There are orgs big enough to swing things their way in the event of they want to take a position somewhere. Even if they dont fill all the slots of the opposing side if they just have 10 of their people log in with low level alts they will cripple the other sides resources enough that they should win a battle for an area.

    None of the rest of this matters until they give a soloution for how this is not going to happen. You cant do a time thing either because people can log in the alts to the area the night before and then log in if they know they are going to take the area the next night.

    The whole concept behind this crowd limiting is really silly. What you need to do is fix the whole PvP system and make the fighting blanced and put more caps on level fighting etc. Limiting crowds in PvP fighting is not going to work. But then again letting 100 people vcrowd into an area and crash your game because it cant handle it is not going to work either.
    Gregg "Tekkor" Alsaqri
    Webmaster of: Agent Sector (Rip 2006)

  11. #51
    Originally posted by Various
    I believe some limit on player density will have to be implemented, but it could be a lot higher than is possible now with engine improvements.





    I agree here 100%. While I have made no attempt to examine the game engine internals, it seems all textures must be loaded before a player can be drawn. In cases where either bandwidth or client limitations (texture uploading) would result in some form of lag, I would like to see generic models used. I think three models, one for each side, would suffice.

    Once these generic models are drawn, the full models should updated only when conditions allow, prioritizing on movement, damage, and effects with bandwidth limitations and frame-rate with client side limitations. Some small percentage of resources though, such as 10 to 20 percent, should be reserved to update models, even in these resource limited situations.

    Excellent ideas. Perhaps the areas for the towers should be treated like dungoens too and require loading when you cross into them so the rendering area is smaller. The idea of generic models in these areas is also another great idea for lag purposes.
    Gregg "Tekkor" Alsaqri
    Webmaster of: Agent Sector (Rip 2006)

  12. #52
    I could live with simpler meshes in pvp-zones.

    Level, breed/gender, side, profession, max hp, current hp and location. What more do we really need? (and damage/combat/movement updates, of course)

    #1 - Less stress on the client. This shouldn't matter much, as any system out there should be able to cope with the player-models in use today (if rendered efficiently), but...

    #2 - Cuts down traffic a lot. When i'm lagged in camelot, i think it's because of bandwidth issues most of the time. (Although the client sure as hell could do things more efficiently as well)


    /Puus
    Last edited by Shadowstalker; Nov 7th, 2002 at 21:32:28.

  13. #53
    Capn Obvious speaks...

    "People have experience what they call "lag" in many situations while playing the game."
    Really...?

    This is a weak fix. You know it and we're bitter about it. You've already got a broken PVP system system here, what next? You now plan to limit it and release jack about the details other then "50/50 and we warp folks out".


    I guarentee that people are going to storm these boards and drop their accts the first time they're in the middle of a large guild battle for a tower and you pull them out of it because of your sloppy fix.

  14. #54
    I kind of like the idea of random kicking from the combat zone. People talk about tactics. People like tactics. People like to think they know tactics.

    I like the idea that the best planned things run into random kicks in the arse by just the way the world works. Being a level 200 shouldn't protect you from tripping over your own foot and knocking you out, preventing you from continuing the fight, any more than being a low level does. Hell it could be explained as a defensive tool of the tower setup, every so often it builds up the power to warp out one of the identified enemies.

  15. #55
    Originally posted by Losobal
    I kind of like the idea of random kicking from the combat zone. People talk about tactics. People like tactics. People like to think they know tactics.

    I like the idea that the best planned things run into random kicks in the arse by just the way the world works. Being a level 200 shouldn't protect you from tripping over your own foot and knocking you out, preventing you from continuing the fight, any more than being a low level does. Hell it could be explained as a defensive tool of the tower setup, every so often it builds up the power to warp out one of the identified enemies.
    Yes, would be nice if an Engineer on the defensive side could execute a subroutine in the Controller tower and have it dismiss an appropriate number of of attackers. Would be quite fitting. I reckon the higher the qlvl of the tower, the bigger the crowd that could be managed by the Controller.

    Keeping your Engineer alive would then be strategic as he's the one who both repairs the tower as it's damaged and operates the machinery to dismiss attackers: Reverse Beacon Warp. Send 'em to their home faction city.

    "Reinforcements for the defense of your towers have arrived in proximity. Tower now charged to repulse 4 attackers!"

    /tower repulse 4


    "3 attackers successfully warped out of combat. 1 resisted"

    /tower repulse 1

    "1 attackers successfully warped out of combat."

    Make it a smidge more tactical.

    /tower report

    "Controller is at 85% health"

    "Executing nanoformula Intricate Repairs: Extended"

    "You repaired Controller tower for 12,850 pts"
    Last edited by Bionitrous; Nov 7th, 2002 at 21:53:46.

  16. #56

    ahh.. yes.. I can see it now..

    Just a few comments...

    AC uses a similar system. It works, but its annoying as all hell.
    AC isn't trying to focus their game on PVP, AO is. Big difference there.

    Notum Wars - Stress the WARS there. This isn't Notum SPORT.
    This is - I have something, you don't want me to have that or want to have it yourself.. then come and get it!!! This isn't a friendly match in some game where you'll recieve points for what you do. You'll be fighting for resources and power.
    You don't defend those resource, you stand to lose them - time spent building them up, creds spent and the time getting the credits, and the bonuses you recieve from the base.. etc. Yes, I'd like a fair fight.. but I won't be counting on it ever. Thems the breaks... there's always someone(s) that'll be bigger, better, and badder then you.

    Just a suggestion.. this is a online rpg, but take a look at Tribes in its pure unmodified form. Offense, defense, bases, and vehicles. Bit similar don't you think? This little blurb really means nothing.. just something to ponder.

    It was mentioned - "will be monitored". By who? When? How do you get this person to be impartial?
    The reasons I ask this are simple.... you can't even fix a invent rearrangement bug. It was around for a long time, then was fixed, and back, and supposedly fixed, and back, and then fixed.. now its back. This is just a one of MANY examples out there of this sort of thing. Gun init being messed up, wrong types of init on weapons, things being deleted from invents or banks, etc.. the list goes on for a very lenghty time. Yes I give you credit, you've done some wonderful things in the past to make this game better. However, I can't think of any of them that didn't have some glitched involved with it. Your track record looks something along the lines of - for every 15 things fixed/added, 3 new bugs are created or are recreated. With this track record your going to put in a new system that is supposed to have a low profile to the players... I can just about image the fallout from this and the amount of time and tweaking to get it working right.. all the while your players having to put up with it, complaining about it, and whining.. etc. I'd like you to prove my hunch wrong on this, but I won't be holding my breath.

    Don't give me any crap about my it being my systems fault on how crappy AO runs on it. I'd like to see a current survey of players and their systems to back your comments up. Since I started playing AO I've increased my overall system power by a factor of 5. The same can be said of my bandwidth. Hardware is pretty cheap on the overall and is improving at a good rate. I'm not rich by any stretch of the imagination.. I pretty much consider myself to be the average online gamer when it comes to hardware. Put the blame where the blame is due..

    Back to the track record.. add another one to the notch on the belt of putting a bandage on it instead of actually fixing it. You sure seem to have a large supply of tweaks, nerfs, and bandages.. but a very small and limited supply of actual fixes.

    As to what I think of this system.. well.. to put it lightly.. it blows. I see more negatives coming from it then positives. I see a huge amount of bugs and potential issues with it that will most likely be addressed by your beta testers and then be promptly ignored by the devs like always.

    Your world is pretty small.. there are places that tend to get crowded.. your adding in a booster that will increase crowding issues... instead of developing other areas, making some of the huge outposts out there into towns that are worthwhile to go to other then for missions.. or adding just blank very basic zones that have just been recently opened to the public from terraforming... instead of having just H2O as a hot spot adding in several other areas similar to it that are clan vs omni hot spots.. instead of having only a very small number of zones with decent reasonable length missions changing it so there's more areas like Broken Shores.. the list goes on and on.. of all the possible things you could do to strain off some of the overcrowding issues.. you chose one thats completely new to the game, storyline, and system... BEFORE trying any other options.. sigh..

    hehe.. I just realized something.. what the hell am I doing? Wasting my time basically...

    Everyones can have a opinion.. this is mine.. if you don't like it, please feel free to say so.. but keep it civil please. 8)

    Luck in the Shadows,

    Datawight

  17. #57

    Read and research before answering

    Many people here have clearly only skimmed the article in anger and others clearly have not read it at all, and are responding to what they are reading from other mis-informed posters.

    First of all: advanced word from beta-testers is extremely positive on the code clean-up that has been done in patch 14.7, which works in conjunction with the booster. So, it very well may be that many more folks will be able to share space smoothly after "boosting". Word is that the frame-rate hit in crowded areas will be much better than it is now. Do some research on this before you lay down too much hostility.

    The first big hot button issue here, the messy code, apparently IS being worked on, so, perhaps, stop harping on that...?

    Second: Read and re-read the article with an objective eye. I don't think it's as bad as you are imagining.
    Ozor: lvl 2(x) - MA - Clanner - RK2
    Qranx: lvl 2(x) - Adv. - Clanner - RK2
    Seclorum: lvl 12 - Agent - OT - RK2

    I'm drunk as hell!

  18. #58
    Well I hope the beta testers for 14.7 are better than the ones for 14.6.

    I'll believe it when I see it.
    Fear is the Mind Killer.
    You wasted life, why wouldn't you waste death?

  19. #59
    Just a quick few thoughts:

    Game-wise, I think it would be more acceptable to allow as many players as want to be involved. This is the absolute fair maethod. Engine-wise just start displaying simple models with limited detail at a certain distance based on how many are around with server side filters so the client doesn't even render the details. This goes for sound etc. as well. Make it a player option. If their system can handle it, let 'em choose.

    Seriously, focus for the human in a confrontation is as limited in-game as in real life. Usually, while a fight is going on only what is in front of them matters so the rest is a haze. The hosts that run the clients to date should be able to handle thousends of simple meshes with little detail which would be fine considering the real life haze.

  20. #60

    Re: Read and research before answering

    Originally posted by Wiccaman
    Second: Read and re-read the article with an objective eye. I don't think it's as bad as you are imagining.
    I have re-read the article, several times. And here is what I found:

    Originally posted by Gaute Godager
    Or we can make a "fair" Crowd Limiting system that tries to understand the PvP situation and balance the fight.
    They can't make a good AI system for their NPC's, how can they make a system that "tries to understand the PvP situation?" They will try to balance the fight, but sometimes fights aren't balanced. If the attackers show up with an overwhelming force, why should they be handicapped by an arbitrary system? Same goes for the defenders.

    Originally posted by Gaute Godager
    As defenders move in, and the sum of characters is above the legal limit - attackers will actually be moved out. (And vice versa.)
    There you go, players will be "moved out." No way to soft sell this one, you get over the limit, and we'll kick some folks out. Forget planning and tactics, just make sure you have 50 folks who are all 200th level, and you stand a chance of keeping enough of them in the fight. Don't worry about being outnumbered, the system will "balance the fight" for you.

    Originally posted by Gaute Godager
    Players trying to gain access to a too crowded area through teleportation, logging in or "zoning" (shifting playfield) will be stopped or get an alternative log-in place.
    So if you aren't even involved with the fight, but you need to get to something in that playfield, or you must pass through that playfield, you'll be told to just chill out for awhile. Go play EQ, we've got more important customers to please.

    Every time this happens people will be forced into thinking about how much they like playing AO. That's not what Funcom wants people sitting around thinking about.

Page 3 of 10 FirstFirst 12345678910 LastLast

Posting Permissions

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