Large nwn PW projects, and Exploding Monkeys!

[sly_1]

Hey all you crazy Avlis party people!

*warning: long, rambling post ahead*

You know, I was thinking about Neverwinter Nights the other day, and I wanted to bounce some things off the Avlis community seeing as how you guys are one of the most knowledgable when it comes to large scale NWN pw projects.

See, here's my take, based off my experience from playing Avlis/COPAP (and to a lesser extent, ALFA). In a nutshell, every large scale wp project for NWN attained mixed success at best. For the most part, any project that linked multiple mods across multiple servers together experienced what I'll call the "hotspot" effect.

The hotspot effect is where, for whatever reason, one or more of the modules/servers becomes the popular hotspot to hang out on, while the rest of the servers/mods languish in relative obscurity. ALFA suffered from this problem in a big way, while Avlis had a lesser case of the hotspots (at least back when I was playing, anyway).

Perfect example was when farrel was first launched. It had a lot of cool things to do, but no players. DM's ran quests there, there were many posts encouraging people to explore the place, but in the end the more established hotspots were were the action was.

Now I dunno if Ferrel ever got populated with players or not, but you can easily extrapolate what will happen if the vision of recreating all of Avlis is ever achieved: 4 or 5 hotspot servers and 50 empty ones (which, as an aside, is almost exactly what happened to ALFA, the other massive nwn pw).

The hotspot issue is also seen in any pw, large or small, within each individual mod. There are always areas in pw's that are there simply to provide a sense of scale when travelling from point a to point b. Case in point: I'd wager some areas in the wilderness of avlis see maybe 2 or 3 players per month, while others see thousands. The simple reason being there isn't any reason to go to many areas: no quests, no npc's, nothing other than (maybe) a spawn or 3 of some sort.

So in a large project like Avlis you have big hotspots (the servers of mikona and elysia) and little hotspots (the markets, the crafting areas, certain pubs)

I think the problem is twofold: for one, there isn't some massive community of thousands of hardcore rp'ers vying for spots on servers. While Avlis has always been considered a major success, it hosts in the ballpark of a hundred odd people at a time, not thousands.

Despite this, the overloaded servers of Avlis seem to show that there certainly are *enough* rp'ers out there to form a vibrant community, But if so, why was Ferrel a desolate wasteland while Mikona was crashing with 60+s on it every 30 minutes and a que of 50 people trying to get online (back in the day, anyway)?

Which brings me to the second problem: travel in a large scale NWN pw isn't very fun. This as opposed to more traditional MMO's that feature crap like world maps with insta-travel, horses, etc. Sure, in avlis you can take a boat or other "portal" to far away lands, but once there you still often have to endure dozens of load screens to get to wherever your going.

Of course, rp'ing along the way was the saving grace of Avlis, but that begs the question: if rp is the focus, why do you need to represent 10,000 square km of landmass? Wouldn't having TONS of areas and TONS of monsters and TONS of respawning boss monsters be more fitting for an action pw than an rp one?

After I thought on the matter, I asked myself a simple question: Would Avlis (or any other large scale pw project) be better if the notion of recreating massive tracts of land were abandonded?

Lets say you simply had Mikona, elysia, t'nanshi and the wilderness. We're still talking about a lot of real estate, here. But lets say these areas, which last time I checked comprised 4 servers or so, were instead spread them out between 10+ servers, each with fewer areas but 96 player caps? And all the effort, coding, time, enegy, and money went into making those bits of the Avlis Universe as immersive and interesting as possible? In other words, what if there were literally NO areas that were there for "filler"? Instead, what if an interesting npc, quest, event, *something* was found on each and every area in what now comprises the 4 "main" avlis servers?

It seems that in the rush to represent massive worlds, rp projects like Avlis and ALFA neglected what NWN is really good at: creating an intimate setting (as opposed to a massive one). This is why I think Ferrel was deserted for so long: Places like Mikona and Elysia were so established, and there was such a community in those spots, that there was no need for people to ever leave. Sure exploring Ferrel was swell, but after one zone of spawn xyz, most people would rather just go back and rp in their main town than see yet another zone with spawn xyz, and another after that.

So anyway, in the end, I think there might be something to the notion that less is more. At least where NWN is concerned. Perhaps NWN2, with it's world map system, will be better suited to massive projects like Avlis and ALFA...






p.s. the exploding monkeys were omitted from this post at the demand of animal rights activists. Damn you, PETA!!!!!

[apandapion]

After I thought on the matter, I asked myself a simple question: Would Avlis (or any other large scale pw project) be better if the notion of recreating massive tracts of land were abandonded?

I wrote this before NWN came out:

http://www.io.com/~shren/.nwn1.html

It predicts both what people would do and the problems that would arise from it. Of course, it gets a lot of things wrong too, but it was before the game was even launched. Nowadays I don't care as long as I'm having a good time.

Lets say you simply had Mikona, elysia, t'nanshi and the wilderness. We're still talking about a lot of real estate, here. But lets say these areas, which last time I checked comprised 4 servers or so, were instead spread them out between 10+ servers, each with fewer areas but 96 player caps? And all the effort, coding, time, enegy, and money went into making those bits of the Avlis Universe as immersive and interesting as possible? In other words, what if there were literally NO areas that were there for "filler"? Instead, what if an interesting npc, quest, event, *something* was found on each and every area in what now comprises the 4 "main" avlis servers?

Some comments in no particular order:

You are neglecting the explorer mindset. Very few people are 100% roleplayer with no desire to experience the other three branches of play.

Have you played on 404? It has a pretty low "empty space" ratio.

I seriously question the premise that dropping geography lets you have 96 players. I would have to see proof - be on a server with 96 people - to believe it.

Lastly, there are always going to be traffic nexi on any graph.

[Silk]

Also, don't neglect the fact that every area in Avlis constitutes a stage that DMs use to tell their stories.

It's not all filler, it's just used when needed.

[spokeydonkey]

I agree with you that hostpots exist on most large PWs. It's a part of NWN's implementation, and probably of volunteer teams working piecemeal on things. A professional paid development team who worked 40 hour weeks would probably make something approaching MMoRPGs.

I don't agree with your assumption that having hotspots makes a PW attain "mixed success." Your post doesn't address this, so my question is: why do you feel that hotspots are bad?

[sly_1]

...
I don't agree with your assumption that having hotspots makes a PW attain "mixed success." Your post doesn't address this, so my question is: why do you feel that hotspots are bad?

I suppose the worst case example was seen in ALFA (A Land Far Away, another massive pw project that attempted to represent the entire forgotten realms).

At one point, ALFA had like 20 live servers, but only 3 or 4 had active playerbases. A project with 3/4ths of its servers standing empty almost all the time is at best a mixed success. When Ferrel was first lauched, I saw it as a mixed bag: a lot of work obviously went into the place (and I thought it was well done, personally), but very few players were enjoying it. (Granted I stopped playing NWN and Avlis a long time ago, and for all I know Ferrel is a hotspot these days. )

So the hotspot problem becomes bad when time and energy goes into building a server that hardly gets used when instead all that effort could have gone into making a lesser used portion of an allready popular server better. For example, there are several small towns here and there in nanshi and m'check that could have been populated with quests, npc's, interconnected storylines, etc.

@apandapion:

I understand the explorer mentality, that's why I mentioned the idea of enhancing the "filler" areas

As for fitting 96 people on a server, I was just trying to make a point. If the mod on the server currently labeled "mikona" was broken into 3 mods and spread across 3 computers, it could theoretically contain 3 times as many players. I used 96 players as that's the max I've ever seen in NWN pw's, but it could just as easily be some other #.

[chamalscuro]

Ferrell actually is a hot spot these days. Too much so, for my liking. Crafting is well-nigh impossible when Ferrell is loaded. And the Ferrell server seems more ...uh... "Fragile" than the others -- i.e. Crash-happy.

Frankly, I am the type that goes to the webpage to check the server levels before logging. I won't voluntarily go on servers with 30+ logged-in. The rubberbanding, and laggy conversations & spells are really irritating for me.

Actually, I suspect Avlis has become more savvy about filling their servers in your absence. The problems you talk about, many I do not see.

edited for %*# typos

[sly_1]

...
Frankly, I am the type that goes to the webpage to check the server levels before logging. I won't voluntarily go on servers with 30+ logged-in. The rubberbanding, and laggy conversations & spells are really irritating for me.

.... The problems you talk about, many I do not see.

Well, in a way, you are illustrating my point. It's like so:

- for the sake of argument, lets say a server has 150 areas and holds 40 people

- 30+ people causes all these intolerable lag + crash problems you describe.

- but as an alternative to opening a new server with new content, you could simply split the 150 area mod into 2 mods w/ 75 areas and 30 people each. Any leftover dev time is then spent making the existing areas more detailed, interesting, immersive, etc.

At the end of the day my question is this: which would be better: a truly massive NWN pw with thousands of areas (but many "placeholder areas" that require a dm to be present to have any meaning), or a more detailed pw that while having many areas, spreads them over many servers to enhance stability?

please note, none of this is meant as an attack on Avlis. I respect what you guys have accomplished immensly

[spokeydonkey]

There's an upside to hotspots. If you have areas that concentrate 90% of the characters on a server, there's a much better chance to bump into people to play with than if the server has all areas equally popular, and players evenly distributed throughout.

[DurGil]

I love every inch of Avlis.. every area has life and vibrance. every turn gives a familiar jump in my chest. Every place has a fond memory. Every New place gives me a renewed excitement. I win at this game each time I open my eyes to see the vast expansive dream that is Avlis.. My eyes see what my mind has imagined. anything less would certainly not be more.. it could only be less. The geography and the people help to make up for my shortcomings as an rp'r.. I'm each day reminded I can be better.. I just look around and Instantly I am "In" this world..

just my rambling for the month.

[Reinstag]

Also, don't neglect the fact that every area in Avlis constitutes a stage that DMs use to tell their stories.

It's not all filler, it's just used when needed.

This is possibly the most important point here.

Those empty patches of land are what a lot of DMs call "Empty Stages". They're on Avlis, they're on Spire, and I bet they're on ALFA. These maps are usually off the main paths of travel, they have few, if any, spawns, and are there simply to "set the stage" for a DM to run an event, be it part of an on going plot or for a single encounter for a small group of players.

And, as spokeymonkey said, hotspots concentrate players. This is good for players and DMs. Players find it easier to get a group together while DMs find it easier to spawn Sereg or dire chickens for maximum damage spread.

There's also the neglected roleplayer: The Explorer. This person usually plays a ranger or druid and spends countless hours wandering the server. While he or she may not be in a group of people discussing the weather in Mikona, they are roleplaying their character's desire to wander/explore. This type of roleplayer needs very little to be entertained, just have a lot of wilderness to wander and they're golden.

And I feel cheated. I wanted some exploding monkeys!

[chamalscuro]

...but as an alternative to opening a new server with new content, you could simply split the 150 area mod into 2 mods w/ 75 areas and 30 people each. Any leftover dev time is then spent making the existing areas more detailed, interesting, immersive, etc.

In some cases, I think you are right. Example:

The Wilderness Server is the glue that connects all the other servers of the south. Though there are exceptions, in general you must portal to Wilderness to travel between the other servers. Elysia to Deglos, or Le'Or to Ferrell, as examples. On a bad night, travelling the Wilderness is enough to induce alcoholism.

Wilderness should be split. There's even a natural demarkation for the split: The M'Chek Wilderness and the T'Nanshi Wilderness.

Why hasn't this been done? I imagine it would be a lot of work; thankless for the coder(s) doing it, as you are bound to piss off as many people as thank you, no matter how logical the change.

In most other cases, I want big servers. I am one of those Explorers mentioned earlier; my main is purpose-designed for exploration. Avlis attracts me because it is so vast.

[Buddha]

And I feel cheated. I wanted some exploding monkeys!

[apandapion]

Wilderness should be split. There's even a natural demarkation for the split: The M'Chek Wilderness and the T'Nanshi Wilderness.

Why hasn't this been done? I imagine it would be a lot of work; thankless for the coder(s) doing it, as you are bound to piss off as many people as thank you, no matter how logical the change.

I think an overland/underdark split is impending regarding wilderness.

[lanrik]

Wilderness should be split. There's even a natural demarkation for the split: The M'Chek Wilderness and the T'Nanshi Wilderness.

From what I read on the boards, if I understand correctly, the Dark City server will contain the underdark areas that are now on the wilderness server. I imagine that will help quite a bit with the server population. Someone correct me if I have this wrong.

[chamalscuro]

Ya, I heard about Underdark being split to the Dark City server. Surely that will help some. But the Wilderness server will still be an unwieldly mess post-Change.

It makes little sense for the M'Chek badlands and T'Nanshi Forest to be on the same server. I mean -- Gah! -- I count 20(!) areas from Entrance of Deglos to Mikona. So a DM event in M'Chek causes lag at Dwarftrade. Swell.

Worse, as I mentioned above, one frequently MUST cross a portion of Wilderness to travel between the other servers.

[Orleron]

I could answer this in two ways: from a game design perspective, and from an Avlis server perspective.

It helps to know the difference, along with some history of why the Avlis servers are the way they are.

If we were a MMORPG with paying customers, Avlis would not have as much filler space. Why? Because too much filler space pisses off paying customers and makes them leave. If we were a MMORPG with real customers, the design aspects of Avlis would get more attention than they do. SOME filler space is a good thing to give some sort of scale to the place and to let DM's have stages to run their stories. But you would not portray the entire continent of Negaria in the scale that we did, OR, you would portray it that way and put many many more portals to get people to travel around. Think of the moon gates in UO.

The reality is we are a free game, which is not nearly as lucrative, but gives us a TON more freedom to do what we want. What I wanted was to put all of Avlis online. I had to pick a scale, and I picked "distance between outside areas = 50 miles". In a continent that is a little bigger than the USA mainland, that's a lot of areas, but it's the scale I chose for the grid because I thought it was doable over time.

A lot of those areas would be filled and fleshed out over time with second passes, and many have been. Those who have been around longer will see that areas in M'Chek and T'Nanshi have much more stuff in them than the newer areas of Deglos which are still on their first pass.

We once thought the Wilderness would contain ALL wildernesses on the continent, from the South to Tyedu. What a laugh that was! We didn't understand the limitations of NWN, lag, and disc space. Now we do.

[downsystem]

I think now these days more of the other servers are finding quite a bit more traffic and lag is always a battle with such big servers. I think having smaller server areas with the ability to have more people on them could possibly be a good thing for immersion aspects. Though wouldnt that still make the server laggy.

Smaller server area + larger population +internet limitations + personal computer limitations +nwn game mechanics limitations= LAG.

Larger server area + smaller population +internet limitations+personal computer limitations + nwn game mechanics limitations = LAG

I havent played any other online fantasy games besides avlis, But has anyone out their played one that didnt have any lag at all?

[pstanton]

Although there are several servers that seem to have very few people on it. Everyone seems to head for Elysia, Mikona or Wilderness.

If you look at LeOr, Ferrell, Deglos, Kuras and Visi they're usually empty, or have only 1-2 people on.


Solution: Make incentives for people to head there, have DMs constantly on those servers. Something like that.

[apandapion]

Although there are several servers that seem to have very few people on it. Everyone seems to head for Elysia, Mikona or Wilderness.

If you look at LeOr, Ferrell, Deglos, Kuras and Visi they're usually empty, or have only 1-2 people on.


Solution: Make incentives for people to head there, have DMs constantly on those servers. Something like that.

I think this is just a function of time. After all, if nobody ever got into the novel, we'd all be hanging out on Mikona 24/7 now. There are a lot of incentives to go, but you have to be in the right mood to deal with the novel.

[sly_1]

Although there are several servers that seem to have very few people on it. Everyone seems to head for Elysia, Mikona or Wilderness.

If you look at LeOr, Ferrell, Deglos, Kuras and Visi they're usually empty, or have only 1-2 people on.


Solution: Make incentives for people to head there, have DMs constantly on those servers. Something like that.

Sorry to ressurect this old thread, but this post illustrates my point. Rather than creating incentives to get players to inhabit these upopular areas, why not make the allready thriving areas able to support more players by splitting them into smaller modules spanning multiple instances, and fill them with even more stuff to do?

Imagine if there wasn't any lag in Mikona and it never crashed, but it was split into 2 or 3 instances. Given the current popularity of Mikona, the 2-3 instances would each be full of players from day one if such a change were made. No dm quests would have to be orchestrated to get people on the new servers/instances, no new exotic crafting materials would have to be placed, etc. All that time and effort can just go to making the 2-3 instances representing Mikona even better than it is now.

Would that be better than adding an entirely new module representing a new part of Avlis? Especially considering that it takes a lot of effort to prod players to populate new areas, and thus new servers will tend to stand empty for weeks or months until they catch on?

[Talwin Hawkins]

Although there are several servers that seem to have very few people on it. Everyone seems to head for Elysia, Mikona or Wilderness.

If you look at LeOr, Ferrell, Deglos, Kuras and Visi they're usually empty, or have only 1-2 people on.


Solution: Make incentives for people to head there, have DMs constantly on those servers. Something like that.

Sorry to ressurect this old thread, but this post illustrates my point. Rather than creating incentives to get players to inhabit these upopular areas, why not make the allready thriving areas able to support more players by splitting them into smaller modules spanning multiple instances, and fill them with even more stuff to do?

Imagine if there wasn't any lag in Mikona and it never crashed, but it was split into 2 or 3 instances. Given the current popularity of Mikona, the 2-3 instances would each be full of players from day one if such a change were made. No dm quests would have to be orchestrated to get people on the new servers/instances, no new exotic crafting materials would have to be placed, etc. All that time and effort can just go to making the 2-3 instances representing Mikona even better than it is now.

Would that be better than adding an entirely new module representing a new part of Avlis? Especially considering that it takes a lot of effort to prod players to populate new areas, and thus new servers will tend to stand empty for weeks or months until they catch on?

Not to upset anyone but that pstanton post is wrong. there are LOTS and LOTS of people on Ferrell. LeOr more and more. Deglos in my opinon is the only one REALLY lacking and thats probably due to the fac that it scares the shit out of everyone