The numbers game

[Nebu]

Terra Nova says what people here have been saying for years.

How can we accurately assess a world’s population size?  This is important for both reasons of social research and commercial viability.  It is important that as MMOs continue to grow as a cultural phenomenon that we neither downplay their impact nor over-inflate their growth.

That was the closing line of the piece.  I don't really have much to add to this, I just thought I'd bring it to the attention of those that don't frequent terra nova.  It's no surprise that companies want to inflate their numbers.  Businesses have been doing that... well... as long as there have been businesses.

[Akkori]

Count the unique credit card numbers or IP addresses, heheh. That would get you a lot closer than anything else. Besides, companies would never tell the truth, since that would mean that they would likely "lose" when compared to someone else.

[Venkman]

As I said over there, one problem is that different numbers mean different things to different companies. Players  like to validate themselves by noting their game has more subscriptions than others. But in the end, it's sophestry. For all the millions in WoW, or the million of GW, or the 550k+ of FFXI, the average player will never see any more than 60 other people at most at any given time, and their entire server popular is measured in the low number of thousands. It's irrelevant to me as a player what happens on SWGs 25 other servers, unless they all close up, and as a result require SOE shut it down of course :)

And this academic debate about using numbers to gauge success for future emulation I see every so often is somewhat limited thinking in my opinion, Red Ocean Strategy as it were. Designing to emulate is designing for yesterday's players. Unless you've got a major license or more money in advertising than any ten developers have as budgets, you're not going to take down the big boys anyway. So do an end run, of course scaling your business appropriately. Eventually you'll be rewarded.

[Nyght]

As I said over there, one problem is that different numbers mean different things to different companies. Players  like to validate themselves by noting their game has more subscriptions than others. But in the end, it's sophestry. For all the millions in WoW, or the million of GW, or the 550k+ of FFXI, the average player will never see any more than 60 other people at most at any given time, and their entire server popular is measured in the low number of thousands. It's irrelevant to me as a player what happens on SWGs 25 other servers, unless they all close up, and as a result require SOE shut it down of course :)

Just points out that, more then ever, you need to pay and play based upon what you actually see, not what they tell you it will be. Because if you play based upon promises of improvement, then the reality of their numbers can effect you.

[Soukyan]

As I said over there, one problem is that different numbers mean different things to different companies. Players  like to validate themselves by noting their game has more subscriptions than others. But in the end, it's sophestry. For all the millions in WoW, or the million of GW, or the 550k+ of FFXI, the average player will never see any more than 60 other people at most at any given time, and their entire server popular is measured in the low number of thousands. It's irrelevant to me as a player what happens on SWGs 25 other servers, unless they all close up, and as a result require SOE shut it down of course :)

And this academic debate about using numbers to gauge success for future emulation I see every so often is somewhat limited thinking in my opinion, Red Ocean Strategy as it were. Designing to emulate is designing for yesterday's players. Unless you've got a major license or more money in advertising than any ten developers have as budgets, you're not going to take down the big boys anyway. So do an end run, of course scaling your business appropriately. Eventually you'll be rewarded.

For the next generation of MMOGs, one needs to look at games like Eve where all players are in one world. Arguably, one could say that what players are doing in zones far away have no impact on them in Eve, but the possibility of all players affecting you is there. But that's a pretty specific example. One MMOG that gave you the chance to interact with more than 60 people at a time was DAoC, but you couldn't communicate with 2/3 of those people by any other means than combat.

[Venkman]

Just points out that, more then ever, you need to pay and play based upon what you actually see, not what they tell you it will be. Because if you play based upon promises of improvement, then the reality of their numbers can effect you.

That's an axiom :)

For the next generation of MMOGs, one needs to look at games like Eve where all players are in one world. Arguably, one could say that what players are doing in zones far away have no impact on them in Eve, but the possibility of all players affecting you is there. But that's a pretty specific example. One MMOG that gave you the chance to interact with more than 60 people at a time was DAoC, but you couldn't communicate with 2/3 of those people by any other means than combat

My point about the numbers of people on a server was only to talk about who's actually relevant to a player at any given time. Whether that's 3,000 concurrent players or 15,000 can't really matter between games because they're all so different. The "strength" of a SWG server economy is not comparable to the "strength" of the Eve galaxy except by RMTers, and those stats themselves aren't necessarily very insightful into the economic conditions ingame.

It's more important to achieve a proper cumulative density of many different playstyles than to strive for a total number of players. The numbers of inputs is based on core game systems and the size and gerth of a virtual space. Just as a business can scale to deliver all manner of experiences for any sort of budget, so does the economy of a game scale to what's actually there. Eve's galaxy is far larger than a SWG server, specifically built as it was to support every player who logs in, and therefore capable of supporting the concurrent inputs of a far larger number of people. But that alone does not make an economy inflationary or deflationary or flat. Those are going to happen in any game. It's just a question of how tight the resolution of the graph is.

[Sky]

For all the millions in WoW, or the million of GW, or the 550k+ of FFXI, the average player will never see any more than 60 other people at most at any given time, and their entire server popular is measured in the low number of thousands.

Planetside again taps the nature of mmo. Sure, you might get a 60-person raid in a traditional pve mmo, but continents are always hitting the 300 person cap in PS, and the bulk of those people are in one or two hotspots, with high potential for everyone involved to get into a huge furball.

I anxiously await SOE's next mmofps, I'll probably block out a year of it up front like I did with Planetside. I put my money where my fun is. Wait, that sounded bad.

[Venkman]

I didn't mention PS only because chances are everyone in that firefight is the entire server population :)

But seriously, you're right. The dynamics of a large-scale fight like that are almost unique to the game. Shadowbane also applies here though, with scores of people laying siege to a town defended by scores of people, and a third showing up to mess with things. I miss those days, but the game just required too much of me.

[Akkori]

Is it really possible to have more than, say, 50 people in your line of sight and not hit .025 fps? Do we have the right technology for it? I dont know if its true, but SOE apprently tried to buy out Dark & Light. As much as I am not terribly fond of SOE, I am even less fond of the positibely glacial progress Farlan is making on DnL. At one point, I believed DnL would be the answer. 1 shard, with up to 10k simultaneous connections, and a hundred or more poeple in your line of sight. Supposedly. It would be cool though.

[Venkman]

Eve is a single shard as well.

I'm no tech, but I did see PS and SB in action enough to know they do work pretty well. It comes down, I imagine, to how much information is truly unique per character versus how much of it is boilerplate. SB was pretty low res, and there's a relative few templates in PS, so it's not as tricky to scale into higher populations as, say, EQ1.

That has always been a problem with a game like SWG too. Every single item is unique, most of them are visible, and as such, each must be treated as a discrete rendered object. That game absolutely chokes with dozens of people in one area, and it's not been since around launch that scores of people tried to gather in one area. The closest would be the Coronet Starport of old, where there'd be maybe a dozen or two people outside for commerce with a dozen or two in the Starport landing area waiting for a shuttle or online for a doc buff. People quickly learned to flatten key graphics settings before entering Coronet.

[Sky]

Is it really possible to have more than, say, 50 people in your line of sight and not hit .025 fps? Do we have the right technology for it?

Yes. I've seen MASSIVE battles in PS, and it's been playable. Remember, though, a lot of people will be sharing a single model, 3 people in a tank, etc.