iheartnyc Posted February 24, 2012 Share Posted February 24, 2012 How can a set of relative population data ever be "complete"? The servers would have had to have close for that to be the case. How can a data set measure relative population change ever get more accurate with time? Unless you can rewind and rerun time itself, it cannot! All that will show is what any relative population change is from Jan 24 to March 24. Just like all it shows from 15 Dec to now is any relative population change from then to now. The data doesn't become any more or less accurate or reliable between the 24th of Jan 24. It won't magically make more people be playing (or less for that matter), it will simply show the population change between those two date points, as any set of data between two date points will also do. You are trying to argue that if a currencies value is at £10 on 15/12/11 and you have daily data for it's value every single day till the 24/02/12.... that actually you have no idea what value of that currency was between the 15/12/11 and 24/02/11.... until you have daily values up until the 24/03/12. It utterly makes no sense. I think you are misunderstanding the OP, and the OP doesn't know enough about statistics to properly articulate what he really means. Statistics is NOT about accuracy or reliability and most certainly not about presentation of facts. It never has been and never will be. It's about interpretation. Collection of data, and the analysis of that data. As an M&A specialist with experience on some of the largest and most high-profile deals (both buyside and sellside), I've had the joy and pleasure of picking apart all forms of "statistical data". And these population "stats" are meaningless. The underlying data set is scewed beyond any measure of reliability. I would guess that the confidence interval of these graphs (don't know what that means? look it up) would be close to 33%. In the M&A world, we would tell the people who produced such a thing to come back after removing all data from December 20-January 3rd and then talk to us. The fact that they left those numbers in, shows that they don't care about skewing or something we call "confounding variables" or "Type 1" or "Type 2" errors. The fact that we are here and trying to prove or dispprove hypothesis based on charts for which we don't even have the data set or can't control for various factors is just foolish. My opinion: These charts are meaningless and should be dismissed. Link to comment Share on other sites More sharing options...
Azumio Posted February 24, 2012 Share Posted February 24, 2012 They need to do a couple of things: If they don't wish to merge (Understandable, it will provide negative publicity) then they need to patch in a Cross-Server tool for all group content (Flashpoints, Ops, Heroics, etc.) I can think of a couple of ways they can do this to keep both the players like myself (Who could careless about Cross-Server ruining 'community' simply because most of the servers don't have enough people for an actual 'community') and players who simply loath it. An easy way, is in options have a selectable setting, so it's: *Local Cluster Only (Own Server) *Cross-Galaxy Cluster (Cross-Server) There you go. Let the playerbase decided what they wish to use. This atleast takes the pressure off merging abit, and provides people on low-population servers an avenue to keep their characters, while still getting to enjoy group content whenever they wish it (Provided it's a success). Or, and this is the one I'm most fond of, but it's also the very, very least likely to happen. Have one HUGE server (For Logins, maybe), and then smaller servers branching off from that. So, in this case, picture a Tree. The trunk would be the huge server, the branches would be the smaller "branching" off servers. These small servers will be spilt up into planets. So: Taris will have 10 'channels' (Servers) that have a capacity of lets say 100. You go to Taris, and are automatically placed in a channel. If channels 1-7 are full, you get sorted into channel 8. Communication between channels is like planetary communication right now, IE: You see chat from everyone on the planet regardless of their channel. Upon joining a group, you have the option of queueing to join their channel (If it's full) or they have the option of joining you on your channel. You can queue to join a full channel, jump from a full channel to a non-full channel seamlessly. You get a pop up "You are not queued for Taris (Channel 1)" and when you get accepted another Popup that says "You will move to Taris (Channel 1) in 10 seconds.." Getting into combat will make it stop (This stops PvPers from abusing it to avoid a death). Stipulations will be put in to make farming rare enemies impossible (Something simple like, if you killed xxx enemy on Channel 1, you cannot kill it again for another xxx time (His respawn time). This makes sure everyone gets a legit chance. Add in a Global LFG chat channel (That is capable of either being auto-sortable by level/class/role or specific Flashpoint/Operation/Heroic), and you are good to go. Both options rid the entire complaint about dead-servers. I'm more fond of Option 2 because it provides the game with tons of players, which is the way I envisioned a game like SWTOR that is about space, and galaxy etc. Link to comment Share on other sites More sharing options...
iheartnyc Posted February 24, 2012 Share Posted February 24, 2012 (edited) They have effects on the population, yes, but that is not the same thing as an outlier. I would like you to explain to me how you believe they are. Charts are based on data sets. Data sets can have outliers. There are rules for eliminating outliers so as to not render a chart practically useless. These charts have failed to do so. They are, but launch and holidays also effect subscription numbers (or are you saying they don't? ), which would make them "outliers" by your and Drew's argument. You can't have your cake and eat it, it is one or the other. Subscription numbers =/= population numbers. Not even close. EA won't even comment on population numbers as these are subjective and would not stand up to the rigor or scrutiny of Sarbanes-Oxley or SEC regulations. Subscription numbers = pulled straight from accounting numbers. Population numbers = affected by subjective factors. Misleading has nothing to do with data reliability. Yes it does. If we are not provided with the data set and the charts are skewed, our confidence interval (a statistical term) goes down. Charts are rated based on their confidence interval, among other hard factors. Confidence interval is a measure of reliability. When a chart is presented without the CI, or the data set, it can be misleading to people who don't know how to interpret the data. You might draw a wrong conclusion from the data, but the data remains as reliable and accurate is it ever was or ever will be. No. Changing the data set can alter the confidence interval (measure of reliability). You should experiment on SPSS or whatever the statisticians use these days. Again no, you cannot make any sort of exact measurement (as I have repeatedly said), but so long as all servers are measured by the same rod (as they seem to be, and it would make no sense for them not to be), it will clearly show overall relative playing population trends. Not if these charts are based on averages, which they are. Averages based on skewed data sets = fail. As I said if overall playing trends are down long term then almost inevitable so are subs, just as if overall playing trends are up then almost inevitably so are subs. That's a hypothesis which you would have to prove using data sets, for which you and I don't have access to. Just so we're clear about confidence interval - it basically says you could take the same assumptions you used to create a chart for a time frame, go to another time frame and collect data again, and you would get the same results/trend x% of the time. A good CI is 95%. So that would mean, you would need to be able to take these charts, replace the data with data from say March-April, and 95% of the time, you would get approximately the same results and/or trend. If that is NOT the case, then the chart is unreliable or the assumptions are flawed or the data set contains too many outliers or the people were lazy and didn't account for variables. Basically, the chart makers aren't sophisticated and you shouldn't make any interpretations based on the chart. Thank you. Edited February 24, 2012 by iheartnyc Link to comment Share on other sites More sharing options...
Lethality Posted February 24, 2012 Share Posted February 24, 2012 ah - you don't have access to their sub numbers. You just believe it to be a case. So when you said it was a fact - you actually meant it's your opinion. Now I understand. No, it's a fact alright. Blizzard have never once in their entire history reported a drop in subs for WoW until last year... it has always been an increase. This is to their investors, and the SEC, by law. So yes... fact. Link to comment Share on other sites More sharing options...
Lethality Posted February 24, 2012 Share Posted February 24, 2012 The graphs - for what they are - are entirely accurate. They take a snap-shot of the server status(light/medium/heavy) and then post a chart. This is entirely correct. The method of measurement, however inaccurate it is, is done the same way every time, so trends within it are relevant. That's all there is to it. Link to comment Share on other sites More sharing options...
Gleneagle Posted February 24, 2012 Share Posted February 24, 2012 ... Why would they continually raise and lower server population caps? Raising them to a level they are sure won't crash, yes. Perhaps even sequentially so, but continually doing so why and how would they do that? And to make any massive difference they've have to be doubling caps and doubling again...... so within a few raises you'd have server caps in the 1,000,000's! Server pops change every minute: if you are paying for your virtual servers to autoscale to demand then they will change size frequently. Just because you are unfamiliar with advances in the technology does not mean the technology does not exist. Just because you fail to grasp the implications does not mean they aren't there. For myself I am done with you, as you are not being intellectually honest, nor are you bothering to read the material others have presented. Your pretense at interest is fraudulent. You are not interested in fact, but effect. Link to comment Share on other sites More sharing options...
GalacticKegger Posted February 24, 2012 Share Posted February 24, 2012 (edited) Are you trying to say that Bioware is shifting the server cap status for each server individually constantly? They are shifting the server cap thresholds which determines server status, so yes. Edited February 24, 2012 by GalacticKegger Link to comment Share on other sites More sharing options...
Sullydog Posted February 24, 2012 Share Posted February 24, 2012 BLASPHEMY! Take your science and math and get out! This is AMERICA and we rely on faith and gut feeling when interpreting things! Facts and numbers only exist to manipulate and confuse! I SHOOT FROM THE GUT and INSTINCT! Link to comment Share on other sites More sharing options...
Kaelshi Posted February 24, 2012 Author Share Posted February 24, 2012 They are shifting the server cap thresholds which determines server status, so yes. I had never thought about this at all - of course I know jack about server technology To clarify for Gortezu though they are not manually shifting the caps - the auto in auto-scale would mean automated I'm assuming So it's an implementation of technology not a Bioware employee sitting at a console monitoring every server and going, oh pops rising BOOM BOOM CAP RISE!!! Link to comment Share on other sites More sharing options...
GalacticKegger Posted February 24, 2012 Share Posted February 24, 2012 (edited) I had never thought about this at all - of course I know jack about server technology To clarify for Gortezu though they are not manually shifting the caps - the auto in auto-scale would mean automated I'm assuming So it's an implementation of technology not a Bioware employee sitting at a console monitoring every server and going, oh pops rising BOOM BOOM CAP RISE!!!Ah yes. If the caps are autoscaling all over the place then "High"could represent 2,000 yesterday, 1,000 today and 3,000 tomorrow. Auto-scaling masks the true mathmatical values behind it. Conversely static thresholds would be the accurate method for displaying actual population numbers (like fill lines on a measuring cup) and ultimately trends; but that appears to not be what BioWare wants. And those server status thresholds can be manually adjusted (there isn't much on the back end that can't be.) Which means a server can be made to look as populated or as unpopulated as they want by adjusting those thresholds. A nice and subliminal server population relocation tool when implemented properly. Edited February 24, 2012 by GalacticKegger Link to comment Share on other sites More sharing options...
DarthTHC Posted February 24, 2012 Share Posted February 24, 2012 You seem like a smart person. Smart enough to know that using logic on the Internet is guaranteed to fail. Good post though. Link to comment Share on other sites More sharing options...
GalacticKegger Posted February 24, 2012 Share Posted February 24, 2012 (edited) You seem like a smart person. Smart enough to know that using logic on the Internet is guaranteed to fail. Good post though.If that was meant for me, then thank you. Fail matters only if winning matters - and it doesn't on the Internet where logic in discussion is concerned as you pointed out. At least to me it doesn't. Won't prevent some like myself from at least trying to help the community though, which isn't a bad thing. Edited February 24, 2012 by GalacticKegger Link to comment Share on other sites More sharing options...
DalrisThane Posted February 24, 2012 Share Posted February 24, 2012 (edited) Deleted. No point. Everyone is TLDR these days. +See Sig. Edited March 31, 2013 by DalrisThane Link to comment Share on other sites More sharing options...
Drewser Posted February 24, 2012 Share Posted February 24, 2012 (edited) No he's trying to say measuring someone's age gets more reliable with time (whether he realises it or not). It doesn't. No, I am not. The measurement of age is entirely different than something like measuring a population. I will give you another example and if you don't get this, you are being willfully ignorant. A baseball team is getting ready to make a long term order to stock and lock in prices for their concession stands. Which way is better to get an average number of attendees? a) The first weekend of the season when they have a 3 game series with the Yankees. Weather was beautiful and the stands were packed attendance was 47k per game. b) Wait until the end of the first month of the season when they had a variety of opponents, weather and it was not opening weekend? Average attendance 38k per game. Even better is if they have years of history for attendance numbers, they will be able to more accurately predict what they will need. But this whole analogy fails as it applies to SWTOR measurements because the stadium would use the same metric for measuring the attendees, whereas with SWTOR, the server statuses are not a consistent. It would work better if we knew the numbers for light, standard, etc were static and the same for each server. Edited February 24, 2012 by Drewser Link to comment Share on other sites More sharing options...
iheartnyc Posted February 24, 2012 Share Posted February 24, 2012 The ownage on this thread makes me laff. Link to comment Share on other sites More sharing options...
GalacticKegger Posted February 24, 2012 Share Posted February 24, 2012 (edited) The ownage on this thread makes me laff.Glad you are enjoying yourself, especially since ownage exists primarily in one's own mind. Edited February 24, 2012 by GalacticKegger Link to comment Share on other sites More sharing options...
wOOOOt Posted February 24, 2012 Share Posted February 24, 2012 I dont need to see other stats on server population. All I need, is to look at my own guild. 250 active players to begin with. Around 200 now. Some dropout was expected. But what concern me MUCH more is the ingame activity. Out of around 200 members, we only see 15-40 online at the same time. Our guild is 10 years old, with 400+ members in various mmo's. And SWTOR average member activity is about half of what we have seen in any other mmo in the past 10 years. Reason. Lack of endgame content. Once you raid operations, all other things to do is a waste of time. MMO gamers basically don't want to reroll new alts all the time. A few is fun, but you want to keep op playing with your Main char. And if that char hits the wall-of-nothing-to-do too quickly, then the player looses the interest in the game, becaue why play another Alt up to have no future with? w Link to comment Share on other sites More sharing options...
GalacticKegger Posted February 24, 2012 Share Posted February 24, 2012 (edited) I dont need to see other stats on server population. All I need, is to look at my own guild. 250 active players to begin with. Around 200 now. Some dropout was expected. But what concern me MUCH more is the ingame activity. Out of around 200 members, we only see 15-40 online at the same time. Our guild is 10 years old, with 400+ members in various mmo's. And SWTOR average member activity is about half of what we have seen in any other mmo in the past 10 years. Reason. Lack of endgame content. Once you raid operations, all other things to do is a waste of time. MMO gamers basically don't want to reroll new alts all the time. A few is fun, but you want to keep op playing with your Main char. And if that char hits the wall-of-nothing-to-do too quickly, then the player looses the interest in the game, becaue why play another Alt up to have no future with? wI remember WoW having no end game for over a year until AQ was patched in. At least TOR has some semblance of one right now. Hopefully Bioware won't take until the game's first major expansion to have a solid end game like Blizz did. Edited February 24, 2012 by GalacticKegger Link to comment Share on other sites More sharing options...
OldBenSmokin Posted February 24, 2012 Share Posted February 24, 2012 (edited) I remember WoW having no end game for over a year until AQ was patched in. At least TOR has some semblance of one right now. Your memory sucks then... Edited February 24, 2012 by OldBenSmokin Link to comment Share on other sites More sharing options...
GalacticKegger Posted February 24, 2012 Share Posted February 24, 2012 (edited) Your memory sucks then...BRS and Ony weren't end game, and that's all there was at launch. ZG came out in the first couple patches and everything up to 1.9 was just dungeons. BGs didn't happen for 7 months. WoW had squat until the Patch 1.8.3 launcher fixed all the crashes. My memory (and record keeping) is fine thank you. Edited February 24, 2012 by GalacticKegger Link to comment Share on other sites More sharing options...
OldBenSmokin Posted February 24, 2012 Share Posted February 24, 2012 BRS and Ony weren't end game, and that's all there was at launch. ZG came out in the first couple patches and everything up to 1.9 was just dungeons. BGs didn't happen for 7 months. WoW had squat until the Patch 1.8.3 launcher fixed all the crashes. My memory (and record keeping) is fine thank you. Perhaps we have differing opinions of "endgame". For me, it's what can I do that's progressive and fun after max level. Perhaps I'm wrong in my definition. In WoW, we had alot. No one was clearing MC a month and half after release. Link to comment Share on other sites More sharing options...
GalacticKegger Posted February 24, 2012 Share Posted February 24, 2012 (edited) Perhaps we have differing opinions of "endgame". For me, it's what can I do that's progressive and fun after max level. Perhaps I'm wrong in my definition. In WoW, we had alot. No one was clearing MC a month and half after release.Some guilds had, but only a few firsts. No different than EV where some have cleared it, while most haven't. Had there been a WoW before WoW and people had todays tools and cheats to power level, then MC probably would have been on farm by this time after WoW's release. Excluding world bosses ... EV & KP operations plus 8 Hardmode Flashpoints versus a MC raid plus a BRS dungeon is what we're comparing for combat-based non-PvP things to do at level cap correct? Edited February 24, 2012 by GalacticKegger Link to comment Share on other sites More sharing options...
OldBenSmokin Posted February 24, 2012 Share Posted February 24, 2012 (edited) Excluding world bosses ... EV & KP operations plus 8 Hardmode Flashpoints versus a MC raid plus a BRS dungeon is what we're comparing for combat-based non-PvP things to do at level cap correct? Yea, I guess I am. It looks great on paper, your viewpoint. The sheer # of FPs outweigh vanilla lvl 60 dungons. But the questlines.. (Prison Break) that led to the those dungons.. that led to MC.. that led to finally downing Rag took much more time, coordination, and effort than it takes to clear all content here. I get it, some people don't think that's a bad thing.... I do, but alot don't. Edited February 24, 2012 by OldBenSmokin Link to comment Share on other sites More sharing options...
Icebaron Posted February 24, 2012 Share Posted February 24, 2012 I'll tell you this much if the population was growing we would see numbers from Bioware or EA. Link to comment Share on other sites More sharing options...
Failstar Posted February 24, 2012 Share Posted February 24, 2012 In WoW, we had alot. No one was clearing MC a month and half after release. Well yeah, because it took people a ton of time to reach level cap in WoW, then you had to get 40 people together with the right roles and such to even have a shot. That was more of a design "issue"..You had to literally slam your head against the endgame dungeons over and over to get the gear you needed to even take a shot at MC, then had to have 40 people coordinated properly so you wouldn't die. If you've noticed, once new content is coming out for WoW NOW the normal modes are absolutely blasted in a night by some of the most hardcore guilds. *Shrug* I couldn't really care either way if population is rising or dropping as long as I'm having fun. It doesn't really matter if you prove a point either way..I mean..it's the internet right? Play if you have fun log off if you don't? Anyone? No? Back to leveling my new BH then. Link to comment Share on other sites More sharing options...
Recommended Posts