Pages: [1] 2 :: one page |
|
Author |
Thread Statistics | Show CCP posts - 1 post(s) |
Taedrin
Gallente Kushan Industrial
|
Posted - 2011.08.26 02:16:00 -
[1]
Each chart compares the peak logged in users of one day of a year with the same day on the previous year. The leap day in 2008 was dropped. Data ends at March 15, 2006
Data was pulled from Chribba's eve-offline TQ status website (hope he doesn't get upset about me using this information - he had it sitting in the page source as XML, ripe for the taking. I don't see why he would complain since this is CCP's data anyways)
SO FOR EXAMPLE:
If on Jan 1, 2010 the peak number of users was 100 and on Jan 1, 2009 the peak number of users was 25, then the 2010-2009 graph shows a growth of 75 for the day of Jan 1.
Data "starts" at August 22nd of the latest year, and works backwards to August 23rd of the previous year. There were some missing data points in Chribba's data. I replaced these missing days with zeros, in case the reason why he couldn't get the data was because the servers were down for an extended period of time. It is also possible that Chribba's page was down on those days while TQ was up.
IMPORTANT: Because the data starts at the latest date in the dataset, and works backwards a year, you need to realize: 1) The left side of the graph represents data closest to the present day. 2) The right side of the graph represents data farthest into the past. 3) Positive slope on a trendline indicates that growth is falling (but does NOT necessarily mean that EVE is losing players!) 4) Negative slope on a trendline indicates that growth is rising (but does NOT necessarily mean that EVE is gaining players!) 5) Since Chribba's data stops part way through 2006, the 2007 minus 2006 graph only looks at the dates of March 15th through August 22nd (since there 6) Now that I look closer at my data, I missed an anomaly in the 2006 data (it's missing April 27, 2006), so the earliest month on the 2007 minus 2006 graph is subtracting the day on 2007 to the day before on 2006.
Without further ado, I present my data: 2011 minus 2010 2010 minus 2009 2009 minus 2008 2008 minus 2007 2007 minus 2006 ----------
Originally by: Dr Fighter "how do you know when youve had a repro accident"
Theres modules missing and morphite in your mineral pile.
|
Taedrin
Gallente Kushan Industrial
|
Posted - 2011.08.26 02:17:00 -
[2]
Edited by: Taedrin on 26/08/2011 02:27:08 You will notice that growth is strong from 2006 to 2009, with 2009 being the best year in terms of growth.
However, growth tapers off in 2010, and has started to just barely dip into the negatives about two months ago.
This shows that Incarna has been BY FAR the least effective expansion - actually causing players to leave/accounts to unsub. On the other hand, Apocrypha was the best patch, being the most effective at bringing in new players/accounts. ----------
Originally by: Dr Fighter "how do you know when youve had a repro accident"
Theres modules missing and morphite in your mineral pile.
|
Nova Fox
Gallente Novafox Shipyards
|
Posted - 2011.08.26 02:46:00 -
[3]
Any plans to line average graph the three articles of data? or possibly a line chart continious from 2006 to 2011?
I mean I'm not trying to dog on your work just that scattered dot data all in one color even though you described that its just a reference of difference between recent and previous year. It be helpful to present data in a more appeasable manner. Its how people present some really good information and wind up looking bad because formatting wasnt 'customer' frinedly though im not encouraging the vice versa of making any bad data look better than what it actually is.
Pre-order your Sisters of ≡v≡ Exploration ship today, Updated 3APR11
|
Taedrin
Gallente Kushan Industrial
|
Posted - 2011.08.26 03:00:00 -
[4]
Originally by: Nova Fox Any plans to line average graph the three articles of data? or possibly a line chart continious from 2006 to 2011?
I mean I'm not trying to dog on your work just that scattered dot data all in one color even though you described that its just a reference of difference between recent and previous year. It be helpful to present data in a more appeasable manner. Its how people present some really good information and wind up looking bad because formatting wasnt 'customer' frinedly though im not encouraging the vice versa of making any bad data look better than what it actually is.
You mean like line up the 2011 minus 2010 graph with the 2010 minus 2009 graph, so that we can see growth over the period of, say, two years rather than just over a single year? I'll start working on that right now. It'll be interesting to see how that changes the trendline. ----------
Originally by: Dr Fighter "how do you know when youve had a repro accident"
Theres modules missing and morphite in your mineral pile.
|
Pax Infinitas
|
Posted - 2011.08.26 03:24:00 -
[5]
Originally by: Nova Fox Any plans to line average graph the three articles of data? or possibly a line chart continious from 2006 to 2011?
I mean I'm not trying to dog on your work just that scattered dot data all in one color even though you described that its just a reference of difference between recent and previous year. It be helpful to present data in a more appeasable manner. Its how people present some really good information and wind up looking bad because formatting wasnt 'customer' frinedly though im not encouraging the vice versa of making any bad data look better than what it actually is.
60 Day Moving Average, 2006 to Present 60 Day Moving Average Comparison, April - July 2011 and 2010
Not the OP, but that's what I've got.
Quick and dirty Excel charts, despite the fact I stated in the other thread that I cba to upload them. Time is flowing in the right direction, too. :)
The MA is calculated centered on the mean so the trend doesn't lag by 30 days.
Ignore the year portion of the second chart. The red line is the time period from 2010 and the blue line is the time period from 2011.
The interesting part is that the big drop this year couldn't have been caused by Incarna, as Incarna hadn't been released yet. I think it's people getting bored. Whether it translates into actual lost accounts or is simply people saying, "meh, cba to log in," I couldn't tell you. Originally I'd thought the crater looking bit in 2009 was Unholy Rage, but that didn't happen until later in the year.
I seem to remember there was a ridiculous period either associated with the server move/upgrade or a horribly botched patch that resulted in several days of downtime? Average is getting skewed by several days of numbers in the hundreds instead of tens of thousands between 5/15 and 5/25/2009.
|
Taedrin
Gallente Kushan Industrial
|
Posted - 2011.08.26 03:51:00 -
[6]
Originally by: Pax Infinitas
Originally by: Nova Fox Any plans to line average graph the three articles of data? or possibly a line chart continious from 2006 to 2011?
I mean I'm not trying to dog on your work just that scattered dot data all in one color even though you described that its just a reference of difference between recent and previous year. It be helpful to present data in a more appeasable manner. Its how people present some really good information and wind up looking bad because formatting wasnt 'customer' frinedly though im not encouraging the vice versa of making any bad data look better than what it actually is.
60 Day Moving Average, 2006 to Present 60 Day Moving Average Comparison, April - July 2011 and 2010
Not the OP, but that's what I've got.
Quick and dirty Excel charts, despite the fact I stated in the other thread that I cba to upload them. Time is flowing in the right direction, too. :)
The MA is calculated centered on the mean so the trend doesn't lag by 30 days.
Ignore the year portion of the second chart. The red line is the time period from 2010 and the blue line is the time period from 2011.
The interesting part is that the big drop this year couldn't have been caused by Incarna, as Incarna hadn't been released yet. I think it's people getting bored. Whether it translates into actual lost accounts or is simply people saying, "meh, cba to log in," I couldn't tell you. Originally I'd thought the crater looking bit in 2009 was Unholy Rage, but that didn't happen until later in the year.
I seem to remember there was a ridiculous period either associated with the server move/upgrade or a horribly botched patch that resulted in several days of downtime? Average is getting skewed by several days of numbers in the hundreds instead of tens of thousands between 5/15 and 5/25/2009.
Yeah, your numbers match up with mine. That period of several days downtime due to the server move was the reason why I substituted each "anomaly" with a day with zero players logged in, instead of just dropping it.
Here are my graphs, which show are the same thing as my OP graphs - they compare a day with the same day on the previous year, except that they go farther. I added a 30 day moving average
2011 through 2009 2011 through 2006
These new graphs make an interesting suggestion, with the 30 day moving average added in:
EVE Online always experiences growth slowdown at about this time of year. However, this year is the first time that this growth actually went well into the negative range. However, there was a rebound after this dip, suggesting that many players resubscribed/started playing again after the CCP/CSM emergency meeting.
----------
Originally by: Dr Fighter "how do you know when youve had a repro accident"
Theres modules missing and morphite in your mineral pile.
|
Cave Lord
|
Posted - 2011.08.26 04:00:00 -
[7]
Nice. You actually did research and work. I wonder if CCP will take this into account since they said they'd only pay attention to what people do.
Here's for hopeful wishing: *crosses fingers*
-Cave
|
Kogh Ayon
|
Posted - 2011.08.26 04:23:00 -
[8]
They took me minutes to understand. It would be a good idea to draw them in linear graphic and put diffrerent graphic inot one file that people can easily compare.
|
Sullen Skoung
|
Posted - 2011.08.26 04:34:00 -
[9]
Originally by: Taedrin actually causing players to leave/accounts to unsub.
OR it shows the downward trend as they ban the **** out of bots as they say they are -------- CCP knows better than the players whats good for their game. SOE knew what was best for SWG too. Better than all those players that left too. |
AnzacPaul
the united Negative Ten.
|
Posted - 2011.08.26 10:30:00 -
[10]
Originally by: Sullen Skoung
Originally by: Taedrin actually causing players to leave/accounts to unsub.
OR it shows the downward trend as they ban the **** out of bots as they say they are
I wonder if CCP discovered say a large number like 20,000 confirmed bots, they would ban them all..... Thats alot of lost revenue...... |
|
Karl Planck
Labyrinth Obtaining Chaotic Kangaroos
|
Posted - 2011.08.26 10:53:00 -
[11]
and of course neither of your guys' data includes anything about the "slow burn" of bot accounts. Until you have data on when those accounts were activated and turned off you data conclusions don't show much at all.
|
Caldari Acolyte
|
Posted - 2011.08.26 11:06:00 -
[12]
It will be interesting to look at those numbers when Diablo 3 & Guild Wars 2 come out close to December, can you say fail cascade?
|
Lilith Ishanoya
Lai Dai Shipments
|
Posted - 2011.08.26 11:24:00 -
[13]
Thank you for this thread, facts and averages make it more clear than the data shown by Chribba.
__ Friendliness goes a long way |
Forum Worrier
|
Posted - 2011.08.26 11:27:00 -
[14]
Originally by: Caldari Acolyte can you say fail cascade?
I could but I wont. Anyone who uses that term...
|
Just Another Killmail
Minmatar Republic Military School
|
Posted - 2011.08.26 13:47:00 -
[15]
Originally by: Cave Lord Nice. You actually did research and work. I wonder if CCP will take this into account since they said they'd only pay attention to what people do.
Here's for hopeful wishing: *crosses fingers*
-Cave
Here here, Hilmar is going to eat those words, when your players are averaging data FOR YOU instead of playing the game, in the hopes to wake them the eff up and reverse course from a catastrophic trainwreck MAYBE they will put the gdamn beer down for 5 minutes and realize their clique of friends and advisors who just want to beat off to girl avatars walking around in the privacy of their own CQ could just open up youtube and quit trashing EVE.
But, if the dev attitudes are any indicator from the last fan meet, I believe the quote was..
"We spent 4 years working on WiS, you are going to fking walk in stations whether you like it or not" -CCP_Douchebag -
If you had powdered water, what would you add? |
|
CCP Loktofeit
|
Posted - 2011.08.26 14:29:00 -
[16]
Originally by: AnzacPaul
Originally by: Sullen Skoung
Originally by: Taedrin actually causing players to leave/accounts to unsub.
OR it shows the downward trend as they ban the **** out of bots as they say they are
I wonder if CCP discovered say a large number like 20,000 confirmed bots, they would ban them all..... Thats alot of lost revenue......
But, in some cases, a lot of server resources freed up for the actual players.
|
|
Ciar Meara
Amarr Virtus Vindice
|
Posted - 2011.08.26 14:56:00 -
[17]
Originally by: CCP Loktofeit
But, in some cases, a lot of server resources freed up for the actual players.
Therefore saving on the energybill for the servers/fans/cooling. - Hilmar getur ekki tala= vi= ¦ig n·na, hann er a= fara f japanska tfskuverslun.
|
Othran
Brutor Tribe
|
Posted - 2011.08.26 15:23:00 -
[18]
Originally by: Ciar Meara
Originally by: CCP Loktofeit
But, in some cases, a lot of server resources freed up for the actual players.
Therefore saving on the energybill for the servers/fans/cooling.
Not really. Its not a colo'd 1U box you know
CCP will be paying for a specified GUARANTEED amperage. If they don't use it all, they don't get a refund - the backup capacity still has to be there in case they use it tomorrow.
|
Sullen Skoung
|
Posted - 2011.08.26 15:25:00 -
[19]
This
CURRENT data please -------- CCP knows better than the players whats good for their game. SOE knew what was best for SWG too. Better than all those players that left too. This is your NGE/CU moment CCP. |
Sun Liping
|
Posted - 2011.08.26 15:26:00 -
[20]
You might want to compare weekdays instead of actual dates. Weekend playtimes certainly are different than working days. 1st Jan 2009 = Thursday, 1st Jan 2010 = Friday, 1st Jan 2011 = Saturday
|
|
Vincent Athena
|
Posted - 2011.08.26 15:35:00 -
[21]
CCP has said in the past that it does not consider botting to be a revenue source.
Originally by: CCP Sreegs The first question is one I haven't FULLY researched but the general consensus is that removing bots actually increases CCP's income because we have more happy customers. The idea that we would want people to bot for $15 is a false one, though I could see why some people would leap to that conclusion.
On the concurrent user plots: Eve has a 7 day cycle of logged in users, so maybe 7 day averages would be best. Also to reduce scatter from down time, remove days with low user counts (relative to surrounding days) and re-average.
I know how to do all that, what I do not know is how to get the data.....
|
Runnin Through
|
Posted - 2011.08.26 15:36:00 -
[22]
Originally by: Ciar Meara
Originally by: CCP Loktofeit
But, in some cases, a lot of server resources freed up for the actual players.
Therefore saving on the energybill for the servers/fans/cooling.
Therefore saving on buying infrastructure becouse they can't handle the server load. Thats a lot more isk than electricity bills.
|
Zaila
Wyo Wolves Confederate Empire Alliance
|
Posted - 2011.08.26 16:10:00 -
[23]
If you could, I wouldn't mind a copy of your data.
I'd like to run a few statistics on it as well as a few other operations (remove outliers). |
Ranger 1
Amarr Imperial Shipment
|
Posted - 2011.08.26 16:41:00 -
[24]
Edited by: Ranger 1 on 26/08/2011 16:42:00
Quote: On the other hand, Apocrypha was the best patch, being the most effective at bringing in new players/accounts.
Indeed. Apocrypha was also by far the most bug filled expansion that CCP has put out to date... but it did have lots of new shiny.
For reference, it is this fact that prompted Hilmars oft quoted (out of context) statement that the data shows that new content is far more effective at bringing in new subscribers than polished content.
This eloquently proves his point, whether we like that fact or not.
===== The world will not end in 2012, however there will be a serious nerf to Planetary Interaction. |
Razin
The xDEATHx Squadron Legion of xXDEATHXx
|
Posted - 2011.08.26 17:01:00 -
[25]
Edited by: Razin on 26/08/2011 17:09:13
Originally by: Ranger 1 Edited by: Ranger 1 on 26/08/2011 16:42:00
Quote: On the other hand, Apocrypha was the best patch, being the most effective at bringing in new players/accounts.
Indeed. Apocrypha was also by far the most bug filled expansion that CCP has put out to date... but it did have lots of new shiny.
For reference, it is this fact that prompted Hilmars oft quoted (out of context) statement that the data shows that new content is far more effective at bringing in new subscribers than polished content.
This eloquently proves his point, whether we like that fact or not.
No it doesn't. This actually proves only one thing: good content trumps bugs. The Incarna 'expansion' proves the same thing. ... Return the Old Hangar Back... for Immersion.
|
Destination SkillQueue
Are We There Yet
|
Posted - 2011.08.26 17:39:00 -
[26]
Originally by: Ranger 1 Edited by: Ranger 1 on 26/08/2011 16:42:00
Quote: On the other hand, Apocrypha was the best patch, being the most effective at bringing in new players/accounts.
Indeed. Apocrypha was also by far the most bug filled expansion that CCP has put out to date... but it did have lots of new shiny.
For reference, it is this fact that prompted Hilmars oft quoted (out of context) statement that the data shows that new content is far more effective at bringing in new subscribers than polished content.
This eloquently proves his point, whether we like that fact or not.
It proves nothing of the sorts. If he had said, that bugged high quality new content is more effective than new lackluster polished content or polishing old features without adding to their gameplay in bringing in new subscribers, I would have agreed with him wholeheartedly. But he didn't say that.
Apocrypha had great content additions compared to any other expansion, so it proves that adding more good and interesting content is what the players, new and old, want. Apocrypha content additions shouldn't be made the benchmark or compared against all "polished" content additions though, since it provided exceptional new content additions in quantity and in quality.
New content only sells if it is interesting for the players. Wormholes where great as an idea and had good implementation. The problem is that most new EVE content work great as ideas, but suffer greatly from lackluster implementation. Too often CCP settles for dull and mediocre and leaves things at that while giving promises of getting back to make it what it should have been from the start. PI being one of the worst examples of a great idea reduced to something uninteresting and boring compared to the initial plans.
I think the danger in thinking in terms of new vs polish is, that you easily fail to take quality into account in both the new and the polish, and can therefore easily come to wrong conclusions about why something was succesful in increasing subs. Boring and low quality new features are bad as far as gaining more subs is concerned, since few want to play them or they offer too little gameplay and can't keep players interested. Polish that doesn't aim to improve the gameplay and quality of the features being reworked isn't any better. It might make the feature less painful for the players already doing it or new ones trying it, but it won't make it any more appealing for the people who weren't already interested in it.
Point being that if anything apocrypha proved that interesting ideas that are implemented well are what the players want. I don't think it matters at all if it is entirely new content or reworked old content as long as it delivers the great gameplay. Polish is a secondary concern only in that initial bugs are relatively unimportant, if the content is interesting and offers great gameplay options, since then bugs become minor annoynces and people can live with them until CCP gets around to squashing them.
|
Ranger 1
Amarr Imperial Shipment
|
Posted - 2011.08.26 19:40:00 -
[27]
Originally by: Destination SkillQueue
Originally by: Ranger 1 Edited by: Ranger 1 on 26/08/2011 16:42:00
Quote: On the other hand, Apocrypha was the best patch, being the most effective at bringing in new players/accounts.
Indeed. Apocrypha was also by far the most bug filled expansion that CCP has put out to date... but it did have lots of new shiny.
For reference, it is this fact that prompted Hilmars oft quoted (out of context) statement that the data shows that new content is far more effective at bringing in new subscribers than polished content.
This eloquently proves his point, whether we like that fact or not.
It proves nothing of the sorts. If he had said, that bugged high quality new content is more effective than new lackluster polished content or polishing old features without adding to their gameplay in bringing in new subscribers, I would have agreed with him wholeheartedly. But he didn't say that.
Apocrypha had great content additions compared to any other expansion, so it proves that adding more good and interesting content is what the players, new and old, want. Apocrypha content additions shouldn't be made the benchmark or compared against all "polished" content additions though, since it provided exceptional new content additions in quantity and in quality.
New content only sells if it is interesting for the players. Wormholes where great as an idea and had good implementation. The problem is that most new EVE content work great as ideas, but suffer greatly from lackluster implementation. Too often CCP settles for dull and mediocre and leaves things at that while giving promises of getting back to make it what it should have been from the start. PI being one of the worst examples of a great idea reduced to something uninteresting and boring compared to the initial plans.
I think the danger in thinking in terms of new vs polish is, that you easily fail to take quality into account in both the new and the polish, and can therefore easily come to wrong conclusions about why something was succesful in increasing subs. Boring and low quality new features are bad as far as gaining more subs is concerned, since few want to play them or they offer too little gameplay and can't keep players interested. Polish that doesn't aim to improve the gameplay and quality of the features being reworked isn't any better. It might make the feature less painful for the players already doing it or new ones trying it, but it won't make it any more appealing for the people who weren't already interested in it.
Point being that if anything apocrypha proved that interesting ideas that are implemented well are what the players want. I don't think it matters at all if it is entirely new content or reworked old content as long as it delivers the great gameplay. Polish is a secondary concern only in that initial bugs are relatively unimportant, if the content is interesting and offers great gameplay options, since then bugs become minor annoynces and people can live with them until CCP gets around to squashing them.
Polished content it this context means relatively bug free.
Apocrypha had interesting content yes, but it was completely riddled with bugs... far, far more than we have with Incarna.
So yes, you just reinforced my point. Nifty new, interesting crap that is hopelessly bugged and breaks half the game is far more popular than smaller, less challenging (but bug free) content.
Hilmar spelled out exactly what he meant in an live dev blog that somehow everyone seems to conveniently forget about.
He also stated in that same blog that they would avoid going that route again as it was a nightmare at their end.
More quality, less shiny overdose per expansion. I can live with that. ===== The world will not end in 2012, however there will be a serious nerf to Planetary Interaction. |
baltec1
|
Posted - 2011.08.26 19:45:00 -
[28]
This clearly shows we are growing.
|
Taedrin
Gallente Kushan Industrial
|
Posted - 2011.08.26 20:26:00 -
[29]
Originally by: Ranger 1
Polished content it this context means relatively bug free.
Apocrypha had interesting content yes, but it was completely riddled with bugs... far, far more than we have with Incarna.
So yes, you just reinforced my point. Nifty new, interesting crap that is hopelessly bugged and breaks half the game is far more popular than smaller, less challenging (but bug free) content.
Hilmar spelled out exactly what he meant in an live dev blog that somehow everyone seems to conveniently forget about.
He also stated in that same blog that they would avoid going that route again as it was a nightmare at their end.
More quality, less shiny overdose per expansion. I can live with that.
This is correct - I should have been more careful with my wording. Apocrypha WAS one of the buggiest expansions EVE has ever had, and most of the Dominion performance problems were partially due to the bugs introduced with Apocrypha. HOWEVER, despite these problems, Apocrypha remains as the most popular expansion to date for EVE Online - causing the strongest growth we have ever experienced even after Unholy Rage.
ANYWAYS, the points I want to make with this thread are:
1) CCP saw slowly accelerating growth from 2006 through 2009. I would also argue that this acceleration of growth started since EVE's birth 2) 2009 was the best year for CCP in terms of growth 3) Since 2009 growth has been slowing down (but still growing) 4) CCP actually experienced negative growth these past few months, but found respite with a moderate rebound back into positive growth again.
ON THE TOPIC OF BOTS:
CCP has reason to hate RMTers due to several reasons: - Every penny that RMTers earn is a penny that CCP could have earned through the sale of PLEX and/or Aurum. - RMTers pay for EVE with stolen credit cards, which get charged back to CCP, ultimately COSTING CCP money since not only do they lose subscription fees that were paid to them, but they also have to pay chargeback fees. - RMTers are more expensive to provide services to, as they are active 23.5/7 and consume more server CPU cycles with their activities than normal players do.
Now, bots in *general* are not necessarily the center of CCP's rancor. A bot run by your average player pays for his subscription just like a legitimate player, so CCP DOES actually make money from them.
The only arguments against bots in general from CCP's perspective are: - bots damage the economy by making various occupations in EVE unprofitable or unrewarding (e.g. high sec mining) - bots cause unbalanced gameplay, with players who utilize bots being able to earn income while they sleep/go to school/go to work. This leads to other players having less fun and thus causes CCP to lose players. - bots are prohibited by the EULA, and thus players using bots are challenging CCP's authority.
If you want to claim that CCP doesn't ban bots in general, you need to argue that CCP's revenue from bots is GREATER than revenue lost from legitimate players who leave the game due to bots. ----------
Originally by: Dr Fighter "how do you know when youve had a repro accident"
Theres modules missing and morphite in your mineral pile.
|
Bodrul
Caldari Caldari Provisions
|
Posted - 2011.08.26 20:29:00 -
[30]
Originally by: CCP Loktofeit
Originally by: AnzacPaul
Originally by: Sullen Skoung
Originally by: Taedrin actually causing players to leave/accounts to unsub.
OR it shows the downward trend as they ban the **** out of bots as they say they are
I wonder if CCP discovered say a large number like 20,000 confirmed bots, they would ban them all..... Thats alot of lost revenue......
But, in some cases, a lot of server resources freed up for the actual players.
yeap, got to keep servers clear for those Bots :)
............ "you dont need a reason or a three piece suit to argue the truth" Game Reviews |
|
|
|
|
Pages: [1] 2 :: one page |
First page | Previous page | Next page | Last page |