Fastest growing grids of 2015

OpenSim gained a record-high 10,013 active users in 2015 — more than double the increase of any previous previous year.

By comparison, last year, OpenSim’s public grids gained 2,189 active users last year, 3,652 in 2013, 2,652 in 2012, and 2,671 in 2011.

Annual active user growth. (Hypergrid Business data.)

Annual active user growth. (Hypergrid Business data.)

OpenSim also gained 80,702 registered users this year, another record high, and 16,493 standard region equivalents.

Fastest-growing grids

OSgrid regained its usual top place in growth after a break last year due to an extended outage.

Gains in active users, by grid:

OSgrid: 3,446
Great Canadian Grid: 1,265
DigiWorldz: 1,148
YrGrid: 647
Island Oasis: 464
Lost Paradise: 390
Logicamp: 370
Littlefield: 311
Virtual Brasil: 279
Exo-Life: 278

Increase in active users for ten fastest-growing grids. (Hypergrid Business data.)

Increase in active users for ten fastest-growing grids. (Hypergrid Business data.)

Gains in standard region equivalents, by grid:

OSgrid: 4,417
Kitely: 3,620
DigiWorldz: 1,900
SimValley: 1,798
AviWorlds: 1,093
Great Canadian Grid: 943
Lost Paradise: 821
Metropolis: 458
Virtual Highway: 271
InWorldz: 232

Increase in standard region equivalents for ten fastest-growing grids. (Hypergrid Business data.)

Increase in standard region equivalents for ten fastest-growing grids. (Hypergrid Business data.)

Related Posts

maria@hypergridbusiness.com'

Maria Korolov

Maria Korolov is editor and publisher of Hypergrid Business. She has been a journalist for more than twenty years and has worked for the Chicago Tribune, Reuters, and Computerworld and has reported from over a dozen countries, including Russia and China.

  • Frank Corsi

    I guess you do not count AtekGrid, since it was active last year. I did not see DigiWorldz was active last year.

    • What were Ateks stats then? This is only a top ten. And correct, Digiworldz didnt open till march 2015. So in actuality The Digiworldz numbers are for the last nine months and not the full year.

      • Frank Corsi

        In December 2014 AtekGrid was listed on the stats here, with 96 land regions, December 2015 having 1984 land regions. In fact if we use the comment that increased over last year, Digiworldz does not qualify, but for sure with almost 1900 more in one year AtekGrid should have been listed. Unless for some reason Maria choose to exclude some grids from her report.

        • Talla Adam

          Does AtekGrid actually publish grid stats anywhere? I haven’t seen them.

          • Frank Corsi

            LOL Now this is funny… On the front page of AtekGrid .com are stats since the day it started! I guess you did not look at the front page of the website.

          • Talla Adam

            Ok, thanks. I’m not sure what’s funny Frank but you’re right I don’t often look at grid web sites and my search scraper only looks at splash screens. Atek doesn’t publish stats on their splash page so there is the problem. I will have to do it manually. I see that the Magnuz Binder scraper does read stats from the Atek web site though but even there the active users data is absent so Atek ends up in the “excluded column” and denied inclusion in his analysis along with some 80 other grids for whatever reason. Anyway, the problem is a lot of grids don’t publish “actives” regardless of whether it is combined Diva stats or not so we will never have an actuate picture of traffic and all we can hope to do is draw a trend from both the unique login’s and the general movement of Hypergrid traffic that is published. And the trend has been upward which is good.

          • Magnuz Binder

            My criterium for full inclusion in my compilation is simply whether grids publish their online figures for concurrency or not. Still, since the compilation shows grid status in near real time, and probably is better updated regarding grid names and URLs than e.g. the Hypergrid Business listing, I realized the listing might be of interest for other purposes, so I follow and list other grids as well, but without full statistics inclusion. It’s still “under construction” though, so in no way any official, complete or even very reliable list.

        • gabegw11

          Every time I have visited Atek Grid it seems dead. I can’t find anyone and also leave messages inworld for staff but no one ever responds. I remember when it first opened I always saw people but lately can’t find anyone online. I hope things turn around and it gets more busier, be nice to have some fun events.

          • I haven’t heard a peep out of ATEK in months (advertising, stats, etc), honestly I thought it had shut down in the summer.

          • I had been to atek grid and once people was there but all dead when people suddenly left grid. I had trouble rezzing my avatar there and shops empty and offline IM don’t work for there either. I assume that they are outdated with opensim version which probably 0.8.0 but most grids are now up to 0.8.2-0.9.0.

          • gabegw11

            I guess that would explain why no one returns IMs or dropped NC. It sounds like maybe they don’t want business.

        • Actually, I think DigiWorldz does qualify, the only difference being that its stats for last year were 0. 😉

        • my oh my how the mighty has fallen 🙂

  • Lani Global

    Thanks, Maria! Your chart graphic presentations are always wonderfully done. The data you consistently research over the years is of great historic value and current interest to all of us.

  • Magnuz Binder

    Sorry if I burst the bubble, but I suspect a major part of the growth of “actives” is rather an increase in hypergrid activity, since at least Diva Wifi, and probably many other stats modules, count hypergridders as unique visitors.

    I am running stats on my Metropolis regions as if they were an independent grid, and there the hypergridders have increased from about 25-30% of “actives” at the beginning of 2015 to 55-60% at present. My regions may not be quite representative for the entire OpenSimulator metaverse, but since they would, as an own grid, rank as number 35 in terms of actives and 14 in terms of land area, compared to the Hypergrid Business compilation, I think they’re indicative at least.

    I am also compiling concurrency stats for my Metropolis regions and for a number of major grids, and they have not shown any significant increase during 2015, save for a temporary and expected “speed bump” because of OSgrid coming back online this spring.

    So, 2015 probably wasn’t the year when the usage of OpenSimulator took off, but perhaps the year when hypergridding took off, thanks to more stable hypergridding, more grids opening up to the hypergrid, OSgrid being down shaking up the usage pattern, and several groups arranging and encouraging hypergrid travel, which I think is good enough.

    On a side note, I helped out a grid owner with her stats recently. She wanted to add hypergridders to her actives stats “because Hypergrid Business recommends it”, and it turned out the query she already used was the same as the Diva Wifi uses, and therefore already included hypergridders. Still, it took a bit of effort to convince her about it, since “Hypergrid Business says hypergridders aren’t included in the stats”. It makes me wonder if there are any more grid owners who double count hypergridders, because they are under the misconception that they aren’t already included in the “basic” stats.

    • Samantha Atkins

      Are you in any way arguing that coming to a grid from another grid counts for less somehow than directly pointing your viewer at the grid? If so I don’t get why you think so. If anything it would be a positive sign of getting closer to a true VW/3D web like experience and the success of decentralization from a focus on a particular grid. Activity on a particular grid does rise from visitors from other grids participating in events and enjoying offerings. In any other business you would count this as a net win.

      • Magnuz Binder

        No, but I think (active) presence, i.e. concurrency, is a better measurement than “actives”. The reason is a user spending in total 10 minutes in OpenSimulator grids is spending 10 minutes, regardless if they do it in 1 grid or 5 grids. But, if a user hypergrids to 4 other grids, besides the grid where they log in, the user will count as in total 5 “actives”, instead of 1 “active” if the user had stayed in its login grid, while it still only contributes in total 10 minutes of presence. That means that increased hypergridding can inflate the “actives” count without increasing the presence and “real” activity. That’s why I prefer focusing on online/concurrency numbers in my own compilation, rather than “actives”.

        I’m happy to see hypergridding increase, linking the metaverse closer together, but I’m not happy seeing statistic compilations disregarding the inflation effect above, when focusing on “actives” count alone. This can give a false picture of OpenSimulator booming, when in reality it isn’t. This can lead both to people skipping needed efforts to engage more people in OpenSimulator “since it’s growing anyway”, and to investors and other interested parties being burned when they discover they’ve been fooled by skewed statistics, and becoming “bad will ambassadors” for OpenSimulator, warning others about it being a bluff. The latter happened both to Second Life, and to the once biggest OpenSimulator grid. I don’t want to see the same happen to OpenSimulator as a whole.

        • yep

          • Talla Adam

            Oh, very graphic but NO, He’s wrong!

          • He is absolutely right when he says it is concurrency that matters, and that the same person hopping around the grids inflates the active users reported overall.

          • Talla Adam

            No, he is wrong because he can not provide a true and accurate concurrency and even if he tries how can you prove it is true when people can inflate the numbers with alts. Don’t tell me no one does it!

            At least with Hypergrid logins, which some grid owners do actually show separately as well, we get an idea of TRAFFIC, which has a lot more meaning that ACTIVE or CONCURRENCY in an infinitely fakeble system.

          • Concurrency is not easy to fake with alts since you would need multiple logged in sessions in that case, and most systems are only capable of having 2 client sessions going at the time unless they use a textual client.

          • Talla Adam

            Are you saying few would attempt to use a textual client like Radegast? Well, I know of people using Radegast in Second Life to maintain alts and bots. It’s not uncommon at all and given the nature of Opensim running on multiple private servers it is just too easy to manage a whole bunch of alts that way in Opensim. But, I concede, we do have NPC’s – unlike SL – that can show presence without the need for alts so much but, seriously, do you really think no one uses a text client?

          • I know people are using text clients of course 🙂
            But I seriously doubt they are used to inflate concurrency as nobody seem to care much about it, when IMO it should be the focus (real logged in personas.)

          • Talla Adam

            Aren’t you now just suiting your argument by doubting if anyone cares about concurrency enough to inflate the numbers?

            My argument is different. I want to know about Hypergrid activity which, to my mind, is real and not a bunch of Alts being logged one after another on rotation to inflate the 30 days actives. I think it would be time consuming and difficult to to run loads of alts Hypergriding about showing presence and interacting so I believe Hypergrid stats more than unique logins. That is what I call Traffic and why I think Magnuz has the wrong idea on concurrency.

          • TBH I don’t care much about someone inflating their activity data be it direct or hypergrid logins.

            What makes SecondLife attractive and what also will make or break OpenSim is if you actually meet someone there. There is of course a (limited) scope for the lonely builder but most people are social creatures that want to interact with someone else. – Which is why concurrency matters.

          • Talla Adam

            Well, you said it, “people are social creatures and want to interact with someone else” which, in my view, is why Hypergrid activity matters most. Concurrency is a blind metric and the easy one to fake. Apart from being open source Hypergrid is Opensim’s strongest feature. All the closed grids are like little tribes of inward-looking fiefdoms with every reason to fake stats although I wont say all do. Hypergrid enabled grids are outward-looking social groups and their grid hopping is difficult to fake.

          • I totally agree that Hypergrid is OpemSim’s strongest feature – without it I would hardly bother.

            I still strongly feel (Hypergrid) concurrency is the most important metric, as it says how likely you are to find others to interact with on the overall Hypergrid.

          • Talla Adam

            Perhaps we have a meeting of minds, Geir which would be a worthy outcome to this debate. And I agree, that the Hypergrid statistic, if it were separated from the active login’s would be more meaningful in so far as it would indicate both growth in Hypergrid activity and tell us where the most people are gathering in Opensim grids. That is why I call it “Traffic” and look for the trend in the statistics.

        • Talla Adam

          So one person log’s in many times in one grid per month and only shows up as one login for the whole 30 days while a person travelling is showing they are moving about and interacting with the Hypergrid community. I’m sure one person logging in once in 30 days is doing just as much as the person who travels but there is no way in Opensim to show that currently unless one attempts to manipulate the figures which I assume you are doing with your own compilation, and, given you can’t get all the numbers and have to exclude many grids, you can never get a true concurrency anyway so all that you say about giving true stats is nonsense. No one can because the Metaverse is not under one roof with common standards of reporting statistics. Least ways, grid owners only publish what they want to.

          Most people who spend their time in the open grids are using Hypergrid and there is nothing about their movements that can be called false. They go to different grids and do stuff. What is false about that? Kitely Market totally depends on Hypergrid activity for trade and many grid owners go to great lengths to attract Hypergrid traffic, putting on events and tours. This, to my mind, shows a great deal of activity – TRAFFIC – and it is totally right in my view to count their movements as logins to each grid they visit.

          The Opensim Metaverse is not huge anyway. We all know that but I find it amazing that some people can’t do enough to make it seam like we have no growth at all – not even an upward trend! Usually, this is what commercial grid owners attempt to do because they really don’t see any money to be made out of a successful Hypergrid they have no control over. I think Maria does a good enough job here and all I’m interested in personally is the traffic and the trends. Concurrency is a false metric invented by Second Life that has never had true figures in all it’s existence.

          • Da Hayward

            Well said Talla, I couldnt agree with you more.

          • Tom Frost

            At the risk of getting flamed at: I do like Magnuz’ stats a lot. They give me a sense of what’s happening right now (yes, a sense of, not actual, hard numbers, but there aren’t any stats in the hypergrid that give actual, hard numbers – it’s all guesswork and hand-waving). That there’s several tens of thousands actives on the hypergrid-enabled grids doesn’t mean I get to meet tens of thousands of people online.

            In reality, there aren’t tens of thousands of people on the grid daily – and that’s what counts for me personally: how likely is it that i will meet someone new? Or meet someone outside of the circle of hard-core avi’s i meet at every event?

            But the bottom line is, neither of the HGB nor the MB statistics are ‘right’ or ‘wrong’ (i wouldn’t know where to find Talla’s overviews btw, anyone can share a link?). They are just there, for anyone to draw their own conclusions. Don’t turn them into a religion please, there’s enough of war already 🙂

            Btw, @disqus_VYiOZ1Awb8:disqus : you are aware that the opensimworld.com stats are very very easy to fake, right? The numbers are provided by scripts running on the regions themselves, free to modify by the region owner. If I wanted I could easily modify it so that it would show, say, anywhere between 6 to 12 people in my region when in fact no-one is there, but switch back to accurate reporting once someone is in the region (to thwart attempts to check the numbers by going there).

            Not that people need that, there’s plenty of chatbots placed just to get a region on the number 1 spot at opensimworld.com. But even discounting that, on average, at peak times, there’s around 80 avatars online in all the regions that have an opensimworld.com beacon. That really is insignificant compared to the tens of thousands of actives reported.

            Magnuz’ stats give a less bleak picture, of about 300-400 online at peak times in hypergrid only grids (personally, I don’t really care about non-hg-enabled regions, and I think for example yrgrid should not be in the hg-enabled category).

          • ” (personally, I don’t really care about non-hg-enabled regions, and I think for example yrgrid should not be in the hg-enabled category).”

            YES YES YES ( but nothing against yrgrid, I thought it was open to HG…I don’t keep up well anymore ) I have been saying for some time now that “hypergrid” business should report only on hypergrid enabled places.

            And if the others cry foul ( or not ) another page done just for the closed non-HG grids.

          • Tom Frost

            Last I heard, yrgrid is only open for hg from some arbitrary list of grids.

          • ah, ok.

          • Da Hayward

            the statistics here personally i think are great, Maria goes to some effort to present them to us and really does deserve a pat on the back. You are so right about not turning them into a religion as well.the statistics may be accurrate or they may not but i feel they do give us an indication on what is happening…So again well done Maria keep up the good work. I for one do appreciate it

          • Tom Frost

            I never said i don’t appreciate Maria’s stats. I like all stats 🙂

          • Talla Adam

            I have not argued for a moment that any of the stats are accurate and because I think they are so inaccurate I think it is very difficult, if not impossible, to use an algorithm to interpret the numbers as Magnuz is trying to do. If he thinks that breaking the Hypergrid visitor numbers down gives a more accurate statistic, fine. But I see more value in the hard numbers Maria publishes because, as I find myself repeating often here, it takes one person to log in and physically move their avatar from one grid to another to affect the Hypergrid stats. That strikes me a real activity or traffic and not a dumb alt that log’s in to game the “logged in now” stats and dose nothing but sit there. Hard numbers are just what grids publish. They are not invented by Maria. What Magnuz is doing is taking the “logged in now” numbers and attempting to interpret them to present what he claims to be an accurate concurrency number. If anything is “guesswork and hand waving” I think that is. What Maria’s stats show is the raw data and it can show growth and decline without dwelling on the accuracy of the numbers. The raw data can show an upward or downward trend which is useful. Hypergid stats can show movements and where people gather most which is why I would like to see it separated from the unique login stats.

          • Magnuz Binder

            I guess some people need to hang on to the simple and convenient “truths” Hypergrid Business serves, that the OpenSimulator metaverse is doing great, that all trends are upwards, so nothing needs to be done to improve things. I’ve been through this once before, with the same kind of people wanting to burn me at the stake for showing statistics and warning about that the then biggest OpenSimulator grid I then called home wasn’t doing all that great. Today that grid has fallen from place 1 in actives and concurrency to place 15 in actives (-94%) and place 11 in concurrency (-98%).

            Actives is a convenient but rather blunt measurement tool for activity, regardless of what believes one holds about it’s relevance. Decreases, but not increases, lags 30 days, and it smooths per definition over 30 days, while concurrency can be instantaneous and in the shape I compile it, running 7 day average to even out the major daily and weekly cycles, it lags by 3.5 days and smooths over 7 days. Just as an example, from the approximately 200 grids I now compile full statistics for:

            2015-11-28 00:00 UTC, total actives were 28916 and total concurrency 376.

            2015-12-28 00:00 UTC, total actives were 29413 and total concurrency 329.

            So, in 30 days, actives increased by 2% and concurrency dropped by 12%. But don’t panick believing concurrency is going down the drain. It’s holiday season, and some strange creatures prefer spending time with real-life kin rather than virtual kin then. (At least I hope that’s the reason!) Concurrency picked it up, actives didn’t.

            There are a few more things with my compilation, even for non-die-hard fans of statistics in general and concurrency in particular. An “OK” column shows which grids were online last check, and checks are made every 30 minutes, and an “O.now” (Online now) column shows how many people were online at that time. It could give a nice hint about where to go if you’re looking for company. It’s still much a work in progress though, so things probably will change.

          • Talla Adam

            Magnuz, really! What are you talking about? Opensim has been massively improved in the past few years and the metrics published by HB has had no influence on that whatsoever. It all happened because the Dev’s wanted to make it happen in response to the growing belief that Opensim is becoming a viable platform for the 3D web. Even you, yourself, delighted in posting regular topics to G+ Opensim Virtual about the recent merging of Avination code to the Opensim main branch. Now you are saying Opensim development is stunted because HB’s monthly stats are false.

            Honestly, I wont say what you are doing with your compilation is a waste of time. I do one myself but I don’t get too carried away on interpretation and I think you do in order to prove your point you want to make. I could just as easily poll grids every 30 minutes as you do to grab the “logged in now” data and analyse a sample of the stats to see what they say, as you are doing. But I am happy to let the raw data speak for itself as Maria does here. If I want to know what is happening on a particular grid I look at Opensimworld who have beacons on many regions (not just grids) sending currently logged in data back every 30 minutes. And I know that is reading region data where people actually are, not grid data that only shows a grid has some people logged in – which could so easily be partially faked.

            In my view Hypergrid traffic can not be so easily faked and it could show where people who use HG are gathering most which is much more useful to know than blind concurrency stats. Opensim is too small to be trying to get accuracy out of the stats on sample data which appears to be what you are trying to do. And I don’t believe for one minute any one takes the HB stats, your’s or mine seriously enough to be swayed either way into thinking Opensim is, or isn’t, worth investing in.

            My experience is that people interested in Opensim for investment will take time to join grids and use Hypergrid to move about so they can do their homework. And then, maybe they will experiment a little with SoaS perhaps and, eventually, make a decision whether Opensim could meet their needs, be it in education, business, government or the wider community of Hypergrid worlds. In that respect Opensim is so very different from Second Life. With SL, Linden Lab are selling access to a dwindling market of people and charging over the odds for it so concurrency is important to their business model. Some of the none-hypergrid grids try to emulate that business model too so concurrency is important to them as well. But for the rest of us concurrency is not so important because it doesn’t tell us where the real activity is. It just gives us an idea and maybe a trend.

            Personally, I think it would be helpful if the Hypergrid stats were separated from the Login stats so we can see where people gather most and leave it to Opensimworld and that kind of system to zero in for a closer view. Stats in themselves can never prove anything for certain but they can show a trend and that is what makes HB stats useful.

  • I think this poll accurately reflects the current market. If some stats are missing it is because those grid owners have either failed to provide them, have a reputation for manipulating stats, or have a system in place that prevents programs from easily picking up statistics. ~Corran Journal

  • Thanks Maria 🙂

  • Shy Robbiani

    Interesting information, thank you for collecting and sharing, Maria!
    But even I don’t agree in all aspects I also like Magus’ thoughts and the many responses to it very much 🙂 Personally I don’t care much, as finally, for me as a user, it’s not the numbers that count.

  • yippeee ( belated ) for the Hyperverse!!