50,000 concurrent in February ??

I don't think SL will get to 50K concurrent for at *LEAST* six months, and I am not too confident that it will reach that number in 2007 at all.

If it were simply a matter of hardware I would be more optimistic, but clearly design is a major hindrance at this point in time.

Yeah, there are a great many people not aware of that, and quite a few that are. And from both of those groups, a very very large percentage that couldn't care less about whether the technology is bleeding edge because from their point of view it simply doesn't work.

Its not the idea how glueless they are, its just normal oif the incorrect PR SL is known for.

There is always aloud mouth trying to do PR which they don`t have the first idea how good PR is done! SL is always know for very poor INWORLD PR. This is what i think is the problem.

Yes, that's a valid point, and I have to keep reminding myself that people without a technical background can get very easily frustrated and have expectations for performance which simply can't be met. And, as noted above, they also believe the PR, which I don't even notice anymore because I'm used to ignoring the vaporware promises of marketing reps.

"Cutting edge" might better be termed "bleeding edge" and believe me the blood loss at Enterprise levels makes SL look like roses. It's all too common for large corporations to drop hundreds of thousands of dollars on new software applications that are riddled with bugs and that bring their product services to their knees. I know, having been on the other end, where everyone in my company is frantically trying to fix a problem that was caused by a new application our programmers don't really understand yet. And by the time they do understand it, the whole damn thing has been "outdated" and the next version is on the way.

The entire computer industry is a scam if you believe that you can simply pull it off the shelf (or sign up online) and experience an error-free use of the product under all circumstances, on any computer. It just doesn't happen that way, ever, unless you're dealing with a very simple, limited application.

Just this last week, I had to rip a third-party vendor component out of a development web site because the "enterprise-level version" was so bug-ridden that it was unsuable. That's the norm, unfortuantely. And no matter how well a product may work now, when all the other applications around it "upgrade" it can easily become a casualty too. As anyone who made the leap to Vista is probably finding out already. Lots of stuff that worked on Windows XP probably won't work on Vista, and you'd better hope that MIcrosoft's attempts to provide backdate mechanisms works.

I know that since I upgraded my computer, I've had far fewer problems with SL than other residents report. Far less lag, and only the most glaring of TP problems, which have been transitory for me. This dependence on client side confirgurations is a nightmare for programmers because there's simply no way you can anticipate all the possible bugs for every possible combination of preferences, graphics cards, computers, OS, monitors, RAM, etc etc.

This heavy reliance on high-end personal computers is a problem, but even that has not stalled SL's growth. Which futher convinces me that it's worth staying in the game.

If I point this remark out. And I don`t mean this in a neg way.

Even high end users are having issues in high traffic areas. If the sim is sitting on a very very bad server there is no hope. But yes slowing the growth will take care of issues

Lets be realistic here. Nothing in SL is "Bleeding Edge", its not even "Cutting Edge". I'll give you innovative and maybe even "Pushing the Envelope" but nothing that LL wants to do here is beyond the scope of readily available hardware and software. Where the problems come from is that from very early in the development process some poor decisions were made. Poor decisions in design, development processes, scalability, version management, QA procedures and customer and public relations to name just a few. If it were merely a matter of hardware and software it would be realatively easy to fix but the problems are more fundemental.

LL needs to take a good hard look at their operation from the ground up. They are a 24 hour a day, seven day a week 365 day a year operation but they don't act it. They are a business service provider but they don't act in that manner either. They are still in the "Grand Experiment" mindset, and while SL has moved out of that phase, the corperate culture of LL has not. Until LL makes some real and substantial changes in the way they do business they are doomed to fail. They need to think like what they have become, a business and not like a bunch of hackers having fun playing with technology.

There are many businesses out there that require 5 nines uptime, 99.999% availablity of their mission critical services and they achieve it. Until LL starts to think of SL being a mission critical service and treating it as such we will be stuck with the sort of thing we've been experieencing lately. LL says SL is not a game but is a world, an environment where we can do and be whatever we want. Until they truly treat it as more than "just a game" we will see a downward spiral.

LL wake up and take the leap to the next level, you are doomed otherwise.

Well said Ravanne........much more concise and to the point than me

But I want to add one more thought..........."behind the 8 ball". Every step of the way. A shame when you think about it. Reactive.........not proactive. And no successful business will ever survive as long as that is the case............none will survive. Cold, harsh fact of life in the competitive world of technology.

Yep! I was also pleasantly surprised to see Robin's reply on the subject (Hi Jesse!!!)

I've been in SL long enough to know I can do a few things that will help make my SL a bit smoother, in rough times.

I make an effort to keep my inventory as trim as possible. No we don't have inventory limitations, but I've always been able to tell a difference when I trim mine back within reason.

When TPs become unstable, I cut back on the amount and prim count of my attachments. And when they are very bad, I make sure all my current fav attachments are in a primary folder for easy access & just do a Detach All, before TP then reattach after I arrive. Heck, half the time when its that bad, we arrive looking unadorned anyway

I've been quite impressed with LL's efforts of late. To log in and be able to work (albeit at 700m) with over 20k online, is still utter amazement for me. My faith has had major swings, I've done my fair share of grumbling.. ok, & whining. But I have truly been quite pleased with the great progress being made since the media has taken BIG notice of SL.

The growth spurt of the past 6mos to a year has been somewhat of a big surprise, I think. And rest assured more are coming! People have started asking me, "Isn't that Second Life you have been talking about all this time?"

Hum. Too many people treat SL like a game they can play anytime, with no excuse for slowdown, lag, bugs etc. Surely you must take a wider look, and think about the sheer scale of what Linden Labs is doing before complaining about your personal problems. I take it those kind of people would rather LL close Second Life for good, seeing as they can't give some of their customers a perfect experience. Jesus, don't be so selfish.

I have full confidence in Linden Labs, but there are only so many hours in a day. Things can't be fixed the very second they go wrong, and as long as something is being done it will be fixed in the future. Despite the problems you are having now, you will be thanful for your extra experience when SL becomes a much bigger part of life than it is today. This is a project that will stretch on for many more years to come (who knows, maybe even Decades...) and things WILL fix. I seriously doubt Linden Labs would let this world come to ruin, even as it's momentum starts to gather on such a large scale.

I think a little example is in order: taken on the subject of RECYCLING...
(Here, pollution = lag, population = residents, waste = assets [inventory, mainly])

Now, everyone bar the youngest of infants knows of the importance of recycling. The governments know, the schoolchildren know. Hell, even African Tribesmen know. However, what percentage of these people actually do anything to solve it? The governments do their bit, but it is the general population that really has the power to change the waste problems. If they throw it away, it stays on Earth and continues to be waste. If they recycle it, it still stays but negates the need for a new one of said item to be made. Therefore, it reduces the waste output. If everyone recycled, say, plastic bottles, there would be no more on the landfills. Thus, the waste amount is culled (although not neccessarily reduced).

The same applies to Second Life, although on a less urgent basis. There are residents I know with 30,000+ items in their inventory, and how many of these do they actually use/need?

"Uh, about 50 regularily, and I need only about 500 of them."
- Anonymous Resident

"Why not get rid of the ones you don't need then?"
- Rob Arten

"Can't be bothered really."
- Anonymous Resident

Now, imagine there were about 1,000 - 5,000 residents like this (of which I am sure there are). Do you realise that getting rid of those 29,000+ items each (Purging, that is. Not just Deleting)would result in a noticable drop in global lag? There are people that complain about lagging asset servers and database load, but how many of these actually do anything about it? I understand that some vendors need to archive their products, but of the 6,000 million+ items (resulting in an average 3,000 items each for the 2 million residents, I'll guess) how many are needed? While inactive accounts may need to be kept for name issues, why not completely purge their inventories?

I've cut down my own inventory from around 20,000 items to about 2,000 in recent days, and surely other esidents could do the same. If not for themselves, at least for the other residents.

Simply deleting everything called "Object" would be a big help.

I work on the principle that if it wasn't important enough for me to give a name to before putting it in my inventory, it's not something I need to keep, or a component part of something I had to take a copy of to duplicate it.

Broccoli

Where are all these people???? The SIM limit is 50, so does SL have like some ghost sims somewhere? 26,000 users = 520 SIMS at maximum usage .

The SIM numbers and the database, are the top of the range limitations. How many indexes can a database support? Each user has an avaerage of 10,000 items in their inventory <coughs> and even more.... We can make new folders and rename them dynamicaly, fill them with all sorts of items, then transfer them in a few seconds to someone anywhere in SL.

What is the slowest thing in SL?.... Textures.... ranging from 1x1 to 1024x1024 in size, even with compression they cause bandwidth loss and FRPS to drop. I have seen an increase in the amount of warnings issued by the client, stating that the mask had been received but the aux source was missing, this goes on for several seconds and even minutes, while the database sorts this out and the problem 'goes away' on top of the load places upon it by downloading 300 different textures everytime you turn a corner. Its a pity we don't have the facility to reduce the level of textures from within the client to suit our systems, other than draw distance and bumpmapping texturing.

Its frustrating to turn off certain 'features' in the client rendering menu, only to have them default to all on, when I relog. While I'm on my soapbox, why did they manage to 'fix' things like IM box fixed position, only to have them default to the bottom of the screen eachtime you relog now... and the inventory openning on a folder you closed 3 days previous, and guestures no longer in alphabetical order (they used to be).. am I asking too much, like a perfect world?? <takes meds> just as well its only a game LOL

Exactly -- many problems with SL are exacerbated by resident and land owner abuses. Bloated inventories, poorly coded scripts, slapdash application of textures, and HUD overload.

Unfortunately, there aren't any mechanisms to force the abusers to stop from bringing a sim to a grinding halt and penalizing the more considerate residents in a sim. But Linden Labs has enough on its plate without having to play sim cop, and it's not really their place to do so when SL is viewed as a platform rather than a formalized game.

Island management has dealt with this problem by creating an SL version of Homeowners Associations, and mainland could use the equivalent of zoning to do the same.

Perhaps a future social development for mainland society is to develop some mechanism for regulating destructive practices. To use your analogy, developing an SL Green movement, if you will. It might be voluntary, but there are ways to create incentives by raising public awareness and making those who flout community standards pay for their anti-social behavior with lost business.

Here's my take on it.


The grid was planned and created several years ago with certain expectations put in place. We expect X% of growth over X time period. At that time, broadband saturation and technical immaturity (CPU horsepower, limited memory, bandwidth bottlenecks) were the prevailing limitations. It took what? 2 or 3 years to go from Alpha to Beta to production?

Fast forward to the decision to open up access. This was a business decision, and I'd be very surprised if any of the "grid monkeys" supported it behind closed doors.

So the floodgates opened. For the large part, they've been able to throw more Sim servers at the problem. The problem hasn't really been the Sims. The problem is the assets that the glut of new residents brought with them. As of the town hall a few weeks ago, there were over 300,000,000 asset items. This includes every notecard ever created, every texture ever uploaded, every script version ever saved. Nothing has been deleted. The asset server turned out not to be something where you could just throw more machines at the problem.

Back when the community was small, they got away with UDP and homebrew messaging systems. As the world got bigger, those systems started to break. Essentially now, they are rewriting the entire grid server structure from scratch.

Yes, there are things they know now that they didn't know the first time around. But it is still going to take time. You can only work so fast. Humans max out at a certain level of productivity. Unlike your avatars, humans do burn out. Humans do need food. Humans do need sleep. And yes, humans do make logistical mistakes they wish they didn't make that don't crop up and bite them in the ass until several years later. But it takes time to deal with these things. They are working on it.

Yelling and bitching and moaning isn't going to help them get it done any faster. If anything, your voices are getting heard and they divert time away from long-term stability to bandaid your problem now.

Is the population growing too fast simply to catch up? At current growth, that, unfortunately, is a possibility. But should that end up being the case, Phil and the VC investors will surely step back and assess the situation. The VCs may see this as a losing venture, and leave. But I'm inclined to believe, there is a much larger line of potential investors just waiting to get a crack at this.

Part of the problem also appears to be a perception gap. So many angry posters on the forum seem to believe that these problems can be fixed within a week or two, or even a month or two, and that if the fix doesn't appear in that amount of time then it means LL isn't working on it or doesn't care.

Many also seem to be under the impression that every employee of LL is equally adept at every single aspect of high-level programming so that even the appearance of re-styled blog brings cries of "Why are you re-designing the blog when I can't TP!!!"

I'm not sure just how Linden can counter that attitude, since I've seen for myself the countless times someone tries to explain that the person who redesigns the blog has no connection with the programmer who is working on database issues and that many LL employees probably couldn't tell a SQL request from a request for a Margarita. But the screaming continues because there's alway someone who didn't read that response, or doesn't even believe it.

There's a limit to how much coding progress you can make by throwing bodies at the problem. After a point, more programmers just create too large a project team to manage efficiently. God knows we've tried to explain that concept to clients who want us to meet an impossible schedule and insist we simply add more people to make it happen tomorrow. As if you could simply add a thousand programmers to a project and complete it in one week instead of six months.

Well, the mainland sim limit is 40. Private islands can go up to 100, I've been in a sim with 60 (and we crashed it!!).

The economic statistics page says there are 3516 islands. So SL could support 140,000 concurrent users. (and wouldn't that be fun!!)

But they would be elbow-to-elbow, and people probably wouldn't be able to get into the sim they wanted. Not to mention search, asset and TP problems which we see at 20,000 users.

I agree with the suggestion to delete/purge extra inventory items. But I'm sure there are people with more UNUSED inventory items than my TOTAL inventory.

Max

Time for every responsible content creator to stop selling copy clothing to keep the asset server happy .

Fine, I was joking, but a Kitty can dream .

You're right that it can't be fixed just like that, but it can be solved literally overnight in a way that's not all that different from fixing to most of us.
There's monumental difference between peak hours and off hours. I've never had any problem with tp'ing, assets, rezzing or whatever during low load. LL can do a whole lot to reduce the number of concurrent users if they wanted to.

You're right that part of the problem is a perception gap, but its not the one you think.

When LL redesigns the BLOG and pats themselves on the back publicly about it while the in world infrastructure is falling apart, they are broadcasting the perception that the one, the BLOG, is more important than the other, the infrastructure. This is a sign of a company whose culture and customer service mentality is woefully out of touch with its customer base. When 20,000 people log on and TPs die, searchs crash, no one can move and instead of addressing this issue with your customers you say, "Look at our pretty BLOG", you are asking to be screamed at. If time after time you declare an issue "Resolved!" and pat yourself on the back for a job well done and your customers can very readily see that nothing has been "Resolved!", prepare yourself to be eviserated in the forums. You can have the best product in the world and if you present the perception that you are uncaring, you will lose your customer base. Conversely, if you give the perception of caring and listening, of trying to resolve the problems, you can generate a customer loyalty that nothing can shake.

One TINY little remark about people using red text................LOSE IT!

For some cultures its very very rude to use red text. Why use it?

Thank you

SL seems to be a kind of addictive thing (as internet surfing for many). So , when you don't have it, the users become a bit upset.
It doesn't happen the same like going to buy a song or a book from an electronic store and it's down due to increased usage. If you can't buy it from Xamazon, then you go to f bookstore for example. The difference is that fortunately or unfortuantely, there is no other place you can find SL. Maybe there are variations out there, whatever you call them, of a 3D world, but SL is SL.

SO i suppose it's something we have to live with it. At the end , many will leave full of dissapoinment and try to find other places for having fun, or socializing , or making business.
That's life in RL also.

I am quite sceptical if SL will have a technical crash at the end , facing scalability and performance problems. Nobody knows what is coming.

What i see is that on November the max concurrent users I could see was 15k and every day there were about 20k new registrations (I know that alt's could be in there but..).
2 weeks ago SL hit 20K users concurrent. Yesterday and today it hit 26k. As a notice, today is much better than yesterday in terms of performance than yesterday (I am the one to feel it?).

I have seen also (after Christmas and New Years Eve break) an accelerated number of new registrations. The last 24 hours the number of new residents was for sure more than 50k. Yesterday also the number was quite increased. What are all these about?

One thing for sure, new people try SL. From them , most will leave, some will stay.
It's a pity that old users decide to leave , because SL has changed (or matured or whatever word you want). It's understandable but time is not going back and things can never be the same in SL as there were before.

The 50k concurrent are not much time ahead. WIth or without inventory, naked or not, with lag or not, the number will show up for sure. WHo will be here to celebrate? Nobody knows. For one thing I am quite sure. From what we discuss here, Linden is (I hope) and should be 1 or 2 steps ahead. They should be planning what are the next steps .At a server technology level (no cheap servers but at the edge of technology if it helps), move to new code or rewrite the whole thing (which can go to a big mess), find a bigger guy to help them out (with more resources) etc etc.

Whatever we say here, I believe at the end that most will come back , at least till the next competitor show up.
Just my pure opinion

Thats 40 avies on mainland

and 100 avies on islands

Even if you use four corner concept you can`t get more then 400 avies on (those 4 islands ) all together.

Very well put Brazil ... an addiction. An addiction for those already bitten by the appeal of SL and also, those attracted by the growing reputation, press and word of mouth. You can visibly see people reacting "OMG!!! The teleport is down again. Aaaagh!"
With or without inventory, naked or not ... does anyone else notice that most of the time, the chat and IM-ing carries on through the lag? That is stunning management. I take my hat off to LL!
Alright, alright, it might not be due to their choice of priority but heck, as long as I can talk to people around the world, I'm not going to give up and go elsewhere.

Hi

My comment was purely subjective, not literal. When I 'browse' the map, hovering over many sims, I see 75% empty, and for the remainder it varies to 1-40 ppls (15 being the average on the mainland). Its hard to imagine therefore that there actually that many people ingame. After all.. do we all believe in what we read?

There is not any official writing rule, but there are some unwritten unofficial behaving rules, that say that for example plain capslock -writing means yelling, and yelling is rude. Likewise red color is quite like bloody screaming, and well, bloody screaming is also kind of wierd behavior.

50,000 concurrent users is quite a lot. But at least then you have more people that you can meet and talk with. Also all traffic counts go higher, because more people are walking around everywhere. That way just getting more traffic does not mean that your place is getting more popular relative to other places. You need to get more new traffic than the other places to beat them. But is there some kind of upper limit for traffic. If the most famous clubs reach the peak traffic, then how do I know which one is the best? And can I enter them, if they are full?