GTA 5 breaks entertainment records but Online woes continue
theguardian.com
theguardian.com
The world of GTA V feels hollow when 1% of the buildings can be entered. When I drive to what I recognize as the "Commerce Casino" in the real world, I want to go inside!
It seems largely that they just need more servers. With the type of income (and expenses) that they've already incurred, setting up a few hundred more servers (a few clicks on EC2 if they have their Chef scripts done right) should do it largely right?
Can you think of other online systems that need to scale to massive numbers of users on day one of it's release? If you can, how many of those systems survived?
With my own game project, I used around 50,000 bots on AWS and attempted to model them as closely as possible to real players (using real player data taken from our Closed Beta). Looking at resource utilisation on the servers, it looked like we should be able to scale even further than those tests. By that I mean, the most overloaded systems were still only using about 35% of the system resources.
Come open beta day, we got concurrency of around 75,000 at peak, higher than the optimistic estimate we had made of 50,000. New problems developed. In fact, new problems developed at around 45,000 that somehow the bots didn't manage to find. The worst problem was a deadlock condition that could take out one of the database servers randomly kicking off ten's of thousands of users at a time. It's hard to describe the feeling of knowing that you are frustrating so many people around the world.
More servers would have done nothing to prevent these problems.
Anyway, I guess what I'm saying is that scaling all at once is really hard. Even if you take lots of precautions, random problems will still develop.
This isn't my line of work but it would seem simulating 50,000 bots in an environment is not quite the same as 50,000 humans from all over the place with their various connections speeds and/or issues.
Plus, you'd have to teach your bots to try to do incredibly stupid things that humans tend to do, willingly or otherwise.
Chat messages per second
Party creations per second
Average party size
Number of zone transitions per second
Average number of open zones per player
Character saves per second
PVP Match list requests per second
And so on, and so on.
Certainly bots are not going to be the same due to the issues you mentioned, but we do the best we can! We didn't simulate all kinds of connection issues a real player might have, but we did match the non-graceful connection termination metric.
The game is fantastic I'm very hooked!
If it worked for gmail, and if it was good enough for facebook, then it is good enough for GTA 5, and any other online game that hasn't yet scaled.
Don't fall prey to the thundering herd problem.
Build invitation fences.
That is totally different from not having something due to some psychological reason (which I don't know). Just like people hate it when a price goes from $0 to $1, but don't care as much when $10 goes to $15.