I'll send this over to Gray. I'm also interested in the status of our travails with the Packeteer.

Do you have an 8800 card?

And routers are cheap, cheap, cheap these days. We've found a handful of routers that just drop packets, and you'll find life is better with other games if you upgrade to something that just works.

Oh, and by the way, it happened to us. On hardcore, enterprise class hardware, we found a port that was droppping outbound UPD packets, but not inbound. Can you guess how long it took for us to track that sucker down?

Clasewitz, are you still having problems? The servers are up and fine.

Using pirates_local.ini to force TCP will work, but it's suboptimal. Make sure you go through the steps in the Knowledge Base, including bypassing the router. You'll have better performance if you use UDP.

Hi, sorry for the trouble, I'll see what can be done.

Here's the status:

1) If you're crashing to desktop, please file a support ticket with us.
2) Slow frame rates - We'll have that corrected in either the next patch or the one right after that. For now, there's another thread (I'll find it tomorrow) where I explain how to download an INI file to speed up your frame rate. Blast these split threads!
3) Nvidia - They promised me on Friday it was their highest priority, and they'd have something for me on Monday.

We have multiple GMs working around the clock and they're looking at every ticket that comes in for the rollbacks.

Well, sorry, in the next patch, he's been muted forever. You wind up standing near him too long to have repeating comments.

The load issues seem to be nailed at this point. We don't see any issues with scaling further. The population wasn't a problem, but exposed another issue.

To elaborate, I have operations tools that show me cluster status, etc. It pulls this data from various databases. Yesterday it took 30 seconds to do stuff. Today, after the fix, it's instantaneous. That delay from yesterday was causing all of the issues we've experienced over the last day or so.

As far as reimbursements, we can tell exactly when you gained or lost items/missions/xp/db so you can claim you had a first rate, but we can prove you wrong, or prove that you lost it last week in a port battle. I believe our general policy is no reimbursements, but we do look at each case, so file the support ticket.

As far as WHY the rollbacks in particular happened, it's caused by the Database cache layer crashing or exiting abnormally without writing out all of the data. Again, this was caused by the issue we fixed this morning.

Obviously we don't like having these issues any more than you do. I'm the Ops guy, so I can't really address what circumstances may warrant a reimbursement, etc. I can just tell you WHY something happened and what we fixed