Servers cant keep up?
- Thread starter BluStar
- Start date
I imagine what the game would be like if the networking worked smoothly. Players can all connect properly, wing up, no rubber banding, orbital cruise drops are instant instead of waiting an age and stopping the progression toward the planet etc etc. Just imagine...wouldn't that be cool. Unfortunately the basic framework for the network is never going to be able to accommodate this. Maybe in 10-20 years when global communications are quantum based ! 
I imagine what the game would be like if the networking worked smoothly. Players can all connect properly, wing up, no rubber banding, orbital cruise drops are instant instead of waiting an age and stopping the progression toward the planet etc etc. Just imagine...wouldn't that be cool. Unfortunately the basic framework for the network is never going to be able to accommodate this. Maybe in 10-20 years when global communications are quantum based !
I think we're more than 20 years from large scale commercial quantum networking, but I agreewith everything else.
been experiencing server issue since a month roughly. Thought it was because i'm away for work and relying on the hotel internet connection. I have changed hotel 2 days ago and having a nice 50Mbit here. Still getting server issues.
It's getting ridiculous.... we have a weekly wing-up (3 of us) all live within 15 mile radius... have tried using both default and port-mapped (non-upnp) options...
Lately ... without fail
1. One or two will get kicked after a few minutes with connection error (matchmaking)
2. One or two will not see a ship in the instance the others are fighting
3. NPC Ships show up solo and then turn into wing of 3 or 5
4. NPC ships show as no-shields then show up 1/2 way through with shields
5. NPC will be 'stuck' in space
6. NPC will appear stuck in space but will actually be attacking one invisibly (similar to 2)
If FDev are coding more features in 2.3 (multi-crew) to have us play in groups more ... how can this work with the networking in the state it is.
Lately ... without fail
1. One or two will get kicked after a few minutes with connection error (matchmaking)
2. One or two will not see a ship in the instance the others are fighting
3. NPC Ships show up solo and then turn into wing of 3 or 5
4. NPC ships show as no-shields then show up 1/2 way through with shields
5. NPC will be 'stuck' in space
6. NPC will appear stuck in space but will actually be attacking one invisibly (similar to 2)
If FDev are coding more features in 2.3 (multi-crew) to have us play in groups more ... how can this work with the networking in the state it is.
It was alright for me a couple of days ago, but I couldn't play at all yesterday due to how long it was taking certain things to open and tonight I managed about an hour before I got fed up with all the mission board failures
I also think the server issues are the reason why I don't get interdicted very often. Every time I jump into a system there's no other ships around and it takes about 5 minutes for the first ones to appear. By that time I'm pretty close to docking, so if any NPC's were heading to intercept me, I've usually arrived at the station.
I also think the server issues are the reason why I don't get interdicted very often. Every time I jump into a system there's no other ships around and it takes about 5 minutes for the first ones to appear. By that time I'm pretty close to docking, so if any NPC's were heading to intercept me, I've usually arrived at the station.
Last edited:
Long story short is Frontier cheaped out on the server solution and are using AWS. I get the distinct impression they are nickle and diming it and the result is what we have now.
duly +1'd.
In one of the threads about this very same issue, there was a post made by a Frontier employee that seemed to state that they manage the scalability of their cloud. Meaning it's not automated, or if it is automated that it still requires some human interaction.
Maybe it's not their cloud server, but how they're managing it? Or maybe there's been a big uptick in users they've failed to account for.
Maybe it's not their cloud server, but how they're managing it? Or maybe there's been a big uptick in users they've failed to account for.
Last two nights I've been talking to a mate in chat and get the 'commander x has gone offline' message before continuing to watch him talk to me in chat with me completely unable to reply. Main menu shows him offline in my friends list, he doesn't appear in the list shown in the second tab of the chat window and even logging out completely (including closing the launcher) and back in doesn't change it. It's only started happening this week and it's driving me mad, not least because it makes me look very ignorant.
Something is definitely afoot with the servers, to the point that assuming FD rent them, it's long past the time that they should be having a conversation with whichever jokers their contract is with. It can't carry on like this, there's been a noticeable increase in complaints about it on the forums and it's becoming utterly shambolic. It definitely doesn't give a good impression of the game either, in particular to newer players who may not be aware that it hasn't always been this laughable.
Something is definitely afoot with the servers, to the point that assuming FD rent them, it's long past the time that they should be having a conversation with whichever jokers their contract is with. It can't carry on like this, there's been a noticeable increase in complaints about it on the forums and it's becoming utterly shambolic. It definitely doesn't give a good impression of the game either, in particular to newer players who may not be aware that it hasn't always been this laughable.
I had no problems last night...but then again I was off at 14:00IGT (so I could watch episode 3 of Stranger Things) before the server update (?)
Frawd
Frawd
Out of rep, exactly right.
As I remember it from many posts I've read over the last two years, ED is hosted on AWS servers that are distributed in multiple locations around the world.
There was a spell not too long ago when you could tell it was the weekend without a calendar just by the server glitches.
Since then, there was quite a bit of improvement after constant forum rants.
I'm not sure if capacity is automatically adjusted to meet demand or done manually, but even if automated, the algorithm parameters may need some calibration.
Then again, they may be trying to sneak in a stealth upgrade.
Either way, it needs to be addressed.
- - - - - Additional Content Posted / Auto Merge - - - - -
Yup, same here. Did me some Mr. Robot episodes I had recorded. Almost stranger than Fringe...
There was a spell not too long ago when you could tell it was the weekend without a calendar just by the server glitches.
Since then, there was quite a bit of improvement after constant forum rants.
I'm not sure if capacity is automatically adjusted to meet demand or done manually, but even if automated, the algorithm parameters may need some calibration.
Then again, they may be trying to sneak in a stealth upgrade.
Either way, it needs to be addressed.
- - - - - Additional Content Posted / Auto Merge - - - - -
Yup, same here. Did me some Mr. Robot episodes I had recorded. Almost stranger than Fringe...
Your instance due to peer to peer architecture is only as good as the weakest member. If they have older comp or bad Internet it can cause lagging.
Try posting your NetLog in the support sub-forum.
You can't "Fix" AWS. AWS is not designed for realtime loads. The only way this gets better is if less people play Elite: Dangerous. When the servers appear to work well, that's because the servers aren't suffering enough stress to start flaking out. If there's too many people playing at the same time, you'll get database blocking issues and eventually timeouts because asking the AWS to handle realtime transactional loads is, and I'm going to repeat a metaphor I used earlier - Like rolling up to the WRC in an agricultural tractor you stuffed neons, a stereo system and a "Best of Skrillex" CD into.
Yes, you'll get to the end of the stage, and probably on all four wheels, but you're not gonna get to the end -fast- or -efficiently- because despite the fact that tractors and rally cars possess VAGUELY similar characteristics (they have wheels, engines and steering wheels), they're designed for vastly different purposes. In the same manner, using AWS to handle realtime loads is like trying to rally with a tractor, it's possible, but you'll be making one hell of a mess doing so, and you'll not be using the right tool for the job.