Sudden judder or momentary freeze

I guess the only way to get this investigated is to upvote the issue on the tracker.


Oh well I guess I can't.

1742809115765.png
 
Started playing again yesterday after a few months away from Elite and noticed these freezes / stutters. At least it sounds like my PC isn’t on the way out!

Have voted for the issue and added my own experience to the list.
 
Last edited:
The one thing I haven't done is to see what my FPS is during a stutter.

Hmm. If I remember, I'll turn on the monitor and see today.

Yesterday, while looking out of the cockpit at empty space in SC, it hit really badly. And it was the usual 05 past the hour.

ETA - once again, thanks to all who upvoted and commented - we're now replete with a full pyramid! \o/
 
Last edited:
The one thing I haven't done is to see what my FPS is during a stutter.

Hmm. If I remember, I'll turn on the monitor and see today.

Yesterday, while looking out of the cockpit at empty space in SC, it hit really badly. And it was the usual 05 past the hour.

ETA - once again, thanks to all who upvoted and commented - we're now replete with a full pyramid! \o/
For me, at least, when this happens there is a sudden and constant drop in frame rate for a fraction of the time. The last time it happened I decided to activate the MSI Afterburner resource monitor to check and this is what I noticed:

1743342312873.png

The spikes in frame response time are to be expected, however, since not rendering that frame directly will cause the response time to increase. What really intrigues me is the pattern of behavior of this issue. The spikes last for around a few seconds to a maximum of a minute and at fixed intervals of one second.

I have not noticed any connection between this issue and external video or hardware settings. I have just upgraded my PC's memory to 32 GB of RAM and my additional settings are perfectly capable of running the game, and I have reset all settings to default several times and tried running tests with various combinations, with no difference in the results, which could indeed indicate that something external, besides the game client, could be causing this behavior.
 
I am of the opinion that the game is polling our PCs for certain telemetry so feeding them with performance / utilisation information. It is far too much of a stretch to think that it is anything other than the game code causing this when it is so prevalent with so many of us players and at such specific times. In the main the 5 min past the hour every hour seems ubiquitous and the other instances at multiples of 5 minutes (I notice it at 25 to and 25 past the hour) seems frequent if not consistent.
 
Last edited:
stutter is now reliably happening every 10 minutes on the 5's
:05, :15, :25, :35, :45, :55 after every hour

it's probably when Inara or another one of the "tools" floods the ED servers with requests for information
rogue API requests gone wild
 
Last edited:
stutter is now reliably happening every 10 minutes on the 5's
:05, :15, :25, :35, :45, :55 after every hour

it's probably when Inara or another one of the "tools" floods the ED servers with requests for information
rogue API requests gone wild
Well some Twitch stream chatter who professed to know what they were talking about called it "the journal issue".

Can't really be Inara as that only updates when you click the "Import Game Data" button in Inara to read the Frontier account/server. E.g. station commerce info doesn't upload until that's done.
 
I was just looking at the issue tracker. If you search for keywords like "freezes", "stuttering" or "sudden", the trackers lists lots of various reports. I couldn't, when searching, find the ticket I raised on page one of this thread.

Hmm... it may be my search powahz, or it could be that his stuttering and freezing is pretty widespread and has been for a while. However, I've only really noticed instances of it since PP2.0 and Ascendancy.

... although I didn't see any reports of the 05, 15 etc. minutes past each hour occurrence.
 
Fdev might be updating the servers on a set timer to get around the maintenance replica backup issues on a Thursday.
 
Yes, multiple extremely disruptive issues. Especially the past two updates.

Game stutters - many ascribe this to particular minutes past each hour, I think there may be confirmation bias in those reports. I'd say every multiple of 5 mins past each hour is likely to have issues.

Also in past couple of weeks, many more Team issues (Teams getting partially set up - the icon at the top of the screen only partially populated). Team issues arriving in journal but not in UI. Team invites visible in UI for a while before being acceptable (accepting them does nothing).
 

Ozric

Volunteer Moderator
I'd like to preface this by asking, please let's not start a "this would be easy to fix" route :) but I present for information's sake something I've noticed.

As has been noted the issue happens at variants of :x5 minutes, there's something in the netlog called GPSS that happens around these times. GPSS seems to basically be a way of queuing transactions to happen at regular intervals, though how it's used in Elite we obviously don't know.

I looked in my Netlog from last night and it was littered with references to it, after the initial run on loading in, they are all at a minute ending in 5. One query for it alone contained a repeated part (probably the actual transaction part?) 106 times.
{18:15:49GMT 38.118s} GPSS AddJobs state change 5->6
{18:15:49GMT 38.135s} GPSS m_systemMapIt != m_systemMap.End() state change 6->5

I did some further digging through old netlogs, found some comparable ones and searched the whole document for GPSS

July last year (pre PP2.0) in a 1 hour 20 minute session there were 26 hits for GPSS
November 2nd (post PP2.0) in a 1 hour 30 minute session there were 345 hits for GPSS
Yesterday (post Colonisation) in a 1 hour session there were 1,664 hits for GPSS

So while doing a similar thing (a few runs of a Buckyball race) since pre Power Play 2.0 there are now 64 times the amount of these transfers being made. And as Colonisation and PP2.0 grow, so will these numbers.

edit
Pre PP2.0 the transaction was only made on log in.
 
Last edited:
I'd like to preface this by asking, please let's not start a "this would be easy to fix" route :) but I present for information's sake something I've noticed.

As has been noted the issue happens at variants of :x5 minutes, there's something in the netlog called GPSS that happens around these times. GPSS seems to basically be a way of queuing transactions to happen at regular intervals, though how it's used in Elite we obviously don't know.

I looked in my Netlog from last night and it was littered with references to it, after the initial run on loading in, they are all at a minute ending in 5. One query for it alone contained a repeated part (probably the actual transaction part?) 106 times.


I did some further digging through old netlogs, found some comparable ones and searched the whole document for GPSS

July last year (pre PP2.0) in a 1 hour 20 minute session there were 26 hits for GPSS
November 2nd (post PP2.0) in a 1 hour 30 minute session there were 345 hits for GPSS
Yesterday (post Colonisation) in a 1 hour session there were 1,664 hits for GPSS

So while doing a similar thing (a few runs of a Buckyball race) since pre Power Play 2.0 there are now 64 times the amount of these transfers being made. And as Colonisation and PP2.0 grow, so will these numbers.

So, what I'm hearing is, FD need bigger hamsters!
 
Back
Top Bottom