Update #6 (1.10.11.0) is now live!

[odourboy 1,289]

19 minutes ago, Honnli said:

Yep I have tried with multiplayer off, live weather off, photogrammetry off, rolling cache off and even tried reducing my graphics settings and disabling Azure ATC but nothing seems to solve the issue. These stutter-freezes (some lasting up to 10-15 seconds where the sim just hangs and the the CPU and Manipulator utilization goes completely idle according to the dev mode FPS display) seem to happen around the same areas in the area I've used as my test area after each update (around the Melbourne CBD area to the south east of Melbourne Airport ICAO YMML if anyone wants to give that a try). Someone on the MSFS forums also mentioned these stutter-freezes around the Melbourne area that's been happening for them since the second patch. So I've basically resigned to the fact that the sim still needs further optimizations and have been flying in less dense areas while eagerly waiting for the next patch or two. But thanks for your suggestion anyway. 🙂

You're not alone. Exactly the same issue with a westbound approach into KLAX.

[MrFuzzy 2,118]

1 hour ago, odourboy said:

You're not alone. Exactly the same issue with a westbound approach into KLAX.

I too have issues in this area. Besides stuttering, a lot of objects popping in and mesh changing shape over time.

I don't know if the region is too complex for our 4 core CPUs or is problematic in itself... I can't wait to upgrade my PC but it's a bad moment to buy hardware now.

[Honnli 69]

2 hours ago, odourboy said:

You're not alone. Exactly the same issue with a westbound approach into KLAX.

Ah yes, I saw your thread previously about the poor performance in large urban centers and it seems the consensus is that certain parts of the world (especially the larger urban areas) are "broken" and cause these stutter-freezes with no way around it unfortunately, even on fairly respectable hardware. I guess we'll just have to wait for further optimization by Asobo.

 

1 hour ago, MrFuzzy said:

I too have issues in this area. Besides stuttering, a lot of objects popping in and mesh changing shape over time.

I don't know if the region is too complex for our 4 core CPUs or is problematic in itself... I can't wait to upgrade my PC but it's a bad moment to buy hardware now.

What's interesting and weird is that this issue didn't manifest itself on my machine prior to the second patch while others are seeing this issue pop up after subsequent patches. Your i7-7700K should still be relevant and fast enough though.

Edited November 12, 2020 by Honnli

[Mark_A 31]

Reducing LOD to 50 has resolved my stutters but the scenery is not looking good.

I get them around Seattle and LAX - started after the last patch.

 

[roi1862 602]

19 hours ago, TravelRunner404 said:

Nah. It’s fine. Does a great job. 

Well still waiting on you... show me snow with live weather... its the easiest thing to simply dismiss someone else by saying "Nah. It’s fine. Does a great job."

[109Sqn 1,567]

13 hours ago, Glenn Fitzpatrick said:

Most of the "hassle" results from people using the default install options and accepting the default location for the community folder which means you need to jump through hoops to save or modify stuff.

If when installing the MS version you nominate somewhere else for the packages location it more or less works the same as steam.

 

Glad you said "Most", because I installed in a custom location.😉 

Gaming Services is a totally flawed app. I've seen posts on various forums (including Microsoft's) where the issue is raised and never fixed, going back a few years.

[JRBarrett 3,077]

23 hours ago, EmaRacing said:

Yes I would, but being on 4G, sometimes the connection speed drops below 50Mbps and the game starts displaying GIANT banners taking the whole screen in flight, instead of just making use of the (reduced) band he has for a couple of minutes.
The only way to avoid that is setting a lower bandwidth limit.
I want to see who thought, at bugsobo, that displaying a giant red text banner was wayyy better than having low res in a corner of the map for a couple of seconds more, or some building popping out...

You should not get the banners if the bandwidth merely reduces. Those only appear if the connection to the scenery streaming server is completely lost. I often monitor all the TCP/IP connections MSFS makes while running using a utility program called CurrPorts. Typically the game is connected to between 4 and 7 discrete IP addresses (remote servers) while running. All the streams are encrypted, so I can’t see the actual contents with a packet monitor like WireShark, but one in particular is undoubtedly the scenery stream as that address has a continuing flow of incoming data which increases when the aircraft is in the air and moving. Whenever I get the red “internet connection lost” banner, it always occurs concurrent with a connection timeout to the specific TCP/IP address associated with the scenery stream. When that happens, I can see the connection being re-negotiated in real time in the monitor program, and as soon as it connects, the smaller black banner pops up to indicate that the connection has been reestablished.

I don’t think a mere slowdown in incoming packet rate is sufficient to generate the “connection lost” message. It appears to require a complete disconnect.

FWIW, two of the other IP addresses the sim connects to while running appear to be associated with LiveWeather, as both of those connections will be dropped if I turn LiveWeather off and use a preset instead. Both of these are completely different addresses than the IP address that supplies the scenery. One of those addresses resolves to Microsoft Azure, which, I suspect, contains the MeteoBlue forecast data. The other addresses resolves to an AmazonAWS server which I think may be the surface METAR data. The latter stream has a much lower packet rate than the Azure connection.

[DylanM 1,069]

^ great analysis Jim. Out of curiosity, have you had to set up port forwarding for the Windows Xbox services? I noticed my NAT type is coming back "moderate" and should probably be open...

 

[Ianrivaldosmith 5,317]

Update regarding CPU optimisations and mixture :-

 

Hi all! I got word from the devs regarding two issues they mentioned were coming soon in the last Dev Q&A: Mixture Bug: Issue where the mixture drops drastically after 90% CPU Optimizations: improvements to CPU threading They held back on releasing in Update #6 as they noticed some risks. As of today they things are looking good for them to be in Update #7, though don’t hold me to that. 

[Tuskin38 6,524]

Source for the above^

https://forums.flightsimulator.com/t/news-dev-note-from-update-6/320090

[icewater5 181]

Is update 7 the NAmerica update

Edited November 12, 2020 by icewater5

[regis9 4,014]

Glad to hear they took the cautious route of holding it back rather than pushing it out as originally planned and potentially causing issues.

[Tuskin38 6,524]

14 minutes ago, icewater5 said:

Is update 7 the NAmerica update

The development roadmap from last week doesn't mention it. It just says Update 7 (while the japan one specifically says it's a world update). It might have changed.

Maybe we'll learn more today.

Edited November 12, 2020 by Tuskin38

[icewater5 181]

Thx🙂

[BijanStudio 896]

17 hours ago, odourboy said:

You're not alone. Exactly the same issue with a westbound approach into KLAX.

Something strange is going on for sure. After the update I had an FPS increase by 10 FPS. The next day it dropped by about 20FPS. Today increase by 20FPS again. No setting was changed.

Just did a flight with a320 ultra settings over LA about 50 FPS. Landing was about 40fps. Yesterday it was 20 fps.