Jump to content
  • 0

Beta 7 - v111 lag?


Weylon Santana

Question

I noticed that the server has new settings

 

"Processing": {
    "CpsLock": true,
    "MinLogicThreads": 2,
    "MaxLogicThreads": 4,
    "LogicThreadIdleTimeout": 20000,
    "MinNetworkThreads": 2,
    "MaxNetworkThreads": 4,
    "NetworkThreadIdleTimeout": 20000,
    "MapUpdateInterval": 50,
    "ProjectileUpdateInterval": 50,
    "PlayerSaveInterval": 120000,
    "CommonEventAutorunStartInterval": 500

 

I would like to know if these settings could be causing the game to lag.
 

I was in a more stable version, 101 (there is no stable version, even 6.2 isn't, but this is not so bad), and this version could handle 20 players calmly. Now version 111 can barely handle 5 players, on the same machine that can handle 20.

This is just a test we are doing. I would like to know why.


The server sends all lag signals to clients, such as:
Walking backwards, monsters suddenly deactivating aggression, when they(npc) return to the spawn point(origin) they fill their whole lives suddenly while in combat, in the middle of the fight they ignore the player, deactivating aggression and return to the point of origin, high ping with people who have a great internet speed, and so on. onwards.

We did a test with only about 5 people to check if version 111 was fine. And it is not. The lag was so much that you couldn't play.
Is there anything I can do in this configuration that can fix this?

Link to comment
Share on other sites

9 answers to this question

Recommended Posts

  • 0

Following up.. new db system saves global variable modifications in real time in v103+.

 

This game had an autorun common event that was updating a global variable for every player every map update (give or take).

 

That led to a ton of waiting for the db to become available, and for disk writing which slowed things down considerably.

 

I will make an edit in a future version that batches those variables changes, and writes them to the db at a specified interval designated in the server config.

Link to comment
Share on other sites

  • 0
24 minutes ago, jcsnider said:

What are the specs of the server?

aac7b13bdce8c5d77703b690c5524ec7.png

 

Edit:
------------------
System Information
------------------
Time of this report: 4/20/2021, 11:30:54
Machine name: DAVI
Operating System: Windows 7 Ultimate 64-bit (6.1, Build 7601) Service Pack 1 (7601.win7sp1_rtm.101119-1850)
Language: Portuguese (Regional Setting: Portuguese)
System Manufacturer: QEMU
System Model: Standard PC (i440FX + PIIX, 1996)
BIOS: Default System BIOS
Processor: Common KVM processor (4 CPUs), ~2.4GHz
Memory: 4096MB RAM
Available OS Memory: 4096MB RAM
Page File: 1315MB used, 6873MB available
Windows Dir: C:\Windows
DirectX Version: DirectX 11
DX Setup Parameters: Not found
User DPI Setting: Using System DPI
System DPI Setting: 96 DPI (100 percent)
DWM DPI Scaling: Disabled
DxDiag Version: 6.01.7601.17514 32bit Unicode

Link to comment
Share on other sites

  • 0

Your host machine has 4 cores and you've assigned 4 cores to one machine while running another. Can you try to shut down the other VM or not massively overprovision your virtual machines?

 

Unfortunately virtualization does not make for a magic cure that suddenly lets your computer do more things at once it'll just make an attempt at trying to keep up and potentially slow everything down if it's doing too many things at once.

Link to comment
Share on other sites

  • 0
6 hours ago, jcsnider said:

What CPS numbers are you seeing when the server is freshly booted vs when there are 10-20 players on? Would be interesting to know both locked and unlocked. 

 

On second thought set your Min and Max logic/network threads to 1. 

I will give a detailed report on the tests you ordered

 

This is the locked and unlocked cps with 0 players.

5eed000d542b3b809261ba24e085032e.pngaaf60f1b826f3a9149cc91980d815db5.png

 

locked, is usually locked at 65, sometimes decreases to 61 and returns to 65. Unlocked is between 2 million and 100 thousand and 2 million and 200 thousand.

 

This is the locked and unlocked cps with 8 players.

45c3b692ac207a088272df19fc5b2413.pnge054645d65e2c75583eb6aec1f85712a.png

 

same as above.

 

unknown.png

 

 

after this test I put the settings you said

 

1f86dba169138afa61d982281cff03eb.png

 

This is the locked and unlocked cps with 5 players. The players said it improved a little, but it was still bad.

 

unknown.pngunknown.png

 

500 - 510 locked, 1 million and 900-998 thousand unlocked

 

After this setup I recorded ingame with 5 players online.

 

 

 

in the end, i have typed exit on server console and a lot of error spam the console

https://www.ascensiongamedev.com/resources/filehost/092609e4fd4322dbdbc04eea4aea8b82.log

 

3 hours ago, Cheshire said:

Your host machine has 4 cores and you've assigned 4 cores to one machine while running another. Can you try to shut down the other VM or not massively overprovision your virtual machines?

 

Unfortunately virtualization does not make for a magic cure that suddenly lets your computer do more things at once it'll just make an attempt at trying to keep up and potentially slow everything down if it's doing too many things at once.

And about your comment, I find it difficult that the machine is the problem, we had no problems to hold up to 20 players in version 101, why would we have in version 111? If the only thing that changed was the version and not the machine. I don't know if I could test what you said because the machine is not mine.

 

Link to comment
Share on other sites

  • 0

So another recent version worked fine? Strange I would have thought a 10 year old cpu that to my knowledge doesn't support virtualization running overprovisioned virtual machines would cause random issues like this on multi threaded applications.

 

Anyway, reading through those errors without looking into them too deeply seems to imply something is wrong with logging? On my phone though so not really looked at the whole thing yet.

 

What's the disk read and write performance like? I remember in Leafling I've had to disable a bunch of logging at some point to stop the server from hanging up frequently due to IO wait times.

Link to comment
Share on other sites

×
×
  • Create New...