• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

Alex_brunius

Field Marshal
70 Badges
Mar 24, 2006
22.404
5.024
  • Cities: Skylines - Natural Disasters
  • Hearts of Iron IV: Expansion Pass
  • Cities: Skylines - Green Cities
  • Stellaris: Synthetic Dawn
  • Hearts of Iron IV: Death or Dishonor
  • Surviving Mars
  • PDXCON 2017 Gold Ticket holder
  • Cities: Skylines - Mass Transit
  • Stellaris - Path to Destruction bundle
  • Stellaris: Humanoids Species Pack
  • Stellaris: Leviathans Story Pack
  • Stellaris: Digital Anniversary Edition
  • Hearts of Iron IV: Field Marshal
  • Hearts of Iron IV: Colonel
  • Hearts of Iron IV: Cadet
  • Hearts of Iron IV Sign-up
  • Stellaris
  • Achtung Panzer
  • Hearts of Iron IV: Expansion Pass
  • Hearts of Iron 4: Arms Against Tyranny
  • Hearts of Iron IV: No Step Back
  • Hearts of Iron IV: By Blood Alone
  • Battle for Bosporus
  • Stellaris: Federations
  • Hearts of Iron IV: La Resistance
  • Cities: Skylines - Campus
  • Prison Architect
  • Cities: Skylines - Snowfall
  • Imperator: Rome
  • Stellaris: Megacorp
  • Cities: Skylines Industries
  • Stellaris: Distant Stars
  • Cities: Skylines - Parklife
  • PDXCON 2018 "The Emperor"
  • Stellaris: Apocalypse
  • Europa Universalis IV: Conquest of Paradise
  • Majesty 2
  • Magicka
  • Europa Universalis III Complete
  • Heir to the Throne
  • Hearts of Iron III: Their Finest Hour
  • Hearts of Iron III
  • For the Motherland
  • Europa Universalis IV: Wealth of Nations
  • Europa Universalis III Complete
  • Europa Universalis IV
  • Europa Universalis III Complete
  • Dungeonland
  • Deus Vult
  • Crusader Kings II
Last night when playing I noticed two bugs.

1.) The old multiplayer sync bug from HoI2 remains. Sometimes when the client researches a technology It's not registered at the host. Thus he is stuck and won't be able to keep researching or using anything the tech unlocks until rehost when he looses the tech entirely.

2.) The delay that prevents you from calling off an attack until a few hours have passed introduced a new client bug!
For me (as a client) it was impossible to call off attacks at all, I got the message "initiating combat" even when trying to call off troops after weeks of battle. This happens every time for 5hours of gameplay.
 
Yes, I have experienced both of these issues, but I have noticed the second quite a bit more. I listed it in the bug thread, but no one commented. Please fix this, the game is unplayable this way in multiplayer.
 
How do i recreate the cannot break off?
 
3. Autosliders wont work correctly with a shared country, nor will manual setting of the sliders stick... leaving the screen resets it to default maximum production queue.


this one is making our game almost unplayable... prolly a sync issue again
 
How do i recreate the cannot break off?
If you have two computers available host a game on one and connect to it from the other (If not I can host a game for you, my skype contact infos are available in my profile if you don't want to ask WyK1NG or MR2 for them). Host actually don't have to do anything, the bug only appears clientside from what I understand.

Also feel free to move this thread to bug reports.
 
I second that.

There seems to be another lag-induced issue: you have divisions in a province, and the enemy launches an attack and an artillery bombardment on this province. Later on, other divisions arrive to support the besieged province: these divisions will not join up the defense.
 
Yea you cant stop attack which can be pretty bad when your trying to retreat strategically. For the most part we can get like 30mins-2hours of gaming till the game crashes for someone or its freezes or something. Use the program gameranger it rocks
 
Research Tems - i think i found the bug, but will need some confirmation.
Reason for bug:
In addition to the skill of the researcher there was a semi-random number given, however the effect was very small.
So basically the host and client could differ a small amount due to different random values. I have removed the random factor for now, as the effect was very small.

Retreat issue fixed - seems the update of combat time was never working in HOI2, but the automatic abort for aviation saved the day. - added new network events for combat time update.

Sharing countries slider issue:
Fixed - the data was correctly transmitted, but the screen would not be updated - when exiting the screen you transmit your data to the other - resulting in obsolete data beeing send and overwriting the changes the other player just has made.
 
Great progress!

Another possible multiplayer bug might be that statistics differ between host & client, we saw very different numbers.

Yea you cant stop attack which can be pretty bad when your trying to retreat strategically. For the most part we can get like 30mins-2hours of gaming till the game crashes for someone or its freezes or something. Use the program gameranger it rocks
Never seen this happen during 10h+ off playing. So It might be unrelated to multiplayer and just a standard crash issue (those should be happening more often since its enough that one crashes for all to have to rehost)
 
Research Tems - i think i found the bug, but will need some confirmation.
Reason for bug:
In addition to the skill of the researcher there was a semi-random number given, however the effect was very small.

The effect Alex is talking about is lag based. What happens is this. When there is a small amount of lag the client thinks that the tech team has completed the tech, and the player then changes the tech team, before the host registers that the tech has been completed. Instead the host thinks that the tech team has been changed before the tech is completed, so it registers as a canceled tech since the team has been removed, as opposed to a completed tech.

This MP bug was among several we MP players posted in a thread on MP bugs a few months back.

What needs to happen here is that the client needs to report back to the host every time a tech is complete.
 
Last edited:
Great progress!

Another possible multiplayer bug might be that statistics differ between host & client, we saw very different numbers.


Never seen this happen during 10h+ off playing. So It might be unrelated to multiplayer and just a standard crash issue (those should be happening more often since its enough that one crashes for all to have to rehost)

Yes Alex, this is one of the consistent problems that causes numerous MP bugs. It is simple enough to see by taking a client save and a host save at exactly the same time, and then comparing the save files, that there is substantial difference between the saves. This is particularly troublesome in the case where the host crashes, and players want to continue playing from the client save, because a great deal of the AI information is lost in the client save.

This creates all kinds of oddities, particularly in the diplomatic frame, where wild anomalies start happening, such as Siam declaring against Japan, because it has been given a generic AI, as opposed to a the Siam specific AI it is meant to have.

I would estimate that 1 in 20 MP games are seriously damaged, or even destroyed by bugs caused by incomplete host client information transfer.

Every effort should be made to make sure that all information is exported from host to client, and from client to host, when players make changes.
 
Yes Alex, this is one of the consistent problems that causes numerous MP bugs. It is simple enough to see by taking a client save and a host save at exactly the same time, and then comparing the save files, that there is substantial difference between the saves. This is particularly troublesome in the case where the host crashes, and players want to continue playing from the client save, because a great deal of the AI information is lost in the client save.

This creates all kinds of oddities, particularly in the diplomatic frame, where wild anomalies start happening, such as Siam declaring against Japan, because it has been given a generic AI, as opposed to a the Siam specific AI it is meant to have.

I would estimate that 1 in 20 MP games are seriously damaged, or even destroyed by bugs caused by incomplete host client information transfer.

Every effort should be made to make sure that all information is exported from host to client, and from client to host, when players make changes.
It could be alot more then 1 in 20 games. Now that you mention it I do remember that alot of wierd AI stuff such as strange DoWs has been going on in HoI2 MP games.

Not long ago I had a HoI2 game where we loaded saves from clients a few times, and before long I (playing historically as USSR) was at war with litteraly everyone, Including the allies and Turkey.
 
Could be more than 1 in 20 games. I have a high tolerance level, and also am a skilled save editor, who has learned to save game after game.

The fact is that I would say that 70% of all end session saves require editing. You have to be a real trooper to play HOI II in MP, and I imagine one of the reasons the game has failed to take off in the rapidly growing multi-player gaming market is because most people are simply not willing to put up with the issues.

If AoD designers intend to build word of mouth support for this game through internet MP gaming, they should make eliminating MP bugs a high priority.
 
Hi all

Can we get a Hotfix because of this Host-Client MP-Bugs with the "non-retreat"?

Or can you say Lennartos when a Hotfix for this Bug is coming?

Because playing MP´s to not make sense this time with the Bug.
 
Hi all

Can we get a Hotfix because of this Host-Client MP-Bugs with the "non-retreat"?

Or can you say Lennartos when a Hotfix for this Bug is coming?

Because playing MP´s to not make sense this time with the Bug.

Tricky..
If i want to release something like a patch, it first needs to be approved by paradox and checked through their Q/A, and then all their partners must be ready (steam etc).

We will have to see how it works out.