Mordhau
 marox — Project Lead
Project Lead
  • Likes received 6974
  • Date joined 25 Oct '15
  • Last seen 9 Oct

Private Message

988 6974
  • 26 Jan '18
 marox — Project Lead

@Havoc said:

@marox said:

@Havoc said:
So the "input lock bug" is actually the "animations and player models are desynced causing very early hits on your client via receiving accels" bug.

Not sure what input has to do with this, can you elaborate?

The input lock bug may just be desync.

I'm guessing you're not referring to an actual general input lock, but a specific case of not being able to parry in a specific scenario which you mentioned last time? Shouldn't be related to that in any way unless it magically goes away on map change. It's only related to time (and by proxy) some animations desyncing between server and clients, and since servers run the logic they decide when hits happen. I'm not sure if it's an issue that would become problematic in actions other than attack hitreg.

988 6974
  • 26 Jan '18
 marox — Project Lead

@rob_owner said:
okay so from what i found, if you join a server that was empty and up for a while (theres probably a reason i dont know for why this happens but those two qualities were both consistent when i tested), theres a chance that the server tracers are not synced with your tracers/weapons, and stabs will hit SLIGHTLY early, so will everything else. This is different from regular instahits (which are already pretty brutal) and it looks like thats whats happenin in some of your clips.

i tested on Nal's server with the commands m.drawtracers 1 and the showauthtracers 1 command, you need to be logged in as admin to see the server tracers, but if you join a server that was up and empty for a while, chances are they will be desynced, it was consistent when i tried it. If you just change the map it will fix it. If you set the server speed with slomo to like 0.2 or 0.5 it becomes extremely overt.

pretty sure devs already know and are working on it among other things

Can now confirm this is an issue, we're working on a fix.

988 6974
  • 26 Jan '18
 marox — Project Lead

@Havoc said:
So the "input lock bug" is actually the "animations and player models are desynced causing very early hits on your client via receiving accels" bug.

Not sure what input has to do with this, can you elaborate?

988 6974
  • 25 Jan '18
 marox — Project Lead

@rob_owner said:
okay so from what i found, if you join a server that was empty and up for a while (theres probably a reason i dont know for why this happens but those two qualities were both consistent when i tested), theres a chance that the server tracers are not synced with your tracers/weapons, and stabs will hit SLIGHTLY early, so will everything else. This is different from regular instahits (which are already pretty brutal) and it looks like thats whats happenin in some of your clips.

i tested on Nal's server with the commands m.drawtracers 1 and the showauthtracers 1 command, you need to be logged in as admin to see the server tracers, but if you join a server that was up and empty for a while, chances are they will be desynced, it was consistent when i tried it. If you just change the map it will fix it. If you set the server speed with slomo to like 0.2 or 0.5 it becomes extremely overt.

pretty sure devs already know and are working on it among other things

Nice! This isn't something we were aware of. Any idea how long they were up and empty? Are the desyncs consistent, do other people get them as well?

Due to how Unreal handles world time and how it computes delta time between frames from that time, this might be caused by deterioration in floating point accuracy. I've feared this for some time after initially looking over the engine source, but couldn't really believe Epic would leave such a rookie mistake in for so long. This would require the map to be running for quite a while, but it would fix itself on map change, so it feels like it could be responsible, I'm just not entirely sure it's the culprit. It would easily be fixed by making the map change itself after a while regardless if the server is empty, but I'd prefer to tackle the root of the problem as well.

988 6974
  • 21 Dec '17
 marox — Project Lead

There are some issues with the servers in this patch, not entirely sure what's up, but we're trying to get to the bottom of it. Might be related to the animation optimization on dedicated servers, might be just our host Vultr acting up.

As for the delayed sound/hit, the effects of the hit are split into two stages that could potentially be delayed. These will eventually be brought together into a single packet to reduce the chance of this occurring. Right now what can arrive split is the flinch animation and in the other stage the sound & hitflash & particles. The health updates completely separately from these, so the health bar is not a reliable indicator in any case, but it might be a good idea to preemptively show it in the same frame even though it doesn't tick down.

988 6974
  • 28 Oct '17
 marox — Project Lead

There is no inverse accel, just a speed cap

988 6974
  • 15 Oct '17
 marox — Project Lead

@Jax said:
@marox while you're at it, could you implement "inverse strike" and "inverse stab" keybinds, which would make the attack come from the opposite side of where you looked?

I know there's already an option to toggle this, but I prefer the default system - however, sometimes when I'm doing successive chambers or trying to track an enemy I run out of mousepad or get too off target.

(I might have overlooked an option in the settings, though.)

This might quickly spiral out of control in terms of buttons, given that one person might want left/right swapped or up/down swapped or both. :(

988 6974
  • 15 Oct '17
 marox — Project Lead

I'll add the 240 system angling after button press as a console toggle (for strikes only since stabs are normally on wheel)

988 6974
  • 15 Oct '17
 marox — Project Lead

Thanks for the reminder, the potential fix will be in next patch, if it occurs in the next patch do let us know.

988 6974
  • 14 Oct '17
 marox — Project Lead

Here's the patch notes: https://mordhau.com/forum/topic/10158/changelog-thread/

What do you guys think of the changes? Discuss them here!

988 6974
  • 3 Oct '17
 marox — Project Lead

Here's the patch notes: https://mordhau.com/forum/topic/10158/changelog-thread/

What do you guys think of the changes? Discuss them here!

988 6974
  • 3 Oct '17
 marox — Project Lead

@Cow said:
No it's fine i played a lot and didn't find this bug again, just one time, i didn't heared sound on my death and my corp disapear instantly. (just 1 time in 4 days)

That's a different bug completely, but it's also on my list to fix!

988 6974
  • 3 Oct '17
 marox — Project Lead

It's a bug, we're on it.

988 6974
  • 1
  • 3 Oct '17
 marox — Project Lead

Yes, use DamageFactor, for example:

[/Script/Mordhau.MordhauGameMode]
MapRotation=SKM_Contraband
MapRotation=TDM_Camp
MapRotation=FFA_Grad
MapRotation=SKM_ThePit
MapRotation=TDM_Contraband
MapRotation=FFA_Camp
MapRotation=SKM_Grad
MapRotation=TDM_ThePit
MapRotation=FFA_Contraband
MapRotation=SKM_Camp
MapRotation=FFA_ThePit
DamageFactor=0.0

PS: you can set teamdamagefactor that way as well.

988 6974
  • 3 Oct '17
 marox — Project Lead

@Ruby Rhoderick said:
Will there be customization options for the Messer in the future? I'm a little bummed it's so bare-bones - there also no alternate stance for it, though I'm not sure what it would be. Maybe a two-handed throw lol?

Yes, this isn't actually the messer, but an alt-skin for it which is the Maciejowski bible chopper, that's why it's lacking the crossguard. Proper 2H messers are coming still in that category, and hopefully some customization options for the chopper too.

988 6974
  • 2 Oct '17
 marox — Project Lead

Here's the patch notes: https://mordhau.com/forum/topic/10158/changelog-thread/

What do you guys think of the changes? Discuss them here!

988 6974
  • 1
  • 29 Sep '17
 marox — Project Lead

If you encounter this bug, tell the person that's not emitting sounds to jump off of a high enough thing to take fall damage, then hit him again, that'd be really helpful for debugging it further. (Yes, I know how weird that sounds.)

988 6974
  • 28 Sep '17
 marox — Project Lead

This was supposed to be fixed by the last patch, is it still there?

988 6974
  • 27 Sep '17
 marox — Project Lead

There was an inconsistent ordering of feint and parry, and it was sort of bandaid fixed this way, by preventing them from inputing on the same frame. What happened was that when you bound feint and parry to the same button (usually rmb), it'd trigger feint and then parry, instead of triggering parry (which would fail) and then feint. And it did sometimes one, sometimes the other for different people, and I never did get to the bottom of it. Ended up patching it like this. Would like to fix it properly though.

988 6974
  • 1
  • 27 Sep '17
 marox — Project Lead

The kick clash cooldown stuff was kind of an oversight, we're going to change it