I found what was causing the stuttering with 0 ping and in singleplayer. I
disabled this code in cbaseentity_shared :


>
> //-----------------------------------------------------------------------------
> // Purpose: My physics object has been updated, react or extract data
>
> //-----------------------------------------------------------------------------
> void CBaseEntity::VPhysicsUpdate( IPhysicsObject *pPhysics )
> {
>     switch( GetMoveType() )
>     {
>     case MOVETYPE_VPHYSICS:
>         {
>             if ( GetMoveParent() )
>             {
>                 Log_Warning( LOG_DEVELOPER_VERBOSE, "Updating physics on
> object in hierarchy %s!\n", GetClassname());
>                 return;
>             }
>             Vector origin;
>             QAngle angles;
>
>             pPhysics->GetPosition( &origin, &angles );
>
>             if ( !IsFinite( angles.x ) || !IsFinite( angles.y ) ||
> !IsFinite( angles.x ) )
>             {
>                 Msg( "Infinite angles from vphysics! (entity %s)\n",
> GetDebugName() );
>                 angles = vec3_angle;
>             }
> #ifndef CLIENT_DLL
>             Vector prevOrigin = GetAbsOrigin();
> #endif
>
>             for ( int i = 0; i < 3; ++i )
>             {
>                 angles[ i ] = AngleNormalize( angles[ i ] );
>             }
>
> #ifndef CLIENT_DLL
>             NetworkQuantize( origin, angles );
> #endif
>
>             if ( origin.IsValid() )
>             {
>                 SetAbsOrigin( origin );
>             }
>             else
>             {
>                 Msg( "Infinite origin from vphysics! (entity %s)\n",
> GetDebugName() );
>             }
>             SetAbsAngles( angles );
>
>             // Interactive debris converts back to debris when it comes to
> rest
>             if ( pPhysics->IsAsleep() && GetCollisionGroup() ==
> COLLISION_GROUP_INTERACTIVE_DEBRIS )
>             {
>                 SetCollisionGroup( COLLISION_GROUP_DEBRIS );
>             }
>
> #ifndef CLIENT_DLL
>             PhysicsTouchTriggers( &prevOrigin );
>             PhysicsRelinkChildren(gpGlobals->frametime);
> #endif
>         }
>     break;
> .
> .
> .
> .
>

However I still have stuttering when I set "net_fakelag" to anything > 0.
And its gets worst the higher the FPS...


On Fri, Apr 20, 2012 at 10:10 PM, Psy_Commando <psycomma...@gmail.com>wrote:

> I'm not using the template, I just removed the asw junk I didn't need, and
> re enabled the player.
> But that idea of using the marine as vehicle is interesting, the only
> thing i'm not sure about is the transition between player control and
> vehicle(marine clone) control.
>
> On Fri, Apr 20, 2012 at 8:32 PM, Tony "omega" Sergi 
> <omegal...@gmail.com>wrote:
>
>> I haven't fully analyzed the code, but i was just thinking..
>> you're using alien swarm.. why use the sdk template for alien swarm for
>> what you want? use alien swarm itself, duplicate the marine and make the
>> duplicate a vehicle that flies instead of walks around.
>> alien swarm already has the 'predicted player controlled other entity'
>>
>> -Tony
>>
>>
>>
>> On Sat, Apr 21, 2012 at 3:16 AM, Psy_Commando <psycomma...@gmail.com>wrote:
>>
>>> Predictions are on. Also I think this may not be entirely attributable
>>> to ASW. I found in one of my old vids that the stuttering in vehicles
>>> wasn't present in my early vids, the hl2 buggy worked before, I guess I
>>> must have broken something somewhere. I wonder what though...
>>> The template SDK for alien swarm by Doplhur seems to have the same
>>> problem, the jeep is doing the same thing.
>>>
>>> I printed the velocity of the client and server to the hud, and they're
>>> identical, cl_pclass isn't reporting the velocity correctly for some
>>> reasons...
>>> Also here's the full code, some of it is from Eternal Silence, its
>>> pretty messy though. : http://pastebin.com/eMcKh1YL
>>>
>>> I used a CPhysicsPropMultiplayer as the base prop, because the other
>>> prop types made even weirder stuff.
>>>
>>>
>>> On Thu, Apr 19, 2012 at 11:40 PM, Tony "omega" Sergi <
>>> omegal...@gmail.com> wrote:
>>>
>>>> I can't remember by reading that output, without seeing your code, but
>>>> the prediction dump is showing that the entity is 0 0 0 on either the
>>>> server or client.
>>>> ie: it's only moving on one, and not both.
>>>>
>>>> That's your jitter.
>>>> -Tony
>>>>
>>>>
>>>>
>>>>
>>>> On Fri, Apr 20, 2012 at 9:41 AM, Joel R. <joelru...@gmail.com> wrote:
>>>>
>>>>> Well, you have to make sure prediction is turned on.  If prediction is
>>>>> off, the NetworkOrigin and NetworkAngles and NetworkVelocity will be
>>>>> updating the Origin/Angles/Velocity, and you end up having this struggle
>>>>> between server updating your position and you updating it on the
>>>>> clientside.
>>>>>
>>>>> On Thu, Apr 19, 2012 at 7:39 PM, Psy_Commando 
>>>>> <psycomma...@gmail.com>wrote:
>>>>>
>>>>>> Well its an entity that exist on both client and server. I'll use it
>>>>>> in mp so making it clientside isn't an option.
>>>>>>
>>>>>> On Thu, Apr 19, 2012 at 8:28 PM, Joel R. <joelru...@gmail.com> wrote:
>>>>>>
>>>>>>> Did you create this entity on the server?
>>>>>>>
>>>>>>> There is a lot of code the client player class does to make
>>>>>>> prediction work.  I'd recommend just creating a clientside entity and
>>>>>>> manipulate that, not the one from server.
>>>>>>>
>>>>>>>
>>>>>>> On Thu, Apr 19, 2012 at 7:22 PM, Psy_Commando <psycomma...@gmail.com
>>>>>>> > wrote:
>>>>>>>
>>>>>>>> Found out why, the server ent returned false for IsPredicted() .
>>>>>>>>
>>>>>>>> The plane is still jittery though, but here is the dump i get while
>>>>>>>> moving : http://dl.dropbox.com/u/13343993/abox0003.jpg
>>>>>>>> It would seem velocity differs, along with a few others.
>>>>>>>>
>>>>>>>>
>>>>>>>> On Thu, Apr 19, 2012 at 5:47 PM, Psy_Commando <
>>>>>>>> psycomma...@gmail.com> wrote:
>>>>>>>>
>>>>>>>>> Wait, it worked with "weapon_smg1" but it doesn't work with my
>>>>>>>>> entity..
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Thu, Apr 19, 2012 at 4:55 PM, Christopher Harris <
>>>>>>>>> char...@resrchnet.com> wrote:
>>>>>>>>>
>>>>>>>>>> It should be like for instance****
>>>>>>>>>>
>>>>>>>>>> ** **
>>>>>>>>>>
>>>>>>>>>> **-          **cl_pclass prop_physics or cl_pclass
>>>>>>>>>> weapon_crossbow****
>>>>>>>>>>
>>>>>>>>>> ** **
>>>>>>>>>>
>>>>>>>>>> Also make sure that developer convar is set to at least 1. ****
>>>>>>>>>>
>>>>>>>>>> ** **
>>>>>>>>>>
>>>>>>>>>> *From:* hlcoders-boun...@list.valvesoftware.com [mailto:
>>>>>>>>>> hlcoders-boun...@list.valvesoftware.com] *On Behalf Of *
>>>>>>>>>> Psy_Commando
>>>>>>>>>> *Sent:* Thursday, April 19, 2012 4:15 PM
>>>>>>>>>>
>>>>>>>>>> *To:* Discussion of Half-Life Programming
>>>>>>>>>> *Subject:* Re: [hlcoders] Prediction Jerkiness with aircraft ?***
>>>>>>>>>> *
>>>>>>>>>>
>>>>>>>>>> ** **
>>>>>>>>>>
>>>>>>>>>> I tried cl_pclass with the class name, but nothing happens.
>>>>>>>>>>
>>>>>>>>>> Also, after a few minutes of messing around I get this assert :
>>>>>>>>>> src\game\server\baseentity.cpp (460) : Assertion Failed: fabs(
>>>>>>>>>> CoordFromCell( cellwidth, cell[0], pOut->m_Vector[ 0 ] ) - v->x ) <
>>>>>>>>>> cellEpsilon
>>>>>>>>>> swarm.exe has triggered a breakpoint****
>>>>>>>>>>
>>>>>>>>>> On Thu, Apr 19, 2012 at 3:12 PM, Christopher Harris <
>>>>>>>>>> char...@resrchnet.com> wrote:****
>>>>>>>>>>
>>>>>>>>>> cl_pclass classname****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> run that with your ship classname and you get a nice easy to read
>>>>>>>>>> debug gui to see what predicted variables are off for all predicted 
>>>>>>>>>> parts
>>>>>>>>>> of that entitytype. I only used it on an entity type I have 1 
>>>>>>>>>> instance of
>>>>>>>>>> so not sure what behavior occurs if you have many of em.****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> *From:* hlcoders-boun...@list.valvesoftware.com [mailto:
>>>>>>>>>> hlcoders-boun...@list.valvesoftware.com] *On Behalf Of *
>>>>>>>>>> Psy_Commando
>>>>>>>>>> *Sent:* Thursday, April 19, 2012 12:37 PM
>>>>>>>>>> *To:* Discussion of Half-Life Programming
>>>>>>>>>> *Subject:* Re: [hlcoders] Prediction Jerkiness with aircraft ?***
>>>>>>>>>> *
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> Well, it would seem this is a Alien Swarm related issue.. Even
>>>>>>>>>> the hl2 jeep is having the same problem.****
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 10:11 PM, Psy_Commando <
>>>>>>>>>> psycomma...@gmail.com> wrote:****
>>>>>>>>>>
>>>>>>>>>> Its called in the shared ProcessMovement method.I assume that
>>>>>>>>>> method is called at the same time on client and server..****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 5:40 PM, Joel R. <joelru...@gmail.com>
>>>>>>>>>> wrote:****
>>>>>>>>>>
>>>>>>>>>> Where are you running this simulation code?  You need to make
>>>>>>>>>> sure it's in a location that is sync'd with the server.
>>>>>>>>>>
>>>>>>>>>> Also, I would not use GetNetworkOrigin, as I have seen it can
>>>>>>>>>> become too jittery, use GetLocalOrigin instead.  ****
>>>>>>>>>>
>>>>>>>>>> ** **
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 4:25 PM, Psy_Commando <
>>>>>>>>>> psycomma...@gmail.com> wrote:****
>>>>>>>>>>
>>>>>>>>>> Doesn't the client interpolate automatically ? I thought that was
>>>>>>>>>> what the Network origin was for.
>>>>>>>>>> Anyways, I'm not really sure what they did but here's the
>>>>>>>>>> handling code I'm using, its mostly code made by the Eternal Silence 
>>>>>>>>>> guys
>>>>>>>>>> modified to fit.
>>>>>>>>>> http://pastebin.com/6kVZJxiD
>>>>>>>>>>
>>>>>>>>>> I think they basically just use vphysics for collisions, they
>>>>>>>>>> rotate the angles and velocity directly to steer the craft.
>>>>>>>>>>
>>>>>>>>>> And also FakeJitter should be 0, I never changed it.****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 4:58 PM, Joel R. <joelru...@gmail.com>
>>>>>>>>>> wrote:****
>>>>>>>>>>
>>>>>>>>>> Also, when you added fakelag to reach about 300 ping, the red and
>>>>>>>>>> green crosshairs were still sync'd perfectly.  This should not be 
>>>>>>>>>> possible
>>>>>>>>>> unless you are extrapolating on the server, but even then it would 
>>>>>>>>>> go way
>>>>>>>>>> off on turns.****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 3:55 PM, Joel R. <joelru...@gmail.com>
>>>>>>>>>> wrote:****
>>>>>>>>>>
>>>>>>>>>> You will need to smooth out the origin and angles.  I'm not sure
>>>>>>>>>> why but the physics simulation cannot be followed without doing some 
>>>>>>>>>> sort
>>>>>>>>>> of smoothing to the origin/angles.  What I did was interpolate them 
>>>>>>>>>> by a
>>>>>>>>>> small amount to smooth out the kinks.  I created a "fakeOrigin" and
>>>>>>>>>> "fakeAngles" that are smoothed to the real values.****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 3:51 PM, Psy_Commando <
>>>>>>>>>> psycomma...@gmail.com> wrote:****
>>>>>>>>>>
>>>>>>>>>> using vphysics****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 4:48 PM, Joel R. <joelru...@gmail.com>
>>>>>>>>>> wrote:****
>>>>>>>>>>
>>>>>>>>>> Is the entity simulated using the physics object or bbox and
>>>>>>>>>> gamemovement?
>>>>>>>>>>
>>>>>>>>>> ****
>>>>>>>>>>
>>>>>>>>>> On Tue, Apr 17, 2012 at 3:44 PM, Psy_Commando <
>>>>>>>>>> psycomma...@gmail.com> wrote:****
>>>>>>>>>>
>>>>>>>>>> I'm having a problem with an aircraft vehicle I've made. I asked
>>>>>>>>>> the Eternal Silence team for some pointers and they've shared their 
>>>>>>>>>> flight
>>>>>>>>>> code with me, but I'm still getting this weird jerkiness, even at 
>>>>>>>>>> 0ms of
>>>>>>>>>> ping. I'm using the Alien Swarm SDK for this.
>>>>>>>>>>
>>>>>>>>>> Here's a video to show the problem :
>>>>>>>>>> http://www.youtube.com/watch?v=zs64WcoD_IQ
>>>>>>>>>>
>>>>>>>>>> I'm really not sure what's the problem, I can post parts of the
>>>>>>>>>> code here if that would help.
>>>>>>>>>> Any suggestion would be greatly appreciated.****
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders*
>>>>>>>>>> ***
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders*
>>>>>>>>>> ***
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders*
>>>>>>>>>> ***
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders*
>>>>>>>>>> ***
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders*
>>>>>>>>>> ***
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders*
>>>>>>>>>> ***
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>>  ****
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>>>>>>>>
>>>>>>>>>> ****
>>>>>>>>>>
>>>>>>>>>> ** **
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>>>> archives, please visit:
>>>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>>> archives, please visit:
>>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>>> archives, please visit:
>>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> To unsubscribe, edit your list preferences, or view the list
>>>>>> archives, please visit:
>>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> To unsubscribe, edit your list preferences, or view the list archives,
>>>>> please visit:
>>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> -Tony
>>>>
>>>>
>>>> _______________________________________________
>>>> To unsubscribe, edit your list preferences, or view the list archives,
>>>> please visit:
>>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>>
>>>>
>>>>
>>>
>>> _______________________________________________
>>> To unsubscribe, edit your list preferences, or view the list archives,
>>> please visit:
>>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>>
>>>
>>>
>>
>>
>> --
>> -Tony
>>
>>
>> _______________________________________________
>> To unsubscribe, edit your list preferences, or view the list archives,
>> please visit:
>> https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders
>>
>>
>>
>
_______________________________________________
To unsubscribe, edit your list preferences, or view the list archives, please 
visit:
https://list.valvesoftware.com/cgi-bin/mailman/listinfo/hlcoders

Reply via email to