Page 1 of 2

Posted: Wed Aug 05, 2015 7:03 pm
by Lewis - A2A
Hi everyone,

There is some odd bug that exists in P3D with all planes, which causes the plane to suddenly jump forward when starting to taxi, and the worst case get stuck on pavement until very high power is applied, then the plane suddenly leaps forward. Our Boeing 377 Stratocruiser suffers from getting totally stuck. This exists on all airplanes on some level, even the included aircraft. The higher the horsepower and heavier the plane, the worse the bug.

In FSX, the exact same planes are perfectly smooth with no hint of this issue.

Here is a video demonstrating it happening even in a light Cessna 172:
https://www.youtube.com/watch?v=mamT3S1iPQU

Posted: Thu Aug 06, 2015 6:03 am
by alaxus
Looking that video I think your trust bug is called the brakes. Recheck all your control assignments to make sure nothing is doubled up. Are you using FSUIPC?

Posted: Thu Aug 06, 2015 8:51 am
by capt_Cronic
this is an old, well known bug,
it DOES happen in fsx at some locations, not all.


Posted: Mon Aug 10, 2015 8:50 am
by Lewis - A2A
Alexus, I made the video to show clearly where the bug is, not the brakes, but on the thrust calculations. Can't you see the sudden thrust jump?

Capt Cronic, not sure about what bug you are referring too, but this isn't it - there is no trace of it in FSX. Our Boeing Stratocruiser literally needs 80% power to move in Prepar3D. It was so bad, during our display at Air Venture / Oshkosh, we had to remove the airplane. This happens on every machine we install Prepar3D on.


Posted: Mon Aug 10, 2015 9:50 am
by capt_Cronic
its as if the wheels are stuck in glue.
it takes full power to break loose then you suddenly accelerate before you can cut the throttle back in time.
this happens on my current install of fsx and p3d.
like i said, its an old bug. google it

Posted: Mon Aug 10, 2015 4:14 pm
by acra24
Do you use Accufeel? This can cause these symptoms in some systems. I did for me but usually only on the Aerosoft airbus.

Posted: Mon Aug 10, 2015 7:01 pm
by Mike Schroeter
This is often an issue with the thrust model rather than the friction, as is often assumed. There really isn't anything too complicated going on otherwise... when thrust exceeds friction you'll accelerate. The friction calculation is pretty straightforward, but often thrust is tuned at the high end but the low end is overlooked, particularly propeller curves.

Hope this helps.
Mike

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Sun Aug 16, 2015 12:12 pm
by Lewis - A2A
I made the video to show the behavior in the thrust is different in P3D than FSX. If you watch the video closely, the thrust jumps at about 1/2 mph. There is no jump in FSX. capt_Cronic, the "old bug" you are speaking of is different - that is related to ground friction. This one is purely related to an immediate jump in thrust at 1/2 mph that is only present in P3D, and I think, this started 1-2 patches ago.

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Mon Aug 17, 2015 5:36 pm
by Mike Schroeter
In your aircraft.cfg, is "low_speed_theory_limit" set to zero or something really low? We did change our assumption that this is at least 1 foot/sec. This prevents sudden spikes and thrust at extremely slow speeds. Normal thrust calculations break down at slow speeds, so this limit is necessary to be set to something reasonable. Most of our stock airplanes are set to 80, but I have set it to as low as 15 with good numbers. A value of zero could potentially cause what you are seeing.

Hope this helps.

Mike

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Tue Aug 18, 2015 7:40 pm
by ScottMG
Mike,

This is Scott from A2A, I've been testing this and thanks for the reply. I checked the "low_speed_theory_limit" and it is set to the default 80, and adjusting this has no effect on this.

I re-tested this with the default Mooney Bravo, using the exact same plane in FSX and P3DV2, and these are the results:


FSX:
1,400RPM required for the aircraft to start moving
At 2,000RPM, at 0 mph, thrust is 419. Releasing the brakes, the thrust smoothly decreases, as expected

P3DV2:
1,800RPM required for the aircraft to start moving
At 2,000RPM, at 0mph, thrust is 158. Releasing the brakes, just after the aircraft starts to move, thrust spikes to 260 then slowly decreases


The negative side to this new code is, all aircraft likely take a lot more power to start moving (even 1,400RPM is too high, let alone 1,800RPM). The sudden leap forward looks odd, and worst for us, this code has ruined the taxi behavior for our entire A2A fleet on P3D, that we worked so hard on for FSX to get just right.

Now, if you say to us "we can't fix this" for whatever reason, we can adapt. We will just have to make some kind of kludge to mask this behavior (something like multiplying the thrust X2.5 when static and have it rapidly lower as the aircraft starts to move). But we don't want to go through this effort, as we'd prefer not to use a kludge fix, especially if down the road, this is addressed or changed on your end.

I hope you do see, the severity of this bug and how, as it is, we are unable to have proper taxi behavior. For me, personally, I would be more than happy if whatever change was made was either just reversed (same as FSX), or, if there is some way we can by way of perhaps a CFG file, have this code disabled for our fleet at least. But, I do hope you see this behavior is not correct and I don't see any reason why anyone would want it this way. It seems like it should be relatively straight forward to fix since it works quite nicely in FSX.

I hope my above post doesn't come across as too critical, it's just this bug has been a thorn in our side for some time now and we just need this to be fixed and behind us.

Best regards,
Scott.

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Fri Aug 21, 2015 2:17 am
by Mike Schroeter
I've changed the low end limit for the next release to be more similar to what it was previously. This is possible as we've since added another limiting function to prevent extremely high thrust.

Note also, that we had added a scalar for low end thrust in a recent release... see "low_speed_theory_scalar" in the SDK.

Hope this helps,
Mike

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Thu Sep 03, 2015 11:58 pm
by freequest
this is actually related to ASN wake turbulence i had this effect in every aircraft in FSX-SE and P3D you pretty much have to turn it off.

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Fri Sep 04, 2015 2:04 pm
by alaxus
I seeing this now as well. I have a new p3d installation and the aircraft that I am using won't even move at full throttle on grass. Once its moving (using slew to get it going) its OK, but that initial movement from a dead stop just doesn't happen.

I have fiddled with he propeller configs and managed to get the plane moving by changing the pitch and gear reduction. Still requires a lot of power though.

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Fri Sep 04, 2015 2:26 pm
by Martyson
alaxus wrote:I seeing this now as well. I have a new p3d installation and the aircraft that I am using won't even move at full throttle on grass. Once its moving (using slew to get it going) its OK, but that initial movement from a dead stop just doesn't happen.

I have fiddled with he propeller configs and managed to get the plane moving by changing the pitch and gear reduction. Still requires a lot of power though.


I would like to test this.

Which default aircraft and default airport/location showed this problem?

Re: Sudden thrust BUG – from stopped to barely moving

Posted: Sun Sep 06, 2015 10:26 am
by alaxus
Martyson wrote:I would like to test this.
Which default aircraft and default airport/location showed this problem?


Nope not a default aircraft. I am using this one
https://flyawaysimulation.com/downloads/files/15765/fsx-fly-synthesis-texan-club-550-lsa/

With ORBX Australia SP4, Pilots MESH 2010, Australia Holgermesh, OZx Australian airfields.

Its just the initial breaking away is the issue.

I should also mention that I am still using 2.4.