I can't say this is really PMDG's fault (maybe more FSX's ridiculous wind modelling at higher altitude), but with any wind shift and resulting change in IAS, the autopilot "very aggressively" uses pitch to get to the desired IAS as soon as possible. Even with a wind smoothing addon like ActiveSky that limits the wind changes to a few knots per second, it can still cause the autopilot in climbout mode to go from 1500fpm to 7700fpm (!!) within just a few seconds, as I saw earlier tonight. That kind of change is easily a couple/few g's and would make a lot of passengers vomit and/or scream. Also in the real world, no 738 would ever climb at 7700fpm.So this isn't so much a bug report as a feature request for SP2: could the autopilot maybe limit the climb/descent rate to 3000fpm or so, or if it's calculating g forces please limit it to maybe 0.9-1.1g's to be more realistic? Also weight things to use throttle more for speed control, instead of pitch? 8000fpm in a climb is not at all realistic for any 737, let alone the autopilot pulling 4 g's in the climb and negative g's in a descent.

I can't say this is really PMDG's fault (maybe more FSX's ridiculous wind modelling at higher altitude), but with any wind shift and resulting change in IAS, the autopilot "very aggressively" uses pitch to get to the desired IAS as soon as possible. Even with a wind smoothing addon like ActiveSky that limits the wind changes to a few knots per second, it can still cause the autopilot in climbout mode to go from 1500fpm to 7700fpm (!!) within just a few seconds, as I saw earlier tonight. That kind of change is easily a couple/few g's and would make a lot of passengers vomit and/or scream. Also in the real world, no 738 would ever climb at 7700fpm.So this isn't so much a bug report as a feature request for SP2: could the autopilot maybe limit the climb/descent rate to 3000fpm or so, or if it's calculating g forces please limit it to maybe 0.9-1.1g's to be more realistic? Also weight things to use throttle more for speed control, instead of pitch? 8000fpm in a climb is not at all realistic for any 737, let alone the autopilot pulling 4 g's in the climb and negative g's in a descent.

something is up in your setup.. mine does not do that.. Are you flying online? If you are and using fsinn make sure the weather on fsinn is shut off.

i doubt the g forces were that extreme. although i've never seen a climb/descent as high as 8000 ft/min i do know that climbing or descending isn't really percieved by the passenger like many think it is. you could be doing a 4000 ft/min descent and passengers wouldnt know the difference if you changed it to 1000 ft/min. i think in the sim since we have no physical reference to what is going on we focus on numbers and the projected image for feedback and that can cause certain things to be exaggerated.

@mikea76 I do use FSInn, and the weather is disabled, but I also see it offline. Regardless of the weather, the autopilot should still be limiting climb and descent rates to some maximum g force, if not for passenger comfort then for structural limitations. 7700fpm is just not realistic in a climb.Akzach -- well the g force comes from the rate of change of vertical velocity, so if you go from 4000 fpm to 1000 fpm in 1 second it will feel a whole lot different than doing it over 30 seconds. I saw my climb rate go from 1500 to 7700 in just a few seconds, which would be a really big g force which is why I'm sure many of my virtual passengers vomited.

@mikea76 I do use FSInn, and the weather is disabled, but I also see it offline. Regardless of the weather, the autopilot should still be limiting climb and descent rates to some maximum g force, if not for passenger comfort then for structural limitations. 7700fpm is just not realistic in a climb.

Your right what you describe is not a realistic climb and what I am saying is it is not a characteristic of the NGX.. I have only seen that type of behavior you describe from the ngx ONCE.. it was after a fresh install of FSINN and I forgot to turn the weather off, the plane was going nuts going from 1800 or so in the climb to 6-7k instantly.. IAS was erratic when this happened too. Other than that time I screwed up with the FSINN setting I get normal climb rates. With a issue like this the first place I would look is something interfering with your weather, thats why I asked about fsinn. so maybe something else is.. Paul Deemer ( I think I spelled his name correctly) has a great write up on how to setup activesky and fsuipc I would read it and maybe it is something in your settings

Your right what you describe is not a realistic climb and what I am saying is it is not a characteristic of the NGX.. I have only seen that type of behavior you describe from the ngx ONCE.. it was after a fresh install of FSINN and I forgot to turn the weather off, the plane was going nuts going from 1800 or so in the climb to 6-7k instantly.. IAS was erratic when this happened too. Other than that time I screwed up with the FSINN setting I get normal climb rates. With a issue like this the first place I would look is something interfering with your weather, thats why I asked about fsinn. so maybe something else is.. Paul Deemer ( I think I spelled his name correctly) has a great write up on how to setup activesky and fsuipc I would read it and maybe it is something in your settings

But weather shouldn't matter. Real world aircraft encounter sudden gust fronts, or microbursts, and their autopilots don't violently pitch up or down to maintain airspeed like this version does. There's a balance between pitch and throttle, and I think the current code just relies too much on pitch. For the next update I hope they scale that back, put more emphasis on throttle, and put in limits on g forces.

But weather shouldn't matter. Real world aircraft encounter sudden gust fronts, or microbursts, and their autopilots don't violently pitch up or down to maintain airspeed like this version does. There's a balance between pitch and throttle, and I think the current code just relies too much on pitch. For the next update I hope they scale that back, put more emphasis on throttle, and put in limits on g forces.

I understand but what I am talking about is FSX weather issues causing unrealistic wind shifts. From your first post you said you have activesky doing the wind smoothing. Maybe some other setting is undoing that.What I have been trying to explain is the NGX does not do what you describe even if I fly the thing into the middle of a t-storm. So this problem you are having is not a characteristic of this add-on. You have a issue on your end and I am trying to help point out something that could be causing it.. such as two programs causing a weather conflict. Might not be the cause of why your vnav is acting up but it's the first place I would look. Maybe I am not doing a good job of explaining the FSX weather bugs could somebody else jump in that is better at explaining it?

I understand but what I am talking about is FSX weather issues causing unrealistic wind shifts. From your first post you said you have activesky doing the wind smoothing. Maybe some other setting is undoing that.What I have been trying to explain is the NGX does not do what you describe even if I fly the thing into the middle of a t-storm. So this problem you are having is not a characteristic of this add-on. You have a issue on your end and I am trying to help point out something that could be causing it.. such as two programs causing a weather conflict. Might not be the cause of why your vnav is acting up but it's the first place I would look. Maybe I am not doing a good job of explaining the FSX weather bugs could somebody else jump in that is better at explaining it?

You're explaining it well, and I totally agree. What I'm saying is regardless of whatever buggy weather and unrealistic wind there is in FSX, and no matter what add-on people have or not, the PMDG autopilot should still be limiting climb/descent rates to realistic levels. Please PMDG, just put in limits!

I'm sure PMDG might comment on this but they might also remind the OP that it is forum rules to sign your posts with your real name.Otherwise they might remove them.There are tons of past posts on this and maybe a good search might help.

How do we sign our posts here? Just put your name at the end?Well I'm Gabriel Keewaydin, mods can you maybe affix that to my posts here, sorry I didn't know the rules.--Gabriel Keewaydin

Hey Gabriel,I have a couple thoughts on the subject:First off, I experience the exact same phenomenon usually at least once per flight, so I'm totally with you on your frustration. I've seen 4500fpm ocilations that would absolutely not occur in real life. However, PMDG no doubt coded the VNAV correctly. If such a windshear condition did occur in the real world, the real NG would preform exactly as you saw in the simulator. Your request for them to impose clime/desent limits, although reasonable, is not true to the real aircraft and doesn't actually address the real issue which is the FSX weather engine. Imposing such a limit may inhibit an intentional senario wherein you need that kind of climb preformance. Activesky is alot better that the default at handling wind changes, but still fails epically at times. I personally would like to see someone develop a smoother weather engine. Anyone. In the mean time, some of the guys suggested that the best option is jointly using AS6/ASE for wind (is one smoother?) and a registered FSUIPC for pressure. That might help you somewhat.From a pilot's perspective, when the wind starts to shift (you can usually tell its coming), engage VERT SPEED mode untill the change is over. This trick makes my flights a lot cleaner. You'll still experience turbulence, but that at least is realistic.I hope this helps you out.-Nate

PMDG no doubt coded the VNAV correctly. If such a windshear condition did occur in the real world, the real NG would preform exactly as you saw in the simulator.

No! They did not code it correctly. That's the whole point. The real world NG does not violently pitch up into a 4g climb when the wind shifts, which would cause real-world passengers to scream and vomit. The code needs to be changed so that no matter what the weather conditions, or however the wind changes, the plane does not violently pitch upward or downward.I agree we can compensate for this bug by clicking out of vnav mode and doing things manually all the time, dialing in the speed and climb rates in order to keep things realistic. I do this myself, but this just isn't realistic. I just hope PMDG can address this problem for the next release!--gabe

I think it's the weather in your FSX that's broken. I use ASE and I've never experienced the violent pitch changes you've described, even in thunderstorms etc.Have you followed the advice in this thread:http://forum.avsim.net/topic/350389-my-weather-config-for-pmdg-planes-part-ii/page__hl__fsuipc

No! They did not code it correctly. That's the whole point. The real world NG does not violently pitch up into a 4g climb when the wind shifts, which would cause real-world passengers to scream and vomit. The code needs to be changed so that no matter what the weather conditions, or however the wind changes, the plane does not violently pitch upward or downward.

I don't think you are understanding this Gabriel.The weather in the real world does not shift like you see in the sim.If, in the real world, "god" decided to introduce a weird bug into the real weather, that caused the same wind shifts you see in the sim... then the real 737 would behave the same way.Thankfully, god [if he exists] is better at generating the real weather than Microsoft are in the sim.The bug is in flight sim's weather, not the NGX.You can't expect PMDG to dumb down their flight model becuse Microsoft screwed up. Better for you to sort out the sudden wind shift issue yourself, like the rest of us have with FSUIPC or other method.For me, using REX weather, and a registered version of FSUIPC, with wind smoothing and turbulence suppression engaged... I have never, not even once, seen the crazy climb rates you mention.Once again, it is not an issue with the PMDG 737 NGX.

Have you tried to turn off thermals in FSX?

I don't think you are understanding this Gabriel.The weather in the real world does not shift like you see in the sim.If, in the real world, "god" decided to introduce a weird bug into the real weather, that caused the same wind shifts you see in the sim... then the real 737 would behave the same way.Thankfully, god [if he exists] is better at generating the real weather than Microsoft are in the sim.The bug is in flight sim's weather, not the NGX.You can't expect PMDG to dumb down their flight model becuse Microsoft screwed up. Better for you to sort out the sudden wind shift issue yourself, like the rest of us have with FSUIPC or other method.For me, using REX weather, and a registered version of FSUIPC, with wind smoothing and turbulence suppression engaged... I have never, not even once, seen the crazy climb rates you mention.Once again, it is not an issue with the PMDG 737 NGX.

I understand your point, but this is an FSX product, not a real 737 avionics product. It should take FSX limitations and flaws into account, I think. All I know is when I look down and see it climbing at 7,700 feet per minute, it's just not realistic. No matter what flawed data is to blame (FSX, FSUIPC, REX...), PMDG's autopilot simply should never be climbing at 7,700 feet per minute, ever. They need to put in limits, that's all. Limit the change in climb/descent rate to keep the g force felt by passengers to realistic levels!