Actually, while I am sure it wasn't the intention; i have learned quite a lot from reading this thread about the FBW system and the 777.  It's been very educational.

Sorry guys....I wanted to do some test flight and post results, but I had not time at all the last few days.

 

I hope tomorrow, otherwise I am afraid I wont be able for another week or so :-(

 

 

Actually, while I am sure it wasn't the intention; i have learned quite a lot from reading this thread about the FBW system and the 777. It's been very educational.

Yes, that's true. This thread really made me re-think a lot about the correlation of elevator and stabilizer during all phases of flight.

Although I did fortunately not have any problems with trim reaction and imbalances...

Especially the differences between C* and C*U laws (Airbus vs. Boeing philosophy) became much clearer to me. 

The use of elevator and (stabilizer) trim are THE most essential elements of flight. They are the actual "heart of the machine", even more than the engines themselves. The latter may fail, but the elevator and trim functionality must be obtainable anytime.

I favour the Boeing philosophy because it puts the pilot's discretion on top of the list...

Short feedback.

 

I did three short flights trying to trim by feel and trying to trim with the trim ref speed displayed.

 

Trim by feel definately works much better.

It is still a bit harder to get the final trim setting perfect in the PMDG than in the real plane though.

 

This is mainly because the fbw trim ref speed jumps a bit inconsistantly.

Sometimes one kt, sometimes two or 3 kt and sometimes not at all after a single click on the trim button (physical button on my yoke).

As Ryan has explained nothing can be done about this because it is due to the specific yoke's trim repeat rate.

I should have tried what happens with clicking the trim switches on the VC yoke, but I did not. They can't have a trim repeat rate can they? So they should work better?

 

Trimming over large out of trim conditions with the trim ref speed displayed shows that the trim ref speed moves as it should with a long press of the trim button.

Holding the trim button you can see the trim ref speed move quite fast but smooth at about those 10kt per second. (FSX trim rate at mid position for my yoke).

 

The other thing that makes it hard to fine tune final trim setting is that the Saitek Cessna yoke I bought is really too light.

You hardly feel anything once close to in-trim so that, combined with the elevator being near normal authority again, makes it hard to feel you are not quite there yet. You dont really notice you are still out of trim untill you get in that dead band zone where the FBW system changes to its mode where it will start to search (pitch up/down) for its in-trim speed.

 

Which is why I have set the CDU dead zone to be large.....like 5-10% for this yoke.

(I am not saying this is the way to do it.....I only did 3 flights with this yoke......but it seems to work best)

 

So all in all trim by feel seems to work now that out of trim bahavior is symmetric (due to SP1b installed correctly now).

To make setting final trim easier I still recommend displaying the FBW trim ref speed untill you get the hang of how the trim ref speed jumps about.

Rob,

 

although I am not using this aircraft, or even FSX / P3D, reading this thread has provided a worth og excellent information thanks to your persistent testing, and the posts by other PMDG users, and of course thanks to the work of PMDG themselves that provided their users with Sp1b and a plausible implementation of the "Boeing fbw" system used in the 777.

 

I believe the next step would be to completely override MSFS's FDM and use an external flight model for the 777. This would include all of the physics modelling of flight, ground physics, and interaction with weather. Maybe one day PMDG considers this approach for their products and we can get an even better 777 :-)

 

 

You hardly feel anything once close to in-trim so that, combined with the elevator being near normal authority again, makes it hard to feel you are not quite there yet.

 

Yes, agree this is a problem; leads to a bit of "hunting" when one is almost in trim. I am going to experiment with calibrating may Saitek yoke in FSUIPC with a slight exponential curve (so that one has to push/pull the yoke harder to get those small elevator inputs), and see if that helps a bit with intensifying the still not quite in-trim feeling.

 

Rob

I believe the next step would be to completely override MSFS's FDM and use an external flight model for the 777. This would include all of the physics modelling of flight, ground physics, and interaction with weather. Maybe one day PMDG considers this approach for their products and we can get an even better 777 :-)

Much of PMDG's coding is already outside of flight sim, it's been this way for quite some time.

 

They may not use NASA flight dynamics software like Majestic do for the Dash, but as I say, external modelling is not a stranger to PMDG.

Yes, I am aware that both PMDG and A2A, for instance, use external processing, with A2A going even a step further in some aspects pertaining, for instance, to prop effects, but AFAIK, PMDG uses native MSFS FDM, which isn't bad at all, but has it's limitations...

 

Programming a model of the Boeing C*U law and FCS must have been challenging for sure, but also a nightmare when you have to sync it with the internal MSFS FDM.  If PMDG could develop their own physics engine, and use FSX mostly as a visuals / weather generator, I believe it would be possible to achieve an even better overall quality in complex products like the 777 NGX, etc...

If anyone is inferring that the MJC way is the way to go, you should really have a look at the deficiencies of that whole avenue.  Don't get me wrong - it has a bunch of advantages, but those advantages come with some weird disadvantages.  The camera effects get a little messed up (everything is treated like "lock" mode using the default camera - no smooth transitions), which to me isn't a huge deal.  The time is off in that one minute of real time doesn't match up with one minute that the plane sees.  While I know there's a generic FSX issue where the time doesn't match perfectly, it's exacerbated by their implementation of external processing - flights take longer than they should, given the same parameters in real life.  Finally, the plane can't really respond to turbulence or weather very well.

 

Basically, you're flying the plane in a hacked version of slew mode.

 

Companies live and die by what they do differently, and some people like how one company behaves versus another.  What works for one company might not for the other.  To each their own, but in my opinion, it isn't the right way to go.  Sure, it works, and it's a great product with a unique strategy, but that's not to say that there aren't some glaring deficiencies.

I don't like the MJC feel, so that was really not my comparison... there's something "strange" about it...

 

It's more like what I have experienced with PSX and VisualPSX very recently. I have a perfect 744 simulation out of FSX, and use FSX, and even the PMDG 744 model, for the visuals...

 

I'm saying that maybe the effort of building a standalone Boeing 777, 787... simulation outside of FSX, but skipping the "out-of-windshield" visuals that for instance PSX still offers too, would probably give PMDG the chance of getting rid of the limitations imposed by the underlying FDM in MSFS / P3D...

 

That could be a future Project for PMDG :-)

If anyone is inferring that the MJC way is the way to go, you should really have a look at the deficiencies of that whole avenue.

 

Huh... don't know where that came from, nobody claimed that.

 

However...

 

 

The camera effects get a little messed up (everything is treated like "lock" mode using the default camera - no smooth transitions), which to me isn't a huge deal.

 

 

Yes, but never noticed any issues at all. I always hit F11 anyway for taxi.

 

 

 

The time is off in that one minute of real time doesn't match up with one minute that the plane sees.

 

 

Not noticed that. But then I recall the clock issue was fixed back in December 2013.

 

 

 

Finally, the plane can't really respond to turbulence or weather very well.

 

 

Not true at all. Responds great to weather. Initially there was no turbulence, Majestic fixed it with a service pack. Fantastic now in turbulence.

 

To each their own, but in my opinion, it isn't the right way to go. Sure, it works, and it's a great product with a unique strategy, but that's not to say that there aren't some glaring deficiencies.

 

 

 

Totally disagree, and I recall Ryan commenting some time ago when the Majestic NASA software approach was discussed, that it was indeed the "right way to go" for them, the best way to simulate "a free-rotating turboprop" , but not for what PMDG do.

 

 

 I see no "glaring" deficiencies at all. Nothing is perfect, but the Dash is the most enjoyable aircraft I've flown. I must ask you, do you actually have the product?

 

 

 I see no "glaring" deficiencies at all.

 

Not to support Kyle...something i don't enjoy doing really  ...but, while i very much enjoy the MJQ400, if you push it outside the comfortable flight parameters it goes very strange.  Try a stall with a severe nose up AoA, it falls like a paper plane, like the FDE suddenly forgets about the mass of the aircraft.  At extreme control deflections it rolls faster than the DCS Sabre...ok slight exaggeration.  It really is an amazing aircraft but it too has it's issues.

Shhhh! Say nothing to RR, but I still look fwd for the day he announces a PMDG add-on for DCS World v2 :-)

Shhhh! Say nothing to RR, but I still look fwd for the day he announces a PMDG add-on for DCS World v2 :-)

 

Me too, might need a slightly larger area though, the PMDG C-5 from Batumi to Sochi.

 

 

 I see no "glaring" deficiencies at all. Nothing is perfect, but the Dash is the most enjoyable aircraft I've flown. I must ask you, do you actually have the product?

 

Why is that even a question...?  I don't usually criticize things that in depth if I don't own them (exceptions for ATC programs, because I know where and how they're going to fail).

 

Definitely don't own it, though.  I'm clearly just parroting what someone else told me:

 

Here's a question in return, and then a request:

Do you have FS2Crew for the Dash 8?  If you do, switch it to the US voice set...you may recognize the voice...

 

Additionally, I'd suggest having a look at the speed/timing issue.  Ian brings up good points, too, though I don't tend to hit those limits to notice it.  To add even more, the "radar" is laughable at best.