What stage are we at with this product?

[hootsmangreen]

I'm just wondering when you think this software will be finished? At first I'm blaming myself, like I'm I not flying it correct or do I have a system on that should be off or the other way around. There are so many faults and things that are not operational yet I'm loosing faith, I've been flying PMDG and LVL D stuff for years and where I might not be doing everything by the book I can certainly go from cold and dark, fly on IVO', Descend, Land, and shut down. So what I'm saying guys is I don't think it is me I think it your  software, now you may think because you are half the price of project magenta that it is ok to put out a product so unfinished but I still class £300 plus for a bit of software as an expensive bit of kit and should be problem free. I'll stop moaning now and try and list the problems I have. My biggest problem is with the MCP
1. say I have SPD set at 300 I can be going along no problems the I look at the N1 Red lining If I now click on C/O instead of showing me a mach number it will be something like 477 knts when this is displayed if i now reduce the A/T SPD to 300 it will come back to three hundred.

2. The banking/turning is too shallow. When directed by ATC I am constantly flying through the ILS and having to come back on it, and as you can imagine guys this is embarrassing for my time spent on ivo' this does not happen to me in any other aircraft on FSX.

3. Why don't you adjust the FMC/Auto pilot to begin the turn slightly before the fix then we stay on the magenta line instead of passing it and having to come back on it.

Look I don't want to be all negative because I love the gauges, the fmc, the fact you can scale, move and network is everything I want. I've bought them all FSxpand I've bought PM RJ and PM General aviation and when you guys came along with a cheaper solution to PM 737 I jumped on board quick style but now I wonder if I should have just waited a bit longer to save the cash for PM. But as I've already invested in your software I'll have to stick with it. I know you say in your PDF manual that we are expected to be able to fly a 737 but maybe if you did print a procedure manual for your 737ng then me and others less experienced with self build could at least be able to judge if we are using your software correctly.

Yours Hangin' in There........
Colin


 

[admin]

Hi Colin,

I understand your frustration about the product not being finished yet.  I'm working continuouesly on this software and there are indeed many things not 100 % working yet.  The last couply of months most of my efforts went into programming the instructor station.  But I'm now almost ready with that so I can redirect my efforst again to the system logic and the Glass Flight Deck. 
I cannot say when the software will be 100% finished, because I don't think it will ever be.  And with that I don't mean I won't finish the product, but it means there will always be little bugs, or room for improvement on the A/P for example.  That's software...  And then again:  do you think PM is working 100% ?  And in real life FMC software has to be upgraded as well to fix bugs...

To answer your questions about your problems:

1)  I don't completely understand your problem...   You are flying 300 kts and then do a C/O to mach ?  What mach number do you get ?  Does the aircraft then speed up ?
2)  I will check this.  Can you give me an airport/runway and a STAR/approach for it to test ?
3)  If you are talking about SID/STAR procedures then the FMC is not turning before the waypoints.  When flying airways it should turn before the waypoints.  In real life normally all SID/STAR procedures are using fly-over waypoints.  There are exceptions and these are described in the procedure in the navigation database.  Because I lack this information I implemented the procedures as always fly-over points.  Maybe I should add a configuration option somewhere to enable/disable this.

[dvincent]

And to add a 4th point of intrest:

each time the computer has to make a change in altitude or lateral position, it is reading data from the flightplan or the nav database. Each time there is a read action, there is a control action of the antivirus program wich has at that time the 100% focus of the computer; that way the datastream to the flightcomputer is interrupted wich result in a double delay of the flightcomputer. If the data or the the flightplan are located on a second computer, that second computer has the same delay, so all that together results in a quadruple delay.
Solution: turn off the antivirus during flights BUT don't forget to switch it back in action on the computer connected to the internet afterworth.

[hootsmangreen]

Thanks for your quick response, and I hate having a go at you guys. It is difficult for me to put into words the problems so when I get time over the next couple of weeks I'll film a flight and post it on youtube so you can see exactly what happens. The last few nights I've constantly been trying EGCC - EGLL : T/O Rwy 23R LISTO 1R  LISTO HON then a BNN4A star for 27L at egll on atc he'll tell me leave BNN at 120 then soon he'll turn me a couple of times then intercept the ILS and it just seems to take ages to turn. Anyway no point writing it down I'll record it.

Cheers

Colin
 

[dvincent]

Hi Colin,

I have taken a look at the BNN4A star & approach charts of London...  you are trying to catch a loc/gs with an angle of 152 degrees... that's impossible.
Two problems: 1: you're too close near the interception. 2: impossible angle => leaving BNN @ 120 degrees to an 272 degrees ILS.

Maybe I didn 't catch your message as I should or there is a lack of information about your flight between BNN 120 and EGLL ILS 27. Both can be 

B737-800 Loc interception speed must be at about 180 kts.

[NAX]

It doesn't mather where it is Dirk I have problems with intercepting ILS with much lower angles, type 30-40 degrees. The problem is that the software turns after or when crossing a waypoint, not calculating the shortest route as it should.
To put it in another way, the FMC shal calculate how to fly as much as it can on the route. Flying from point to point and turning after the WP gives less time on the magentaline. Bart states the 737 flies from (and over) point to point, but what I have experienced (in FS) is that the aircrafts never fly over the WP unless it's a straight or almost straight route. If we get this calculation done, we have a flyable software we can use online. Then there is the VNAV, but I can live with that.
As said before, this is advanced and complicated computing. We will get there eventually.

[dvincent]

Hi Nax,

thx for the info. I did not follow this thread so far. Makes it more clear to me now.

[admin]

Hi Nax,

What I said is this:

SID/STAR procedures are usually fly-over points. Normally navdata stipulates for every point when it is fly-over.  Because the lack of this information I implemented it this way.  I think I will add a configuration option to disable this.
Normal airways and direct to points are pass-by waypoints, meaning the aircraft will turn before the point.  This is also implemented this way, only calculating the distance before the point to make a smooth turn is not easy.  Also when the course change is larger than 135 the CDU will also instruct a special path where an overshoot will happen in the route !!!!

From the Smiths CDU Manual:

[NAX]

Hi Bart.
You don't have to shout hehe
I see you got it all covered and that I have misunderstod or not understod at all some vital information.
Anyhow, the path the aircraft flies to intercept ILS 01L from SIGxx STAR at ENGM is not in any book I'm sure.
That is the STAR I have flown most times during testing.

I know you got a heck of a job to get these things right Bart. I will surely be here and  I hope we keep mr. Hotsmangreen with us as well  during further testing and all the way to a fully competent flightmodel.

[themax222]

Hello Bart,

In a beta version, you implented smooth magenta line turns, LNAV worked nice. Any idea of when you will be putting it again in the beta's ?

Thanks for the great job.

Maxime.