Microsoft FS2020 confusion

[Richard McDonald Woods 168]

I remain confused about future flight simming with MSFS 2020. The function split between FS2020 and other commercial flight simulation developers is quite unclear to me.
We have now had reveals regarding scenery (including elevation data), aircraft models, aerodynamics and cockpits, apparently leaving 3rd party developers to develop only aircraft models, airports and associated functions.
There can be little doubt that Microsoft intends to support 3rd party developers by giving access to a new SDK.
So it appears to me that Microsoft could take one of the following three routes in their development:

1. Develop their own aircraft simulators largely to demonstrate and test the new SDK
2. To provide a library of all required simulation objects that developers require for embedding in their own aircraft simulators
3. To develop their own fleet of simulated aircraft in partnership with at least Boeing and Airbus.

Can anyone caste light for me?

[chapstick 2,649]

Why are so many people overthinking this? A new Flight Simulator is coming out. Add-ons will be made for it. People will exchange money for those add-ons.

Done. 

Edited November 28, 2019 by Chapstick

[b737800 470]

Microsoft are effectively just upping the bar for the content that 3rd party developers need to create, while giving them a bigger box of tools to do it with.

In other words, they are saying - this is the quality of content that we can provide out of the box. Now go create something even better and people will buy it.

Got to be good for all of us.

[Krakin 4,930]

4 hours ago, Richard McDonald Woods said:

We have now had reveals regarding scenery (including elevation data), aircraft models, aerodynamics and cockpits, apparently leaving 3rd party developers to develop only aircraft models, airports and associated functions.

Microsoft are not going to make every plane on the planet so there will be plenty room for 3pds to make all sorts of cockpits.

[Claviateur 789]

I can`t wait to explore the SDK but I don`t think because the engine is advanced and many great features are there, out of the box, we should worry about add-ons.

I think we are not used to see that many quality features by default in a simulator title....So it's simply a relative thing. I am confident Addons will be able to push the details and quality to higher levels than the default features that are already wonderful as they are...

This applies to aircraft indeed, but from a scenery developer point of view, I look forward to have user friendly and flexible SDK tools to push the scenery details even further... For example, to remodel the terrain components on a granural level. This will make the simulator have a great transition from an already rich and convincing global data coverage by default to highly detailed customized areas by 3rd party.

Even if the scenery is detailed by default, developers can finetune or regenrate even more sophisticated and customized procedural areas on small or regional level (ex: finetune architecture types per area inside a specific country), as well as adjust the terrain elevations and components for a specific small area (ex: airport elevations, tune the roads leading to the airport or transform an area that was a forest to a suburb...).

Again, it's all relative and the same goes for aircraft. If MS/Asobo raised the bar in this version in terms of the details and quality of the default aircraft you get, then the addons developers/teams will simply have to raise that bar even higher.

All in all, I feel the same tradition of the old MSFS titles continues (SDK and open for developement) but not we have a lot of value and quality already out of the box...

 

 

Edited November 28, 2019 by Claviateur

[Claviateur 789]

19 minutes ago, Krakin said:

Microsoft are not going to make every plane on the planet so there will be plenty room for 3pds to make all sorts of cockpits.

Exactly and perhaps adapt old FSX models to the new dynamics and 3D standards...

[FishermanIvan 110]

1 hour ago, Claviateur said:

Exactly and perhaps adapt old FSX models to the new dynamics and 3D standards...

If the 748 is anything to go by, I wouldn't spend a penny on anything that was in FSX.  The bar has been raised.  Time for people to meet it.

[Claviateur 789]

4 minutes ago, FishermanIvan said:

If the 748 is anything to go by, I wouldn't spend a penny on anything that was in FSX.  The bar has been raised.  Time for people to meet it.

Yes ideally models must be created from scratch for the new platform. 

I am one of those who can wait for addon aircraft to be created from scratch with the capablility of the new simulator dynamics... I just don't want to get converted fancy FSX models with a so so flight model conversion and be back to the "fly on rails" experience I can`t stand at all after spending 4 years with X-Plane 11... 🙂

Edited November 28, 2019 by Claviateur

[HighTowers 1,407]

Wasnt too thrilled about hearing a legacy mode.  We have been through that enough.  Anything from the past could very well cause performance issues, bring in more bugs, and just doesnt take advantage of all the cool new features imbedded in the sim.  Further, I hope it doesnt give an excuse for bare minimum ports and then charge for it.  If the old FSX/P3D addon gets ported, it should be held to a higher stand that clearly MS/AS has set.  

[Lorby_SI 1,972]

13 minutes ago, HighTowers said:

Wasnt too thrilled about hearing a legacy mode.  We have been through that enough.  Anything from the past could very well cause performance issues, bring in more bugs, and just doesnt take advantage of all the cool new features imbedded in the sim.  Further, I hope it doesnt give an excuse for bare minimum ports and then charge for it.  If the old FSX/P3D addon gets ported, it should be held to a higher stand that clearly MS/AS has set.  

Let us hope that the community and the reviewers will quickly distinguish the chaff from the wheat. For me, it doesn't really matter if it is a legacy product causing trouble or just a badly made native one. 

We have been in this exact situation with P3D all the time. Yet, performance issues were always attributed more to the simulator itself than to the fact that legacy code, models, animations, gauges, textures, terrain, etc, etc, etc were being used. Quite the contrary, people were told to disable the content error logging that the sim provided, so they didn't have to look at the thousands of errors in the addons. I never really understood that kind of "ostrich" reasoning...

Best regards

Edited November 28, 2019 by Lorby_SI

[n4gix 4,899]

There are two major "flaws" with P3D's "ContentError.txt" feature:

1. It doesn't limit itself to only the current aircraft being loaded. It generates error messages from every aircraft the user has installed, sometimes none of the errors pertain to the aircraft being loaded!
2. Often the "errors" are broken into many entries instead of just one, which only makes the list longer for no good reason.

[Lorby_SI 1,972]

Just now, n4gix said:

There are two major "flaws" with P3D's "ContentError.txt" feature:

1. It doesn't limit itself to only the current aircraft being loaded. It generates error messages from every aircraft the user has installed.
2. Often the "errors" are broken into many entries instead of just one, which only makes the list longer for no good reason.

Hello Bill,

does that really matter? The content error log is supposed to be totally, completely empty. For my programmer's heart, one entry is already too much, since I cannot fathom what the consequences are. Let alone more than a thousand, like in my own simulator installation - that just can't be good.

Turning the log off and then complaining about micro stutters, crashes or low FPS is IMHO somewhat contradictory.

Best regards

[n4gix 4,899]

1 minute ago, Lorby_SI said:

Hello Bill,

does that really matter? The content error log is supposed to be totally, completely empty.

I probably should have listed as a 3rd "flaw" when the log lists an actual "non-error" as an "error." I've seen this more than once. Or, more often the "error message" is so vague that it's impossible to find, such as ".... a missing space". Just how is one to scan the modeldef.xml for a "missing space?"

[Lorby_SI 1,972]

5 minutes ago, n4gix said:

Or, more often the "error message" is so vague that it's impossible to find, such as ".... a missing space". Just how is one to scan the modeldef.xml for a "missing space?"

That is a good example. The missing space character is in the RPN animation code right? So the sim has to deal with an exception when the model is being shown, which causes a tiny delay in execution?

How do you find it? Using your eyes and a lot of time of course. I'm sorry that you don't like doing that. I have had to do this very same thing as a programmer more times that I can count, and unspecific error messages are the essence of a programmers life too. The customer will still not accept bugs, even if it is a major effort to find them (I was in individual enterprise software development where there is only one customer who has to pay for the whole thing. Millions usually.)

Best regards

Edited November 28, 2019 by Lorby_SI

[n4gix 4,899]

21 minutes ago, Lorby_SI said:

That is a good example. The missing space character is in the RPN animation code right? So the sim has to deal with an exception when the model is being shown, which causes a tiny delay in execution?

How do you find it? Using your eyes and a lot of time of course. I'm sorry that you don't like doing that.

The "missing space" could be anywhere, but quite often it can be as simple as "...)}" which is one example that is easily found with a "Find" query. A space is required between the closing parenthesis and the closing bracket. Fortunately I use Visual Studio for all my dev work as it is very XML friendly.

I also build a custom modeldef.xml so where only entries actually used in the specific project are present, thus eliminating multi-thousand lines of "chaff." The aircraft I'm working on at the moment has 29,607 lines. If I had left in the 'default' XML entries it would have been closer to 80k lines!

However I think that my main gripe is being missed here. Even though my current project has no errors at all, the ContentError.txt file still lists errors that are in every aircraft currently installed, and not limited to the specific aircraft being loaded. Why I need to know that another's aircraft has errors is a complete mystery...