An SDK user said he found the calculated 1000 air points are emtpy variables. Only that his SDK version might  be empty, doesn`t mean that Asobos build is also empty. To conclude now even further is way too far.

 

Nice screenshot of the progress report.

Sorry to say, but must it really take 5 years to realize that something is wrong with inertia and damping?

32 minutes ago, BigDee said:

An SDK user said he found the calculated 1000 air points are emtpy variables. Only that his SDK version might  be empty, doesn`t mean that Asobos build is also empty. To conclude now even further is way too far.

Kinda hard to figure out what's actually going on, isn't it? 🙂 

I think everyone underestimated the amount of work that would go into MSFS.  While I don't think Asobo is in over their heads yet, I have to wonder how close they are.

In any case, MSFS is amazing, and it's likely to get better as time goes on.

Hook

30 minutes ago, BigDee said:

Sorry to say, but must it really take 5 years to realize that something is wrong with inertia and damping?

The problem is that they have been creating a sim of an unreached scale within 5 years. It looks very amateurish, but such things can happen when you have so much to do. I bet some of the planes were only done the past few months.

But considering the madness over Alpha invites, I guess most simmers are happy to get a sim that still needs improvement over waiting.

 

34 minutes ago, BigDee said:

An SDK user said he found the calculated 1000 air points are emtpy variables. Only that his SDK version might  be empty, doesn`t mean that Asobos build is also empty. To conclude now even further is way too far.

Yeah, such reports should not be taken too serious. Especially with rising madness before release.

3 minutes ago, LHookins said:

I think everyone underestimated the amount of work that would go into MSFS.  While I don't think Asobo is in over their heads yet, I have to wonder how close they are.

The most interesting part is how those surfaces are calculated. Lookup tables? Functions? Airflow calculation? I didn't hear anything specific about that yet.

 

22 minutes ago, tweekz said:

 .

 

Yeah, such reports should not be taken too serious. Especially with rising madness before release.

For the record, the information comes from the webmaster of the main French flight simulation site, Pilote Virtuel, an alpha/beta tester  and freeware developer who has the SDK in hands. 

I do not particularly vouch for him but don't dismiss him too lightly. This report motivated my question on Twitter to Sebastian W.  Let us stop calling mad hatters (or rather tinfoil hatters 😁 ) those who bring unpleasant news. They might be right, they might be wrong but look first at the background of who is talking. 

See RudyB24's post on page 22 ot this thread where he refers to the French post and gives a translation

Edited August 14, 20205 yr by Dominique_K

37 minutes ago, LHookins said:

I think everyone underestimated the amount of work that would go into MSFS.  While I don't think Asobo is in over their heads yet, I have to wonder how close they are.

They are getting pushed way too hard for the last days, I feel sorry for the programmers.

21 minutes ago, tweekz said:

The most interesting part is how those surfaces are calculated. Lookup tables? Functions? Airflow calculation? I didn't hear anything specific about that yet.

I've never seen a "lookup table" in anything other than pre-calculated trig functions back before math coprocessors.  Everything you see is table driven one way or another, with numbers from configuration files feeding functions, even with blade element theory and "airflow calculation."  

Oh yes... the earliest COBOL compilers implemented multiplication with lookup tables according to someone who was there at the time.

Robert mentioned the word "emulation" earlier, and that's what it ALL is, no matter how it's calculated.

Just fly.  If it looks and feels and acts right, then it IS right, no matter how it's calculated.

Hook

1 minute ago, LHookins said:

I've never seen a "lookup table" in anything other than pre-calculated trig functions back before math coprocessors.  Everything you see is table driven one way or another, with numbers from configuration files feeding functions, even with blade element theory and "airflow calculation."  

Oh yes... the earliest COBOL compilers implemented multiplication with lookup tables according to someone who was there at the time.

Robert mentioned the word "emulation" earlier, and that's what it ALL is, no matter how it's calculated.

Just fly.  If it looks and feels and acts right, then it IS right, no matter how it's calculated.

Hook

Amen.

6 minutes ago, BigDee said:

They are getting pushed way too hard for the last days, I feel sorry for the programmers.

It is always like that when release day approaches.  One of the best things working for Warbirds was that there were no announced release days, new versions were pushed out when they were ready, not according to some artificial schedule.

Kind of like how A2A does it.  And one reason they make superior aircraft.

"We're releasing it Wednesday.  Have it ready."  -- actual quote heard at Warbirds concerning the boxed release.  Made my blood run cold.  Thankfully I wasn't involved in that part of the project.

Note:  To avoid confusion, I need to mention that the quote was said by a manager from iMagic to one of their programmers brought in to make the box version.  Normally Warbirds was a download.

Hook

Edited August 14, 20205 yr by LHookins

55 minutes ago, LHookins said:

Everything you see is table driven one way or another, with numbers from configuration files feeding functions, even with blade element theory and "airflow calculation." 

Psst... don't say that too loud. 😄

 

55 minutes ago, LHookins said:

Robert mentioned the word "emulation" earlier, and that's what it ALL is, no matter how it's calculated.

Sure. Every model is wrong. But some are more useful than others.

If you look at professional engineering software, they also use quite a lot of streamlines to optimize their aerodynamic design. Not saying that level of detail is needed in flight simulation, but if it is possible - why the hell not?

I think there are also quite some people who are a little mad that they have to relearn everything in terms of modeling.

Edited August 14, 20205 yr by tweekz

59 minutes ago, LHookins said:

I've never seen a "lookup table" in anything other than pre-calculated trig functions back before math coprocessors.  Everything you see is table driven one way or another, with numbers from configuration files feeding functions, even with blade element theory and "airflow calculation." 

The alternative to table interpolation is to use polynomial functions to compute parameters, at least where the functions are continuous.  For a single airplane this might make sense...for a flight dynamics engine that needs to emulate a wide variety of aircraft, tables provide an easy way for FDE designers to approximate and fine-tune the necessary functions, and are much easier to manipulate than having to do curve fits to define function parameters.

The terms in use are somewhat confusing to me..."surfaces" implies to me that aerodynamic forces across the airfoil are being computed based on flow and geometry and then applied, where "points" suggests that moments, probably derived from tables, are being applied at various locations on the acft geometry.  Both methods can produce a good flight model--or a bad one.  I tried several of the early versions of XPlane (last one I tried was version 9 I believe), which use the "surfaces" approach--aka blade/finite element based dynamics, and the planes behaved like they were on the ends of a puppetmaster's strings, with no sense of momentum/inertia when control forces were applied.  A good methodology, but poorly applied, and in the opinion of both myself and many of my r/w pilot pals not nearly as convincing as the old FSX engine with a well designed set of FDE parameters.

The one recent MSFS video that caught my attention and raised my eyebrows was a plane flying in bad weather, and the nose was hunting about with rapid, nearly instantaneous movements that sure looked like what I experienced with that XPlane of old.  I'm reserving judgement for when I get my own hands on the release version, but I know that what I was seeing in that video wasn't how a plane behaves in turbulence and gusty winds.

5 minutes ago, tweekz said:

Sure. Every model is wrong. But some are more useful than others.

"Wrong" implies a binary set of possibilities..."right" or "wrong". 

Better to say that all models are inaccurate to some extent.  The question is whether that inaccuracy, or deviation from reality, is significant enough to render it materially nonrepresentative of what the model is intended to emulate.

1 minute ago, w6kd said:

The alternative to table interpolation is to use polynomial functions to compute parameters,

Polynomial function:  A x squared plus B x plus C.

A, B and C come from tables.  Please, Please, PLEASE don't tell me those are hard coded in any sim! 😄 

I think MSFS still has some fine tuning to do.  If I'm not happy with what we get out of the box, I'll do my own fine tuning.

Hook

I re-watched the aerodynamic video. It was interesting and seeing those points on all the surfaces was pretty cool. It looked like each point hit by airflow interacted to the points beside it to some degree. It was also said during the live stream that some of the parameters will be tweekable by the end user to some degree to fine tune the feel of that aircraft. I have hope this will, overtime, be something that feels more accurate for each simmer.

Lets hope for the best! 

11 minutes ago, w6kd said:

"Wrong" implies a binary set of possibilities..."right" or "wrong". 

Better to say that all models are inaccurate to some extent.  The question is whether that inaccuracy, or deviation from reality, is significant enough to render it materially nonrepresentative of what the model is intended to emulate.

An inaccurate models representation of the real world is... wrong. 😛

 

17 minutes ago, w6kd said:

The terms in use are somewhat confusing to me..."surfaces" implies to me that aerodynamic forces across the airfoil are being computed and applied, where "points" suggests that moments, probably derived from tables, are being applied at various locations on the acft geometry.

Doesn't BET do both of it? Reading in geometry of surfaces and creating data for the sim how to interact with contact points in certain situations?

If I remember correctly I've also spotted a lot of curves in the airfoilmanager. So those functions seem how XP passes those information to the sim?

I just had a short look on those things, so feel free to correct me.

Edited August 14, 20205 yr by tweekz