A similar statement was made on a French forum. I posted it on page 22: link.

12 minutes ago, RudyB24 said:

A similar statement was made on a French forum. I posted it on page 22: link.

As far as I understand, that information was from a beta tester,and he is saying that the 1000 points only exist on paper right now. He also has a tinted view, if you ask me, when he continues to say that the sim probably will have the features perhaps in 10 years... 

On 8/7/2020 at 8:56 AM, RudyB24 said:

It's mentioned on the Feedback Snapshot, Asobo is working on it:

Insufficient Inertia on Some Planes: IN PROGRESS

This is a post on a French forum by a forum moderator and MSFS tester that was found in another thread here in AVSim:

(Translated by Deepl, maybe Dominique can help out).

 

 

 

Just see your post, sorry and the French post too. Good translation.

On 8/7/2020 at 12:39 AM, dilore said:

Do I understand this correctly, it's basically the FSX/P3D flight model we're getting?

Not the P3D model enhancements, the FSX model..

2 hours ago, robert young said:

If you read in detail their last press release, you learn that they have not in fact achieved that at all.

I have not read that as of yet...if you would have the link to that I would surely be interested in reading it through, because if the dynamics are still fixed/canned according to tables that are expected results instead of actual interaction between moving air and the airfoils then I'd most certainly not bother buying this and I'd rather just stay with what I've got currently with XP11.

15 minutes ago, hangar said:

I have not read that as of yet...if you would have the link to that I would surely be interested in reading it through, because if the dynamics are still fixed/canned according to tables that are expected results instead of actual interaction between moving air and the airfoils then I'd most certainly not bother buying this and I'd rather just stay with what I've got currently with XP11.

If the stall spin test video is based off lookup tables I would be really surprised, it looks extremely well modelled, so unless they have some non public build they are withholding for some reason I find that rumour hard to believe... 

Edited August 13, 20205 yr by High_Alpha

Well, Rob should know, I'm assuming...and he says it's not a new physics engine, so...dunno what to make of that TBH but coming from him it makes me VERY nervous. Was this discussed at all in today's live stream (I missed it)?

Edited August 13, 20205 yr by hangar

38 minutes ago, hangar said:

Well, Rob should know, I'm assuming...and he says it's not a new physics engine, so...dunno what to make of that TBH but coming from him it makes me VERY nervous. Was this discussed at all in today's live stream (I missed it)?

Yes it was discussed in today's (13 Aug) livestream, and if you listen closely starting at about 16:30, it's pretty clear from the discussion on how they were tweaking climb rates, speeds etc that MSFS is using a parametric flight dynamics engine rather than blade-element modelling (the ESP-based FDE in FS/P3D also uses parametric models).  At 17:05 Sebastian Wloch, in describing acft performance modelling, says that "...there are all sorts of parameters in there...", and that they wanted to afford users the ability to tweak some of those parameters.  What was said sounded pretty consistent with what Rob Young posted above.

https://www.twitch.tv/videos/709153754

Now, that said, parametric flight dynamics have served us pretty well over the years...but it sounds like those expecting to see the fine nuances of flight dynamics (slips, adverse roll/yaw, spins etc) based on the advertised 1000-point blade-element modelling may be in for something of a surprise on release day.

 

1 minute ago, w6kd said:

Now, that said, parametric flight dynamics have served us pretty well over the years...but it sounds like those expecting to see the fine nuances of flight dynamics (slips, adverse roll/yaw, spins etc) based on that 1000-point blade-element modelling may be in for something of a surprise on release day.

Well I'm more than a little confused now and also mildly disappointed... Was their original comment about 1000 data points misinterpreted this entire time? Are they using a lookup table system with 1000 reference points? Based off the videos I've seen with stalls and spins it seems like the amount of processing being done is much more complex than is available in the current "legacy" systems... What am I missing?

9 minutes ago, w6kd said:

but it sounds like those expecting to see the fine nuances of flight dynamics (slips, adverse roll/yaw, spins etc) based on the advertised 1000-point blade-element modelling may be in for something of a surprise on release day.

VERY disheartening for me, personally...will surely be waiting until after release now before buying into this. Will wait it out and see how it's received. Thank you so much for this info, much appreciated!

What method they use is irrelevant, frankly. The highest fidelity sims on the planet use lookup tables. The proof is (and will only ever be) in the pudding.

11 minutes ago, High_Alpha said:

Well I'm more than a little confused now and also mildly disappointed... Was their original comment about 1000 data points misinterpreted this entire time? Are they using a lookup table system with 1000 reference points? Based off the videos I've seen with stalls and spins it seems like the amount of processing being done is much more complex than is available in the current "legacy" systems... What am I missing?

I'm not sure at this point.  I think their goal was (perhaps still is at some future point?) something like what XPlane tries to do with a real-time blade-element model.  Stalls and spins can be done with the right kind of parameterization...just because it's parametric doesn't mean it's limited to behave the same as FSX etc.

I won't surmise that FDE fidelity is any better or worse than ESP with Simzilla's default planes...in 4 days when we get our hands on it much more will be known.  I've heard opinion both ways from alpha testers.

 

8 minutes ago, w6kd said:

I'm not sure at this point.  I think their goal was (perhaps still is at some future point?) something like what XPlane tries to do with a real-time blade-element model.  Stalls and spins can be done with the right kind of parameterization...just because it's parametric doesn't mean it's limited to behave the same as FSX etc.

I won't surmise that FDE fidelity is any better or worse than ESP with the Simzilla's default planes...in 4 days when we get our hands on it much more will be known.  I've heard opinion both ways from alpha testers.

 

I fully understand this is all complete speculation, but does this mean my guess that 3rd party payware aircraft will have access to a more refined real time aerodynamics model than the stock included aircraft is a reasonable assumption? I know the SDK is still under NDA (as far as I know). But comments from multiple 3PD's that are considered to be of the highest tier have exclaimed that they have access to a plethora of unique development features that are unavailable in any other simulator... I understand their primary focus is to sell their products, but I find it incredibly hard to believe that they would oversell their upcoming software by such a drastic metric...

Edited August 13, 20205 yr by High_Alpha

18 minutes ago, w6kd said:

Yes it was discussed in today's (13 Aug) livestream, and if you listen closely starting at about 16:30, it's pretty clear from the discussion on how they were tweaking climb rates, speeds etc that MSFS is using a parametric flight dynamics engine rather than blade-element modelling (the ESP-based FDE in FS/P3D also uses parametric models).  At 17:05 Sebastian Wloch, in describing acft performance modelling, says that "...there are all sorts of parameters in there...", and that they wanted to afford users the ability to tweak some of those parameters.  What was said sounded pretty consistent with what Rob Young posted above.

https://www.twitch.tv/videos/709153754

Now, that said, parametric flight dynamics have served us pretty well over the years...but it sounds like those expecting to see the fine nuances of flight dynamics (slips, adverse roll/yaw, spins etc) based on the advertised 1000-point blade-element modelling may be in for something of a surprise on release day.

 

They never had Blade Element Theory to begin with. That's just Austin's obsession. In fact BET is very limited in its inputs. Also, all models require parameters so that alone is not an indication of what flight model they are using. You can use simplified versions of CFD formulas to create a flight model

2 minutes ago, High_Alpha said:

I fully understand this is all complete speculation, but does this mean my guess that 3rd party payware aircraft will have access to a more refined real time aerodynamics model than the stock included aircraft is a reasonable assumption? I know the SDK is still under NDA (as far as I know) But comments from multiple 3PD's that are considered to be of the highest tier have exclaimed that they have access to a plethora of unique development features that are unavailable in any other simulator... I understand their primary focus is to sell their products, but I find it incredibly hard to believe that they would oversell their upcoming software by such a drastic metric...

I wouldn't even try to guess.  I don't think the FDE models in the current batch of advanced P3D add-ons like the PMDG/FSL/Maddog etc leave a tremendous amount of room for improvement, anyway.  Now with sailplanes, singles/light twins, etc the nuanced differences with a good BET model would be quite noticeable to the trained eye.  Turn coordination and engine-out behavior, for example, are simply hideous in the ESP sims.

I have high hopes and somewhat lesser expectations of Simzilla...at least initially.  I'm going to leave everything I've heard--good bad, or indifferent--at the door and form my own first impressions next week.