I have a question about that. I am not an IT person. I read that DX12 will dramatically increase the number of draw calls that can be made? I assume this is related to object batching. Assume 64 bit is coming as well. If so, does this mean that the simulator will be able to call more scenery in a given time frame with the same resources? Then, does that also mean that it will be easier for a developer to create a happy medium between increased memory use; and increased draw calls, to limit the need to access unreasonable large amounts of avaiable memory space? If the reply would be too technical for me, then it's OK...take a pass. I am just fascinated by what I see as some revolutionary changes that may be coming to the most demanding software I have ever used.

 

Yes, that should mean precisely that. Current problem with the CPU is high number of draw calls. In P3D, they have been optimized, or so they say, however when you check how much FPS does autogen cost you and then compare which component is being loaded (CPU), it's pretty clear the problem is still large number of objects. As I understand it, the object batching in FSX is from one standpoint very clever, not batch tree after tree, but in masses - hence popping. P3D solves that differently, however there are still very high numbers of draw calls. And they don't only come from trees and scenery objects, but also from the VC and objects therein. Giving the sim the DX12 ability, it would most likely lower the work CPU has to do, and finally move the load seriously towards the GPU. Granted, you would have to have a monster-GPU to cover for GPU effects (shadows et al), along the high number of objects pushed from the CPU through the new API, but hey, given the ability you can have SLI, hopefully correctly implemented, the doors are open that way... unlike CPU.

That's as far as my technical understanding goes.

It's a fact that flight simulator is probably the most demanding "video-game". But the future seems to hold some really awesome technical achievements. Let's just hope P3D can use them AND the guys at LM have the skills to implement them.

 

 

As a mathematician, I have no reason to prefer one model over the other: final performance compared to the real thing it the only thing that counts.
 
What really bothers me is what we mean by "improve the physics" of the flightmodel. I am sure that the guys at LM know something about the physics of flight, its simulation, and behaviour inside and outside the envelope.
 
But the real question is: what are we actually asking for?

 

Andrea, agree ... and to further add to that complexity is how does one represent that in a 2D coordinate system emulating a 3D space.  There MUST be and will be accuracy lost in that translation, just no way around it.

 

 

I'm confused, I saw a post last week about P3D not going 64bit, thought that weird and now you mentioned 64bit version.
So P3D will go 64bit?

 

Fabio, I deleted your duplicate post.  You saw a post from one person relaying information that was 5 months old.  I'm not going to get into details beyond that because I've been asked not to and as such respect that.

 

 

 

 

A few items I'd put forward for consideration are: Taxing and ground friction.

 

Absolutely encourage everyone provide LM feedback, using me as source will not make it more or less likely to happen.  LM's feedback seciton is a good start: http://www.prepar3d.com/forum-5/forum/prepar3d-development/prepar3d-feature-requests/ .  The more details you can provide the better ... no guarantees one's feedback will get implemented, but don't take that to heart.  Be detailed, be accurate, don't make demands.  Items most likely to make it would be ones that don't cause massive changes (read don't break compatibility) and don't have a huge performance impact.  But I certainly think it's a viable request to allow more flight dynamics option via the SDK.

 

Cheers, Rob.

 

 

Let's just hope P3D can use them AND the guys at LM have the skills to implement them.

 

Thanks for the information. I think they will. I made the change from FSX to P3D. I am sure that it was the correct decision for me. I added the qualifier, as I do not want to convey criticism towards FSX/SE/XP. They are all good in their own right. I just think the direction LM has gone to this point, is where I want to be. I don't know where it's going, but I don't think LM would pass up the opportunity to provide a better experience for their customers and developer partners, provided it pays the bills. Rob has shown some class by divulging what he can and making it clear that he will not betray his associations. I won't degrade that by asking questions that he won't answer.

Rob has shown some class by divulging what he can and making it clear that he will not betray his associations. I won't degrade that by asking questions that he won't answer.

There is an axiom that is pretty much universal among the world's military that goes, "Never give an order you know won't be obeyed."

 

Similarly, another axiom that may just as well be articulated is, "Never ask a question you know cannot be answered." :Applause:

Not clear on this, to be exact, surely the simulator FSX/P3D/XPX is  a 3D coordinate system and thus the dynamics are 3D. Therefore the vaunted laminar flow and so on, with aircraft reacting for example to cross and headwinds and the 6 degrees of freedom.   Where the 2D comes in is the representation of that 3D space on a 2D medium like a monitor?

Real world flying in a light aircraft is MUSH easier than in a sim. I think it is all because of full 3D wraparound visuals (yes, at as high "frame rate", but I don't think that's the reason), plus the feel, the full seat of the pants understanding of what the aircraft is doing, what it is experiencing. I've tried all sorts of addon gadgets for the latter, to no real avail, but the former is now a waste of time because my RP has reached the stage where I have such narrow tunnel vision that 3D and wrap-around vision doesn't even really apply in the real word, let alone in a sim! ;-)

 

Pete

It's funny you mention this Pete as the real life Q400 pilot we got in to do the Majestic Dash 8 training films couldn't fly the sim version of the aircraft he'd flown perfectly well *earlier that very morning* to save his life. In fact, it became a running gag that we suspected he wasn't a real Dash 8 pilot at all, but had just swallowed the FCOM.

 

Jokes apart, the reality is that although the Q400 mirrors the bank angles and pitch and power settings of the real one, the way it feels in the sim is totally different. His verdict was that flying the real one was easier in every way, from the simple manipulation of the controls, to the peripheral vision outside and the fact that the seat of your pants cues are totally absent in the sim.

Going to have to agree with alot of this. Unless we are real world pilots, then how can we demand changes for flight dynamics. Real world flying is just as much about the "feel" of the aircraft as it is visual if not more so. I experienced a commercial full motion DC10 simulator way back when flying IFR conditions and the what you feel is profound.  Im quite certain we are not getting pneumatic and hydraulic kits anytime soon in our homes, so the feel of the aircraft is pretty non existent in a PC based simulation. I think the best is LM listen to feedback from real life pilots and from 3PDs on how things could be implemented with a foundation for FDs and 3PDs having most of the control for it.  How often did we hear its a bug in FSX or its hard coded , nothing we can do about that.

 

LM is a professional aviation company and more and no doubt have the databases of data in order to draw from to use for flight dynamics. The question is really can it  be implemented properly within the confines of programming AND having existing PC hardware able to cope with it in a fluid manner.

Im all for breaking compatibility in order to implement new technologies. Weve been sitting in this foundational box for a long time now, and in order to move forward we have to accept that fact as hard as it may be to accept after spending so much on addons. Call it the necessary evil.

 

Addons are getting more and more sophisticated and so they should otherwise how does a company progress. This means 64bit is paramount as the sophistication will need more ram plain and simple. However being that prepar3d is a commercial corporate sim, perhaps addons are not exactly of utmost importance to LM. Time will tell.

 

Things that give a better immersion might be our requests.  This day and age, why we dont have rain effects on the windshiield with it obscuring visibility and only cleared with wipers is mystifying.  Having AI that are not stupid, particularly ground traffic running through your aircraft everywhere.   Having a better ATC engine would help alot for immersion. These things sound like secondary but if your going to have a serious sim as LM wants to create thru corporate emphasis, than these things I would think need to be fixed and addressed by whatever means.

 

But alas these are just my thoughts coming to mind. Im no programmer what so ever. Just hardware experience. So its purely subjective.  Im just really really happy we have an enthusiastic developer that is very active in making our PC simulation better and better and to include us as part of it, whereas a few years ago it was a more dark and gloomy picture.

There is an axiom that is pretty much universal among the world's military that goes, "Never give an order you know won't be obeyed."

 

Similarly, another axiom that may just as well be articulated is, "Never ask a question you know cannot be answered." :Applause:

 

Thanks Bill. A bit of class from your end...much appreciated.

 

 

Where the 2D comes in is the representation of that 3D space on a 2D medium like a monitor?

 

Yes, that's where you lose the precision.

 

BTW, the same applies to auto racing simulators ... they are not the same as real world even when I can operate them at 200+ FPS, but I still used auto racing simulators to keep muscle training, situation awareness, and eye hand coordination active.

 

Simulations are still very good tools and they don't "need" to be exactly the same as real world ... maybe some day they will be that accurate with holo decks/rooms.  But I will always encourage simulations to keep pushing the goal of "as real as it gets" ... it's been a long long long road from basic lines/geometry to what we have today.  I'm glad I've had the privilege to ride that road from day 1 on the personal computer.

 

Cheers, Rob.

Sure it's a good thing to have high fps, but I've never been into fps, it needs balancing with complexity of other things. Anyway, after any advance in processing speed, fps is coming down again since we always go for more of the other stuff (sliders to the Right). Improved control and accuracy of simulation is always key. On that note, racing sims seem to be well catered for, but one thing surprised me last time I tried a bike sim, the direction did not respond to counter steer.

 

 

but one thing surprised me last time I tried a bike sim, the direction did not respond to counter steer.

 

Bike as in Motorcycle?  I'm not aware of any motorcycle sims that are actually trying to be a sim.

 

Cheers, Rob.