I agree with much of what you say towards the end, however purely on a flight dynamics point I cannot agree with you - I actually find that the current situation is intolerant of anything other than a perfect airliner ‘shape’ : as in my earlier posts in this thread, the flight model is geometry-based but the geometry cannot be defined accurately - any fuselage has to be a perfect cylinder for its full length, and I also showed a screenshot comparing what the simulator defined as the vertical tail to the actual correct geometry input. Nothing alike, which instantly raises questions about the core flight model.
If all of the bells and whistles of MSFS (the scenery, the weather etc) had been created and the core flight model from FSX retained as a proper starting point, but using realistic coefficients and damping levels, you would really have had something special. Something which could have been developed carefully to allow multiple points whilst retaining the accuracy that went before. As it stands, as a developer, an ex-pilot in the real world and most importantly a simmer (!) I find the current situation ‘gimmicky’ and frustrating, feeling like a rushed job to try and be half FSX and half X-Plane, whilst making the job bloody difficult for those building in it.
If I want to enjoy the scenery or weather, MSFS every day of the week and no arguments. If I get time to fly for relaxation, it’s still P3D all the way, using specific aircraft!
I’ll go the opposite way - I have been ‘in control’ of a glider that was climbing with the airbrakes fully out and sideslipping just about managed to level it off!
100%. All I want, as a developer, is to be able to use realistic, calculated figures and not have to stray too far from them to get a realistic ‘feel’. Currently, this simulator cannot deal with real figures or accepted aerodynamic conventions.
“But yes, you can build a flight model correctly, using the calculated figures and accepted aerodynamic terminology. You could even use the figures from, for example, Professor Jan Roskam’s ‘Airplane Aerodynamics and Performance, part 1’ and then “frig around” with the weather!”
I think you completely misunderstood my original post, hopefully that makes it a little clearer.
Of course it made a long way and i understand the comparison with the old flightsims back 20+ years ago. Im out of that era too, Ace of Aces someone? Or Chuck Yeager’s Advanced Flight Trainer and many others from that time?
The problem is that we have other sims out by now too and that makes backsteps possible. MSFS is amazing in many aspects but its still suffering from early day bugs. And the focus towards worldupdates, while we still have the weather bug for example, or other problems that are there since the release, make some people worry imo.
I was heavily into P3D and some XP11 and went on to msfs too but im not the one who wants to say that everything is amazing here. The scenery aspect is great and it makes steps forward but there is still a lot to do until the flying aspect is as far as with P3D or XP.
Sure, why not simply stay in those sims then, one could say. And i dont want to argue on that. I just understand both sides. What i dont understand is constant ranting though and that is still present, in many communities. Those probably will always rant so theres little to do to please them.
My opinion: because we care about this one. If we didn’t care we’d either stick with P3D or XPlane, or we’d say “whatever, this is what it is, why bother?” A lot of us hope that FS2020 can become good enough to fully replace the previous generation of sims, and that’s why we’re here, voicing our opinion.
I think so too, no worries. Thats just what i see the “other side” bringing as argument always. If your not happy just stay in XP or P3D.
But like mentioned, i see both sides, there is also no reason to simply accept every bug or weird behavior. We have to let them know that things dont work. As everyone can see, a beta team is not enough for that. Without the reports here from people that have experienced this or that the si would be half as far by now.
Both sides need to exist if a product wants to be better. The ones demanding and the ones who accept everything.
Example: I’d like the features promoted/promised in all the official pre-launch videos to work, and to work as expected (eg weather, nav database, gps).
That is not a binary choice. It will be both. Only difference is, the customer (we) will pay for the hardware side of this story. When you want realism, it will will cost FpS and you do NOT want stutters for cockpit updates, so any difficult addition to make tubes more realistic will cost hardware too. In fact when you would like extreme realism in an Airbus, your screen won’t do anymore, you’ll need controls. In 8 years, projected MSFS end of life we’ll all have video hardware that is capable of DX-12 Ultimate and ways of control, VR we can’t imagine atm. But to facilitaite that realism, many will have to replace their current board and current PC (power supply) or hop on the X-Box train. Both choices involve needed hardware and if we’ll invest… that choice will be ours to make.
Look, we are talking about a particular topic: adding realism to FS2020. The whole argument started because the assumption was made that computers today do not have enough power to run FS2020 with realistic addons.
P3D (which has similar, if not higher system requirements in practice as FS2020) proves that to not be the case. There are plenty of very realistic addons available, and computers run it just fine. You can have a true to life Garmin GPS unit running in your true to life GA aircraft, flying over ORBX true to life hand placed landscape, using true to life ActiveSky weather. And FS2020, bringing you true to life weather and true to life landscape manages to do it in a more efficient manner than P3D, delivering better performance with better visual quality. So I don’t see how anyone can claim that bringing in the rest of the equation will somehow require a super computer.
That is why I cited you. I think we disagree about that assumption. It may be true… Avionics fans ask the ultimate goal, being “realistic”. That’s good. But I do wonder, if that can be done with current average hardware. In other topics there are already complaints about stutters, when network traffic is involved. Now suppose they would add in much more traffic realism in these avionics tools… Maybe you need faster internet ? Faster servers ? When there are 100s of active gauges in a cockpit, and a realistic tube has, you’ll need a good CPU to keep up with life updates. Hardware does play a role if you require “realistic”. Maybe I’m talking mid-long term you’re talking short term ? Current PC’s should be able to do current addons ?
It can. Like i have said multiple times, here and elsewhere, FS2020 looks much better than P3D on my system, and it gives me more FPS than P3D. And my configuration is similar to the recommended one. Since i could run complex addons in P3D, I don’t see how someone with average hardware today would have any issues running realistic addons in FS2020.
No addon i can think of had anywhere near the levels of bandwidth requirements FS2020 has. ActiveSky only occasionally would download METARs. VATSIM is probably the most intensive one, and it was pretty low.
If a 5-6 year old CPU can do it while also running the performance hog that P3D is compared to FS2020, most anyone can do it.
I’m talking about the lessons we learnt about realism in P3D and in FSX. Hardware has always been an issue with those sims because their cores were never optimized to run on multiple cores. Even P3D, where a lot of effort has been spent to optimize it, is nowhere near as efficient as FS2020. So even on a PC that is being strangled by P3D or FSX you can still run complex addons. I can’t see how you’d suddenly have issues with a better optimized sim.
I lack the knowledge to say something meaningful about these “realistic addons” for other sims you seem to know so well (you call them “realistic”), but I wonder how you deduce that there is no relevant CPU load for additional controls and a more complicated aircraft model. E.g. those addons interface via simconnect, not using CPU or RAM bandwidth. They sit on top of Windows network infrastructure. Most of these controls are programmed in high level interpreted languages like Javascript… and you seem to assume that can all be shown on 1 page (your cockpit). Being a developer in RL, I can tell you additional renderings will always take time when you have to leave part of the screen to a shader in DX. That cockpit is overhead too, not only the scenery. There are slow aircraft and faster aircraft, even smaller aircraft I like to fly, e.g. TBM and DA-62 are noticably slower than other types. An SDK-version of DA-62 is even slower. I see the difference in FpS when I put cockpit update speed low. I run low end hardware, i17/16G… that’s why I notice…
I know that, other sims can link in DLL’s, but in MSFS it will at least require Wasm (web assembly) translation, that is working via CPP in a sandbox… I “tutorialized” an SDK example here when I started out with SDK … how-to-change-a-compass-sdk-gauges-aircraft-example but MSFS supports some kind of CSS+Javascript method too… that makes development a lot easier than CPP, but it comes with a cost. Even Wasm is much slower than native CPP. In a way you’re looking at a very complicated web page, allowing you to navigate instantly left, right, up, down and everything has to be shown (synchronized) real time. A browser needs big time for that, if you compare it to refresh rates in MSFS…
Folks, we are straying way off the OP’s topic here. @CristiNeagu is simply stating that with the release of the new, more complex add-ons we will be getting to better place. Serious flight sim hobbyists have been waiting years to have the next generation of simulators and Asobo/MS willing this looks like it could be it.
@ArcanePython931, in the past we have had very “realistic” add-ons that have had very little impact on the base sim performance. We have also had minimalist aircraft with bad paintjobs and even worse cockpits that would bring the sims to there knees. How much “cost” there will be for the “realism” will be determined by the team (or individual) that programs the addon.
Let’s put this argument to bed. There is no reason that we can’t load our sims down with hyper realistic aircraft and avionics, no matter what level of hardware we have. Those with marginal systems will have to sacrifice some fidelity or maybe fps. Those with high end systems will likely need to make some adjustment to their expectations as well. This is what adding to the system load means. It does NOT mean we cannot have a realistic sim running realistic airplanes with realistic avionics and systems in realistic weather on today’s computers.
Thank you for your summary (genuinely), but i don’t think we are straying off topic, not nearly as much as you’re implying. Fact is that there are plenty of users asking for these kinds of realism-focused addons, the kind with which we’ve grown accustomed to from FSX and P3D. If @ArcanePython931 were right that present day hardware is simply not enough to run it all, then this request from users would really be asking for too much. It would be unreasonable to ask Asobo to fix what only time and advances in technology can fix. So this is still very much an argument about whether users are asking for too much.
Depends on your perspective. I wonder how many MSFS users originate from that P3D world you and Willis are constantly referring to, to show MSFS “should be able” to run addons like for that platform. In my above answer I tried to explain why MSFS addons are (structurally !) slower than DLL’s in native CPP. Somehow that point seems irrelevant ?
And in fact willis is also right all this is for a large part off topic. There may be some good news about their intentions, but I expect there will be limits in the funds MS and Asobo will be prepared to spend on these “realistic addons”. In a sense, it may indeed be asking for too much… When MS and Asobo would be required to partially rebuild their server infrastructure, allowing more bandwidth/response AND same time rebuild the SDK to facilitate DLL’s safely only to serve, say, 15% of MSFS users, that is not going to happen imho. Ask too much, you won’t get it for free.
Fair enough. I guess we can see the technology argument as an extension.
If @ArcanePython931 was to look deep into the developers bible, they would see that there is a verse in there much like the book doctors use. “Do No Harm”. Simply put, the goal of any programmer worth their salt is to create code that reduces excessive routines in favour of efficiency. There will always be some resources used. Can’t get away from that.
To stay with the “asking too much” theme, I have to task this goal to the addon developers. The onus falls to them to ensure they are not frivolous with the limited resources we have. We, the end users, are wanting multiple aircraft and tools to add to our experience. All those tools and adds need to share equitably or Arcane will be right. Our systems will crumble under the weight. Asobo needs to do their part to make MSFS as efficient as possible. We know that is their ultimate goal. We can only hope that the aftermarket guys are focused on the same thing. Is that, asking too much?
I think I can be pretty confident that they will spend next to nothing to achieve those realistic addons. They will take a cut from the sale of those adds through the marketplace. The adds will be the responsibility of third party developers. The more realistic they are, the more they will charge.
