Sign in to follow this  
Von Target

Those at the Conference please tell LR...

Recommended Posts

That XP10 wouldn't be XP10 if it wasn't for...

 

--> Murmur's solution to the infamous "torque bug";

--> OSM and W2XP developments, as well as the alternative Meshes by Andras and the tools for creating photo scenery;

--> The NOAA weather plugin much more plausibly interpreting METAR data specially for winds with a variability and gusting group, and adding turbulence and temperatures aloft, long before these were re-introduced ( with bugs ) around XP10.45 if I'm not wrong...

--> Ventura Sky, which is looking to me as another reference;

--> The recently introduced static aircraft at airports, and the "AFCAD"-like info.... ( thx to LR );

--> The improved GNS, even if few probably really use all of it's features ( thx to LR );

--> Extended DSF ( thx to LR );

 

Please also tell him that the Moon and the Sun are out of sync with reality in his plausible World. If he could just make it as plausible as, at least, MSFS and others in this aspect - heck, it shouldn't be that difficult, and unfortunately it isn't an easy task for a plugin because there are no associated datarefs - must be hardcoded ( ? )

  • Upvote 2

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

That XP10 wouldn't be XP10 if it wasn't for...

 

--> Murmur's solution to the infamous "torque bug";

--> OSM and W2XP developments, as well as the alternative Meshes by Andras and the tools created for creating photo scenery;

--> The NOAA weather plugin much more plausibly interpreting METAR data specially for winds with a variability and gusting group, and adding turbulence and temperatures aloft, long before these were re-introduced ( with bugs ) around XP10.45 if I'm not wrong...

--> The recently introduced static aircraft at airports, and the "AFCAD"-like info.... ( thx to LR );

--> The improved GNS, even if few probably really use all of it's features ( thx to LR );

 

Please also tell him that the Moon and the Sun are out of sync with reality in his plausible World. Please make it as plausible as, at least, MSFS and others in this aspect - heck, it shouldn't be that difficult ?

 

I would settle for weather....unfortunately rumour has it that the 3 layers of wind and 2 layers of turbulence restriction is here to stay. I truly hope this is not the case.

Share this post


Link to post
Share on other sites

 

 


unfortunately rumour has it that the 3 layers of wind and 2 layers of turbulence restriction is here to stay.

 

Where do you have this rumour from?

Share this post


Link to post
Share on other sites

as well as the alternative Meshes by Andras and the tools for creating photo scenery;

That does not need to be said, as they (especially Ben Supnik) knows very well what I have done with their tools over the years :smile: .

 

You need to know, that my work and Laminars default Global Scenery work live in a very close symbiosis. I use (and in small parts improved / hacked) the Laminar scenery generator code, which I learned to work with while I did all the Global Scenery rewamping for them before XP10. After that, I branched off my own projects, but they still get back all my improvements (and essentiaslly use them, whenever a new official scenery is cut). For example they - in the mean time - have a completely new set of raw data, as over the years I have made a lot of improvements to the landclass data (which you have often seen in HD / UHD Mesh - and is now completely replaced / updated in comparison to the original XP10 Global Scenery data), but also OSM data has seen a few update cycles (the latest data I sent to Laminar is from August 2016).

 

So, you see ... my role is maybe a bit different from the classic developers (because of my closer ties to Laminar) ... but nevertheless, thanks for mentioning.

  • Upvote 2

Share this post


Link to post
Share on other sites

I know Andras, the same applying to Phillip regarding the GNS implementation :-)

 

Anyway, if it wasn't for your efforts, XP10 wouldn't be as good as it is, IMO :-)

  • Upvote 2

Share this post


Link to post
Share on other sites

@jcomm - I fully agree with your statement.

 

In fact, I believe it would be great for X-Plane if Laminar pulled an Apple, and absorbed some of these features and hired some of these plugin developers. Native features would somehow just appeal more to a new audience than having to use a plugin (although there's the adage 'there's an app for that!), giving newbies the idea: 'huh? The sim doesn't already do that?'

 

Just imagine if XP11 shipped with a weather injector (forget the triple layer depiction for a while) as complex and powerful as the NOAA plugin, and with a scenery generator like W2XP, not to mention an option for HDv3 and UHD meshes.

 

Then XP10 would have to ship physically on a HDD :P

Share this post


Link to post
Share on other sites

I would settle for weather....unfortunately rumour has it that the 3 layers of wind and 2 layers of turbulence restriction is here to stay. I truly hope this is not the case.

 

I think it's 3 layers of turbulence as well? However as you said, from the video about X-Plane 11 presentation, the weather layout seems unchanged from the current one.

 

Actually I don't think the 3 layers restriction is a decisive limitation, after all it allows the depiction of two cloud layers + a high altitude cirrus layer, that is more than enough for most weather situations.

 

The real limitations of X-Plane weather engine lie elsewhere: lack of cloud types, lack of vertical development clouds (other than stratus), poor and static cloud structures, clouds not moving with the wind. If all of these issued would be resolved, it would be a decisive improvement.

Share this post


Link to post
Share on other sites

jcomm, on 21 Sept 2016 - 01:53 AM, said:

 

That XP10 wouldn't be XP10 if it wasn't for...

 

--> Murmur's solution to the infamous "torque bug";

--> OSM and W2XP developments, as well as the alternative Meshes by Andras and the tools for creating photo scenery;

--> The NOAA weather plugin much more plausibly interpreting METAR data specially for winds with a variability and gusting group, and adding turbulence and temperatures aloft, long before these were re-introduced ( with bugs ) around XP10.45 if I'm not wrong...

--> Ventura Sky, which is looking to me as another reference;

--> The recently introduced static aircraft at airports, and the "AFCAD"-like info.... ( thx to LR );

--> The improved GNS, even if few probably really use all of it's features ( thx to LR );

--> Extended DSF ( thx to LR );

 

Please also tell him that the Moon and the Sun are out of sync with reality in his plausible World. If he could just make it as plausible as, at least, MSFS and others in this aspect - heck, it shouldn't be that difficult, and unfortunately it isn't an easy task for a plugin because there are no associated datarefs - must be hardcoded ( ? )

Jose,

 

Neither Austin nor Ben visit these forums, so posting here will not necessarily get the information to them. And, if someone "conveys" your message to them, that message will likely not have all the same content that you originally intended and be "watered down" in the process.

 

The absolute best way to assure that your suggestions and complaints get through to Laminar is to contact them directly, either by e-mailing Austin/Ben or, even better than that, by filing a feature/bug report. This way it is all in writing and no one has to "remember" the details of what someone said to them.

 

Use this link to file a feature/bug report...

 

http://dev.x-plane.com/support/bugreport.html

 

Regards,

Jim

 

PS. Any bugs/features reported, especially with regard to the flight model (Austin's responsibility), should be well documented in as much detail as possible.

  • Upvote 1

Share this post


Link to post
Share on other sites

I think it's 3 layers of turbulence as well? However as you said, from the video about X-Plane 11 presentation, the weather layout seems unchanged from the current one.

 

Actually I don't think the 3 layers restriction is a decisive limitation, after all it allows the depiction of two cloud layers + a high altitude cirrus layer, that is more than enough for most weather situations.

 

I disagree, because modeling "most" weather systems and not the most dangerous ones like CB's is not enough for a supposedly modern, world-class flight sim. A simplistic 3-layer model can't model vertical convection, or discrete weather systems in the distance, within view of the plane. I suppose it's a valid question as to whether resources should be spent to model weather systems pilots are supposed to avoid, However, vertical airflow is still part of the environment a real-world pilot flies in. Sometimes you can't avoid these systems.

 

And even if you *do* want to avoid flying into a CB, changing your flight plan to steer around them, you still need to see where they are! VFR pilots do this all the time. Airline pilots steer around them too, using radar, ATC or visual guidance, because the tops of these suckers can reach or exceed the flight levels. Why can't we do it in a flight sim?

 

If there is a way for 3rd parties to inject discrete 3D weather systems within X-Plane's primitive 3-layer model that's great. But I haven't seen anyone accomplish that yet. It may not be possible at all, unless Laminar rebuilds the weather system from the ground up.

Share this post


Link to post
Share on other sites

I disagree, because modeling "most" weather systems and not the most dangerous ones like CB's is not enough for a supposedly modern, world-class flight sim. A simplistic 3-layer model can't model vertical convection, or discrete weather systems in the distance, within view of the plane. I suppose it's a valid question as to whether resources should be spent to model weather systems pilots are supposed to avoid, However, vertical airflow is still part of the environment a real-world pilot flies in. Sometimes you can't avoid these systems.

 

I don't see why modeling a single convective system would need more than one cloud layer per se. For example, in MSFS you can model a cumulonimbus/thunderstorm using only a single cloud layer. Of course it's not perfect, and using 3rd party addons improves the end results.

 

Let's imagine a weather engine where, in the case of vertically developed clouds, the inputs for each cloud layer are:

.cloud base

.cloud type (cumulonimbus/stratus)

.cloud top

.cloud coverage

.intensity of precipitations (rain, hail, etc.)

 

If the weather engine is capable, it can build believable convective systems (or stratiform systems) even using one single cloud layer. This is basically how the MSFS system works, and infact even if you limit yourself to using only one single cloud layer, MSFS can depict a convective system much better than X-Plane.

 

It's also a matter of priorities. I prefer to have X-Plane improve its depiction of vertical developed clouds, and having moving and dynamic clouds, rather than having 10 different cloud layers.

  • Upvote 1

Share this post


Link to post
Share on other sites

I don't see why modeling a single convective system would need more than one cloud layer per se. For example, in MSFS you can model a cumulonimbus/thunderstorm using only a single cloud layer. Of course it's not perfect, and using 3rd party addons improves the end results.

 

It's not about increasing the number of vertical cloud layers. It's about the need to move from a 2-dimensional weather system to a 3-dimensional one.

 

Visualize 3 nice fluffy pancakes stacked on top of each other. That's what we have right now, with the plane dead center in one of those three layers. There is no way to model "one pancake over there at 9:00 o'clock through the viewscreen, and another pancake over there at 3:00 o'clock... we'd better steer between them."

 

See? The current weather engine isn't a 3-dimensional system that can model discrete weather systems, and not just what's surrounding our plane. What's needed is a 3-dimensional box grid that can place one weather system over here, and another one over there, instead of everything at once in a single horizontal layer surrounding our plane.

 

Aside from a more accurate visual representation -- we would actually see things like CB's and approaching weather fronts at a distance -- it would help solve the problem of weather changing abruptly because it happens with our plane at the center. We could actually fly into and out of weather systems like real-world pilots, instead of the primitive "all at once" setup we have now, that requires smoothing and interpolation to avoid a sudden drastic change. The smoothing is okay when it works (using various add-ons), but we still never see discrete weather systems in the distance that we're about to fly into.

Share this post


Link to post
Share on other sites

It's not about increasing the number of vertical cloud layers. It's about the need to move from a 2-dimensional weather system to a 3-dimensional one.

 

On this I agree 100%

 

Visualize 3 nice fluffy pancakes stacked on top of each other. That's what we have right now, with the plane dead center in one of those three layers. There is no way to model "one pancake over there at 9:00 o'clock through the viewscreen, and another pancake over there at 3:00 o'clock... we'd better steer between them."

 

See? The current weather engine isn't a 3-dimensional system that can model discrete weather systems, and not just what's surrounding our plane. What's needed is a 3-dimensional box grid that can place one weather system over here, and another one over there, instead of everything at once in a single horizontal layer surrounding our plane.

 

That's not 100% accurate. The X-Plane weather system is bidimensional if using the base weather settings screen, but it has some tridimensional features if using real weather or random weather patterns. Here's an example, where I manually set two thunderstorm with different cloud bases over Boston and over Providence:

 

D7celSl.jpg

 

Here's another example where 3d features of the weather system can be seen (I think I used random weather patterns for this):

 

ISM2q3p.jpg

 

Yes, the first thunderstorm shot does not look very good, but the reason is the poor representation of vertical cloud structures in X-Plane, not the 3 layers limit per se. Infact, I'm not even sure that when using real weather, X-Plane is limited to 3 cloud layers, or if it's onlly limited when using the base weather setting screen.

 

 

Share this post


Link to post
Share on other sites

Murmer, on that first shot, were you "painting" cloud layers on the weather page? As far as I can tell (correct me if I'm wrong), all that does is draw puffy clouds in what looks like the middle cloud layer, so you can't get a towering CB that crosses two or more of the layers. 

 

I just tried it, drawing the most intense-looking weather I could get ahead of my plane on that "paint weather patterns" menu. Lots of nasty red inside. All I get when returning to the plane are thin puffy clouds at a middle altitude layer. There's lightning being triggered, I suppose from the red areas, but it looks silly in bright daylight coming from a thin, puffy white cloud. Also some added turbulence flying nearby, so X-Plane apparently thinks it's a CB. But the visuals just don't cut it, and there is no vertical convection effect on the plane.

 

I take your point that this does indicate some degree of 3D positioning of discrete "weather," but it's apparently just placing clouds within the existing layers, and not capable of simulating an approaching weather front or a chain of CB's on the horizon.

 

Share this post


Link to post
Share on other sites

Yes I agree with you, the end result is extremely poor. I hope they improve the depiction of vertical developed cloud, along with the other issues (static clouds, etc).

 

I'd be pleasantly surprised to see a revamped weather engine in X-Plane 11, but I'm not holding my breath. Heck, they could take "inspiration" from MSFS. Here's what FS2004 _default_ weather could do 13 years ago (and that's just a single cloud layer):

 

0LoiwW0.jpg

Share this post


Link to post
Share on other sites

All this about cloud representation is great, but if the performance issues  that come hand-in-hand with dense cloud cover isn't solved, then to me it's all for naught. You could have all the layers you'd like, intricate cloud systems up the wazoo, but if no-one has the hardware to display them, then why?

 

 

That said, the original post could be largely summed up by "XP10 wouldn't be XP10 if it weren't for the time and energy given by countless community members."  Now that... I'm pretty sure they are fully aware of that part. :smile:

Share this post


Link to post
Share on other sites

All this about cloud representation is great, but if the performance issues  that come hand-in-hand with dense cloud cover isn't solved, then to me it's all for naught. You could have all the layers you'd like, intricate cloud systems up the wazoo, but if no-one has the hardware to display them, then why?

 

Very well said, I forgot about performance. That would be number one priority for me.

Share this post


Link to post
Share on other sites

All this about cloud representation is great, but if the performance issues  that come hand-in-hand with dense cloud cover isn't solved, then to me it's all for naught. You could have all the layers you'd like, intricate cloud systems up the wazoo, but if no-one has the hardware to display them, then why?

 

 

I agree in principle. However, it raises the question of what "performance" means, and what hardware we're talking about?

 

Obviously we'd like to see X-Plane deliver high frame rates. For me, that means anything over 30 fps, and that should be achievable on legacy hardware, but not necessarily with all eye candy enabled. And what about new hardware? Shouldn't we have new features that will tax the higher-end hardware available now, to allow for future improvement on the hardware side?

 

For example, I just upgraded my rig -- i7 6700 4.0 GHz, 32 Gig RAM, GTX 970 video. Room for improvement on the video, but better than my previous computer in CPU speed and RAM.

 

Today I flew a 2 hour flight in FSEconomy with broken cumulus (SkyMax Pro with RWC), pretty good cloud cover everywhere. I haven't changed any of X-Plane's settings since I finished setting up this new machine, so I'm still using the old settings. I enabled the FPS display, and I was getting an astonishing 90 fps during the flight, touching 100 fps once in a while. I guess I can start dialing in more autogen!

 

I know a percentage of X-Plane users are having performance problems with heavy clouds, but I wasn't noticing it that much even on my older i7 rig with a GTX 750 ti. And this new one just breezes past it.

 

All due respect to those with older hardware, but X-Plane needs to look forward with new features that can run on newer hardware, while still supporting older rigs by dialing back the new stuff. It happened when they introduced HDR, and it would happen the same way with more intensive weather modeling.

Share this post


Link to post
Share on other sites

Obviously we'd like to see X-Plane deliver high frame rates. For me, that means anything over 30 fps, and that should be achievable on legacy hardware, but not necessarily with all eye candy enabled. And what about new hardware? Shouldn't we have new features that will tax the higher-end hardware available now, to allow for future improvement on the hardware side?

 

Very good points, and the questions raised in regards to 'future proofing', enabling the platform to have features that people might not be able to take advantage of with average hardware yet but might someday, are definitely worth talking about.

 

But since development resources are scarce, it becomes an increasingly difficult balancing act, and one that might have to be weighed in terms of it's potential benefit to the average user.

 

And of course, 'average' could mean a lot of things. After all, there's a good amount of users that can't afford the performance hit of HDR while using 1080p. You can even point to the surprising popularity of the latest thing to come up, 'Ventura Sky' - one of the tricks it does (or at least did in beta, I haven't followed it's development in a bit) is limit visibility to something like 16nm at low altitude in order to boost performance. A ton of my time is spent in XP flying no more than 2500 AGL, and I can't imagine wanting to only have 16nm in front of me just so I can enjoy good performance. Yet there are people that need that...

 

There's absolutely no question that the clouds are in need of attention - I would love my XP skies to look like my FSX or P3D skies do. But I would only love it if I could not only make the sky look similar, but also have it look that way an equivalent level of performance. And that's the catch - it's not enough to have great looking clouds with the knowledge that *someday* hardware will be able to take advantage of them. The platform needs at least equivalent cloud depiction to it's competition, while achieving equivalent performance. Without that performance factor, they just become a talking point in my opinion.

 

But you know what they say about opinions... :smile:

Share this post


Link to post
Share on other sites

I think performance means looks good on todays hardware (with today I mean the day that the software is released).

Clouds especially have been lacking since the beginning, they started to look good with the Nvidia 9xx.

 

To x-planes defense, none of the flight simulators I have were able to depict clouds without a framerate hit, at the day of release.

That was true for every MS-FS release, the flight unlimited series, Flight! and also for X-Plane.

 

Looking back, I mostly flew with blue skies on my pc :'( .

 

So if Laminar is able to deliver great skies at the day of the release of X-plane 11, that would be quite an achievement !

Share this post


Link to post
Share on other sites

Even in AS16 and ASCA I set max cloud layers to 3, because of the constraints imposed by my rig.

 

In meteorology a "chaotic sky" as we designate a situation where cloud classification becomes almost impossible, is something I believe would kill any of our present day simulators :-)

 

As a pilot, and also from a meteorology pov, but also from the perspective of a simmer with limited hardware to run it's sims and looking forward not to have to spend fortunes upgrading to run a new version of X-Plane, three cloud layers are more than enough, provided that factor mentioned by all of you is looked at for the next version of X-Plane - vertical development.

 

Even in FSX and using the best weather generators, CBs are rather limited in area. There are certainly no super-cells or even moderate size Cbs, although AS does it's best to achieve the closest possible to it, and makes the huge difference there is between a thunderstorm depicted in MSFS vs XP...

Share this post


Link to post
Share on other sites

yes José, I have almost the same flying background as you (I fly hang gliders).

IMO, 2 cloud layers should be enough for a good depiction, 3 is sufficient for the more complex weather.

For me, what is lacking in the default x-plane cloud system: 

- sharper edges to the clouds, I think that would be a game changer, even if nothing else changes.

- vertical cloud development (which would also allow for CB).

Moving clouds would be cool, but is less important then the cloud shadows on the ground :-) .

Share this post


Link to post
Share on other sites

yes José, I have almost the same flying background as you (I fly hang gliders).

 

Then, you have a much better background, and one that I was never able to tackle simply because I lack the guts to fly in a hang glider... But that's even closer to the best experience of flight only our bird friends really have :-)  And that brings yet another set of features we would like to see even better represented in XP, although I agree it Is not the purpose of this platform, and, apparently, Condorsoaring 2 is now really confirmed as being under active development....

Share this post


Link to post
Share on other sites

From a developer point of view, without entering in the end-users context, I'd say that there're even more things that should be fixed and/or implemented both in Plane Maker and the SDK... I sent various but reports so I really do hope that at least something will be considered for XP11.

Share this post


Link to post
Share on other sites

I agree in principle. However, it raises the question of what "performance" means, and what hardware we're talking about?

 

All due respect to those with older hardware, but X-Plane needs to look forward with new features that can run on newer hardware, while still supporting older rigs by dialing back the new stuff. It happened when they introduced HDR, and it would happen the same way with more intensive weather modeling.

Even people with high end rigs always complained about the performance of X-Plane 10 clouds, even more so comparing end results to those of msfs. Msfs clouds may be relatively performance hungry, but X-Plane 10 clouds have been worse while giving worse results. Moreover I don't see why allowing lower settings for clouds for those with low end rigs, would limit those with high end rigs.

Share this post


Link to post
Share on other sites

 

 


To x-planes defense, none of the flight simulators I have were able to depict clouds without a framerate hit, at the day of release.
That was true for every MS-FS release, the flight unlimited series, Flight! and also for X-Plane.

Looking back, I mostly flew with blue skies on my pc :'( .

So if Laminar is able to deliver great skies at the day of the release of X-plane 11, that would be quite an achievement !

 

As true as that is, the issue is that now the playing field has changed. What is going to be expected at any new simulators release is going to be viewed through the lens of experience of what else is offered these days, in both depiction as well as performance.

 

 

 


Even people with high end rigs always complained about the performance of X-Plane 10 clouds, even more so comparing end results to those of msfs. Msfs clouds may be relatively performance hungry, but X-Plane 10 clouds have been worse while giving worse results.

 

Precisely. The point I was trying to make is that the end result, of which we all want improvement on, is actually a sum total of both graphical depiction and performance. If you improve one without the other, the net result is still the same.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this