X-Plane 10 HD Mesh v2 - PREVIEW

[alpilotx]

I tend to get annoyed with things I do not understand and furthermore go to extended distances to figure things out. In some cases it will in fact be to annoy the heck out of people like you

and Ben with question about things I want to figure out or do not understand.

If you want to learn more about the X-Plane scenery, then the "best" way is (at least for me it was) to read the documentation (which is not so great and uptodate ... but still better than nothing ) and disassemble a few DSF files (DSFTool is your friend) and try to understand how its internals work (the above mentioned script on my website might be a help in this process too) ... But be prepared! You mind might get twisted quite a few times ...

 

 

So how come X-Plane only have 90 meter elevation data where simulatators like FSX and Prepare3D claim down to 10 meters, 38 by default

Quite simply: why bother with higher res raw mesh data, if the basic triangle mesh won't really utilize it (at least not before we get tessellation). The point is also, that X-Plane works very very differently here than FSX! FSX is rather like a GIS tool, which does lots of the raw data interpretation and transformation "on the fly" (when it loads the data) while X-Plane has most of the scenery really pre-built in the DSF (thats what the above mentioned RenderFarm tool does). Both approaches have their advantages / disadvantages. The FSX approach is far better for scenery developers, because they only need to add much lower level GIS data to the scenery and FSX does the rest of the work BUT at the cost, that FSX can't spend ages on bringing out the best of this data (or the users would need to wait 10 minutes before the scenery comes up) .... On the other side X-Plane does a lot of the complex data interpretation and mesh generation beforehand (RenderFarm), can spend a lot of time to try to bring out the best from the raw data (this makes it possible to generate an irregular mesh etc.) and only need to load almost "finished" data from the DSF. Of course, the downside of this is, that the data, already transformed in a final triangle mesh etc. etc. is far less accessible to 3rd parties (its not realy meant to be edited ... even if its possible and a few devs out there have tinkered with that ).

 

Also remember, that the irregular mesh of X-Plane has the advantage, that - simply put -  it "spends" more triangles where there is interesting stuff, and less, where the land is flat ....

 

And than, hopefully one day we will get tessellation in X-Plane which would combine the best of all worlds ... it wouldn't be necessary to make the triangle mesh more detailed than necessary in the DSF file, while X-Plane would refine it (via tessellation) at load time and adapt it to the underlying elevation DEM data in much better quality (and then, it would make sense to put in higher res DEM - where available).

 

 

Well there are many tools out there that fixes SRTM data (patching holes and filling in like above N60) i.e SRTMFill http://3dnature.com/srtmfill.html and SRTMVoid Killer http://www.dgadv.com/dgvk/

But thats not the point!

Yes, of course. ... But why "waste" time on this (and one can waste extremely lot of time on this), when others have already done it? And not just algorithmically fill holes, but even replace the data with better raw data where ever possible (and its possible in  alot of regions of the planet). Neither would your approach the problem of entirey missing SRTM90 data above 60N .... And viewfinderpanoramas.org already does all of this .... And their final quality in overall is everywhere at least the same as SRTM90 and in many places far better (believe me, I know it is this way .... have studied so many GBytes of elevation data over the years )

 

What i am not sure about is if it even exists better data for the world than SRTM90 3arc collected by Endeavour back in 2000 (except SRTM30 for the US that is).

Or if they did in fact to a new dataset collection on the last shuttlemission.

Yes, for many many place of the earth you can get better quality DEM (not just interpolated) .... an no, SRTM is by far not the only project to map the planets elevation. Many local governments or international organizations have smaller / large patches of elevation data done on their own (with different mapping approaches .... sometimes free, sometimes commercial etc. etc.). You might get a very small overview from here:

http://vterrain.org/Elevation/global.html

 

 

Could you do something about the mountain texture only? Looks like someone has sneezed on it with their now full of cocain LOL..

https://lh6.googleusercontent.com/-VS52wJqGyrQ/UjTnfb6e3pI/AAAAAAAADHU/W6xlVI27TTI/w1597-h832-no/FA-22A_41.jpg

Aaah, yes, thats one of the newer textures I think ... and indeed, there is one which seems to become too blurred. Don't know why ... And no, I am not working on textures, but the guy who does them is a good friend of mine (I will let him know).

[alpilotx]

So what if I had access to 10meter DEM data (i do but not sure if it can be publicly used yet), could it be used in XP?

PS

I think Austin should buy Outerra and merge them for sure

 

Yes, and no ... in theory X-plane could use higher res raster data, but with the current triangle density it wouldn't make sense. With tessellation it would make sense ..... Or if you use Pntadera (I already pointed you at this - they have a script which increases the DSF mesh resolution artificially):

http://forums.x-plane.org/index.php?app=downloads&showfile=16417

 

The lobbying for Outerra is very old .... still its a  much much more complex topic than most out there imagine (its definitely not as simple as buy it, drop it in, and be happy). And neither can Outerra do everything you might want in X-Plane (but X-Plane already has it) ... But don#t expect me to comment more on this issue (and no, there are absolutely no plans to use Outerra).

[alpilotx]

All this tools just interpolate the elevation data. Most of this data are based on SRTM4 data sources. So even if you use 30m or 10m raster layer it will give only more smooth surface but not much details. Outerra creates this details based on fractal algorithm which looks good but is not really true.

Another things is if you find real HI resolution elevation data. I think there are some and Alpilotx probably can tell more about it. (NZ for example).

I have answered this above in my long post for Tom .... yes there is such data, but no, at the moment it makes not so much sense to use it (but will quite probably make sense in the future) .... Only exception is the Pintadera approach, which artificially increases the mesh resolution in the DSF itself (and increases its size of course - thats what a "post processing" approach like tessellation will avoid): http://forums.x-plane.org/index.php?app=downloads&showfile=16417

 

 

1) can RF use HI resolution DEM to create raster layer for XP10 DSF?

Again: in theory yes, but it makes not much sense until now ... with tessellation, this will change. And just to put it in perspective: I would say, that my HD Mesh Scenery v2 is approximately at the detail level, where SRTM90 is used to its full potential (and then look at the size of HD Mesh DSFs ... so really, tessellation will be the solution in the future as it helps to avoid the otherwise necessary boundless growth of DSF file sizes).

 

 

2) Is it possible to combine two or more elevation data to create one DSF? Or the raster layer is a single instance with fixed resolution for entire DSF?

No, only one raster layer per DSF is supported. But you can "manipulate" that raster layer to your liking (in theory) .... and merge as many different sources in it as you would like (but of course, the final layer needs to have the resolution of the highest res source data you used in the merging ... otherwise you would lose detail).

 

 

3) Is it possible to add some data to the RF, that it will add XP9 style mesh only for specified area (i.e. airport fields) and rest will remain as XP10 2D triangulation. (I know that now some points in DSF still have an elevation value - i.e. river and lakes borders).

Good observation   ... Yes, indeed its true, that XP10 is not using the raster DEM data only, but also still has some 3D points in the triangle mesh. This is needed, to make sure that lakes are made really flat (as their edges are calculated in advance via RF and fixed at the correct height). This is also an important part of my knw tiny algorithm which sanitizes the tiny rivers in narrow valleys (if we would only rely on fitting the triangle mesh to the raster data in XP, then that would introduce lots of water going steeply up and down on slopes etc. ) ... So, yes, XP10 allows a combination of both ... where the mesh has only 2D info, it will be adapted to the raster DEM data. Where it has 3D info, it will keep to that fixed elevation.

[alpilotx]

And X-Plane engine is good as well, but not really adjusted.. Terrain textures (texture itself and way they are combined in pattern) by default is what I can't imagine even in my worst dream. Sorry.

This is a sentence (especially the last one) which I don't completely understand (what you mean) .... ???

[Rhard]

This is a sentence (especially the last one) which I don't completely understand (what you mean) .... ???

I meant the default terrain in XP10 looks not good:

[alpilotx]

I meant the default terrain in XP10 looks not good:

Baron_58_1.jpg

Baron_58_2.jpg

Well, it extremely depends on the landscape too (and as so many users tend to look primarily at their home region ... you often have people who are disappointed and a lot of them are very happy ... depending on how "interesting" their landscape is)! If the raw data - landclass / elevation / water features - gives you a diverse environment (like usually is the case in the mountains etc.) ... then its much easier to generate an interesting landscape out of it. Whereas when the raw data does not give you much variety (for example the landclass says "grass" for 1000 square miles ... and the elevation is flat like a pan) then there it is very hard to make such a scenery look interesting (especially when you need to generate it automatically - which is the case when you model the entire planet). For these cases we have quite a few shader tricks (and even some tricks in the scenery generator) ... like a shader which tries to "mix"/alternate textures via some noise function (to make it less repetitive) ... etc. but even those can't make a big / flat / single landclass area "interesting". This is a complex topic which we work on since quite a long time ... but until now, we didn't have better solutions (yes, some better ideas we have ... but they are something for the future).

 

In you first screenshot there are two of those "dull", repeating patterns ... the city ... which is really bad until now (but will be improved with 10.25/ 10.30), and the forest (I don't know how much tehy will improve ... but ok, its at least looks like a forest).

In the second screenshot you complain about the "crop" textures ... again, a very complex issue, because the pattern of crop patterns can be extremely varying and its almost impossible to make this "perfect" by only using a few textures. So here we have to rely on some kind of "plausible" approximation but not have perfect approach .... Still, you can clearly recognize it as agriculture area. In this area there are some ideas for the distant future .... like using the pattern of roads (if they exist in the source data) to derive some more realistic agriculture patterns (but this is nothing you should expect in a year or two).

 

Still, you always have to remember, that with a Global Scenery style "generic" approach, you always have to simplify and work with generic textures ... which can never result in a perfect representation of real world (like a photoscenery). On the other hand, the better (the more diverse) your raw geo data is, the better can this approach work ... (again, I recommend to compare some of the mountainous areas ... there I wouldn't say that the default terrain is "bad"    ).

 

PS: aah, and I see that your screenshots are from Ukraine! There we have another - raw data - problem .... until now I don't have detailed landclass data for that region and have to rely on the more coarse, 300m resolution global data. Whereas in USA, Europe, Alaska, NZ and now Canada I have 100m or better resolution landclass data .... which makes a very big difference!

Edited October 9, 2013 by alpilotx

[Rhard]

Well, it extremely depends on the landscape too (and as so many users tend to look primarily at their home region ... you often have people who are disappointed and a lot of them are very happy ... depending on how "interesting" their landscape is)! If the raw data - landclass / elevation / water features - gives you a diverse environment (like usually is the case in the mountains etc.) ... then its much easier to generate an interesting landscape out of it. Whereas when the raw data does not give you much variety (for example the landclass says "grass" for 1000 square miles ... and the elevation is flat like a pan) then there it is very hard to make such a scenery look interesting (especially when you need to generate it automatically - which is the case when you model the entire planet). For these cases we have quite a few shader tricks (and even some tricks in the scenery generator) ... like a shader which tries to "mix"/alternate textures via some noise function (to make it less repetitive) ... etc. but even those can't make a big / flat / single landclass area "interesting". This is a complex topic which we work on since quite a long time ... but until now, we didn't have better solutions (yes, some better ideas we have ... but they are something for the future).

 

In you first screenshot there are two of those "dull", repeating patterns ... the city ... which is really bad until now (but will be improved with 10.25/ 10.30), and the forest (I don't know how much tehy will improve ... but ok, its at least looks like a forest).

In the second screenshot you complain about the "crop" textures ... again, a very complex issue, because the pattern of crop patterns can be extremely varying and its almost impossible to make this "perfect" by only using a few textures. So here we have to rely on some kind of "plausible" approximation but not have perfect approach .... Still, you can clearly recognize it as agriculture area. In this area there are some ideas for the distant future .... like using the pattern of roads (if they exist in the source data) to derive some more realistic agriculture patterns (but this is nothing you should expect in a year or two).

 

Still, you always have to remember, that with a Global Scenery style "generic" approach, you always have to simplify and work with generic textures ... which can never result in a perfect representation of real world (like a photoscenery). On the other hand, the better (the more diverse) your raw geo data is, the better can this approach work ... (again, I recommend to compare some of the mountainous areas ... there I wouldn't say that the default terrain is "bad"    ).

 

PS: aah, and I see that your screenshots are from Ukraine! There we have another - raw data - problem .... until now I don't have detailed landclass data for that region and have to rely on the more coarse, 300m resolution global data. Whereas in USA, Europe, Alaska, NZ and now Canada I have 100m or better resolution landclass data .... which makes a very big difference!

Yes, I understand that the algorithm is not really simple and is landklass dependent. I also see (not only for this issue) that Ben and Co did a very good job to create global areas. But some how I have an impression (and not only me), that this algorithms not really adjusted to show us all possibilities of X-Plane scenery/render engine.

Just my ideas:

On the second screenshot we can calculate an area of  " "grass" for 1000 square miles" and tell X-Plane make a texture tile size be proportional to this value and make direction of such fields not so chaotic. If you look Ukrainian fields in flat areas they are really big. Also if you put a single forests lines between some fields it will look much better!  Random forests and another staff is not what I see in such "empty" areas as well. 

If we have no data, then we need to create it..

 

P.S. Water fields is another problem. Why don't smooth the corners and long edges? It's shouldn't be so difficult with shape-files. Even without besier curves it can look better.

 

So, i think I am finish here with critic..

Edited October 9, 2013 by Rhard

[alpilotx]

Yes, I understand that the algorithm is not really simple and is landklass dependent. I also see (not only for this issue) that Ben and Co did a very good job to create global areas. But some how I have an impression (and not only me), that this algorithms not really adjusted to show us all possibilities of X-Plane scenery/render engine.

Yes! of course the engine can do a lot ... but its very complex to artificially create a lot of content (we already do much more than you might imagine ) ...

 

On the second screenshot we can calculate an area of  " "grass" for 1000 square miles" and tell X-Plane make a texture tile size be proportional to this value and make direction of such fields not so chaotic. If you look Ukrainian fields in flat areas they are really big. Also if you put a single forests lines between some fields it will look much better!  Random forests and another staff is not what I see in such "empty" areas as well.

As I told in my previous posts: we have ideas about this too ... but none of them are trivial, nor will they happen in the near future (like including the road network structure etc.). Forest lines are a similar idea (do you know my "treeline farms" add-on scenery for Europe/USA ?? You find it on my website) ... which improves exactly this problem ...

 

And another thought: you tell "If you look at Ukrainian fields" .... here you might see the problem. If I (I as a human!) look, then I see that ... but now tell a computer: these are Ukrainian fields, and they look this or that ... and now tell this for 100 different regions ... and define for all of them a unique look (create textures etc.). You see .... this is the problem with a Global Scenery where you want on the one side make everything look as good as possible (well, at least plausible) while on the other side you need to take care of complexity .... and balance the need for human power. If we had a staff like GTA5 has (many hundreds of artists etc.), then anything would go ... but we have to work with what we have at hand ...

 

 

If we have no data, then we need to create it..

We try (and as I told you, we already do more of that than many of you might imagine ...)

 

 

P.S. Water fields is another problem. Why don't smooth the corners and long edges? It's shouldn't be so difficult with shape-files. Even without besier curves it can look better.

You mane water features (water fields ??? )  like lakes etc ... Then you don't know the other implications of this. Water is until now part of the base triangle mesh! And the more you refine the coastline, the more (and smaller and smaller) triangles you induce into the mesh .... which all "cost" extra. You will see, that in my HD Mesh Scenery v2 I already improve this (having used less vector simplification on the data) ... but even there you won't see super smooth coastlines (but definitely better than in the Global Scenery).

[Tom Knudsen]

Seems to me that the triangler mesh is both a bottelneck and an annoying layer that you always need to do battles against Al (mind if I call you that)

Anyway, there are 3 key items I would love to get some feedback on if possible

 

* Repetable texture as Rhard described

- I guess this is a blend of the triangler mesh and coding? Would the performance impact be less or better with i.e a typology mesh (not sure if this is even a factor) or even tessellation mesh

 

* Waterclass

- I have noticed that with the latest update 10.22 there are many missing lakes and uncorrect "shorelines" in the default update. Have not done any comparrasing to earlier versions,

but I did notice this when making OSM scenery (country). OSM scenery is not correct data such as typology raw data from shuttlemissions or similar, but they do try to be accurate.

I went kinda away from it as I did notice a sport among 12-14 year old youths that they willingly mess up thinks like Openstreetmap and wikipedia sites..

 

* Terrain Textures or Landclass

- I understand now that triangler mesh tends to brake up (correct?) a texture tile in order to fit it around i.e a mountain peak.

One can see this if you look closely at a mountain and then streight down from lets say 3000 feet.

 

In XP I noticed that there are many mountains that have for example a top consisting of several texturebits rotated in several different directions making the rock texture look plausible, but far from realistic.

I guess plausible is the Austin way, but i feel this is degrating the overall look and feel of X-Plane. I must add that X-Plane has the most beutiful and realistic default textures in comarasing to other simulators.

But it has as Rhard pointed out, several errors.

 

Repeating textures are one annoying factor, missplaced textures are a strong contender.

 

 

Anyway, I know you have a lot to do and I would imagine everybody els knows this as well. But my question to you is dedicated to this

 

1. Will there not be a huge effort for X-plane to egange more people to do collobration work?

2. Is even X-Plane adaptable to 3rd party vendors that i.e make ground textures like f.eks. "Ground Environment" that aims to start with X-Plane development

3. Are there possibilities for me to add lakes and small waters into x-plane that are missing?

[alpilotx]

Seems to me that the triangler mesh is both a bottelneck and an annoying layer that you always need to do battles against Al (mind if I call you that)

No, no battles here with the triangle mesh .... I have accepted it with all its pros and cons .... and it is anyways the usual way - in computer graphics - to represent terrain. So, nothing really unusual. It only seems annoying if you didn't completely understand how it works, why it works the way it works, why computer graphics likes to use it and why its still not as bad as you might think !

 

* Repetable texture as Rhard described

- I guess this is a blend of the triangler mesh and coding? Would the performance impact be less or better with i.e a typology mesh (not sure if this is even a factor) or even tessellation mesh

Texturing happens "on top" of the mesh and is not necessarily directly depending on it. The texture repetition is an absolutely basic "feature" of how texturing works in computer graphics. And not making it so repetitive is done by different shaders, which "procedurally" change something in the way the textures get "painted" on the mesh (for example by mixing two different textures ... or using tiling etc. ... we have both the COMPOSITE and TILING shader technique in use ... where applicable).

 

Maybe you should read a little bit about how texturing in general terms works (in computer graphics, not just flight sims). That might help to better understand what is happening here.

 

 

* Waterclass

- I have noticed that with the latest update 10.22 there are many missing lakes and uncorrect "shorelines" in the default update. Have not done any comparrasing to earlier versions,

but I did notice this when making OSM scenery (country). OSM scenery is not correct data such as typology raw data from shuttlemissions or similar, but they do try to be accurate.

I went kinda away from it as I did notice a sport among 12-14 year old youths that they willingly mess up thinks like Openstreetmap and wikipedia sites..

If a lake is missing or not has nothing to do with the version of X-Plane (I just want to make this clear, so people don't mix it up) .... but only with what is in the DSF scenery files. When you don't change them, then their look (at least what water features are "in" it) should not change. This is a plain data issue! All DSF Mesh scenery 8either the Global Scenery or my HD Mesh v1) is all based on old OSM data from summer 2011. So it only contains water features which were present at that time ...

 

Now the new HD Mesh Scenery v2 (or any other, newer DSFs done by Laminar) will have at least my current / latest OSM data import from 2nd of October (2013) .... And any lakes / rivers / coastlines which have been added until that date, will show up (definitely in my HD Mesh Scenery v2). AND, to answer you other question ... this is the - supported - way how you add lakes / rivers etc. to X-Plane scenery. And I have told this many times in my forum threads over the last months so people had time to edit OSM during that time ... (and a few users did a lot of additions - a big thank you to them!) ...

 

And whether OSM is good or bad ... thats another discussion. From my point of view, its still a fantastic source (even if it has its quirks etc.) So its quite sure, that it will remain one of the main data sources for X-Plane in the coming years.

 

 

* Terrain Textures or Landclass

- I understand now that triangler mesh tends to brake up (correct?) a texture tile in order to fit it around i.e a mountain peak.

One can see this if you look closely at a mountain and then streight down from lets say 3000 feet.

I can only repeat what I told above. Please read a little bit about texturing in general to understand that at least in its basic for its nothing X-Plane special, but a very basic "operation" in computer graphics (and even using shaders to improve the way how textures are applied is nothing special nowadays ... every game does it billion times). Understanding this makes it clear, what happens on the mountains (and everywhere else in the scenery) ...

 

 

 

In XP I noticed that there are many mountains that have for example a top consisting of several texturebits rotated in several different directions making the rock texture look plausible, but far from realistic.

I guess plausible is the Austin way, but i feel this is degrating the overall look and feel of X-Plane. I must add that X-Plane has the most beutiful and realistic default textures in comarasing to other simulators.

But it has as Rhard pointed out, several errors.

 

Guys, I really don't enjoy to repeat myself ... There are things which are done in a given way because its the economic and feasible way to achieve a halfway plausible look. There are always other possibilities, but none of them will happen by magic (and always needs a lot of work in some way). So, for example the decisions to use the rotating shaders in a way we use them is born from a necessity (as it makes mountains look much better than with non rotating shaders!) and from possibilities you are given by shaders (and what the target hardware can handle with good enough performance! ... this is something which many forget or can't even understand!).

 

As a contrast, you can use photoscenery .... but you might know, that photoscenery has its own drawbacks too (so there is no super perfect solution .... orat least each solution has its own costs, advantages and disadvantages which need to be weighed up!).

 

 

Repeating textures are one annoying factor, missplaced textures are a strong contender.

Again ... you can avoid misplaced textures when you create scenery by hand .... but now do that on a Global Level   Good luck ...

 

 

Anyway, I know you have a lot to do and I would imagine everybody els knows this as well. But my question to you is dedicated to this

 

1. Will there not be a huge effort for X-plane to egange more people to do collobration work?

2. Is even X-Plane adaptable to 3rd party vendors that i.e make ground textures like f.eks. "Ground Environment" that aims to start with X-Plane development

3. Are there possibilities for me to add lakes and small waters into x-plane that are missing?

Yes, you guys have taken away quite some time from me today ...

 

And I think I have answered most of those question above.

 

About the "collaboration" .... well ... its a good thing. See OSM! Its a massively collaborative work, which helps X-Plane in a really good (and big way) .... And there are other projects in the X-Plane world, where you can collaborate with other users. its just that most people only talk about this - a lot - and rarely really do something about it.

Edited October 9, 2013 by alpilotx

[Tom Knudsen]

 

Maybe you should read a little bit about how texturing in general terms works (in computer graphics, not just flight sims). That might help to better understand what is happening here.

 

I study 3D design and animation right now, this is part of why I ask so much and are so keen to understand everyting.

 

 

So, for example the decisions to use the rotating shaders in a way we use them is born from a necessity (as it makes mountains look much better than with non rotating shaders!) and from possibilities you are given by shaders (and what the target hardware can handle with good enough performance!

 

Do you have controll over those rotating shaders or is it auto. I know about rotating matrix in 2D but not sure if you are talking about the same thing or not.

 

 

If a lake is missing or not has nothing to do with the version of X-Plane (I just want to make this clear, so people don't mix it up) .... but only with what is in the DSF scenery files. When you don't change them, then their look (at least what water features are "in" it) should not change. This is a plain data issue! All DSF Mesh scenery 8either the Global Scenery or my HD Mesh v1) is all based on old OSM data from summer 2011. So it only contains water features which were present at that time ...

 

Well these lakes has never been in X-Plane, well at least in v8-9 and 10, but they are large and absolutly there in IRL, I have also adjusted this in Openstreetmap so I will be happy if they

apperare in HD Mesh v2 for sure (Thanks for doing that btw)

 

 

So to summen up this discussion

I tend to believe there is nothing that cannot be done, it just take a bit longer and cost a whole lot more.

But to say certain things cannot be done is just and indication that the person does not have the knowledge or is able to see the final solution (with non perticular in mind)

 

 

“The world is moving so fast these days that the man who says it can't be done is generally interrupted by someone doing it.”

― Elbert Hubbard

 

Good luck on your HD Mesh Alpilot, and once again thank you for contributing to the community

Edited October 9, 2013 by Tom Knudsen

[alpilotx]

Sorry ... but I have spent now so much time on this (and even in PM with Roman) .... I need to take a break here (and now). I think ,most of the things are already more or less answered here (or "between the lines" ).

 

PS: about the rotation: that is "automatic" and done by the shader (but of course, even a shader can be reprogrammed )

[alpilotx]

After some sleep ... some "final" notes

 

Tom, I still have the feeling that you have some way ahead of you in understanding some basic things of computer graphics (the way they are in 2013). Especially when I see a comment like this on Bens blog: "I guess the bottleneck if one can see it like that is the 2D triangle mesh right now. It seems that prevents more details." ... its not a bottleneck ... its the way current graphics hardware understands what you want it to do ... (heck, every other flight sim works - effectively - this way at its core! And with tessellation, it will be even less a bottleneck ... but I have written about it so much here, maybe you should re-read) ...

 

... but I definitely applaud, that you are study "3D design and animation" ... this should lead you in the right direction. I see that you are a visionary guy, who has a lot of great ideas and indeed, those ideas can move a lot of things forward in the long run. But this way of thinking is sometimes a bit hindering when you need to do real work in an existing environment, with given set of possibilities. And with "give possibilities" I mean things like: how does 3D graphics work at its core, and what can you do with it in a meaningful (and well performing way)? What does a given platform - like X-Plane - provide you (with all of its pros and cons), what can you achieve with it, and what is outside of its scope ... Then one has to always evaluate: if I want to improve / change something, does it makes sense withing the constraints of the platform? Can I improve the platform itself? How much does it "cost" to make that improvement? Will that improvement work on current and future hardware? Is that change something which is "nice in my head" but something which is very hard to "tell" a computer (or a graphics card!) without killing it (and having 0.1 FPS as a result).

 

These are all things, which a developer (especially Ben Supnik) needs to consider all the time .... otherwise his work would easily become "Art for art's sake" (aka ''l'art pour l'art'' ... http://en.wikipedia.org/wiki/Art_for_art%27s_sake) .... where you create something beautiful, but something that just doesn't works on anyones computer. BUT ... this doesn't mean that Ben and Austin don't innvoate in their engine. They do it all the time (and I am part of the team for quite a few years and could see this happen behind the doors) .... but they cant just innovate all the time in one area, and forget a lot of other things they need to do. So they (and me too, by the way  !) always need to allocate their resource wisely ... which can easily lead to the impression, that "they are not doing anything about this or that" (from ones own perspective) .... while in reality they are working on other important things which please another 1000 users (but maybe a feature which is less important to you).

 

So, I agree, that there is "nothing that cannot be done" ... BUT, in real life it always means much more than just coming up with fancy ideas (which we all have more than enough) ... its always about weighing pros and cons, (work) resource allocation, cost benefit calculations, feasibility studies (what makes sense at the moment) etc. etc. etc. ...

 

---

And yes, if you have added your missing lakes in OSM ... then you will see them very likely. Because THIS is the only way (well or any other digital form) to tell a computer what is there in real life! Because you can argue as long as you want with X-Plane, telling it "hey, show me that lake ... its there in real life!!!" .... it won't understand you. The only language it understands is data, data manipulation, algorithms etc. (but not "human language" arguing) ...

---

Edited October 10, 2013 by alpilotx

[riccardo74]

Sorry guys I can't read entire topic, will be this a free upgrade?

[Cameron]

Sorry guys I can't read entire topic, will be this a free upgrade?

 

Remarkably, yes.

[alpilotx]

Sorry guys I can't read entire topic, will be this a free upgrade?

Sorry, riccardo ... but a few questions / comments brought this thread "a bit" off topic ...

 

And of course it will be free (just a s free as all my other work - like the previous v1 - was) ... I just like to call it donationware    (but that means effectively: free).

[riccardo74]

Thank you very much guys, I've red it would be available from summer 2013, logically an estimated time release, so any news on possible release?

[alpilotx]

It has been postponed a few times for "different" reasons, but in the meantime things are developing quite well ... so, at the moment I expect the release to be in approximately 3-5 weeks.

[riccardo74]

Thanks so much.

[Tom Knudsen]

 

... but I definitely applaud, that you are study "3D design and animation" ... this should lead you in the right direction. I see that you are a visionary guy, who has a lot of great ideas and indeed, those ideas can move a lot of things forward in the long run. But this way of thinking is sometimes a bit hindering when you need to do real work in an existing environment, with given set of possibilities. And with "give possibilities" I mean things like: how does 3D graphics work at its core, and what can you do with it in a meaningful (and well performing way)? What does a given platform - like X-Plane - provide you (with all of its pros and cons), what can you achieve with it, and what is outside of its scope ... Then one has to always evaluate: if I want to improve / change something, does it makes sense withing the constraints of the platform? Can I improve the platform itself? How much does it "cost" to make that improvement? Will that improvement work on current and future hardware? Is that change something which is "nice in my head" but something which is very hard to "tell" a computer (or a graphics card!) without killing it (and having 0.1 FPS as a result).

 

 

I think you summened up all things very nicely with this section sir, well said. I guess this is the major fault with us consumer. We do not see the backend, just only what our own eyes see and know what our heart desire.

We see cool things show up now and then, and read i.e what Ben writes and make conclusions and wishlist longer than our own noses reach.

 

It is very difficult og limit our needs and wishes if we do not know how things work, the end user only care about product cost and we tend to forget new feature as soon as they have been implemented and used.

Think I speak for everybody when I say we appreciate all the effort designers and developers do in order to sell and refurbish their product. As X-Plane is an ongoing product and we the fans are contributing to its

success, well we do hope the developer listen to our needs and do what is possible to fix, amend or implement new features.

 

Its a visious circle he he..

 

 

 

And yes, if you have added your missing lakes in OSM ... then you will see them very likely. Because THIS is the only way (well or any other digital form) to tell a computer what is there in real life! Because you can argue as long as you want with X-Plane, telling it "hey, show me that lake ... its there in real life!!!" .... it won't understand you. The only language it understands is data, data manipulation, algorithms etc. (but not "human language" arguing) ...

 

Could one edit the default DSF file and add water to it via i.e polygons?

[alpilotx]

 

Its a visious circle he he..

Amen

 

Could one edit the default DSF file and add water to it via i.e polygons?

In theory yes (as its always possible to edit data structures), in practice no (and I really mean NO ).

[alpilotx]

UPDATE: I have good news. I am now on the "short final" before landing .... Most of the tests are finished (the big regions like USA, Europe and most of Canada ... Alaska and some other regions still need to be visited) and I have even started packing up the ZIP files today. I also plan to start uploading the first big ZIP files tonight ... as the whole upload process will take many days (we are talking about 55,6 GBytes of data) anyways. In parallel I will do the final preparations (website etc.), tests, screenshots ... so, with a little luck, I might be able to officially release within the next 5-10 days (hopefully ...  there are still some potential "road blocks" ahead, but I hope they wont be too numerous).

[cabnz1]

Yay - Thank you for the update

[AtomicFrawg]

Just go ahead and upload it directly to my computer. 

 

I can't wait for this.

Edited November 19, 2013 by AtomicFrawg

[alpilotx]

By the way guys ... after I have packaged most of the data, my script did also generate the final coverage map (it will be a KML overlay helping in the download process) which looks like this (only the northern most tiles DON'T extend that far up ... only a few degrees instead of 10):