Jump to content

RWC and cloud coverage shape


Havner
 Share

Recommended Posts

Quoting the manual:

"The clouds shown don’t match perfectly with X-Plane’s weather radar.
Our clouds are positioned very precisely, often even more precisely than X-Plane’s own default clouds. As a result, X-Plane’s built-in weather radar display will show the same general shape of weather patterns that RWC displays, but the correlation won’t be perfect."

See the screenshots. Every time I encounter an overcast, the edge consists of perfect lines. Like it was composed from perfect squares of clouds.

I'd say that it's RWC that shows some "general shape". Very general.

How is this more precise? Is this supposed to be like that? It looks plain ugly.

c4_1.jpg

c4_2.jpg

Edited by Havner
  • Upvote 2
Link to comment
Share on other sites

Another thing that RWC doesn't seem to work well with is the full overcast layers. I would be approaching clouds and then all of a suddon there is a full overcast layer below me. Also, as I'm flying, I should be over an overcast layer, but RWC puts scattered.

Edited by poodster
Link to comment
Share on other sites

4 hours ago, Havner said:

I'd say that it's RWC that shows some "general shape". Very general.

I'd actually say that you're guessing when you make this statement. Just like X-Plane's default weather engine guesses and approximates.

4 hours ago, Havner said:

How is this more precise? Is this supposed to be like that? It looks plain ugly.

Precision is a matter of being...precise! Placing clouds more precisely where they are compared to approximating is very different. Just because you feel something is ugly does not mean it's not precise. For what it's worth, I think the Airbus series of aircraft are ugly, but sadly they're real. :( My opinion means nothing on their reality and shape.

Keep in mind the X-Plane weather map screen is NOT a real weather radar depiction. It is the approximation of Laminar's algorithm to say "this probably looks something like this".

I could take a pencil and draw a straight line on a piece of paper, but that would be my approximation. The ruler will keep my line precisely straight. SkyMaxx Pro has a high resolution grid, far more capable and precise than that of X-Plane's default weather engine. Add in the ability for five cloud types and double the cloud layers and it starts to not even be comparable, especially when you come across such areas of weather using things to the full potential!

3 hours ago, poodster said:
Another thing that RWC doesn't seem to work well with is the full overcast layers. I would be approaching clouds and then all of a suddon there is a full overcast layer below me. Also, as I'm flying, I should be over an overcast layer, but RWC puts scattered.

It's actually a selection of particle clouds attempting to create an overcast look and once you get far enough into it it will transition to solid or broken stratiform. This is covered in the manual, by the way. You should check it out if you haven't done so already to get a better understanding. X-Plane's own default system can make some rather awful scenes when things should be overcast if not enough cloud puffs are present (I was actually just witnessing this today...and I do mean AWFUL).

All of this said, Real Weather Connector is in its infancy at version 1.0, and as we've always shown, our commitment to the weather implementation is more than just surface level. SkyMaxx Pro is now in its third iteration (with many updates in-between). We have continued to better it over time, and Real World Connector will get the same treatment.

Sorry if you guys aren't pleased! That's too bad, but thanks for the criticism anyhow. :) Consider going out and getting some fresh air. I've read a few of both your guys' posts in various places in recent days, and the negative nancy side shines at times. You know it's bad when I interact with thousands of people over a couple of days during release times like this and a select few stick out in memory. It feels almost like you enjoy compulsively complaining or stirring, but reality is there's no need to be so mad at every little thing always. Try to see the positives in life too. We're not going anywhere, and neither is Real Weather Connector. Things will always improve.

Just for grins I'll add this customer photo I saw on Facebook by Andreas Mueller. It looks pretty great, if I do say so myself. I, for one, am very excited about Real Weather Connector!

13048162_10208544774432105_3530320390770

Link to comment
Share on other sites

33 minutes ago, Cameron said:

I'd actually say that you're guessing when you make this statement. Just like X-Plane's default weather engine guesses and approximates.

Precision is a matter of being...precise! Placing clouds more precisely where they are compared to approximating is very different. Just because you feel something is ugly does not mean it's not precise. For what it's worth, I think the Airbus series of aircraft are ugly, but sadly they're real. :( My opinion means nothing on their reality and shape.

Keep in mind the X-Plane weather map screen is NOT a real weather radar depiction. It is the approximation of Laminar's algorithm to say "this probably looks something like this".

I could take a pencil and draw a straight line on a piece of paper, but that would be my approximation. The ruler will keep my line precisely straight. SkyMaxx Pro has a high resolution grid, far more capable and precise than that of X-Plane's default weather engine. Add in the ability for five cloud types and double the cloud layers and it starts to not even be comparable, especially when you come across such areas of weather using things to the full potential!

It's actually a selection of particle clouds attempting to create an overcast look and once you get far enough into it it will transition to solid or broken stratiform. This is covered in the manual, by the way. You should check it out if you haven't done so already to get a better understanding. X-Plane's own default system can make some rather awful scenes when things should be overcast if not enough cloud puffs are present (I was actually just witnessing this today...and I do mean AWFUL).

All of this said, Real Weather Connector is in its infancy at version 1.0, and as we've always shown, our commitment to the weather implementation is more than just surface level. SkyMaxx Pro is now in its third iteration (with many updates in-between). We have continued to better it over time, and Real World Connector will get the same treatment.

Sorry if you guys aren't pleased! That's too bad, but thanks for the criticism anyhow. :) Consider going out and getting some fresh air. I've read a few of both your guys' posts in various places in recent days, and the negative nancy side shines at times. It feels almost like you enjoy compulsively complaining, but reality is there's no need to be so mad at every little thing always. Try to see the positives in life too. We're not going anywhere, and neither is Real Weather Connector. Things will always improve.

Just for grins I'll add this customer photo I saw on Facebook by Andreas Mueller. It looks pretty great, if I do say so myself. I, for one, am very excited about Real Weather Connector!

13048162_10208544774432105_3530320390770

I've actually only posted this about RWC in one place only, which is here. The other event was talking about another subject which you know ;)
And to be clear, I'm more than pleased! This addon is fantastic. I was flying today, and it went from a crystal clear day to a few clouds here and there and then finally a massive snow storm. It was very very cool! I actually want to fly in a cloudy area with wide ranges of clouds now because of this addon. Transitions are very smooth, only one time it lagged for 1 second because it was updating the weather but that's minor and I probably won't notice it that much anyway. (1 second of lag during a whole flight is a massive improvement I can't thank you and the devs enough for)

I'm not trying to be negative nancy as you put it lol. I'm just puting some criticism here since It helps with future development. And to be honest, those 2 criticisms i put in this thread are the only 2 I have so far. The rest of my thoughts are very positive thoughts.

I'm sorry if i seem angry or negative. I did not mean to be cold hearted yesterday on facebook. It's exam week and I guess the stress and anger of possibly failing an exam and stressing over 4 others got to me. Sorry cameron and I hope you forgive me for that. 

 

Thank you for this amazing addon! Best addon of 2016 so far (until IXEG comes out ;) ) and I can't wait for furthur development. 

 

Link to comment
Share on other sites

7 minutes ago, poodster said:

I'm sorry if i seem angry or negative. I did not mean to be cold hearted yesterday on facebook. It's exam week and I guess the stress and anger of possibly failing an exam and stressing over 4 others got to me. Sorry cameron and I hope you forgive me for that. 

Best of luck on your exams, dude. Study hard, and play harder...just don't tell your parents.

Link to comment
Share on other sites

12 hours ago, Cameron said:

I'd actually say that you're guessing when you make this statement. Just like X-Plane's default weather engine guesses and approximates.

Then please answer me a simple questions you didn't. Guessing aside.

1. Is this supposed to look like this?
2. Is RWC really putting specific types of clouds in squares within the max area set in settings?
3. Can something be done about that?

Quote

Keep in mind the X-Plane weather map screen is NOT a real weather radar depiction. It is the approximation of Laminar's algorithm to say "this probably looks something like this".

I don't care about X-Plane's weather map. Just posted it for reference. And X-Plane's "approximation" is not causing any rough edges. It has other problems, sure, but at least the edges of weather fronts look plausible.

I don't get your negativity towards X-Planes weather system. And this is what I remember from you everywhere I look. If by approximation you mean it following METAR data, then maybe, I haven't parsed them. But if you look at X-Plane's ability to put small "clouds" or small coverage with different sizes in several places it is actually better. It cannot do vertical variation of the thickness of the clouds though.

Quote

I could take a pencil and draw a straight line on a piece of paper, but that would be my approximation. The ruler will keep my line precisely straight. SkyMaxx Pro has a high resolution grid, far more capable and precise than that of X-Plane's default weather engine. Add in the ability for five cloud types and double the cloud layers and it starts to not even be comparable, especially when you come across such areas of weather using things to the full

Negativity aside, you didn't really said anything constructive to what I showed you.

I don't really care about perfect precision. I won't ever see exactly the same weather I can see outside my window. I want a realistic look with fairly similar weather that exists in real life. What I posted before is not exactly realistic.

And telling me that you remembered my negativity while this particular issue is the only thing I've been negative about is plain rude. Please, check my posts again.

I'm not saying in any way this is a bad product. But with your overall praises about it when I saw that edges I was completely shocked that you would call that realistic.

Please, don't be this place (those forums) where only praises are allowed. I saw something completely unrealistic and (sorry, even if subjective) ugly. And I posted about it. As a paying customer I feel I have a right to do that.

Edited by Havner
Link to comment
Share on other sites

9 minutes ago, Havner said:

And telling me that you remembered my negativity while this particular issue is the only thing I've been negative about is plain rude.

Nope. I remember you clearly from a post at another website. This particular issue is just "in addition to" another post(s) I have seen from you. :)

9 minutes ago, Havner said:

Negativity aside, you didn't really said anything constructive to what I showed you.

Actually, I said this:

"All of this said, Real Weather Connector is in its infancy at version 1.0, and as we've always shown, our commitment to the weather implementation is more than just surface level. SkyMaxx Pro is now in its third iteration (with many updates in-between). We have continued to better it over time, and Real World Connector will get the same treatment."

Link to comment
Share on other sites

26 minutes ago, Havner said:

Yes, I posted about this issue on "another site". I said "this particular issue". Not "only my posts on this site". Whatever.

Not just this issue. Speaking about another thread as well about SkyMaxx Pro.

26 minutes ago, Havner said:

And again you didn't answer my questions.

I have zero desire to engage in discussion with you. Perhaps if you had approached this situation better and "constructively" things would be different. Instead, you're out with words like:

"it's an eye-candy, nothing more." with regards to SkyMaxx Pro v3 on another website.

or

"I'd say that it's RWC that shows some "general shape". Very general."

or

"It looks plain ugly."

So, yes, excuse me while I call you out for your negativity as of late. Thankfully we have thousands of others not acting the same way you are, and if they have something constructive to say, they say it in a constructively positive manner.

I've already answered you anyhow. You apparently don't like hearing that RWC is at version 1.0 and will of course get better. Such commitment has always been shown in the SkyMaxx Pro series. So, in your own constructive words: "Whatever."

Link to comment
Share on other sites

RWC allows for a much tighter grid of data points.

These data points are so small that you can see the cellular structure of the data above.

X-Plane has a larger grid and is asking for a more general rendition of "scattered clouds".

RWC is attempting to provide precision placement of cell structures over a very wide area that can be flown around.

What is essentially needed is some form of 3D volumetric cloud anti aliasing between SMP's already complicated cloud structures.

This may seem incredibly simple in theory; "Why the hell can I see blockies!" ... it's actually quite a complex problem when you think about it in the context and detail required above.

 

I hope that helps shed some light on why version 1.x might look a little rough here and there.

( And, also... Frank from Sundog wrote it, not me. So he might just tear my theory down in a few hours. :D )

Edited by Ben Russell
  • Upvote 1
Link to comment
Share on other sites

2 minutes ago, Cameron said:

"it's an eye-candy, nothing more." with regards to SkyMaxx Pro v3 on another website.

It's taken out of context and I don't consider this negativity. I wrote that Skymaxx Pro could only display uniform weather. Which objectively means it's eye-candy when you're after realistic weather (which means weather fronts) depiction. Sorry you feel bad about it, but obviously you had a similar feeling. That's the reason you invented RWC after all.

2 minutes ago, Cameron said:

"I'd say that it's RWC that shows some "general shape". Very general."

"It looks plain ugly."

The thing on the screenshot looks ugly. And I stand by it. Never said that the RWC (as a whole) looks ugly.

And the rest of the quotes are from this very thread. Completely ignoring all the rest of stuff I wrote.

Sincerely sorry if you felt offended. I thought that people can handle a little criticism.

My second post was more toned down and constructive but it seems you didn't feel that way. I also didn't want to get into a discussion with you. That's why I summarized what I meant in my second post. I'm sorry I couldn't get you to discuss this in a normal matter.

I read that you want to improve on RWC. What I wanted to hear is whether you consider this particular issue an issue at all. Which gives me an indication if it will improve in the future.

Best of luck in the future with improving RWC! I'll be waiting for future patches. Take care.

 

Link to comment
Share on other sites

12 minutes ago, Ben Russell said:

RWC is attempting to provide precision placement of cell structures over a very wide area that can be flown around.

What is essentially needed is some form of 3D volumetric cloud anti aliasing between SMP's already complicated cloud structures.

That's what I wanted to hear. Thank you. Blurring the edges (of the grid map) would obviously solve it. I just don't know whether SMP can actually display such resolution that would fix the issue. I also didn't know where the aliasing came from in the first place. Whether METAR data is limited, RWC processing is or SMP ability to actually place clouds in high resolution grids. All those things where my intent. To know them.

Quote

This may seem incredibly simple in theory; "Why the hell can I see blockies!" ... it's actually quite a complex problem when you think about it in the context and detail required above.

I completely understand that I _can_ see them. I'm a programmer by occupation and did my fair share in 3D graphics. I just wanted to know whether this is actually normal now and if so will it be improved upon.

Quote

I hope that helps shed some light on why version 1.x might look a little rough here and there.

( And, also... Frank from Sundog wrote it, not me. So he might just tear my theory down in a few hours. :D )

Cheers!

Edited by Havner
Link to comment
Share on other sites

8 minutes ago, Ben Russell said:

Except generated dynamically in 3D, on the fly, between whatever cloud structures there are.... are you sure you want to eat that kind of performance penalty?

Well, if SMP is just following a grid provided by RWC majority of the performance is eaten by actually rendering the clouds. Anti aliasing a map can be done _very_ rarely (like every few minutes for a new map) and is not a costly operation.

Edited by Havner
Link to comment
Share on other sites

4 minutes ago, Havner said:

Well, if SMP is just following a grid provided by RWC majority of the performance is eaten by actually rendering the clouds. Anti aliasing a map can be done _very_ rarely (like every few minutes for a new map) and is not a costly operation.

Sundog Software is very experienced in this domain. They sell an entire array of Weather SDK's.

It may sound trivial to just AA a bitmap, but think about it, to have any real effect at reducing the saw tooth the cell data tracking is going to go up by a factor of at least four.

If not all you're going to do is end up with blurred data from nearest neighbour cells which gives you the exact same saw tooth cell resolution with weaker weather conditions in each grid position at the fringe..

It's all a trade off...

 

History has shown the commitment to continual product improvement and customer loyalty programs that these products come with... if we can find a way I'm sure you'll hear about it.

Edited by Ben Russell
Link to comment
Share on other sites

12 minutes ago, Ben Russell said:

It may sound trivial to just AA a bitmap,

In terms of dealing with the edges it actually is. There are very good algorithms to deal with this particular issue. Like HQ3X:

Test_nn.pngTest_hq3x.png

(Images courtesy of Wikipedia)

Quote

but think about it, to have any real effect at reducing the saw tooth the cell data tracking is going to go up by a factor of at least four.

Yes, this might be an issue. That's why I wrote it might be down to SMP not being able to actually process such a high resolution map. But I wouldn't call a present state of it realistic.

Quote

If not all you're going to do is end up with blurred data from nearest neighbour cells which gives you the exact same saw tooth cell resolution with weaker weather conditions in each grid position at the fringe..

Things like HQ3X actually solve this very well.

Quote

It's all a trade off...

Of course. Like I said, I just wanted to know whether this will be worked on.

Edited by Havner
Link to comment
Share on other sites

Hi. Guy who wrote Real Weather Connector here.

Yes, it's a known issue that the gridded nature of our weather data can become apparent from some viewpoints and weather conditions. But solving it is a hard problem; it's not just a matter of coming up with an algorithm to smooth out the corners, it's a matter of translating that into individual clouds in individual cloud layers, and doing it so fast that the "stutters" we've been so harshly criticized for in the past don't come back.

It is something I plan to work on for future revisions, but for version 1.0, it was a conscious trade-off in order to get a fast-performing and stable initial release out.

 

  • Upvote 2
Link to comment
Share on other sites

3 minutes ago, Ben Russell said:

That was a very poorly selected example bitmap that leaves a massive amount of unused room.

Play fair.

C'mon. HQ3X is used severely in up scaling low resolution games nowadays. Firstly, my screenshot from original post was even a simpler situation. 2 colors (2 cloud types, overcast and lack of anything).
Just wanted to show what it does. And here we are also limited to a low number of colors (6+lack of clouds). It would handle well.

Link to comment
Share on other sites

Just now, Havner said:

C'mon. HQ3X is used severely in up scaling low resolution games nowadays. Firstly, my screenshot from original post was even a simpler situation. 2 colors (2 cloud types, overcast and lack of anything).
Just wanted to show what it does. And here we are also limited to a low number of colors (6+lack of clouds). It would handle well.

You're over simplifying the problem and cherry picking solutions that are irrelevant to the actual problem at hand. This is much closer to 3D texturing than anything remotely approaching retro gaming.

Link to comment
Share on other sites

8 minutes ago, Ben Russell said:

You're over simplifying the problem and cherry picking solutions that are irrelevant to the actual problem at hand. This is much closer to 3D texturing than anything remotely approaching retro gaming.

I'm not cherry picking any solutions, just showing that they exist. I wouldn't propose this to anyone as they will do as they please. And seeing how SMP and RWC behave, the "map" of types of clouds do exist and looking at it's not a high resolution. I'm not talking about actually rendering the clouds but telling SMP where to render them. It is similar to upscaling retro games in my opinion.

And like we both said, whether SMP can actually handle that it's completely another matter.

And it was you that brought retro gaming in the first place :-)

Edited by Havner
Link to comment
Share on other sites

8 minutes ago, sundog said:

Hi. Guy who wrote Real Weather Connector here.

Yes, it's a known issue that the gridded nature of our weather data can become apparent from some viewpoints and weather conditions. But solving it is a hard problem; it's not just a matter of coming up with an algorithm to smooth out the corners, it's a matter of translating that into individual clouds in individual cloud layers, and doing it so fast that the "stutters" we've been so harshly criticized for in the past don't come back.

It is something I plan to work on for future revisions, but for version 1.0, it was a conscious trade-off in order to get a fast-performing and stable initial release out.

 

Sorry, I missed your post while replaying here.

Thank you for your answer. I think that sums up what I wanted to know for now.

Hopefully this can be fixed at some point. Will be looking for future releases.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...