We got a new video from World of Aircraft: Glider Simulator and are asking for some alpha/beta testers! Check this post.

 

Jump to content
Sign in to follow this  
Peter Lürkens

CCS slope-file content

Recommended Posts

Don't know how to deal with the fact that the existing slope-files vary significantly in their characteristics, depending on their origin.

What I mean is not the dependency on the installed mesh when scanning with FS9 or FSX (yes it works, if you set a sufficient delay, even minimized), but the differencies between "scanned" slope-data and offline calculated slope data, coming from SRTM-data.

The difference is, plainly spoken, that CCS-scans produce much better lift, than the slope files from SRTM-data, which are almost all coming from Kris Feldmann and Bill Bomball.

I believe, the reason of this is, that they understood (and me, too) the slope files carrying the steepness of the slope in a measure proportional to the angle vs. the horizontal plane. As depicted in my posts in the CumulusX-thread, with two distinct exception CCS2004 behaves so, indeed (and CumulusX too). The slope-data files scanned by CCS actually do not reflect the slope-angle, rather than a more sophisticated function, which strongly emphasize flat slopes. You can check it yourself by using the debug mode of CumulusX. You will find that the indicated angle is much steeper than the slope you are on.

Different to this the slope files converted directly from SRTM data usually contain the correct measure of angle. These alson contain larger areas, usally.

The problems are now:

1) Combining "Scanned" slope files with calculated ones result in sudden change of the overall strength of the slope lift at the boundaries

2) The "Scanned" slope-files generate unrealistic good lift conditions

3) Further scanning with CCS proliferates this behaviour furthermore.

What could be done is:

a) Makeing a new scanning or conversion module that is consistent with the larger amount of existing data

B) or a conversion program that converts the scanned files into the calculated type, with a bit reduced differentiation of the various strength, or vice versa

c) ?

best regards,

Peter

Share this post


Link to post
Share on other sites

I just got done yesterday scanning the Dillingham, HI area with 10m terrain mesh in FSX. From what you are saying, this may have been for nothing but fun flying I guess. This is going to set the program back a ways if we want it to depict as close to reality as possible, but I'm all for it if it can be done. I'd rather that, then to have something unrealistic. How unrealistic are we talking about though? Was it unrealized by Eric or was it not worth the effort to correct the problem?

Another thing to think about is, if there is a different method to come up with making the slope files, it should probably only be in a format that CumulusX! can read possibly. Because in a contest situation, a person who is less honest may feel no harm in using files created directly from CCS2004 to get an advantage then.

sf4JC

Share this post


Link to post
Share on other sites

I believe my vote would be to convert the scanned lift files to the more accurate calculated from SRTM files. I think we would be better off in the long run. I was affraid we would run into this problem. Also, I know Kris had to do a lot of cleaning up of the SRTM data. There were a lot of holes and spikes to correct before he could process the data. Actually, we may just have to redo the "scanned" files if we want them to be accurate.

Don

Share this post


Link to post
Share on other sites
I just got done yesterday scanning the Dillingham, HI area with 10m terrain mesh in FSX. ....

... Was it unrealized by Eric or was it not worth the effort to correct the problem?

.... Because in a contest situation, a person who is less honest may feel no harm in using files created directly from CCS2004 to get an advantage then.

As both CumulusX and CCS are working with a resolution of 3", a 10m mesh does not bring additional advantage. Actually, the result tends to be some more rough than with a 90m mesh.

I can only gues, if it was intentional of Eric or just a bug. Indeed, it makes slope-lift much more effective even at low windspeed. Maybe that Eric, who used to be hang-glider flying, found it more convenient. Indeed, NZ ridge "race" 2 is a _lot_ more fun with the original slope data from Andrew's preparatory package, yet not very realistic (sigh).

The difference to what I consider "real" is roughly factor of 2 flatter slopes (c.f. diagram below).

Contest and cheating opens up a full new universe. To be honest, it will be very difficult to exclude cheating reliably, because there are so many influential factors, scenery (think of shortcuts!), aircraft.cfg, .air-file, WX, advantage of pausing, tapping additional information sources (think of "debug" mode) and more. You may even consider another simconnect-client that helps along. That's all the problem with an open system. I must admit that the restrictive system of Condor has a fundamental advantage here. Still, one could consider a sort of watchdog, that takes care of authenticity of .slo-data and other major factors.

I believe my vote would be to convert the scanned lift files to the more accurate calculated from SRTM files. I think we would be better off in the long run. ...

... Actually, we may just have to redo the "scanned" files if we want them to be accurate.

That's what is my preference, too. Transformation of scanned files into SRTM-type is better possible than the other way round.

best regards,

Peter

post-176-1194773251.gif

Share this post


Link to post
Share on other sites

Hopefully no one is using this program to an advantage in the FSX Multiplayer environment. I joined a Minden race as an observer and could not keep up with the lift that the person I was following was getting, although I was in a different glider so that was probably why maybe.

sf4JC

Share this post


Link to post
Share on other sites
Contest and cheating opens up a full new universe. To be honest, it will be very difficult to exclude cheating reliably, because there are so many influential factors, scenery (think of shortcuts!), aircraft.cfg, .air-file, WX, advantage of pausing, tapping additional information sources (think of "debug" mode) and more. You may even consider another simconnect-client that helps along. That's all the problem with an open system. I must admit that the restrictive system of Condor has a fundamental advantage here. Still, one could consider a sort of watchdog, that takes care of authenticity of .slo-data and other major factors.

During all the years that I performed the duties of Contest Director at SOAR, cheating was a serious issue with almost every race we sponsored. I never publicly blew the whistle because I could never prove it. The problem was so serious with the Southern Alps Race that I just gave up on it. That is one reason I never ran Race 3. I think MSFS open system is wonderful for personal enjoyment and for creating one's own little world, and informal online flying with friends you know. It also works fine (in my opinion) for "certificate" flights for personal achievement. However, if you are going to host a competition where pilots are going to compete against each other, my vote would be for some kind of closed system that cannot be played with.

Don

Share this post


Link to post
Share on other sites

I guess I was really getting my hopes up. So my presumption may be correct I guess, that it is possible for someone to already be using CumulusX in the FSX Multiplayer environment for his/her own gain. I'm totally heartbroken! Peter has a program that quite possibly has the makings of becoming one of the most downloaded programs for FSX glider pilots ever.

sf4JC

Share this post


Link to post
Share on other sites

Scott,

I have been thinking for several years now about how it would be for SOAR to have "chapters" all over the world consisting of 3-4 members in small communities, and up to 30-50 members per chapter in large cities. Then one could hold "in house" races on individual computers networked together in the same room. Monitoring individuals and race progress would be quite doable and MSFS I believe would work fine in this envirnment. Also there would be socialization among SOAR members which at the moment is sorely lacking.

Cheers, :smiles:

Don

Share this post


Link to post
Share on other sites

I like that idea Don. I am a member of an online racing community for NR2003 which holds a few of those a year. Only, I think it's for everone that is able to join them. So it would be like having many Chapters at the event. I've never been to one and most likely wont ever be able to, but from the pics and the conversation that results, it sure boosts morale amongst members.

sf4JC

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
Sign in to follow this  

×
×
  • Create New...