Jump to content

CumulusX!


Peter Lürkens

Recommended Posts

  • Replies 178
  • Created
  • Last Reply

Top Posters In This Topic

Hey guys, I've been thinking about making a thread with basic instructions on how to convert the slope files for use with CumulusX!. I just don't want people to think it's a whole big process to make them .slo files.

sf4JC

Scott,

The method you describe works with me too. I added an earlier post about vista-users who want to convert .dat to .slo to your sticky-wannabee.

Can you also write a sticky-instruction on how to convenrt the 2002 tour the suisse files to fsx?

Bert

Link to comment
Share on other sites

I'll see what I can do Bert. But for now, just place the two files with the same name (usually: one is a .WX file and the other is a Microsoft Flight Simulator Application file) inside the ...\My Documents\Flight Simulator X Files folder. Then place the gps-nav.dat (may have the same name as the other two files, and if so, rename it "gps-nav") inside the main FSX folder, assuming you have all the files you need to run it already installed to FSX.

Then run FSX and load it from the FreeFlight Load option. Depending on which FS you downloaded from, you may have to change the weather. In this case, I chose real-world weather to get a real weather file, since I'm not a weather man, I don't know how to create a realistic one. I would say if you download an event/contest that was used in CCS2004, the weather should be ideal but I cannot say for sure.

sf4JC

Link to comment
Share on other sites

I forgot to mention that after you hit 'Load', you will get an error message about a certain aircraft (whatever aircraft they were flying at the time) could not load. Then after you enter the flight, usually in pause mode, resave the flight with the aircraft you want to fly.

sf4JC

Link to comment
Share on other sites

I'll see what I can do Bert. But for now, just place the two files with the same name (usually: one is a .WX file and the other is a Microsoft Flight Simulator Application file) inside the ...\My Documents\Flight Simulator X Files folder. Then place the gps-nav.dat (may have the same name as the other two files, and if so, rename it "gps-nav") inside the main FSX folder, assuming you have all the files you need to run it already installed to FSX.

Then run FSX and load it from the FreeFlight Load option. Depending on which FS you downloaded from, you may have to change the weather. In this case, I chose real-world weather to get a real weather file, since I'm not a weather man, I don't know how to create a realistic one. I would say if you download an event/contest that was used in CCS2004, the weather should be ideal but I cannot say for sure.

sf4JC

Thanks for the explaination, but am I right that I also have to make a flightplan to get my GPS and NAV-equipment to work? And do I use the tdsmap.jpg for that?

Bert

Link to comment
Share on other sites

  • 2 weeks later...

Hi folks,

a new small step for CumulusX, not much about functionality, rather than customization. Odd behavior during pause and slew should be mostly fixed now.

And it has a MenuStrip now.

best regards,

Peter

Edit: Attachment removed because of final release

Link to comment
Share on other sites

Well, the good news is it works good as always. The bad news, is for myself, I have to start reading the readme again. I haven't read it since version 04, so needless to say, I couldn't get CumulusX to work for me at first this time around. So if any of you have not read the readme for a while, believe me, you should. And I'm sure Peter would agree that everyone should read the readme, no mater what version it's in and how many you've read before.

So, I want to say sorry to Peter for not doing what I should've been doing every release. I really want this program to work for the health of the FSX soaring community. So for me, LESSON LEARNED, Don't neglect the readme!

sf4JC

Link to comment
Share on other sites

  • 3 weeks later...

Cumulus 0.7 and public beta approaching.

What you see below is a sneak preview on the upcoming release 0.7 (no releas yet because of some more testing). It shows the result of importing a CCS-Script file. Thermals are basically showing and producing lift, using the cloud models from airobjectvisuals.xml. What is currently open:

- correct CCS profile

- leaning

- widespread sink, inversion and weak surface layers

- drift

Basically, FSX default lift is becoming obsolete. Hope, I get most of it before Xmas.

On the roadmap (no schedule)

0.8 Lift in coincidence with FSX builtin weather system and its standard cumulus clouds

0.9 Public beta

Cheers,

Peter

post-176-1196723815.jpg

Link to comment
Share on other sites

  • 2 weeks later...

Here is CumulusX! 0.7 with script-file import and AutoThermal mode.

From the readme:

With Version 0.7 CumulusX! can now import and execute Multiplayer-Script-Files from CCS2004. There should be sufficient compatibility to allow for mixed sessions. The software switches automatically to AutoThermal mode if the script-file is not found.

In AutoThermal-mode similar thermals are created on the fly while flying around. These thermals align with an existing cumulus cloud layer, if that is below 20000 ft or 6096 m. Otherwise, the ceiling settings are taken from the configuration dialog. The same is for the script file.

There are some subtle differences in the parameters compared to CCS2004. Mostly it is that these are metric. Nevertheless, the scriptfile parameters are converted during import.

A more relevant difference is in the meaning of the "Lean"-parameter. As in CCS it indicates the amount of leaning in relation to the vector of an airpacket climbing with the thermal lift and displacing with the windspeed, whci has the factor of 1. Different from CCS, the displacement is now multiplied with the lean scalar so that a lean scalar of zero leads to absolutely vertical thermal. This was not possible with CCS2004 where always a certain leaning was obtained. I felt that this could make it a bit easier for beginners, because staying under the cloud means staying in the lift.

Im almost sure that there are still tons of bugs in the code (which by the way has become a pretty mess meanwhile). If you find one please give me message. I hope to fix with next minor steppings.

Cheers,

Peter

Edit: Attachment removed because of final release

Link to comment
Share on other sites

Peter,

Great job you have done !! I ran it for a little while near LOWZ with slope.files active. And also in an area with no slope file active. Both times in unscripted mode.

Some observations/questions

- I had no problems running CumulusX!0.7 under FSX SP2 and Vista.

- The first two parameters in de debug-window act very nervously. It looks like the data-readout of those two parameters is set to much more times per second than the other parameters.

- Am I right that hhen a slope file is active, Auto thermal mode is also active at the same time? And thermal lift adds up to slope lift at the same time?

- Am I right that I have to set a cumulus cloud layer in the advanced weather options of FSX to get the AutoThermal mode to produce thermal lift?

- Am I right that there are more thermal clouds visual then there are thermals active at that time?

Bert

Link to comment
Share on other sites

Hi Bert,

nice to hear it running under Vista/SP2. Meanwhile I have also SP2.

- The first two parameters in de debug-window act very nervously. It looks like the data-readout of those two parameters is set to much more times per second than the other parameters.

They are nervous, yes, but the update rate is the same. The difference to before ist, that the first two lines are displayed first, than the rest is added along with the first two lines again. Apparently, my debug window management is not best.

- Am I right that hhen a slope file is active, Auto thermal mode is also active at the same time? And thermal lift adds up to slope lift at the same time?

Yes, it is. Lift types are always summing up. Unless you set the script-file to something meaningless and reduce the lift parameters for autothermal to 0 you will always have thermals.

- Am I right that I have to set a cumulus cloud layer in the advanced weather options of FSX to get the AutoThermal mode to produce thermal lift?

No, you don't have. Maybe, it's a bug. If you start with a clear sky and run CumulusX!, clouds and lift should appear around you with the defined parameters. Indeed, I know already of a first bug with cloud altitude in absence of a cumulus layer. However, to my finding this lead only to a disregarding of the preset cloud ceiling and results in clouds at 20000 ft. It happens, if you had initially a cloud layer and by some reason it is turned off later on.

- Am I right that there are more thermal clouds visual then there are thermals active at that time?

It should not for the time being. Maybe it's a bug, too. Perhaps you could check by slewing under a cloud and using at debug window. The second line should show the total thermal lift, i.e. inlcuding widespread sink.

There is currently not yet a native mode for Autothermals or Scriptthermals, so it should behave mostly as CCS2004. In this, lift is never produced outside a window from 10:00 to 18:00 LT, even regardless of the setting in the scriptfile. In addtion, a thermal has a 2-minute build up and decay period at beginning and end. The lift starts to decrease shortly beneath the cloud base and clouse to ground. Eventually there may be also an inversion layer with a reduction lift.

There is currently no explicit management of adjustment of the time of day by the user. This can result in CumulusX! loosing a bit track of the thermals lifecycle. I'm not sure about all consequences at this time. But you have the chance to reset CumulusX! either by shortly disconnecting or by going to the settings page and just clicking "OK".

regards,

Peter

Link to comment
Share on other sites

Peter,

I played around with the settings in 0.7 for several hours now to test the auto thermal mode and I must say, 0.7 works like a charm for me. I see no bugs or strange behaviour so far and it is nice to see that I can now actually soar long tracks over my home-country (Holland) in thermal-conditions, like I can expect in real-life here. This will keep me sharp until real-life soaringseason begins again in april.

Greets,

Bert

BTW: I wrote Vista in the other post, I meant Vista 32-bit.

Link to comment
Share on other sites

Hello everybody,

this is very likely my last post before Christmas: CumulusX! 0.8, and still no manual.

What's new in Version 0.8

Note: This version is compiled with the SP2-Simconnect-Dll. It may happen that it does not work if you run pre-SP2 version of FSX. In this case, please consider update to SP2 (ok, another 170 MB, I know).

After being nearly lost in a mess of code for settings, I decided to rework the ever-growing settings management entirely. As a side effect it is now possible to save and load CumulusX! settings to and from a file, for later use, or for exchange with friends. Still it merely an option and not mandatory. Though, multiplayer will work statisfactorily only in Scripted mode (still) and with properly synchronised weather condition and clock among the participants.

Changes in the control panel will no longer become effective unless you press OK, and are completely discarded in case of "Abort".

In addition, the over-day timing model of the thermal lift was modified. Now, thermal activity starts still at 10.00 LT as before, but the number of thermals increases over the first hour from zero to the average defined in the settings. Lift grows from zero to the rated values over three hours, but in a square-root function, i.e. in the beginning it increases sharply, later on more gradually. In the evening, lift does not decay anymore, but the number of thermals decrease again over two hours down to zero, starting from 18.00 LT. That allows late flights in dying evening conditions with rare, but still good thermals. These modifications are not effective in CCS Compatibility Mode.

Basic messages as "Connect to FSX", "Disconnect", etc. were removed from the debug window, which now only shows information about the lift situation.

Merry Christmas!

Cheers,

Peter

Edit: Attachment removed because of final release

Link to comment
Share on other sites

  • 3 weeks later...

Sneak Preview:

CumulusX! 0.9 with AutoStart and FSX menu integration. Lot's of bug fixing, minor performance improvement, code cleanup and better overall consistency. No relevant changes in primary function, though. Slope Probing is only in the user interface prepared and waiting for Ian's module. There is no code behind it yet.

I will need some testing still.

Cheers,

Peter

post-176-1199752661.jpg

Link to comment
Share on other sites

well done Peter that looks absolutely outstanding.

Over the past few days I've just been tweaking and testing the main formulae in the calculations and evaluating real slopes by moving over the landscape in the sim. Otherwise the mechanics of the code have been stable for a while. I'll give you the source and executable for both the sim_probe and the sim_probe_monitor that is currently reading out the client data.

I am *sure* we'll want to adjust the weights with the sim_probe connected to CumulusX!, mainly to get the *average* amount of lift for a given slope at an agreed figure.

My tweaks of the formulae are giving diminishing returns and the general shape of the lift produced is converging near the image below, which gives the lift generated by sim_probe (in blue) compared with my calculated 'sine of local slope angle' value (in red) as an approximation to what CumulusX! would currently build from the slope file. The ground terrain is the green line.

post-5237-1213361979_thumb.jpg

With multiple samples we *should* be able to get an improved simulation of lift compared to the immediate local slope, and the image shows the sim_probe approach is less susceptible to the spikes you see around 2500 meters and 5500 meters on the chart where there are local dips.

By the end of the week you'll have a version of sim_probe to test with.

Regards - Ian

post-16-1199796772.jpg

Link to comment
Share on other sites

Still tweaking...

Just taking a look at how the ridge lift should change with altitude. The general approach of Peter's formula works great for the *upwind* side of the ridge (the lift is strongest if you tuck in close to the hillside) but the situation is *not* symmetrical on the downwind side. I already have the idea of how to tweak the formula (the presence of high ground upwind, i.e. from one of my probes, will mean the sink extends higher.)

Here's a picture of the kind of analysis I'm doing (nearly there, just the sink/altitude tweak to do) - created from real samples taken within FSX - it took a while to work out how to get Excel to draw this, trust me. The chart is about 9km across and the highest ground is about 400 meters, i.e. very compressed horizontally so altitudes are exaggerated.

2008_6_13_9_28_49_796.BMP

(note EDIT - I've just swapped this image for better one)

B21

post-16-1199966056.jpg

Link to comment
Share on other sites

You guys are making great progress here!! :respect: :respect: :respect: Do you think it will be possible to get a lift program to read the FSX mesh directly or will we always have to import .slo files from CCS2004?

Don

Link to comment
Share on other sites

Do you think it will be possible to get a lift program to read the FSX mesh directly

Hi Don - the entire thrust of my work with FSX is to calculate the ridge lift directly from the mesh, and the graphs and charts I've printed here are derived from flights over the FSX terrain with probes on the surface monitoring the landscape at set distances upwind and downwind from the aircraft and an algorithm computing the appropriate lift factor.

This compliments the work Peter is doing with CumulusX! to place thermals with dynamic behaviour and mark them with cumulus clouds. The intention is to make my 'sim_probe' ridge lift available inside CumulusX as an alternative to the slope files.

I've only been working on the sim_probe technique a couple of weeks so the main effort has been to confirm that it's actually doable. We've got to that point now so it's more a case of integration work with CumulusX! but Peter and I have agreed a way to do it.

There are 2 main possible advantages with the sim_probe technique: (1) with five probes testing the ground altitude at a variety of distances the lift can be more accurate and detailed, (2) you can fly anywhere in the world without a slope file.

Ian

Link to comment
Share on other sites

Hello dear soaring pilots !-)

without problems flying with CCS2004 in FS9 with slo data for my homebase LKKR and nearest mountain Jeseníky.. Perfect !-) Wave as strong ridge works so little too .-)) but not so much realisticly.. .-)

Now i would like to try CumulusX! My problem is that i have installed only XP 64b. and Vista 64b. .-(

Clear installation of FSX + SP1 + SP2 + FSUIPC... Installed CumulusX! as wrote in readme..

After start FS X and load saved fly in window mode startup CumulusX! and after few seconds give this reports:

xp64.jpg

vista.jpg

Is here some problems on my side? or CumulusX! is still not compatibile with 64bit..?

Thx for answers, glider pilot Sněhulák (SnowMan) from Czech FS community ;-)

As my "infoadd" to your great work here is real proportion of sing and lift in WAVE after KRKONOSE MAPS HERE mountains as computed by metheorological model of Czech Best Metheorologer Jan Horák :-) with you can find here: http://www.gliding.cz/Meteo/horac/vlna click in place you want to show diagram....

VLNOREZ = in eng. cross-section of Wave flow

1575cross2.gif

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Privacy Policy & Terms of Use