Jump to content

CumulusX!


Peter Lürkens
 Share

Recommended Posts

Hi folks,

CumulusX! 0.93 comes now with Multiplayer compatbible AutoThermals. It was major effort, so I hope there are not too many bugs in it. Basically, AutoThermals should now coincide anywhere on the globe, among different instances of CumulusX!. This is not the complete package, rather than a patch. Read below:

// 0.93 02.02.2008 Bugs in AutoThermal-Timing fixed, lift mode selection is saved in settings file and registry,

Multiplayer copmpatibility of AutoThermals

What's new in Version 0.93

AutoThermals are now multiplayer compatible, i.e. each instance of CumulusX! produces the same set of AutoThermals, provided that the same program settings are used.

As a consequence, AutoThermals have no CCS compatibility mode anymore. In addition the new timing scheme is required for the synchronized thermal timing an positioning, which allows multiplayer session without prior exchange of a scriptfile.

In addition, the state of the checkboxes for the various lift types is now made sticky between sessions and is also saved to and read from the settings file.

best regards,

Peter

Edit: Attachment removed because of final release

Link to comment
Share on other sites

  • Replies 178
  • Created
  • Last Reply

Top Posters In This Topic

THANKS Peter! :yahoo: I hope it's a wonderful improvement. I believe I'm going to go soar on FSX Multiplayer now. Although, I was still kinda hoping a contest session between FS2004 and FSX could still be done, but you've come a long way with CumulusX!. MANY, MANY, Thanks!!!

Scott (sf4JC)

Link to comment
Share on other sites

Peter,

On the first look CSX! 0.93 works fine under Vista and FSX SP2.

One observation though: When I tried to load one of the setting-files you included in previous version, CSX! gave me a warning that the settings are not correct for this version.

Is it your intention to make new setting files?

Bert

Link to comment
Share on other sites

Hi Bert,

indeed, in the full package the updated settings file will be included.

Nevertheless, you can safely ignore the warnings and load your current settings files and save them again. The will be amended with the additional settings and a new header which will stop CumulusX! from complaining about the version.

regards,

Peter

Link to comment
Share on other sites

0.93 working well for me with Acceleration. Remembers settings fine.

In mid summer, so there are a lot of thermals @ 45 degrees lat (Blairstown, NJ, USA), the effect is the CumulusX! clouds are quite close together so you get maybe 6/8 cloud cover. Is that what others see? I think the density of thermals would be better maxing out a bit less, and increase the size/strength instead of the count above that level. I might have got it wrong somehow though.

Thanks,

Ian

Link to comment
Share on other sites

Peter,

Just a thought I have for days now. Could it be possible to make the thermal coverage (1/100km^2) setting in CSX! to be dependent on the cloud type and cloud coverage settings in FSX->settings? That way in multiplayer everybody has the same amount of themals as the host's setting.

Bert

Link to comment
Share on other sites

Is that what others see? I think the density of thermals would be better maxing out a bit less, and increase the size/strength instead of the count above that level.

You can reduce the coverage setting to lower value. I have a suspicion, that there might be a bug around, because sometimes I found it increased to the maximum value (15/100km^2) without remembering to have it changed. My usual setting is 5.

Could it be possible to make the thermal coverage (1/100km^2) setting in CSX! to be dependent on the cloud type and cloud coverage settings in FSX->settings? That way in multiplayer everybody has the same amount of themals as the host's setting.

Again, unfortunately, this is not so straight forward as it seems to be. Btw, there is currently no communication between different instances of Cx!, so it is essential, that everybody has the same settings. It would be really nice, if ther would be a chance to abandon the settings at all, but I don't see a chance how to do this, in view of the widely varying conditions of ceiling, strength, coverage, and size of thermals. That brings me back to the previously mentioned approach of "Template Settings" with a regional structure, similar to the slope data base.

Actually, I'm working already on picking up more weather information from FSX. At this time I have a version under test, switching off thermal activity entirely, when there are non-thermal WX situations, e.g. with stratus as lowest cloud layer. Also I turn off the CumulusX! clouds at overcast cumulus or sky clear conditions. In the first case, you will have an FSX cumulus overhead anyway, in the second you'll get blue thermals.

The problem is, to rebuild the situation in a repeatable way, when a change occurs, which often happens quite discontinuously. It can be managed, if thermals are turned on or off entirely, however if there are gradual changes, it becomes difficult.

An example: The only easy-to-get weather information is at the users aircraft, however, to fill the sky with a reasonable picture, you have to create the clouds also ahead of you. As you fly along, it happens, that you enter an area where the cumulus altitude is different. In previous versions I kept the characteristics of a thermal simply constant, resulting in a situation that already existing clouds stay as they where created until they expire, while new clouds were created with the appropriate weather conditions. If the overlapping area is small enough, the inconsistency is not too bad, because also in in reality there can be such situations for a limited time. Another advantage is, that you are not cheated while approaching a nicely high cloud, by dropping it down by surprise to a useless altitude.

In multiplayer situations that is no longer sufficient, because if two planes are meeting from different sides, the first might have created the clouds in the meeting area with conditions from his original location, which could have been different from the conditions the second had. The current workaround is now, that the thermal parameters are adjusted everytime the FSX cloud altitude has changed, which is sort of compromise, because all of a sudden, the Cx! cloud layer jumps to new altitude. Even probing would not solve the problem because it means that one has to make sure that the information from the probes are taken at the same moment in time for all participants.

Nevertheless, I have the further adaption of thermal situation on my roadmap anyway, and I will see what I can do.

regards,

Peter

Link to comment
Share on other sites

Hi Peter,

could it be possible that the "pumping" of parking airplanes is back with version 0.93? Ian and I encountered a problem (crashes) yesterday when we tried to leave the runway after landing. We were really slow when we rolled onto the grass and after long thoughts about that I remembered that I had some crashes on the ground while there was that pumping.

Never mind... it's running fine otherwise.

Dirk.

Link to comment
Share on other sites

could it be possible that the "pumping" of parking airplanes is back with version 0.93? Ian and I encountered a problem (crashes) yesterday when we tried to leave the runway after landing. We were really slow when we rolled onto the grass and after long thoughts about that I remembered that I had some crashes on the ground while there was that pumping.

Not really, there aren't any changes w.r.t. that. But it reminds me, that usually I have turned off crash detection because of some problems with certain add-on scenery. Apparently that needs definitely more attention.

Could you describe the situation more in detail? I may be, that an add-on aircraft has the notorious "spring" behaviour. Nevertheless, at the end of the day, the software should deal with it properly.

thanks for the info,

Peter

Link to comment
Share on other sites

Not really, there aren't any changes w.r.t. that. But it reminds me, that usually I have turned off crash detection because of some problems with certain add-on scenery. Apparently that needs definitely more attention.

Could you describe the situation more in detail? I may be, that an add-on aircraft has the notorious "spring" behaviour. Nevertheless, at the end of the day, the software should deal with it properly.

thanks for the info,

Peter

All right Peter,

I'll try to describe it a bit more detailed.

We were both flying the FSX-DG808s, in my case with your compensated-vario-panel (don't really know anything about Ian's panel).

Ian landed at an airfield near Sion (I'll look that name up in the evening as I'm in the office now) just before me (I was on a 2nm final when he touched down) and the moment he left the runway it looked (for me) like his DG sank into the ground for a couple of inches / or centimeters, doesn't matter here.

As usually after crash was detected he disappeared (my point-of-view) and I continued my own landing. After almost rolled out (speed was 2-3kias MAX) i turned around a bit to leave runway (used side rudder and kept wings levelled) and got a crash myself. As usually I was "thrown out of the plane" imediately and had the short look at my crashed DG (before FSX kicked me out to rejoin the session) and had the impression i did land with retracted gear - after an instace the plane hopped up a few times and I was able to see me gear, so that I'm quite sure I really had lowered it while on final...

If you like, I can offer a longer test with multiple landings with multiple plane types on multiple airfields in the evening for a better description of that issue.

I'm a lucky guy today... my boss shortened my working day hours from 16 to 8 about an hour ago: Let's hit it - later on...

Always blue side up,

Dirk.

Link to comment
Share on other sites

I realized now I failed to mention that in an earlier version, that once in a while after landing and coming to a stop (even if on runway if I remember correctly), I would get a jumping movement after a short period. I am also using no crash detection. I have yet to get these situations in the current build, but will surely make a video of it if it happens again for better description.

sf4JC

Link to comment
Share on other sites

I rolled off the runway onto grass at the same level, moving at maybe 2 knots (I only just managed to clear the runway) and sank maybe half a metre and recorded a crash.

Ian

That's exactly what it looked like from here.

BTW: No strange heights towards ground level today... till now.

Dirk.

Link to comment
Share on other sites

It looks like the jumping happens when the plane is on the ground under a thermal. Yesterday and today I started and landed several times at La Palma (Spain), where the jumping happened on the runway. ot so much that it lead toa crash,so Icould exminethe situationsome more. In top down viewmode I could see there was a thermal right above me. I then slewed a few hundred meters away and got out of slew mode. The jumping was gone. Back to the startup spot and the jumping was back.

Another observation: while jumping a little on the runway I turned off CLX! and the jumping was gone. Turning CLX! back on and the jumping was back (and the thermal bove me).

Bert

Link to comment
Share on other sites

It looks like the jumping happens when the plane is on the ground under a thermal.

I'm not sure this is related to my aircraft recording a crash as I left the runway, although it does seem a nice improvement to CumulusX! would be to force to zero the lift/sink effect below some threshold AGL. I'm guessing the 'jumping' is related to the necessary granular adjustments CumulusX! makes to your aircraft height to apply the incremental affect of lift/sink. At any normal flying altitude these aren't noticeable (although on aerotow I think you can detect the vertical movements relative to the tow-plane and maybe on the ground even a small adjustment is very noticeable. Just a theory and we await the diagnosis from the great man.

I need to roll off that runway a couple more times with/without CumulusX! to confirm whether there's any link at all to CumulusX!.

Ian

Link to comment
Share on other sites

What I found out so far, is that I do not modify SimProbe lift at all, and that it leads to pumping, when your landing spot is in lift and to the sinking when your spot is in sink condition.

I will rework the code for the y-management, so that all sort of lift are always reduced to zero close to ground.

However, it does not explain the crash. Could you specfiy the exact place where it has happened and the weather conditions?

regards,

Peter

Link to comment
Share on other sites

However, it does not explain the crash.

Don't read too much into the crash as yet - it could be completely unrelated to CumulusX!. We were flying multi-player in winter (I was hosting) so there was little or no CumulusX! thermal activity, although there was plenty of ridge lift. I just rolled off the runway, sank half a meter or so, and FSX recorded a crash. I need to try it again with/without CumulusX! running.

cheers

Ian

Link to comment
Share on other sites

I think Ian is right. Tonight I tried to reproduce the situation at La Palma I examined yesterday, but I could not create any jumps anymore. So I withdraw my conclusions of yesterday for now. Some more testing has to be done to find the cause.

Bert

Link to comment
Share on other sites

Anyway, at least the issue with the "ground lift" may be still a point. Formerly it prevented me even from take-off behind the Maule because the place was in a lee side situation and I simply couldn't get enough angle of attack to take off from the runway.

In the meantime I fixed also some other things, in particular a stupid bug with multiplayer thermals, which is reason enough to spawn another version. So there can't be more than 5 more beta's before 1.0 is achieved. :-)

// 0.94 09.02.2008 Major bug in settings management, bug in AutoThermal timing, enhanced matching of thermal

situation to sky conditions, global lift decrease at very low AGL altitude, better performance

What's new in Version 0.94

A new function in this version is an enhanced matching of the thermal situatiuon to the sky conditions. Thermals activity occurs now only, if there is a blue sky, or the lowest cloud layer is cumulus and there is no overcast stratus layer, or there is not only a stratus layer. If stratus is the lowest or the only layer, or a stratus is under a cumulus layer, then there will be no thermal activity.

In case of a clear sky, blue thermals are created, in case of overcast cumulus, thermal activity exists still, however extra cumulus clouds are not created anymore. There is an override of these "Blue" conditions by a new "UnBlue" checkmark under "Info", which makes the thermals visible again by the usual clouds. This can be also considered as a cheat key.

Two major bugs were removed. The first destroyed the settings from the previous session in some cases, the second affected the starting time and ending time of the AutoThermals, which were created relative to current time rather then their intended begin of life, even when filling up a thermal tile starting from the past. This led to an unnatural high density of AutoThermals at the beginning of a session. After the last of this initial set of thermals had expired, the desired situation established, but usually this resulted into an inconsistent thermal situation among the participants of a multiplayer session during the initial phase of a session.

When using SimProbe as source of ridge lift, a pumping effect of the plane on ground reappeared, because the reduction of lift when very close to ground was handled individually in each source of lift. As SimProbe creates the plain lift figure, without corrections for FSX/CumulusX! constraints, this contribution was not always zero for a plane on ground. Therefor a new global lift reduction for AGL altitudes less then 3 m was introduced that affects all lift sources equally. This height should be also sufficient for larger planes.

Performance of one part of the AutoThermal routines was substantially improved.

best regards,

Peter

Edit: Attachment removed because of final release

Link to comment
Share on other sites

But with me!

I was tracking a bug in 0.94, and maybe before, (happens if you switch forward from winter into daylight saving time period, then you get masses of thermals), while parking next to the RWY of LFNA, basic FSX scenery, while in bird's eye view. After several minutes I got a crash message, without changing anything.

I couldn't reproduce ist either.

BTW, do you have FSUIPC installed, and if, which version and is it fully licensed?

regards,

Peter

Edit: Overall it may be an FSX problem. I found this thread:

http://flywestwind.com/communityserver/for...ShowThread.aspx

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share


×
×
  • Create New...