Jump to content

Cumulus X lying on the ground


Hodge7

Recommended Posts

I have done some detailed testing on this cumulus x issue:

1. A careful look at the debugger indicates everything working perfectly, about 325 thermals and all other indicators acting as I would expect.

2. I have set the user defined weather for free flight to cumulus and cloud base to 6500 with cloud top at 7000 "AND" it makes no difference - that is this issue is bypassing any settings for missions or free flight.

3. Just reiterating - the problem exists in both missions and free flight.

4. Pattern is as follows: Starting out in a mission or free flight everything is normal clouds in the correct place etc then at about 2 minutes into flight irrespective of whether I am slewing or not the clouds which were at 6500ft are replaced by clouds at 3400ft. This originally gave the impression of 2 cloud bases but that was not correct. It was simply the upper level leaving the screen and the lower level being generated over a period of perhaps 30 - 45 seconds.

5. I tried this using 3 missions (Day 1, Day2 and Day 3) and the pattern was the same. and the pattern is also the same in free flight.

I am now going to test some of the Austrian missions and also will test to see if there is lift above the clouds once they have established themselves at 3400ft base.

Ive tested b21s Austrian missions and they react the same way.

Ive tested the default Austrian mission and could get no cumulux clouds at all (the default Austrian mission has schematic thermals built in).

I then tested the Swedish Championship course (also a default fsx mission) which has no built in schematic thermals until you switch them on using the weather thermal schematic setting. This course had 2 permament layers of Cumulusx one at 5500 and another at 3700 so my conclusion that the two layers simply being a transition from one to the other appears incorrect because in this mission they stayed there.

When I slew up above the lower cumulusx clouds the lift disappears.

I am trying to be a bit more scientific about this so that we can nail the problem.

thanks again for all your help

Jeff

Link to comment
Share on other sites

  • Replies 50
  • Created
  • Last Reply

Top Posters In This Topic

I think this is getting towards a fairly specific description of the problem, which seems completely repeatable. I would recommend sticking to the following factors to minimise complexity:

1 - a simple WX file with a single layer of Cu at 6500

2 - free flight daytime / summer / Mifflin

3 - stay in slew mode at 4000 feet but remain stationary and just watch what happens (I assume the problem will recur)

4 - keep the debug window open (I'm assuming nothing significant changes there after 2 mins)

and wait to hear from Peter...

Ian

Link to comment
Share on other sites

Hi Jeff,

normally, there should only be a single layer of CumulusX! clouds, except during transitions from one ceiling to another. Actually, when CumulusX! detects that cloud ceiling has changed, it eliminates existing clouds and whith a short delay recreates them at the new altitude, but at the same place. As this is not done one-by-one, it seems as if there are simply new clouds, rather than clouds hopping up or down (it took me a while to find out how to avoid this effect).

If FSX does not provide a cumulus layer, then the cloud ceiling of the CumulusX!-settings is used. This can explain, that two different ceilings occur, depending on the coverage situation. As I saw from the previous part of the discussion, there is not common sense, which settings file to use for a contest. So, in general, I would propose, to include a specific CMX-file in the race package, to make the choice of CumulusX! settings explicit.

Nevertheless, there are two more supplementary ways to create thermals in FSX, yet which are less realistic then CumulusX! thermals. They add to the total thermal situation and potentially produce extra cloud layers which look exactly the same as the CumulusX! thermals.

One is generated by FSX' built-in thermal engine. This creates thermals to little extent depending on season and weather conditions, but strength, size and height are always the same. There is an original FSX-file "thermaldescriptions.xml" in the FSX root directory, which controls the appearance of built-in FSX-thermals. These simply coexist with the CumulusX! thermals, and I know there are some replacement versions of the file available at AVSIM and others. Usually, I rename this file, which effectively disables FSX native thermal engine and leaves only CumulusX! thermals.

The next one is scenery thermals. I'm not sure, but I think these use also the same visual model as CumulusX! does, so they look the same. There is a number of freeware sceneries for numerous areas of the world from Vargas which produce static thermals in the individual regions. If there is such a scenery installed, there may be extra cloud layers.

Eventually, the last option, I know from, is mission thermals, which are pretty flexible (because following a predefined script), but not responsive to weather conditions at all. These would be defined within Ians mission files, but, afaik, he did not do so, so this should not be a reason for the occurrence of extra layers.

Is the "laying on ground" effect still there?

regards,

Peter

Link to comment
Share on other sites

Hodge,

Just a wild guess: in the FSX->weather->settings there is an option to set the rate at which weather changes over time. On Peter's advise of Peter IMy set that rate to "low". While testing earlier versions of CumulusX I experienced cumulus-clouds disappearing from time to time and noticed that my rate was set to "high". Peter then advised me to set the rate to "low" and since then I experience less problems (almost none as a matter of fact).

Bert

Link to comment
Share on other sites

Peter, I understand each point in your mail and I'm sure Jeff will check to see if he forgot he installed Vargas's scenery thermals, but the possibilities are really narrowing down at this stage. There are absolutely no mission thermals used and the problem repeats outside of a mission so it's not those. FSX default thermals are pretty sparse and I don't think they cycle do they? But Jeff will need to check his thermaldescriptions.xml if he's changed that. The Wx in the mission and the Wx tested outside has a single Cu layer at around 6500 feet, so there shouldn't be an issue of CumulusX operating without a Cu base layer.

My guess is something is causing CumulusX to believe the Cu base has changed to 3400 feet (or zero feet in other situations) and it's re-drawing the Cu's at that height. Is there any way for Jeff to work out what CumulusX thinks is the current cloudbase ?

Ian

Link to comment
Share on other sites

My guess is something is causing CumulusX to believe the Cu base has changed to 3400 feet (or zero feet in other situations) and it's re-drawing the Cu's at that height.

Probablyit's nothing else than the altitude setting on the CumulusX! settings pane which cause the cloud layer shift to that altitude, when there is no FSX cumulus layer. If that setting is very low, it can well produce the "laying on ground" appearance.

regards,

Peter

Link to comment
Share on other sites

Peter - I have reinstalled Cumulusx twice and have renamed the "thermalescriptions' file to "thermaldescriptions copy"

The only scenery I had for the area was FS Genesis but I have unticked that in the scenery database so that I am flying with default fsx only, I have no other tricky stuff like the Vargas you mention.

My overall installation is XP + SP3 +FSXdeluxe + Acceleration + Cumulusx + Simprobe + b21 te vario + netto vario + downloads of various Missions such as day 2,3,4 Austria by b21 and the new Mifflin series.

In my testing I have not had the laying on the ground effect lately just the lower level of cumulusx - I will do some more testing to try and reproduce that.

Ian - I will follow your thoughts about staying stationary in slew mode etc and watch events from there.

Bruin thanks I will check that out.

Peter - this "altitude setting pane", is this something I can access to check out?

regards

Jeff

Link to comment
Share on other sites

"altitude setting pane"

Peter actually wrote "the altitude setting on the CumulusX! settings pane" which I screenshot earlier (here).

I think he is referring to the "lift ceiling(m)" parameter. If so you could test this by setting the range to 1372-1372 and seeing if the cloud layer moves to 4500 feet. If so there will be a quick hack for all the Mifflin missions if you set the range to 1981-1981 (i.e. 6500 feet).

In free flight started with Wx = Cu @ 6500 you could also check the weather after the Cu's move via the FSX flight menu to see if it still says that.

Ian

Link to comment
Share on other sites

We may have beaten this issue!

For the record this is the multilevel cloud situation: I thought the images would have been larger I must have selected the wrong size however as you know just double click on the image and they will come up full size.

The higher cumulusx are at 9000ft and the lower at the same height as the glider, approx 5400 as is the small cumulus nearby.

This is a general shot showing those cloud layers.

multilevelcloudsez4.th.jpg

This shot is the clouds lying on the ground:

maxedoutconfigln5.th.jpg

This one is using b21 cumulusx config:

b21configxnz4.th.jpg

And finally this is using the default config:

defaultconfiglt8.th.jpg

So I think we have this problem by the you know whats: As a workaround it seems that all I have to do is set the Min lift ceiling to b21's suggestion of 1981m and then when it tries to produce more clouds at a different level - low and behold they are at the same level.

I still dont understand why knowone else has had this issue because Im sure other pilots have not had to set their config to 1981m anyway if it works thats fine with me! Yay!!!

regards

Jeff

Link to comment
Share on other sites

Hi Jeff,

the "laying on ground" is now clear. The cloud altitude range in the settings pane starts at 100 m. CumulusX! selects a random altitude from the range for all the clouds, rather than individually for each cloud, if there is no other cumulus layer from FSX. To get it the same in multiplayer, the random generator is initialised with the time and date of your current situation. By incidence, in your case this led to a cloud altitude close to the lower limit of the range, producing the weird "on ground" appearance.

The dual layer is not that clear to me. It seems the the clouds in your neighbourhood are substantially higher than those far away, which seem to be roughly at the same altitude as the FSX cumulus near to you. This should not happen, except, when there is a transition from an initial cloud altitude to a new one. E.g. this can happen when initially there were no FSX clouds and CumulusX! generates the clouds according to the settings pane, then FSX generates a cumulus layer, and CumulusX! re-adjusts the cloud altitude (or vice versa). If so, then after a few minutes this dual layer appearance should vanish.

best regards,

Peter

Link to comment
Share on other sites

I think we've found it, and I'm the culprit.

Jeff - I think the WX file in the Mifflin missions is the problem - others weren't impacted because they had a default minimum 4500 foot (i.e. 1500m) cloudbase, but yours was set to 100m for some reason.

Can you test this patch ASAP (all it does it put a new WX file in every Mifflin mission folder, plus the prior update for the day 3 mission xml file from a previous patch):

http://carrier.csi.cam.ac.uk/forsterlewis/...in_v9_patch.zip

I *think* the Cumulus settings in the mission WX files are not applying globally, although I don't know how that could have happened.

Thanks,

Ian

Link to comment
Share on other sites

Thanks guys for your regular input on this, Im starting to get exited that it is getting close to problem solved. At first I thought it would be a workaround to solve the issue but from your comments Ian its sounds like it may be able to be fixed.

Peter - the screenshot next to the cumulus showing the lower cumulus x in the distance is simply how it starts. First those clouds start appearing and over 30-40 seconds they completely replace the upper layer. A very surprising thing I found was that if I just sat still in slew mode nothing would appear to happen. However in a mission I would slew closer to the first turnpoint and at about 13nm to go suddenly the lower level would start to appear - now the next thing staggered me! because when I then backed up away from the turnpoint half a nm, the lower level clouds would disappear again, go closer again and there they were again building up at the lower level. I found if I let them build up completely and completely replace the upper level then I couldn't reverse the process. At one stage I had a colour A3 blowup of the mission area and was trying to map the areas where they exhibited this phenomenon however I could not discern a particular pattern.

I also noted with interest that the cloud count rose significantly from eg #335 up to eg. #380 when the lower level cumulus layer was being generated, eventually this settled back to the original count once the level change was completed.

Ian - this WX file - seems strange that I downloaded from your site like everyone else and how would I end up with a different setting? I will load that patch tonight and give it a go. Boy I sure am looking forward to a couple of good nights sleep.

Thanks again for your help Peter and Ian, I was getting pretty low for a while there because it was looking like I would not be able to fly any of the Mifflin missions which relied on thermalling.

Jeff

Link to comment
Share on other sites

Ian,

Regarding Cumulusx and the Mifflin missions, it strikes me that all pilots should be flying using the same cumulus x configuration. Your original posted times and mission development were done using a particular config which I call the b21 config, and its likely that some of the other pilots are soaring using the default config. Even though the differences may not be great, I think it is still significant, eg the lift is slightly weaker but there are 3 times as many thermals per unit area in the b21 config so the sky looks much more crowded with thermal clouds.

Id say use the b21 config and post it on the Mifflin site, what do you think?

Jeff

Link to comment
Share on other sites

Peter,

I have clarified exactly what height the cloud layers are in the first screenshot above by editing that posting, sorry I didnt define them better the first time.

regards

Jeff

Link to comment
Share on other sites

Jeff to clarify in response to your questions:

1) Everyone does indeed have the same Wx file in the Mifflin missions (and the same Wx file is used in all five missions) BUT that Wx file is not working as I intended - what was intended to be a global Cu cloudbase layer is *not* global - perhaps it is affecting the nearest weather station to Mifflin only. Depending on the direction you fly, you can quickly fly to an area without that Cu layer.

2) With the *default* settings of CumulusX, in the absence of a defined Cu layer (i.e. when you fly out of the local Mifflin area) thermals will be created to maybe 6000 feet (because default has the range starting at 1500m). These pilots don't really notice much of a difference except it probably goes blue.

3) You, on the other hand, had 'default' thermals with clouds created to maybe 800 feet, i.e. ground level, because of the 100m altitude base in the CumulusX settings. This exposed the error in my Wx file.

You're right about 'everyone should fly with the same CumulusX settings' but rather than distribute a CumulusX settings file I would rather simply tell people to 'Reset'. The downside of this is it introduces everyone to the CumulusX settings panel that I'd rather avoid unless we have a mechanism of checking the settings. For my purposes it would be helpful to confirm the settings are default.

I'm flying the missions now with the default CumulusX settings (and the new Wx file) and they're working as I intended so I don't need the additional complexity of a .cmx file. My times might be *slightly* different than with my earlier settings but so far the range is closer than the difference between me and other pilots.

Hopefully Jeff we'll get you sorted - a discussion I can hear coming of 'how do we make sure everyone has the same settings' we can keep for another thread.

Ian

Ian

Link to comment
Share on other sites

Finally, after three days of repeatedly "killing myself" trying to get back to a ridge after reaching tp1, I managed to finish Mifflin day3. With the new wx-file installed and the standard CX!-setting (and the standard FSX-cheat: schematic thermals. But hey, I had to see if the thermals were any better now, didn't I?). Much more thermal activity now around tp1. I managed to round that at about 5000ft and caught some more thermal-height on the way back to the ridge to set off for tp2. My time: 2h22m. I think it can be flown faster, but my goal tonight was to stay alive for a change and finish.

Great job Ian. It's doable now, but it still needs all the flying-skills available and a lot of coffee and cigarettes (yeah right, coffee :cheers2: ).

Bert

Link to comment
Share on other sites

My understanding was that schematic thermals in fsx and CumulusX did not coexist and that Cumulus X is only initialised when toggle is set to "natural". Is that correct?

Jeff

Link to comment
Share on other sites

Ian, - I downloaded Mission patch for wx file and tested, unfortunately my hopes were dashed as lower level clouds still appeared as before. Maybe Ill have to move to the Northern Hemisphere. Ill try flying a mission with ceiling height set to 1981m and see how that goes.

Jeff

Link to comment
Share on other sites

Cumulus X is only initialised when toggle is set to "natural"

No that's just to show the Cu model atop each thermal. As per my top-view screenshot the CumulusX thermals will display as spirals if schematic is selected.

Jeff please note this bit:

I'll be a lot more comfortable if you absolutely confirm you have the same issue in free-flight summer daytime Mifflin with global scattered Cu @ 6500 and wind 17 knots from 311. *Don't* copy the mission Wx file to get that weather or you won't eliminate that source. Last time you had no Cu layer and your settings had the default min at 100m so that didn't tell us anything.

For CumulusX! in general, I've discovered the default settings actually make it extremely difficult to thermal up from 2000-2500 feet with the weather set as above, and I think Jeff is right that the lift improves much higher up and is generally weak at that level.

Ian

Link to comment
Share on other sites

My time: 2h22m

Well done Bert - the thermalling part is a bit harder than I planned :whistling: so Day 3 is probably the hardest mission in the series. I flew the missions originally with different CumulusX settings (by accident, I picked up earlier defaults) and am now duty-bound to fly them again on the 'standard' settings. So far I've improved my times though, but for Day 3 I very much doubt I'll do that.

Ian

Link to comment
Share on other sites

So just to clarify the weather settings issue. Is sounds from your post Ian, that the user defined weather settings for freeflight are a relevant factor when flying the mifflin missions? I hadnt fully realised the relationship and have been assuming that free flight and missions were separate.

Jeff

Link to comment
Share on other sites

CumulusX and sim_probe just read the *weather* and decide which thermals and ridge lift to place in the landscape - neither cares whether a mission is loaded or not.

We need to replicate this problem in FREE FLIGHT, so we can stop referring to the missions which may be an irrelevance.

So I am not suggesting weather in free flight somehow affects a mission - it doesn't - I'm just asking you to:

* make sure CumulusX settings are 'reset'

* go into 'Free Flight'

* 'Location Mifflin'

* 'Weather 17 knots wind from 311 degrees, single layer of scattered Cu at 6500 feet'

* 'Date/Time Summer/daytime'

* and then slew up to 4000 feet and FLY that free flight and confirm you see similar problems there.

Whereas I know exactly what I put into the Mifflin missions, Peter doesn't, and that leads to a lot of second-guessing (although the fact is the mission has just a FLT and WX file that affect CumulusX) - we need to be 100% sure you have tested your PC with a different WX file, hence the free flight instructions above.

My understanding is you've only tested free flight with weather set to no clouds and CumulusX set with a minimum thermal height of 100m, so the fact that you then had Cu's at ground level (800 feet @ Mifflin) didn't tell us anything was wrong.

Ian

Link to comment
Share on other sites

Ian- I have followed your request as accurately as possible with global scattered (2/8) Cu @ 6500msl base and 6700 msl tops and wind from 311 @ 17kts. Free flight 1.00pm in July Mifflin. To apply this globally I clicked "apply to all weather stations" in the advanced weather settings.

I set Cumulus x to "unblue" and configuration to "default"

I can confirm that everything started out correctly with both the Cumulus x and the scattered cu at 6500msl.

I slewed to 4000 and then headed SW and after approx 2 min in slew mode in this direction Cumulus x began replacing all the cumulus x clouds with clouds at 9000msl until all that was left was Cumulus x at 9000 and the scattered cu still at 6500.

Remembering that the default config shows min lift ceiling of 1500m and max at 3000m (which is close to 9000ft) it seems that these config settings are overiding everything else but only when moving some distance SW geographically.

regards,

Jeff

Link to comment
Share on other sites

YAY awesome. Thanks a lot Jeff - this might not sound much but it was still unclear to me whether something in the mission was screwing things up or not so now we know it's definitely not that. The bad news is that means my 'v9 patch' for the mission was never going to fix things for you.

It sounds like you 100% created the flight and the weather correctly, and if you saved it that will give us a repeatable sample to test with.

First you should check the weather is indeed *global* (I made this mistake with the prior version of the mission). On the weather/customise page, click on any weather station that is a long way away from Mifflin:

post-16139-1213863051_thumb.jpg

The 'apply weather to' area should change to 'a specific weather station'.

Then click 'Advanced weather'

post-77-1214301282_thumb.jpg

and you should see the 'global' weather you created. There is a possibility the weather here is different, if so you might poke around to the SW of Mifflin and see if it's different there too which would give a clue.

*edit* continued....

There is a weather station to the SW of Mifflin (you'll see it on the 'Weather' map) - I *think* it's called Altoona-something - this would be the critical one to make absolutely sure it has your global 6500 foot Cu setting, or you could expect the behaviour you are seeing as you move towards it. On the earlier release of the Mifflin missions this somehow got screwed up and the wind is global but the Cu's aren't - I have no idea how this could have happened - to keep things simple don't test any more with the missions, even if you've applied the v9 patch.

The whole sequence where you move in a particular direction and the weather changes just seems like the settings of an individual weather station to me, so we really need to find out what the situation is in your free flight scenario.

**another edit...

In fact, to be sure about testing, you're better off with a distinctive altitude range in your CumulusX settings, e.g. 650-650 metres (i.e. 2000 feet).

Ian

post-16-1211991206.jpg

post-16-1211991323.jpg

Link to comment
Share on other sites

Just checking that instead of checking the box "all weather stations (settings applying to individual stations will be lost)

I should be checking the box for an individual weather station that is a long way away (like in the next state)?

Jeff

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