Using Active Sky Next with CumulusX! to create a virtual soaring environment

Related to this post is a YouTube video featuring one my virtual cross-country soaring tasks in the Aerosoft Discus X. The video can be accessed at the end of this post.

This is the third cross-country flight in a series of test flights using real-world weather add-ons for FSX.  This series of flights explores the notion of using a real-world weather add-on to augment the virtual soaring environment provided by CumulusX!.  The methodologies I chose to test this notion are not rigorous. The test flights were as much about enjoying a cross country task based on some newly acquired Orbx scenery for Laufenselden airfield as they were about discovering some potential new ways to create a virtual soaring environment.  Nevertheless,  I  am going to take the liberty of drawing some overall conclusions, from the anecdotal evidence acquired in these experimental flights.

In this specific test case I used Active Sky Next (ASN) to capture and save a historical snapshot of  ‘real-world weather’ in FSX.  To capture the weather snapshot, I waited until the initial weather update had been downloaded and installed into FSX and then I saved the current flight in FSX.  I then shutdown the ASN client application to prevent any further weather updates.  As demonstrated in my previous YouTube video, periodic weather updates are undesirable when using CumulusX! because some changes in reported weather can remove all CumulusX! thermals in an abrupt and unrealistic way.

I made one slight concession towards the principle of  ‘comparing apples with apples’ in that I used the same cross-country task as my previous tests. It is a 63 nm task with three turn points based at Laufenselden in Germany. It was not practical to use the same date and time as any of my previous test flights because ASN did not produce a useful thermal distribution until I selected  a new virtual date and time in FSX.  In fact, it was difficult to find any dates or times in the ASN historical weather records that would produce a reasonable thermal production from CumulusX!. This issue may turn out to be an ongoing frustration in trying to extract good thermal conditions from ASN.  I should point out, however, that I  only tried the first few weeks of August, plus I skipped some days and also I chose to rule out any clear sky days because I do not like flying cross-country in conditions that only produce blue thermals. After much trial and error I settled on 12:45 UTC on the 9th August 2014.  This weather snapshot produced a promising distribution of CumulusX! clouds around the Laufenselden airfield.

For this test I also set ASN Wind Options for updrafts and downdrafts to zero so that lift and sink would be controlled exclusively by CumulusX!.

So how did it go?

The following Plan-G screen capture shows my flight path overlayed on the task flight plan:

Flight Path ASN test

The first leg of this task from Laufenselden to Selters began well.  I was even able to stay close to cloudbase by ‘dolphining’ and only stopping to circle in two thermals.  As I crossed north of the A3 autobahn at Montabaur, I noticed that the northern region of this task course was devoid of any CumulusX! clouds for as far as I could see.  Moreover, as I progressed further northward,  a patchy stratus cloud layer began to replace the abundant cumulus layer, although some cloud textures looked more like stratocumulus.  CumulusX! will not produce thermals where a stratus cloud layer is dominant. CumulusX! will not even produce blue thermals in these conditions. Faced with this desperate situation I chose to fly at a very conservative (slow) speed to maintain glide efficiency and I abandoned any hope of a more direct route to Weilburg by backtracking to the south after passing the first turn point. This strategy gave me a reasonable potential to find some thermals and avoid a land-out. After backtracking to the south-east, the first cloud I tried was a dud, but I quickly found another and managed to claw my way back up to cloud base.   I was then able to proceed eastwards until within reach of the second turn point.  At that point I wasted a lot of time climbing as high as I could in a mediocre thermal before making a dash for the Weilburg  turn point.   Having reached Weilburg with sufficient height, I was then able to return safely to the thermal producing region on my south west track towards Diez.

The necessity of the off-course diversions (zigzags) to intercept thermals in the last two legs of this task are debatable. Nevertheless,  I arrived at the finish line with just enough height to make a slightly truncated  left hand circuit and land safely at the airfield.


You might be tempted to think that I should be dismayed by the uneven thermal distribution around this task.  To my mind, however, that is  precisely what made this task challenging and thus more interesting, even realistic by real-world standards. Nevertheless, CumulusX! still appears to be rather too pedantic about eliminating all thermals under a broken stratus cloud layer – a few blue thermals in those conditions might be more realistic.

The weather transitions were relatively smooth and realistic compared to any weather variability achievable with default FSX weather.  The light and fluky wind variation in the saved weather snapshot was well controlled and convincing.  I was surprised that this variable wind appeared to be maintained even though I had shut down the ASN client application. Thus I wonder if the ASN server component is still active in circumstances like this? Moreover, this rather nice feature of the ASN  weather snapshot was evident even if after I re-booted the PC  and re-started FSX with the saved flight. So far I am impressed!

On the other hand, I suffered much frustration in trying to find a date and time that would result in reasonable thermalling weather from ASN with CumulusX!.  The problem here could be twofold: firstly the cloud reporting might be inaccurate or limited in the weather information resources used by Active Sky Next and secondly CumuluSX! has an extreme aversion to any dominant stratus cloud layer, even in areas where that cloud layer is not overcast.    In contrast to ASN, the FS Global Real Weather add-on has an option to give preference to cumulus clouds and thus is a better option for working with CumulusX!, even though the weather transitions are less well controlled.

In an overall sense, CumulusX!  is a fabulous FSX add-on for virtual soaring enthusiasts but it is a pity that its absolute aversion to any dominant stratus cloud layer could not be watered down a little or even made tunable in the configuration settings.  Then we could better utilize the outstanding weather control of the Active Sky Next weather engine to provide some more challenging and realistic cross-country soaring tasks.


9 thoughts on “Using Active Sky Next with CumulusX! to create a virtual soaring environment

    1. James Allen Post author

      Sorry about that. I am getting spammed often enough to warrant filtering and moderating all comment posts. I am also wanting to keep this a commercial-free site, but some spammers cleverly disguise their advertisements as comments or advice. So in some cases it is difficult to decide what passes and what does not.

  1. Harry O

    Great to have found you, I’m a CumulusX + ASN + FSX pilot myself. I’ve got a great way of designing cross country tasks. Do you know about ( If you create an account, you can download IGC files for real world cross country flights and then run those IGC files in Simlogger along with CumulusX/ASN combination. The first few tries at it have produced a better than 60% rate at producing very convincing virtual cross country flights in ORBX scenery with a real world glider to follow along with in the simulator. It is a lot of fun.

    Let me know if you want to know anything about it or get together for some virtual gliding. I’m an Aussie. Merry Christmas to you!

    1. James Allen Post author

      Thanks very much Harry,

      I will definitely look into your interesting suggestions – am just tied up at the moment trying to get FSX Steam edition working in parallel with my existing FSX installation and having some problems.

      I wonder if our main task designer in the UKVGA might be doing something similar because some of the UKVGA tasks are based on real-world tasks but are often shortened a bit to make them more suitable for our on-line multiplayer events. Moreover, he often tries to match up the weather manually in FSX to give a reasonable representation of the prevailing weather conditions for that task. The UKVGA use manual FSX weather for their multiplayer events because ASN is not good for multiplayer gliding tasks using CumulusX! unless all players have it and have identical settings so that soaring environment can be synchronised.

      Nevertheless, I am also interested in off-line tasks (or limited multiplayer flights where all participants have the same scenery and weather engines) using weather engines like Active Sky’ Next. It is the CumulusX! – Active Sky Next combo that I find most vexatious. ASN seems to prefer stratus clouds more often than not and if these are dominant or lower than any cumulus layer it kills all thermals in CumulusX!


      1. Harry O

        Hi James
        I’ll try to get my act together and join in multiplayer at some stage. The time zone difference kills it for me (need sleep!) On the topic of ASN/cumulus and stratus, yes it is a pain! As well as what you mentioned about CumulusX being overly pessimistic about stratus and thermals, ASN also seems to have a bug. One in every two or three times, ASN will load in some stratus instead of all cumulus. When that happens (and I know there should be no stratus), I reset the flight in FSX and try again. Sometimes, the results are spectacularly good, sometimes not. I guess we can’t complain considering how good FSX and gliding has become!

        1. James Allen Post author

          Yes, the tyranny of different time zones is a big spoiler for international multiplayer events. I am not not attending the UKVGA multiplayer events much these days unless one or more of the members flags a possible late attendance in the forum so that I might have someone to fly with. I just don’t function well that early in the morning and you need maximum concentration to race gliders cross-country. Moreover, early starts don’t fit in well with my family life, plus I often have clashing appointments on weekday mornings. A few years back the guys in the UK used to adjust the event times a bit later to suit me but now that we have members in South Africa they can no longer do that.

          So you think the ASN inconsistent cloud choices might be a bug? Interesting – I never considered that possibility. I could be wrong but I got the impression that because metars don’t always specify cloud types that ASN was making a random choice of cloud type and possibly layers as well, but not just at the start of a session , the program seems to make a random selection with each weather update!! But I could be confused by the variation due to interpolation changes as I fly between weather stations. Whether or not it is bug or an arbitrary choice by the programmer, it is a shame that ASN doesn’t always play well with CuX because in other respects it provides some very convincing air texture and even a few blue thermals. Also on those occasions when they are working well in combination, it seems very realistic to me.

          Also I think CumulusX! would be more realistic if it produced some blue thermals if the conditions were right and the stratus layer was not full overcast. In the real world I have encountered thermals with stratus as long as there are sufficient gaps to let sunlight through and create some instability. Unfortunately there doesn’t seem to be any practical way in FSX to associate thermals with specific ground features other than sun facing slopes and broader land classes such forests. Thus any task based purely on blue thermals is nearly impossible in the sim.

          Recently I tried to create a task based at Benalla in Victoria for my YouTube channel using ASN and CuX in real time with ‘real’ dynamic weather. It looked like fantastic conditions as I was setting up for launch but it was a disaster – a sled ride to an out-landing after an aerotow launch to 3000 feet. So I am going to retry it after Christmas but with manual FSX weather based on the ASN metar data for the day.


  2. Harry O

    Thanks James. I’m an Aussie up in Queensland. As you have noticed, I think what is happening in ASN is that its “advanced interpolation” adds some randomness with stratus, which is a curse for us pilots in CumulusX. I think it is a bug because it is not consistently random either. The way I have learned to handle it is to observe the clouds ASN injects right at the beginning of the session by looking around with TrackIR. I select “Flights – Resets” until the clouds I observe match what I expect. It can take a few resets sometimes.

    I haven’t yet observed the ASN “bug” with each new weather update but you could be right! At one point I was turning down the updates to the slowest possible but recently have returned it to default and achieved an amazingly good cross country flight in the sim on default updates.

    I know the simulated weather is performing well, when I can fly the same cross country course as the real world pilots at the same time as them, using their own flight recording data from OLC injected via Simlogger. If I notice that they are topping out at roughly the same altitudes in a thermal, tracking across the ground at roughly the same speed and maintaining the same speed and height across ridges, things are going well! It doesn’t happen all of the time (because advanced interpolation is pretty rough) but when it happens it is amazing. Success rate is better than 50% so far though. I’m always on the OLC site checking for the amazing flights the real world pilots have done the day before. Sometimes there are up to 25 flights to choose from in cross country just in Australia alone.

    I think there is also a bug with the interaction between CumulusX and ASN during updates. Sometimes when ASN goes from patchy cumulus clouds to clear, CumulusX doesn’t seem to follow and keeps cloud cover when it should switch to blue. What I do, when I suspect that something is wrong (for example, the real world pilot is thermalling at higher altitudes when I am limited by the cloud tops), I see what ASN is reporting and if it seems different to CumulusX, I save the flight and reload and maybe even reset the flight until it matches up as I expect. It can be frustrating, but the results are great when it works. Pretty stunning.

    Two days ago I followed a real world cross country flight across the Mount Beauty ranges. The real world pilot used an ASK21 but in the sim, although the wind direction was about right, at lower altitudes the ridge lift dropped off from (18 knots to 5 knots some times). This forced me to switch from the ASK21 to the ASW22BL and with the much better glide ratio, could easily keep up with the real world pilot and sometimes improved on his times between way points. It was a flight I will never forget. If you want the save game and files for it, I would happily send it to you so that you can try the course out for yourself. Next attempt, I’m going to take off from Mount Beauty airfield and attempt to snake up to altitude using the same ridges as the real world pilot. I suspect it is possible in the ASW22BL.

    Since reading your blog, I have found Plan-G and will work with that when entering in the flight plan from the OLC website. Also interested in trying your recommendation of the DG-800. Since it is Christmas, I have seriously got my hand on the mouse to click on the 33% sale for the ORBX FTX scenery for north and south islands of NZ! The cross country there is amazing when you look at the flights on OLC. I just worry that ASN advanced interpolation is going to be too rough to handle New Zealand, but they say they have made progress on it.

    Once I really nail what the problems are between ASN and CumulusX, I’ll try to put in the effort to inform the developers about the issues. (Peter Lurkens and the developers for ASN via their ticket system).

    Have a very happy Christmas James. Stay in touch. Other than one other person in WA which I have unfortunately lost touch with, I think we might be among the tiny number in this country of 23 million, who love soaring in the sim! I’d be happy to try multiplayer online sometime! But I’m clueless on how to do it and would have to be walked through. I’m an offline specialist because of being so isolated here in Australia.

    Cheers, Harry

    1. James Allen


      I will get in touch again after Christmas. If you do get the NZ Orbx scenery, maybe we will organise a MP flight at Omarama on the South Island. Otherwise, Mount Beauty is also good for me – I have the OzX airfield scenery, which includes the Mount Beauty airfield, as well as the Orbx FTX Australia regional scenery. I had my first two real-world glider flights with the Mount Beauty Gliding club and I have some relatives living at Porpunkah which is not far from Mount Beauty, so I know that area very well. The Victorian alpine region is a very challenging area to fly in FSX because of the way CumulusX! interacts with terrain – likewise at Omarama in New Zealand, particularly as you track towards Mount Cook and the Tasman Glacier.

      Actually the current task I am trying to get done for my YouTube channel is from Benalla to Mount Beauty. And yes I would be interested in getting your saved flight files for that Mount Beauty flight based on real-world flight log.

      You mentioned my review of the DG-800 in your last reply but I thought I had better mention that it is was actually the DG600-17 from Wolfgang Piper that I reviewed.

      Merry Christmas,

  3. Pingback: Flying the same OLC gliding course as a real pilot over Mount Beauty Australia using FSX | harryosh

Leave a Reply

Your email address will not be published. Required fields are marked *