Dear All,
As there is currently no other forum for international discussion on safety implementations and IGC Annex A rule changes applicable to WGC/EGC, I will try to initiate such a discussion here. Hopefully, this will work.
I initially attempted to do this via email, but that approach is complicated, involves only selected participants, and is difficult to follow for those who join the discussion later.
The Cylinder Start method is a very good solution; however, if it is not properly understood and practiced, it risks being rejected simply because pilots are not prepared for it—either practically or mentally. We still have a few pilots in Poland who complain about it, usually those who prefer flying in gaggles. Interestingly, our junior pilots like it and none of them complain—probably because they do not have the "it was different in the past" mindset 😉.
From my experience, it looks like this: if you approach the Cylinder Start with an open mind and a willingness to understand it, you will appreciate the safety improvements, simplicity, and tactical benefits it offers—especially if you prefer an individual flying style. If, however, the attitude is "another crappy IGC novelty," dissatisfaction and complaints are almost guaranteed. This is why an educational approach is essential.
First of all, please familiarize yourselves with the Annex A rules regarding the Starting Cylinder.
Attached you will also find a short presentation explaining why it was introduced.
Please use only English in this discussion.
Regards,
Karol.
I received one email with some valid questions, please see the answers below.
In reply to your email, we have been using this procedure in Poland for several years now. Of course, Polish nationals are not WGC or EGC, but the system has been well tested in various conditions.
First of all, I do not agree that a waiting time in the current PEV start method is good. In my opinion, it is very bad in some cases, although it may work in others. Every PEV variation has its advantages and disadvantages. The cylinder start was designed with safety as the first priority; all other aspects are secondary. However, I also believe that from a sporting point of view the cylinder is better than a line start.
This cylinder start procedure works, is tested, and has been very deeply and carefully analysed for many different situations. Most elements have been positively validated in real flying during various contests. You can check this on SoaringSpot, as most competitions held in Poland have been using this procedure. It would be a shame if someone tried to change it based only on intuition rather than real experience—it is very easy to destroy the integrity of a procedure.
Everything depends on the weather conditions. On very specific days, in extreme cases, we can see weaknesses of the cylinder start, but in exactly the same weather a line start would be just as problematic.
Regarding your specific example:
Now, if you wait and circle in the forward part of the circle you can see others fly past heading for the first TP. You wait for like 3 to 5 minutes and follow. There is no 8 minutes you have to wait, but can follow straight away.
One of the beauties of the cylinder is that it is a huge area. In fact, you never really know whether gliders have already started the task—unless the altitude difference is set too low. We saw this during our last nationals, where the Competition Director did not accept that the altitude difference should be around 200–300 m below the expected thermal height at start opening. In that situation, with cloud base at 2000 m and the altitude difference set to 1000 m, pilots who had already started were easily recognisable.
With a start line, there are usually one or two "best" places from which everyone starts the task. With a cylinder start, you have many more options. Even on a blue day, you can fly backwards—opposite to the first turnpoint—start with PEV at the beginning of the cylinder, then turn back to the thermal where others are circling, use it, and fly for the task or wait for a better moment. No one can tell whether you are actually going for the task or just faking.
The beauty of the altitude difference is that it is not beneficial to start from the very top of the thermal. Gaining the last 100 m on the inversion makes no sense, so there is no fear of faking the start. Finally, since the best starting place is unknown (unlike a start line), pilots tend to choose their own place that seems optimal. The tactical game starts before the start moment, which significantly helps to scatter gliders across the whole cylinder (weather-dependent, of course).
I fear that in blue weather we end up with one of the older problems; nobody wants to start first.. But these are assumptions.
You cannot eliminate this problem with any PEV procedure. The only real option is to set long tasks correctly, so that waiting increases the risk of outlanding. PEV waiting time definitely does not eliminate this issue, and I could give examples. "But these are assumptions." - Exactly.
It would be a wonderful system if it is possible to add a waiting time.
I do not agree. Waiting time sometimes works opposite to the intentions—it groups gliders because you cannot use the best moment to start. For example, with a cylinder start and altitude difference, you can start your task in a thermal, use it to the top, and then fly away. When you add waiting time, this is no longer possible. Instead, you are forced to circle on airbrakes in the thermal while waiting. I have experienced this several times already: I couldn't start, had to wait, other pilots noticed me, joined my thermal, and we all started together—disaster.
It still seems a nice system because of its simplicity ( except for the calculation part that is)
The calculation part is easy if you have navigation software that supports this procedure—currently LXNav and XCSoar—and it becomes much easier than a start line after some practice (not only in my opinion). Scoring is easy with Polish add-ons. Unfortunately, SeeYou does not want to cooperate for some reason, which is very frustrating. They promised to support it, but nothing has happened so far. My assumption is that they see no benefit in adding it to their scoring software.
...but I don't see why it would prevent gaggle flying or why it would make it more difficult to wait until good pilots leave and then follow.
In practice, it works because it is quite difficult to spot other gliders—however, only when FLARM is used in stealth mode and with no track options, as we do in Poland.
Perhaps I'm totally wrong, it might be a good idea to try this in the 20m class alone. On the other hand, I 'm a bit afraid a WGC is the not the most suitable competition where we test this method until it is watertight and proven.
It is watertight, but it has to be practised by pilots before major competitions. That is why it has been in Annex A for almost two years—there was time for it. If it is not implemented at a WGC, it will probably die a slow death, as no one will be brave enough to use it in competitions. I am pretty sure that once pilots see how it works, most of them will find it beneficial—especially those who prefer individual flying, which is disappearing rapidly, mainly due to spying devices like FLARM and OGN.
Thank you for questioning.
Cheers,
Karol
For context and additional explanation, I include PROPOSAL FOR IGC PLENARY 2023 and 2024. This explains a bit why this procedure has been introduced.
Please find the attachments with the proposal forms.
Proposal 2023:
It is proposed to introduce following starting method into Annex A in addition to existing ones:
Task starting area is defined as a cylinder of 10 km radius. Task has specified maximum allowable
height loss between start and finish. Being anywhere inside t he starting area the pilot is declaring
the task starting point by effecting Event Marker (PEV). Pilot's individual start point exact time,
coordinates and altitude is used to calculate marking distance and task average speed. If height
loss between start and finish is greater than the specified on task sheet pilot will receive penalty
points.
- For safety purposes ground speed during starting moment should be limited and
experience shows that for ballasted gliders it shouldn't be less than 170km/h.
- To avoid leaching and gaggles, Event Marker interval of minimum 5 - 15 minutes should be
required.
- Height loss of 1000 m works well in usual European weather conditions however it should
be possible to adjust it in order to assure adequate separation from the cloud base/thermal
height during task start.
Reasons supporting proposal:
In the opinion of many competition pilots, task start is one of the most dangerous moments during competition flight. With the present start line, we frequently observe gliders piling up in very small area and at similar altitude. As all the pilots are trying to start the task as high as possible it usually takes place just below the cloud base and on a blue thermal day at the inversion level. Additionally gliders which have already started the task are passing on opposite courses those who are flying towards the start line and again usually just below the cloud base with reduced visibility.
The idea behind the proposal is to separate/scatter gliders over a bigger area and using maximum altitude loss condition make pilots starting the task on different altitudes. Usually, with such a starting method, task start happens in the thermal with relatively low speed or flying straight towards the first point.
Additional advantage of maximum height loss is that pilots without being penalized can set their target finish altitude by adjusting starting task altitude. This could be a huge safety benefit when flying in difficult terrain or when pilots are not comfortable/familiar with the airfield.
This starting method has been tested in Poland for the last year on most Polish competitions and general opinion from pilots is very positive. For the main disadvantage, lack of navigation/scoring software support has been pointed out, however it seems fairly easy to overcome. Survey after the 2022 flying season has been published online and 75% out of approximately 70 pilots supported the idea. Survey results are attached to the proposal.
Beforementioned Survey is available at https://forms.gle/GWfutByLfUeBr9ot7
From experience, starting area cylinder radius of 10km works best, 5km radius seems to be too small to give pilots the opportunity of finding their own starting position. In certain circumstances bigger radius could do better however in such a case extension of time between last take-off and start opening should be considered.
Reasons in Year two proposal 2024.
The proposal to introduce a starting cylinder along with a maximum height loss is an attempt to create safety-focused solutions. Current rules do not prioritize safety; on the contrary, they promote risky behaviours. An example is the current Line Start, where it pays off to be as high as possible at the moment of task start and as low as possible on the finish ring. If a pilot tries to be safer by maintaining a greater distance from clouds or approaching the finish line at a higher altitude, they automatically lose out. Assuming that they will descend 100m below the track at start and arrive at the finish line 100m higher than they could and result in a loss of around 15 points - multiplied by 7 tasks, totalling about 100 points at the end of the event. The current altitude limit at the start doesn't improve much; while pilots avoid flying in clouds, all gliders are at the same maximum altitude when they depart on the task. The minimum altitude limit at the finish means that pilots focus on the altimeter readings rather than external observations.
In the case of the Starting Cylinder, the goal is to spread out gliders over a large area by allowing start from any point in a cylindrical airspace of ample volume. The Maximum Height Loss gives pilots the option to choose a different start altitude for each. This means that gliders are separated both laterally and vertically. One never knows if another pilot has already started the task, reducing leeching and gaggle forming risk. An additional advantage is if someone wants to arrive at the airfield at a higher, safer altitude, they won't be penalized for it. With this method, it doesn't matter for sporting result whether a pilot departs 100m higher or lower on the task.
There are some doubts regarding scoring.
SeeYouC still does not support, nor calculate, the start point from a cylinder using PEV according to Annex A, so there is no automatic results calculation as there is with a standard start line. This is despite the fact that Andrej Kolar from Naviter stated that it would be implemented. Unfortunately, it has been taking several years now. It is somewhat strange that the official scoring software does not support an Annex A–approved start method.
That said, thanks to our excellent scoring experts, there is a workaround using add-on software, and publishing results during Polish competitions works quite efficiently; recently it was almost as fast as scoring with a classic start line. In my view, scoring is the smallest issue when it comes to implementing a cylinder start at the WGC.
How much of an advantage do you think pilots who have tested this start method over the course of the past two years have over pilots who get exposed to it for the very first time at this WGC?
Because, let's be honest: the first group is mostly Polish pilots, the other group mostly "rest of the World". We have to be prepared for the accusation that the organizing NAC chooses a start method that almost nobody else has had the chance to get familiar with.
I am not fully objective, as I am Polish and was the main designer of this procedure, but this is my honest opinion, and I am trying to be fair with both myself and with you.
In my view, trying this method during a training flight at your aeroclub should be sufficient. The only area where prior experience may help is proficiency in using navigation equipment. In the case of LX90xx and XCSoar, this method is already implemented and well tested—there are no surprises. It is essentially a "shoot and forget" operation.
As for tactics, they cannot really be practiced in advance; everything depends on the weather on a given day. Start tactics involve observing the conditions inside the cylinder and choosing the best line or area based on the weather, while also considering standard AAT factors such as timing, distance, and task optimization.
An analogy might be this: Pilot "A" has only ever used a start point to begin a task, and then flies a competition where a start line is used. Would pilots experienced with a start line have any real advantage over Pilot "A"? In my opinion, not really. The most important factors remain decision-making ability, reading the sky, and general competition skills.
With a start line, tactics are relatively limited—you want to be as high as possible and positioned under a cumulus just before or just after the line. If you are there, you will usually get the best start. With a start cylinder, however, no one truly knows what the best place to start is, and that is its beauty in terms of separating pilots. The more individualistic the pilots, the more dispersed they will be—unless the situation is very obvious, such as a simple cloud street where everyone knows they should use its full length, starting at the very beginning of the cylinder, or something similar (e.g. a single strong CU inside the cylinder).
Additionally, due to the loss of height at the start, there is no advantage in being as high as possible. There is no need to fly or start just below cloud base, which we often see with a line start when no altitude limit is imposed. With a cylinder start, there is also no need to impose an altitude limit—pilots choose a safe and appropriate altitude for themselves.
Karol.
Due to the different arrival heights the landing procedure has to be adapted from a simple "make all straight-in landings" to a more sophisticated "choose between straight-in or traffic-circuit landing". My experience with the simple straight-in solution during mass arrivals in past competitions was: excited discussions on the radio between pilots about who is the one with the least energy, and how to sequence if the next glider is right behind you. In commercial aviation safety is increased with vertical separation. In gliding it seems to be different...?
Regarding arrival height: based on our experience and in real operations, this issue has not occurred during our competitions even once, at least as far as I can recall.
Most pilots start the task at an altitude that results in an arrival at the finish ring (3–4 km) with about 100–200 m of height. Cases where pilots deliberately start high in order to arrive with 500 m or more—to use a higher TAS strategy—are very rare. To summarize, final glides and arrivals look exactly the same as during classic start or AST tasks.
One important aspect, however, is proper adjustment of the finish ring diameter to allow high-performance gliders to dissipate energy safely before landing. Since Częstochowa is a very large airfield, it is necessary to set a finish ring of at least 5 km when its center is located in the middle of the airfield.
Karol.
Dear All,
Please find a visualization from one task of the Polish Nationals.
The day featured relatively good weather with blue thermals, and in my opinion, on blue days an AAT should always be used.
It takes about four minutes to see how it works on a blue day: an AAT with a cylinder start. As you can see, toward the first turn area there were two or three groups of gliders. Later, due to different speeds and tactical choices regarding where to make the turnpoint, the field became more scattered, only to group up again somewhat on the final glide. I am extremely curious how this would work at WGC/EGC level, where pilots are generally more competitive and strongly score-focused.
An interesting aspect is the starting moment itself. Most pilots chose to start near the beginning of the cylinder in order to use thermal spots they had been checking beforehand. In practice, however, this is not necessarily a good strategy, as the best lift may already be used by pilots who have started earlier. On that particular day, thermals were changing position frequently—if you found a good thermal in one spot, after 15 minutes there was often nothing there anymore. Interesting weather and an interesting task.
As a reference, here is a six-minute visualization from another Nationals day with (partially) blue thermals, a classic start line, and an AST task. The difference in behavior is clearly visible.
Cheers,
Karol.
I don't have a strong opinion on whether the start cylinder is a total game-changer, but it is definitely an improvement over the traditional start line.
With a start line, it's often obvious when someone is about to cross. Which also encourages "fake starts" to mislead others. This leads to unnecessary maneuvering in the most congested areas before the race has even begun.
In contrast, the cylinder allows pilots to spread out over a slightly larger area. While it's still possible to follow other gliders, the leaders' intentions are less obvious, and there is no single line acting as a bottleneck that triggers tactical games. This increased dispersal is where the magic happens - it makes the start much safer and significantly improves competitiveness.
Really good presentation BTW. Haven't seen it earlier.
New presentation regarding Cylinder Start concept and use.
https://drive.google.com/file/d/1cllKlR9XOpHJPvKMJixpwcQYXyN-Dv80/view?usp=sharing (https://drive.google.com/file/d/1cllKlR9XOpHJPvKMJixpwcQYXyN-Dv80/view?usp=sharing)
XC Soar with Cylinder Start implementation:
https://play.google.com/store/search?q=xcsoarpev&c=apps&hl=pl