RC madness

On Tuesday, December 22, 2015 at 3:49:02 PM UTC-7, Sean Fidler wrote:
UH,

Its ALL RC aircraft between .5 lbs and 55lbs. A different catagory of rules apply for 55lbs.

The AMA has negotiated drones only, and expected that. This came as a huge shock to many. The FAA, in their wisdom, changed to scope to included ALL RC aircraft between .5 and 55 lbs at the last moments. All limited to below 400 ft. which is comical for RC sailplanes and most high performance jets and electric aircraft.

Sean

The 400' rule is the problem at this point. The FAA it seems is trying to make it mandatory in the registration process. You can not register unless you agree to the 400' limit. If you do not agree, you can not register, and can be fined and prosecuted. I am told that this is in direct conflict with the existing "FAA Modernization and Reform Act of 2012 HR 658", which states that the FAA can not place additional restrictions on model aircraft.

Mike

On Tuesday, December 22, 2015 at 10:14:22 PM UTC-5, Mike C wrote:
On Tuesday, December 22, 2015 at 3:49:02 PM UTC-7, Sean Fidler wrote:
UH,

Its ALL RC aircraft between .5 lbs and 55lbs. A different catagory of rules apply for 55lbs.

The AMA has negotiated drones only, and expected that. This came as a huge shock to many. The FAA, in their wisdom, changed to scope to included ALL RC aircraft between .5 and 55 lbs at the last moments. All limited to below 400 ft. which is comical for RC sailplanes and most high performance jets and electric aircraft.

Sean

The 400' rule is the problem at this point. The FAA it seems is trying to make it mandatory in the registration process. You can not register unless you agree to the 400' limit. If you do not agree, you can not register, and can be fined and prosecuted. I am told that this is in direct conflict with the existing "FAA Modernization and Reform Act of 2012 HR 658", which states that the FAA can not place additional restrictions on model aircraft..

Mike

This just in:
http://amablog.modelaircraft.org/ama...odel-aircraft/
AMA says hold off till appeal
Now we really drifted a thread.
UH

On Tuesday, December 22, 2015 at 7:50:19 PM UTC-5, jfitch wrote:
On Tuesday, December 22, 2015 at 2:15:16 PM UTC-8, wrote:
SNIP Not trying to be difficult but I must be missing something. If they're that close to you (i.e., within 2 km), they're visible even with Stealth.


From the Flarm documents, V6.0 page 19 (maybe this has changed?):

"Targets with enabled “Stealth Mode� are only displayed..... if they meet at least one of the following requirements:

- target is a threat

- target is within 100m horizontal and 50m vertical
- target is within 2000m horizontal and 300m vertical and within ±45° of own flight track."

So the guys going your way - the ones ND is going to pull up sharply into - you don't even know are there. They aren't a threat because they are paralleling your course. Even if they are close enough to appear, their relative altitude is intentionally wrong.

Now I am going to repeat for the 20th time or so, I don't believe Flarm is a huge increase in safety. It is a big sky and most accidents are spin/stall, not head on. I don't even call my Flarm an anti-collision device, I call it an in-flight entertainment system. But certainly beyond a doubt, it improves situational awareness always, and particularly in the scenario described. On The White Mountains and the Sierra convergence lines, I don't think there have been any head-ons than I can recall. Its a big sky. But plenty of people have had to change their underwear at the end of the day, I can assure you. It was enough of a concern that a rather elaborate procedure was devised in the area, reserving a radio frequency and involving reporting points etc., all of which seemed pretty ineffective, while non-stealth FLARM pretty much solves the problem completely and with no distraction.

i'm not sure.... but think i could miss a glider as i pull if i'm looking right at him dude... you said one glider high left, one low right, so how am i in a position to hit someone if i go high right? especially if i look up in the direction of my pull?? you can't win at hypthetics, because the antagonist can always say, "ok that's fine, but what if..." i'm gonna bow out of this stupid debate now. have the last word if you like. see you dudes! (hopefully)

"no brainer"... Typical.

On 12/22/2015 2:37 PM, Dave Walsh wrote:
The answer is that many glider pilots did not survive; read
Bruno Gatenbrink's article.
Installing Flarm is a no brainer.
Seat belts, crash helmets, air bags, NCAP ratings.. modern
nonsense eh?


--
Dan, 5J

No agreed to scenario?

*14 CFR 91.113****Right-of-way rules: Except water operations.*

(e)/Approaching head-on./When aircraft are approaching each other
head-on, or nearly so, each pilot of each aircraft shall alter course to
the right.

Some people are writing their own rules when the rules are already in
effect. So if one of you turns to the left because that seems the best
course, and the other turns to the right because that's what the FARs
require, what happens?

Winter seems the time for everyone to come up with outrageous
scenarios. And BTW, it's snowing here, too.

On 12/22/2015 5:15 PM, Andy Blackburn wrote:
On Tuesday, December 22, 2015 at 2:14:18 PM UTC-8, wrote:

Almost everybody, when faced with the head on scenario will likely do what you mentioned. Sooooo- a better scenario is to push and no banking as it likely will increase your projected cross section.
Literally it's no different than if we just flipped coins at the start of the contest and divided into "pull" pilots and "push" pilots. Your odds are 50-50. If you could get everyone else to agree to push it would be good - for you.

Matching the other guys bank is best obviously.
I'm pooping my pants just thinking about that one! Thanks goodness it happens rarely.

Seriously, there is no agreed to procedure for head-to-head because there are too many scenarios where doing something by rote only makes matters worse. You should assume that in this situation you will not see anything until it is too late. The closer to collision course the less likely you will be able to pick it up. That's just the biology of the human eye. Best to not rely on panic maneuvers if you can avoid it.

9B

--
Dan, 5J

On Wednesday, December 23, 2015 at 8:23:24 AM UTC-8, Dan Marotta wrote:
No agreed to scenario?



14 CFR 91.113*Right-of-way rules: Except water operations.



(e)*Approaching head-on.*When aircraft are
approaching each other head-on, or nearly so, each pilot of each
aircraft shall alter course to the right.



Some people are writing their own rules when the rules are already
in effect.* So if one of you turns to the left because that seems
the best course, and the other turns to the right because that's
what the FARs require, what happens?



Winter seems the time for everyone to come up with outrageous
scenarios.* And BTW, it's snowing here, too.




On 12/22/2015 5:15 PM, Andy Blackburn
wrote:



On Tuesday, December 22, 2015 at 2:14:18 PM UTC-8, wrote:



Almost everybody, when faced with the head on scenario will likely do what you mentioned. Sooooo- a better scenario is to push and no banking as it likely will increase your projected cross section.


Literally it's no different than if we just flipped coins at the start of the contest and divided into "pull" pilots and "push" pilots. Your odds are 50-50. If you could get everyone else to agree to push it would be good - for you.



Matching the other guys bank is best obviously.


I'm pooping my pants just thinking about that one! Thanks goodness it happens rarely.

Seriously, there is no agreed to procedure for head-to-head because there are too many scenarios where doing something by rote only makes matters worse. You should assume that in this situation you will not see anything until it is too late. The closer to collision course the less likely you will be able to pick it up. That's just the biology of the human eye. Best to not rely on panic maneuvers if you can avoid it.

9B





--

Dan, 5J

And if there is already an aircraft to your right?

I'm starting to wonder if Andy will be able to resist disagreeing with every thing I say.
Quite some time ago I was part of a group discussion on the very topic. The pilots were asked what they would do in just this situation without knowing what the others would say. Virtually every one answered that they would pull and turn.
The fighter pilot in the group said he would push because almost everybody will pull out of instinct.
It ain't 50-50 odds.

We agree on most things, just not all the ones that get the most airtime... :-)

I was simply expanding the fighter pilot analysis to include everybody else since their lives matter as well. You are correct in the scenario as described. The odds are quite good for YOU if everyone else pulls and you push - 100% in fact.

However, the odds for everyone else are poor - 0%. They will all pull together and their fate will be unchanged - at least if that's all they do. Anything else they try, absent good situational awareness, will be similarly randomly distributed between things that help and things that hurt matters. The best you can do to improve this sad state of affairs is 50-50 - by splitting the field in half - between push and pull. Then on some random encounter for a random pilot the odds will be 50% that his maneuver will be complemented versus neutralized by the other pilot. Even those odds suck, so best to come up with something that works for everyone.

9B

On Wednesday, December 23, 2015 at 11:41:38 AM UTC-5, jfitch wrote:
On Wednesday, December 23, 2015 at 8:23:24 AM UTC-8, Dan Marotta wrote:
No agreed to scenario?



14 CFR 91.113*Right-of-way rules: Except water operations.



(e)*Approaching head-on.*When aircraft are
approaching each other head-on, or nearly so, each pilot of each
aircraft shall alter course to the right.



Some people are writing their own rules when the rules are already
in effect.* So if one of you turns to the left because that seems
the best course, and the other turns to the right because that's
what the FARs require, what happens?



Winter seems the time for everyone to come up with outrageous
scenarios.* And BTW, it's snowing here, too.




On 12/22/2015 5:15 PM, Andy Blackburn
wrote:



On Tuesday, December 22, 2015 at 2:14:18 PM UTC-8, wrote:



Almost everybody, when faced with the head on scenario will likely do what you mentioned. Sooooo- a better scenario is to push and no banking as it likely will increase your projected cross section.


Literally it's no different than if we just flipped coins at the start of the contest and divided into "pull" pilots and "push" pilots. Your odds are 50-50. If you could get everyone else to agree to push it would be good - for you.



Matching the other guys bank is best obviously.


I'm pooping my pants just thinking about that one! Thanks goodness it happens rarely.

Seriously, there is no agreed to procedure for head-to-head because there are too many scenarios where doing something by rote only makes matters worse. You should assume that in this situation you will not see anything until it is too late. The closer to collision course the less likely you will be able to pick it up. That's just the biology of the human eye. Best to not rely on panic maneuvers if you can avoid it.

9B





--

Dan, 5J

And if there is already an aircraft to your right?

then i will beg our father for mercy and forgiveness in my remaining moments for being a dumbass who didn't give myself an out....

i told you, i am done arguing about hypothetics. remember this from my last post? "you can't win at hypotheticals, because the antagonist can always say, "ok that's fine, but what if..."

so i wont respond next time bruh.

On Monday, December 21, 2015 at 6:27:51 PM UTC-8, Papa3 wrote:
Wait, was this the analysis where you compared results across two contests in different years in completely different geographies with a commonality of about 15 percent in the participating pilot population and using the the highly nebulous "PRL to Performance Metric"? The one where you claimed it was "only 259 data points" (when one could as easily argue it was 2 data points; contest A and contest B). Where I think you used "Bayesian" in a sentence... a tactic I thought only an economist would resort to?? Yes - just poking fun at your expense.

But seriously, the statistics (lies, damned lies) you are citing strike me as only one aspect of the analysis. I think the other is either unmeasurable after-the-fact or would need a much more controlled environment to study conclusively. All I can tell you is what I've observed in ACTUAL contests using Flarm Stealth and not using Flarm Stealth and... more importantly.... what is very achievable in the realm of software development and user interface development in the near future.



I'll summarize.

There were three analyses I did across four different regional, national and international (pan-American) contests.

1) Looking at overall results across two 15M Nationals (Elmira and Montague) where stealth was used versus not.

There was plenty of Flarm at a distance in use at Montague - and a protest on the final day that included Flarm use at a distance as supporting evidence and an explicit part of the strategy that was being used - which was to attempt exactly what one poster described - running down the person behind you in the standings and sticking with them. It didn't work out - because it is hard to run people down from a distance even if you can see them - using their thermals rarely helps so you have to catch them on your own. Good pilots aren't run down easily. In this case the last day was a MAT so it should have been an easier task to keep in range.

But I digress. The analysis compared 3-year PRL numbers for each contestant to the PRL component generated from each of the two contests. It was a simple attempt to see if use of Flarm at a distance generated skewed results versus what you'd expect based on the three-year average pilot performance as indicated by the 3-year PRL score. The figure of merit was mean absolute error - that is, the average of the absolute value of the single contest PRL number minus the three-year PRL number for each pilot.

The result was Elmira had a much higher divergence from the expected ranking than did Montague - despite the fact that Montague had standard class gliders flying in 15M without benefit of handicaps. Now both contests has some weather inconsistency - Elmira's was arguably worse and that can account for some of the differences. However, even if you drop the days that forced top pilots to the ground for both contests, Elmira still has more unexplained performance different across pilots than Montague.

You can say this is a single, contest-level example or you can say is is a summation of a bunch of pilot-level data. Regardless of you you want to characterize the volume of data, the conclusion I draw is that there are other factors that create far more chaos in the scores than Flarm leeching - principally landouts based on weather conditions that catch top pilots. Arguably weather-based randomness translates to unfairness since it's the sort of thing we devalue in the scoring formulae.

So (putting on my Bayes hat) - what did we learn? At minimum, there is no strong evidence that Flarm leeching is upsetting the results at the highest level. Also, other factors that we consider luck in our approach to scoring are much more powerful influences. I went looking for Flarm viewing at a distance (and we know there was a lot at Montague) and found none. In talking to pilots at Montague I can say that some were obsessed with trying to get benefit and others were ignoring it almost entirely, so I'd be very her-pressed to say everyone was gaining equal share of the potential benefit. It's a limited dataset based on maybe a dozen contest days, so it doesn't prove that Flarm leeching never benefits anyone, but it does say on average it doesn't make a big enough dent to come anywhere close to overpower all the other random variables that affect how contests turn out different than just the skills of the pilot. In engineering terms, we can say the signal to noise ratio is very low.

2) There was a strong argument made half a dozen threads ago about the main way Flarm leeching works which is basically seeing someone ahead of you who has already gone the the time and effort to find and core a thermal and just popping right into a perfect climb, gaining valuable seconds or minutes.. So I looked at every climb for every pilot on a couple of contest days to see pilots who found their own thermals got better climbs that pilots who took thermals with a market in them. Turns out that the more you borrow other people's climbs the lower your average climb rate is on the same course, same day. Conclusion - there is selection bias in when a pilot stops to climb on his own (you try to stop for above-average climbs and pass up below average climbs). When people borrow thermals from other pilots they tend to ASSUME that it is an above-average climb, but frequently thermal strength varies higher and lower across the vertical profile and if you are following by a great distance you will can easily come in at a different altitude and miss the bubble that is the whole reason the first guy stopped.

3) Not satisfied with averaging across pilots for an entire day, I looked at the problem thermal-by-thermal. Here I plotted climb rate versus entry time into the thermal - first pilot is t=0. For multiple thermals the trend was monotonic decline in strength, E.g. 3.6 knots for the first pilot, 3.4 knots for the second, then 3.2, 2.9, 2.5. 2.4 - you get the picture. The difference between a great climb and a mediocre one can be just a few miles - like 3 or 4. Is that true all the time? No, some thermals are AMAZING and seem to be consistent all day long. Whether pilots find these by skill or luck or local knowledge varies by site and day. If you go to Montague and don't know about Duzel Rock, God help you. If you go to Parowan or Nephi and don't know about Monroe Peak or the convergence lines that set up over mountains or valleys and when, God help you. Local knowledge is far more important than stealing thermals. The simple point is that it seems very difficult to follow someone at a distance if you want to leech - you will likely get dropped. If you want to leech 1/4 to 1/2 mile and the same altitude is preferable.

4) I looked at examples in the thermal-by-thermal analysis where pilots deviated to reach a thermal with other gliders in it and sorted for when the deviation took them to a thermal that they would not have state with stealth mode on (a deviation of greater than 2km and 300m altitude). Again, the results were mostly negative - making a big deviation to get to gliders in a thermal is more often a sucker bet than a good one. There were only a few examples of big deviations so the amount of useful information being acted on from long-range Flarm is quite limited. That in itself is useful data.

Yes, I can imagine people will get better at being selective and not getting suckered into poor climbs like sheep, but guess what happens when you do that and pass up the leeched thermal you can't connect with? You end up pushed out in front of the pack, probably lower. Sean Fidler made an interesting point about stealth and how it could interplay with human behavior. If you are intent on following someone you will need to do it much more closely with stealth mode on. You will need to be more diligent about exploring the start cylinder. This was hardly ever a big issue before Flarm - there's usually a limited number of good places to start so you ought to be able to find people just like before - maybe not quite as well, but the minnows will be glued to the big fish. The difference with stealth on will be closer proximity in following and fewer leeches kicked out in front of the gaggle to mark the next thermal. If they lay back and take the weak climb them may be able to avoid that fate, but their speed will suffer and they will get dropped by the faster pilots.

In any event, the data we have pretty consistently points to the paranoia about Flarm leeching being mostly that - paranoia. Every single datapoint seems to indicate that there are too many variables at play to make Flarm leeching a highly successful strategy and the forces at work would lead me to conclude that seeing even further ahead in time and distance - as you suggest with better computer and communications technology - will be information that you best ignore most of the time because most of the time conditions will be different by the time you get there.

I'm sure people will think up other ways Flarm at a distance can be useful (e.g. being saved from an outlanding - yup, could happen, but I think of that as a good thing as landouts are the biggest random variable in racing, or, e.g. finding the convergence line - yup, could happen, so local knowledge matters less - is that a skill?). All these anecdotes happen once in a great I am sure - it doesn't change the outcome at the top and I doubt it does much in the middle either. Over time seeing the rest of the field may show newer pilots where they made their mistakes when they go one way and another pilot goes another way, sort of the way SeeYou bug racing allows pilots to learn - except in the cockpit. I've watched a group of pilots 3 or 4 miles away on a different line make ground on me - it's too far to do much about most of the time and you never know when fortunes might be reversed so you store it away for next time.

If you give me a scenario of advantage I will try to find a way to detect it in micro or macro-level performance, but if you can't find pilots catching up and you can't find a difference in the expected scores at the end of the day, or week, then I am at a loss what purpose we are trying to serve other than emotion and paranoia.

Sorry for the long post on how I did my homework.

9B

On Wednesday, December 23, 2015 at 8:23:24 AM UTC-8, Dan Marotta wrote:
No agreed to scenario?



14 CFR 91.113*Right-of-way rules: Except water operations.



(e)*Approaching head-on.*When aircraft are
approaching each other head-on, or nearly so, each pilot of each
aircraft shall alter course to the right.



Some people are writing their own rules when the rules are already
in effect.* So if one of you turns to the left because that seems
the best course, and the other turns to the right because that's
what the FARs require, what happens?



Winter seems the time for everyone to come up with outrageous
scenarios.* And BTW, it's snowing here, too.




On 12/22/2015 5:15 PM, Andy Blackburn
wrote:



On Tuesday, December 22, 2015 at 2:14:18 PM UTC-8, wrote:



Almost everybody, when faced with the head on scenario will likely do what you mentioned. Sooooo- a better scenario is to push and no banking as it likely will increase your projected cross section.


Literally it's no different than if we just flipped coins at the start of the contest and divided into "pull" pilots and "push" pilots. Your odds are 50-50. If you could get everyone else to agree to push it would be good - for you.



Matching the other guys bank is best obviously.


I'm pooping my pants just thinking about that one! Thanks goodness it happens rarely.

Seriously, there is no agreed to procedure for head-to-head because there are too many scenarios where doing something by rote only makes matters worse. You should assume that in this situation you will not see anything until it is too late. The closer to collision course the less likely you will be able to pick it up. That's just the biology of the human eye. Best to not rely on panic maneuvers if you can avoid it.

9B





--

Dan, 5J

Correct Dan - I was talking about when you don't know exactly whether the glider is slightly right of your course or slightly left or straight and you have 10 seconds to figure out whether you have room to get to his left instead of turning blindly right into each other. It's easy if you have situational awareness, less easy if you only have an approximate indication and little time to make a visual ID that you have only a 50% chance of successfully doing. The Flarm guys don't recommend turning because of the long wings of gliders - everyone turning right can be the exact wring thing to do close-in. That's why we don't automatically say everyone turn right.

As Hank said, pushing and turning gently right is probably not a bad idea, unless too many guys read RAS, in which case you might want to pull and turn right...or left...depending. Probably right.

Bring my brown trousers...

9B