What the hell is Speed-to-Fly!? - BANDARRA

[@JanPBtest]

1:44 This is a great book but this chapter is incorrectly describing Reichmann's speed-to-fly. The actual method is to leave the thermal when its lift is equal to the anticipated lift at the bottom of the _NEXT_ thermal, NOT the _current_ one! (it's easy to draw a diagram demonstrating why the method described in the book doesn't work but it's hard to put it in words here.) Then we fly between the thermals with the airspeed dictated by the in-between air mass movement, as in the classic MacCready speed-to-fly. The specifics of that "dictation" are determined by the wing's polar curve. BTW, there is an interesting study by the Hungarian Academy of Sciences on the Internet somewhere: they tested how close migrating birds fly to the MacCready method. They did it by comparing their flight patterns with those of XC paraglider pilots who were asked by the researchers to follow the MacCready speed-to-fly. The graphs comparing the two matched almost exactly. Obviously birds don't use any calculations, they do it by memorising certain air patterns through _tons_ of practice :-)

[@MikeKing]

Impressive Andre, you did a great job building that little game/simulator in short time with unity. Pretty cool demonstration as well.

[@ObyrneJohn]

Great work! It would be great to highlight the keys that are pressed (or show when a user is using speedbar or breaks) and have a mode where you can record the initial cloud setup, game keystrokes and give it a name to highlight different scenarios where speed bar is better VS worse for example. These could be used as presets shipped with the game and would be watchable. It would help a lot newbies (like me) understand all the different scenarios and best practices of paragliding.

[@AndreBandarra1]

Cheers John, some good ideas there! You can tell what flight regime the player is on by looking at the speeds on the top left and right corners but yeah I agree that the keys lighting up would make it easier/better. About pre-set scenarios I think that's really cool and would be useful for sure. I've just started this game dev journey so will have to look into that. Thanks a lot!

[@PorLasAlturas]

My pleasure!😚 You're doing a great job Andre and you're learning so fast🤓...I'm pretty sure this little project you're starting is gonna flight very high!🪂 Cheers!

[@AndreBandarra1]

Cheers dude, thanks for all the help! 🤗

[@samtolton]

This is a great idea and the visual aspect explains the speed to fly dilemma far better than I've seen it before. One thing that I would add is a way to specify how strong the thermals will be (perhaps min and max settings?) so that you can also look at which thermals you should take to the top and which you should only do a few 360s in.

[@AndreBandarra1]

Good idea Sam, I’ll definitely consider it for the next version!

[@henriquecarcereri8369]

Wow!!! Very nice!! Thanks for showing your job! Cheers from Brazil

[@AndreBandarra1]

Obrigado! Valeu

[@larsbojohansen5117]

Well done - Andre! Fun simulator 🙂

[@FH-cj6ev]

Awesome demonstration!!! It visually explains the concept really well. Maybe I finally got it now …

[@ParaKiteGliding]

I tried the game on your page.... funny challenge for a moment to try to thermal both gliders solo by keyboard. 😂

[@AndreBandarra1]

ahah, yes! sometimes Gemma helped me but often I would test the game on my own and it's a tricky thing to control both players well at the same time :D

[@markmcgoveran6811]

I'm pretty certain that the speed to fly thing is very speculative and it's absolutely clear what you should have done after you lost. Considering that you can't see thermals and there's no mathematical rating of a very much it would be very difficult to calculate which is the best strategy ahead of time

[@AndreBandarra1]

The speed to fly algorithm (or MacCready) has been used for almost 100 years and has a proven track record. However you use it constantly on the go and adjust, rather than take a random guess at the start and stick with it. The more thermals you take the more data points you have and the better your guess is. en.m.wikipedia.org/wiki/Speed_to_fly

[@markmcgoveran6811]

@@AndreBandarra1 this is one of the problems we have as human beings it's a quantitative question. If we had all the information the mccready approach would win every time. Unfortunately we can't get all the information. The more data points you have the better the "guess" you can make. If you had all the data points you have a certainty.

[@instrumentenfreak]

Looks very good to visualise that speed bar is not always better. Multiplayer would be fun xD On my computer it runs to fast. It there a way to slow it down?

[@AndreBandarra1]

I think I fixed it! Just need to fix the bug that you sometimes get stuck at the beginning

[@AndreBandarra1]

Fixed that too, having a good day! :D

[@grejen711]

Hmm.. would this not depend on finding thermals that are actually in line with the intended destination? What if there's a closer but weaker thermal slightly off course. Or what if that stronger but further thermal is quite far off course. .. It's 3D. I suppose it would also depend on your glider's efficiency drop on speedbar vs it's thermalling effectiveness

[@AndreBandarra1]

Absolutely, but understanding it in 2D really helps with 3D after. And once you understand 3D you understand 4D a little better, because it also varies with time. That strong thermal you see and fly towards might not be there by the time you get there. They call it sky chess right? That’s probably an understatement 😅 it’s fun though!

[@hunzaskywalkParagliding123]

Can you please send me a link where I can download this book.

[@AndreBandarra1]

xcmag.com/shop/product/fifty-ways-to-fly-better-bruce-goldsmith/?v=7516fd43adaa

[@fwkb2]

Love it! What a great idea for a project to learn Unity. I too would be interested in taking a look at the code if you do decide to open source it. Maybe I could even contribute some code. I do c# for work and have done just a handful of Unity tutorials. Have you seen the Find the Core game on Android? It's a simple thermal coring game. It is also 2D but on the top down view plane rather than the side scroller plane. I wonder if there would be a way to add something like that to your game, making it so there's a second skill component there, if you know what I mean. You'd need to both core a thermal well and know speed to fly.

[@AndreBandarra1]

Cheers Mark, we'll see what the next version entails but yeah it would be great if I could somehow integrate community code too. I haven't seen that app as I don't have Android but I suspect from the release date that they might have been inspired by this one kzbin.info/www/bejne/bX7aaJylqNRsj6c

[@AndreBandarra1]

By the way, I didn't mention it on the video but we do have a discord server for all things Paragliding Sims including this project :) discord.gg/Xa5EjJPD

[@fwkb2]

kzbin.info/www/bejne/hpOUloCtopyZiZo ... but Discord tells me the invitation is expired

[@fwkb2]

Ah, never mind, it worked after Discord updates. I'm in!

[@crislighter]

nice work andre! is it maybe possible to change the controlls to W,S,A,D? as far as i know its the same on all keyboards. S,X,Z,C is a pain to play on german keyboard 🙈

[@AndreBandarra1]

You're right Cris, I should do that. It was an oversight on my part wanting them to be on the same level as the arrows, but it works a lot worse than WASD lol

[@AndreBandarra1]

Done. Enjoy not getting finger cramps! :D

[@aphlyp]

Nice work! Sailplane pilot here whose been following your videos for a few years. You should check out G Dale's "The Soaring Engine" series of soaring books. They're targeted toward sailplane pilots, but I'm confident many of the same XC soaring principles can be applied to paragliders as well, including speed to fly techniques.

[@AndreBandarra1]

Thanks for the suggestion, I’ll have a look!

[@martincarter4114]

Dude, your just normal n I love it, REALLY enjoy all you vids (both of you), understandable for a normal bloke like meee 🙂 whoop 🙌

[@GodzillaGoesGaga]

Very cool!

[@carlosleon7683]

Exciting news, Andre. I have so many questions I don't even know where to start 😂 1. Are you planning on releasing the source code of the game? a.k.a. make it open source (there are licenses that allow you to open open source while keeping the copy right, right to monetize, etc.) If you make it open source you increase the chance of getting volunteers to contribute to whatever aspect of the game development they'd be most comfortable with. 2. Have you given it a thought about the architecture of the game? I can imagine right now it's very much a PoC but moving forward it would be great to structure it in such a way that core concepts are developed separately and then put together to make the game come alive. For example - Each aircraft and its physics should be defined separately. In that way the game can start with paragliding and eventually extend to any other soaring aircraft (hang gliders, sail gliders, rc gliders, etc.) that defines its own look & feel and behaviors like input controls, stall behavior, etc. Same for the thermal engine. Right now you have already done the rough basics with lift and sink. You could have a thermal engine that based on whatever parameters (combination of user input + randomness) it could generate a day very much like they do in Condor Soaring Simulator (massive simulator in the sail gliding world with world cups happening every year even), with thermals placed at random, mechanical turbulence, cloud street formation, etc. Quite a big beast but if you set the foundation from the beginning you don't have to develop anything until way down the line :) 3. Online playing sounds like a lot of fun. As I mentioned in the previous point, Condor hosts competitions all the time. There is an option in the game to either join an existing server or spinning up your own to host your private competition with your mates. Not sure what your thoughts are on this but I wanted to share what I already know. 4. Do you see in the future community-contributed maps? I can imagine people familiar with a specific area would like to contribute to make the maps more accurate. This of course imposes a challenge when it comes to storage (where do you host the maps for people to download? storage in the cloud is expensive). I'm going to stop now because I can go forever hahaha.

[@AndreBandarra1]

Hey Carlos, thanks for the enthusiasm, that's great! 1. Maybe, I've thought about it and it's definitely a possibility for the future. 2. Yes, I've already started impletementing a completely different architecture to be able to do multiplayer and like you say have different independent modules to make it easier to build more stuff on top of it and expand the sim. 3. Yes this is pretty much my plan for the future. I think Multiplayer competition is where all the fun is. 4. Yes I think in the future once concepts and basic architechture and functionality is in place it would be great to have community contributing with many different types of content, not just maps

[@fwkb2]

Condor is a lot of fun! Just wish we could fly paragliders in it :)

[@AndreBandarra1]

@@fwkb2 One day :)

[@Hemersonr]

Great video as always... Eu penso que speed to fly puro não funciona tão bem para parapentes como funciona para planadores... mas a proposta de voar otimizado é bem interessante (a segunda ilustração). Gostei do game vou testar... ótimo para mostrar para colegas e clientes

[@AngeloMDias-ny8gw]

YOU ARE THE BEST.

[@AndreBandarra1]

Espero que seja útil Angelo!

[@marekppg4638]

Nice jobs👏

[@aslamlsayyed]

Wow so you are also a developer

[@AndreBandarra1]

ahah no, just playing with it

[@AntonMadness]

Ha really nice! Next level calculations if you can get the PC calculate optimal performance so you can see how you did against a perfect run :-) Oh software.... the danger of endless possibilities :-)

[@AngeloMDias-ny8gw]

We are waiting for tou in RJ.