Jump to content

Kilrah

FPV Experimenter
  • Content count

    2,089
  • Joined

  • Last visited

Community Reputation

2 Neutral

About Kilrah

  • Rank
    RC-Cam Mentor

Contact Methods

  • Website URL
    http://www.andrebernet.ch

Profile Information

  • Location
    Dubai
  • Interests
    Modelling, electronics, computers

Recent Profile Visitors

1,352 profile views
  1. Kilrah

    Members only FPV forum

    Of course, what I mean is that when I see that I get the "That's what I want to do" feeling - then it really gets pleasant once I do. That's what we have on our "imagery" machines, stabilised gimbal but with AHI on the OSD. To me it doesn't work at all, it's like trying to fly FPV while also looking at the model, you end up not being able to follow either. Either I'm looking at the image and the AHI makes no sense to me as it's completely contradictory to what I see, or I concentrate on the AHI and then don't even see the image anymore. Maybe with a lot of training... There are some guys who fly H quads with a 2 axis gimbal in front, need to find a video again. AFAIK they don't use an OSD but simply fly in Attitude mode with altitude lock i.e. in the "shooting" style, and rely on the quad's attitude equalling their stick input. Optical flow does the rest to provide a (delayed) feedback that allows to fine-tune the trajectory. At least that way there are no contradictory visual cues, and that I can see myself doing even if it's not really my cup of tea. No way to fly that in manual/rate mode though.
  2. Kilrah

    Members only FPV forum

    An example I just received, apparently the French have announced a new parachute requirement for 2015 for the "involved" usage cases, that none of the existing systems satisfies. So if you want to build a machine for that use case now where a parachute is already required you'll have to shell out a good $1000 for your chute system, but that's only for 4 months as it will not be compliant anymore then. And then you'll have to stop operating until someone finds a way to build a system that satisfies the requirement...
  3. Kilrah

    Members only FPV forum

    Of course not, the Phantom is an awesome machine. And yes things will eventually settle for good I'm sure, but in the meantime it causes a lot of mess and uncertainty. DJI forcing things by putting millions of quads into all hands does both good and bad, good because it pushes for clarifying things, but bad because they've now started a 15-year period of negociations, regulation rounds, and general unstability in the field that nobody wants to have to go through and that was avoidable if people were a bit responsible. See the whole mess with the FAA lately, in one year UAV operations have gone through pretty much every status between OK and totally illegal, and it doesn't look like it's going to settle any time soon. So much for trying to live from what you like in that field when you just bought 20k of equipment and flying it is suddenly illegal again, and you have absolutely zero say. Then you have to do paperwork. Then it goes back again. Then you'll probably have to get a licence at some point... I.e. you end up with more overhead than actual work again. Not talking of the guys who start going out with guns wanting to shoot at multis because they've seen in the media that if they seen one it's probably spying on them. And not either talking of FAAs latest endeavour, yes, aiming at the whole RC hobby, that has just popped up. Where I am I can't say I care much nor see trouble trouble ahead now, but I definitely wouldn't want to have to deal with it the way you'll have to in the USA Absolute opposite to me, the view following the aircraft's attitude is to me 90% of the "feeling of flight" i.e. feeling of really being in the thing and going under that tree with movements that are tightly connected to my actions like in real life, as opposed to jsut looking at something happening on a screen. Then tilting and pitching is only as fast as you make it It's all a question of piloting. I absolutely hate those videos where they fly in Attitude mode i.e. angular speeds are controlled by the FC and are as fast as it can make them, giving highly "harsh" movements. But fly rate mode, with low throws, add a lot of anticipation and good trajectory prediction skills and you've got beautiful flights like the one below. Highly dynamic, free and interesting, yet very smooth. (not my video, but my favorite out there in this category) Same as on a plane really, when I FPV an airplane it's to feel in there. Wouldn't happen if I had no view of the aircraft and a fixed angle. When I fly with a stabilized gimbal I just "steer a camera" that could just as well be on a cable, mast or crane that happens not to have movement constraints. It's excellent too, just not the same thing. I like both, but for different purposes. One makes me fly, the other gives me a great image and perspective but it's just a camera angle. I'm looking at a pretty picture but not flying. (quite old now, before 3 axis BLGs) I hate rumours, but you'll find the known facts here FrSky in Chicago! http://www.opale-paramodels.com They do the chutes and mounting/release systems, not sure if they also sell the automatic trigger devices yet, I know they were looking into it. Have seen some in other places (some guys have asked about configuration for them on the Taranis threads). There are other options, often more expensive. DJI is apparently coming up with their own too I've read, but the descent speeds I saw were scary, hopefully they're wrong. EDIT: Hmm, forum seems to only parse 2 youtube links per post.
  4. Kilrah

    Members only FPV forum

    I've been following all the RCG discussions about the Lightbridge since when they started in January, and it was impressive to see the number of people who found the price outrageous. Several things to point: "Drone people" have no clue about the technology leap involved, how long we'd been hoping for such a thing etc, and often don't even understand the benefits involved They're seemingly incapable of simply looking up data and understanding that the closest existing alternative would cost 4x more. And it's already kind of a moot point as like said above they probably don't even know why they'd want that capability. Anything they feel they can't own because the price is too steep for them is crud and useless - out of frustration of course. Then once it started hitting people's workshops, it became obvious how much these "drone people" rely on precise directions and step by step instructions, can't grasp simple things like what the inputs and outputs of a system are and how you can interconnect those to other systems around it. They don't know how anything works nor what's available in the field, and as a result many buy the thing without understanding much then find no way to do what they want in the manual, and are lost even if all it takes to solve their situation is a simple standard off the shelf part. It's not part of the package and not offered by the supplier, so they're in a dead end. But things are far from being dark all around. The above shows that a significant number of people with no clue and hobby goals have bought a $1500 video transmission system to put on their Phantom. And then the LB seems to sell very well to professionals, those are the vast majority of customers, who have no problem using it, take it out of the box, wire it in and go fly, they don't spend their life saying how cool it is. On the forums you'll never hear of them, you only get the whiners... Same in every field. The market for good things is there, the only problem is that you can't see it, and what you do see is (wrongly) discouraging you from even starting. Just need to find a way to get over it. Can't wait to see how FrSky's next radio, targeted more to high end customers will do.
  5. Kilrah

    2011 FPV Product Predictions

    It's being done step by step with the big picture in mind, just takes time. And yes, the platform is perfectly adapted. The lua interpreter that was introduced in 2.0 is the perfect foundation for the configuration UIs, by having the ability to execute scripts that are basically 3rd party apps anybody can design an UI that is best adapted for a particular type of onboard device without needing any firmware changes. Now what was still much needed is a complete overhaul of the telemetry processing layer. It was initially hardcoded to the finite number and types of sensors of the D FrSky protocol, and was hastily patched for basic support of the new smart port system but without allowing to use its flexibility to the fullest. For example a flight controller will be able to send down attitude angles, there are no such identifiers existing in the FrSky system, so we'd have had to add them to the hardcoded list, which was everytime a complex process involving changes in many parts of the code. It also didn't support multiple sensors of the same type, which is stupid as it's possible with the hardware. That's completely being reworked right now for 2.1. Once that big thing is done, it will only take a few more finishing touches to get there. About which controller to support, we've been quite happy with the ARM-based multiwii-derived boards (Naze32 et al), and an OpenTX user is actually the main contributor of the "cleanflight" firmware that runs on them. He's been improving the original "baseflight" firmware, adding a few things like alternative telemetry protocols including smart port, so I guess the easiest will be to start there. Once this is done as example it should be easy for others to port code.
  6. Kilrah

    2011 FPV Product Predictions

    Now for 2015 what I want to see is a trend towards integration of all the various digital systems on a vehicle. We have an R/C system, with a radio that has a graphics screen and buttons We have onboard flight controllers or other "intelligent" airborne devices We recently got a telemetry downlink data channel, unfortunately usually independent of said devices that would have a ton of useful info to give (getting better though, there have been quite few "bridge" solutions developed this year to connect some flight controllers to some telemetry systems) Some manufacturers implemented the complementary uplink data channel, but it's still unused in most cases... And I still stupidly have to take my PC out anytime I want to change a setting on my FC. So I'd really want to get to a point where the FC can be connected to the RC system's bidirectional data channel, feed telemetry and status data in addition to what "normal" telemetry sensors supply, and with an UI on the radio that allows configuring anything that might need configuring on the FC without any external device. It's a bit stupid to still use a servo channel to set gyro gains remotely when everything's there to just send the digital value... Obviously this has to be done in the DIY / open source world to trigger wide adoption as big brand radio manufacturers will never bother supporting anything beyond their own offerings, and devices that would benefit most from this aren't made by them.
  7. Kilrah

    Members only FPV forum

    Hi all, nice to see you back here Why not. Some kind of "peer review" board where elaborate things could be discussed before public involvement... But IMO for general discussion away from the "noise" the public board here is already quiet enough. For FPV flying as its own goal, the 250-size mini quads and associated signature flying style that have developed tremendously in the past 1-2 years. It's what I currently find the closest to the pure flying freedom. As a rotorcraft they have all the related advantages, but also the capability of fast forward flight and great maneuverability, and if you want to fly them like an airplane you can too so they cover nearly every facet of flying. And the robustness makes anything forgiveable, so you can actually dare doing "the fun stuff". IMO it will in no time become a whole category of RC flying in itself, and is the first thing that will bring FPV flying (i.e. not as a purpose to look a the screen to frame a shot à la drone operator) to a larger group of people in a good light, and will push the fun aspect of FPV beyond a few isolated guys flying alone here and there. People are already starting organising "leagues" with different activities like star wars pod-style racing. Can be done in a controlled secured area, small enough to avoid the common RC and video range problems, is incredibly fun and can easily be done as a show for spectators. Brushless gimbals are awesome for the AP/AV side of things, incredible tool to create images which simply revolutionized the whole field but certainly not what you want to feel like flying when every movement is perfectly erased Depends... I seem to recall that in the old years pretty much everybody liked sharing ideas and getting some amount of feedback from the relatively small but technically inclined community on here, but then we'd still compile that together and build something to one's taste, not necessarily trying to make everybody happy. I still think this is a good way of doing it, when you have too many people it becomes RCGroups and just drowns in the middle of the noise. But as said there isn't much noise on here, so unless there's a larger project with clear business intentions that has to be protected or "sensitive" items to discuss then we can likely talk in the open forum. My company I'm not so much involved with anymore uses multicopters professionally, and current offerings just do all the job we need. Safety systems are there already if you want them, parachutes are available, independent release systems that will trigger deployment automatically in cases like angular rate or accelerations exceeding preset limits (i.e. flip) too. So HD downlinks, or digital SD ones with high reliability and penetration. I have a Lightbridge and a set of Sony 720p goggles, and I can tell you an FPV flight on a calm stable platform just looks awesome. For stuff like mini quads things go so fast that HD is probably useless though, so there something smaller, cheaper but that works better than the age-old analog would be welcome. Yep, that. I have 2 of those that I ordered some months ago in one more of these "let's get to it" moments that vanished after a week of tinkering when real life caught up and made it evident I'd never manage to secure enough time to dedicate to it. It blows the RPi in pieces in terms of performance, has a hardware video encoder, composite video input, "camera" video input that would accept an HDMI decoder, powerful CPU, countless interfaces etc. The chip is easily available alone for a dozen of $ in qtys of one should an optimized, miniaturized system be targeted, Etc. But the amount of work is huge and specialized and getting to anything decent would likely need a "real" skilled development team. We were a rather small bunch of people aiming to do something similar, at times where DIY was simply the only solution. It was all supposed to be fun, and the sharing was part of it. Nowadays as most of the stuff is readily available as you can see on RCG the discussions are not anymore about "what could we build to make it better", but "which of the 35 similar existing products should I buy". As a result I haven't visited the FPV section of RCG since 2008, and I've been doing a lot less DIY as the number of occurrences where I needed or wanted something that didn't exist AND building it was within my capabilities has quickly fallen down to about 0. Also incentives for developing something have dropped as the huge wave of new buyers don't recognise or know about quality, what to expect and what can be done. An example is the simple servo stretcher I made looong ago with a nice 1us resolution... Pretty much all of the Chinese stretchers you'll find in every FPV shop nowadays have a crappy resolution, but nobody complains. A couple of times a year I'd receive an email by someone asking if I still had some to sell as they had tried everything on the market but nothing was nice and smooth... but it's not by selling 10pc a year to those few people who recognise the difference that you'll make a living Haven't completely dropped the "DIY" thing though as I've been quite active on the OpenTX open source radio firmware project since I've dscovered it in 2012, with associated mods around the Turnigy 9X radio it was initially running on, and further developments that led to the FrSky Taranis radio shipping with it since last year. R/C radios were something I had always been frustrated with and revisiting that was one of the items on the list of things that were too much for me, so finding an already well shaped related project I could take part in, initially on the software side but then also a bit on the hardware side thanks to the great collaboration with FrSky was a perfect opportunity to bring something to that field.
  8. Kilrah

    2011 FPV Product Predictions

    This one would actually be a pass for me, at least partial because it might not always do as much as expected, but in the right direction... several manufacturers (Spektrum, Graupner come to mind) have released receivers with integrated 3-axis gyros for at least "wind shake" stabilisation. Then there are heli FBL controllers with integrated receiver that might actually be closer to the definition, even if they are more a "controller with integrated receiver" rather than "receiver with integrated controller". Oh and the DJI A2, even if haven't yet heard about anybody actually making use of the built-in FASST-compatible receiver. Only about 2.5 years late
  9. Kilrah

    Sky Drone HD FPV, Crowdfunding

    Got 4 of the DJI units on order, can't wait to get my hands on one
  10. Kilrah

    "FPV" is being trademarked!

    Hey, cool to hear from you again Yes probably too late for trademarking - it's already been used in the entire world for a good 6 years now, both casually and commercially and even became the standard in other languages... I think it's still possible to find challenges in niche fields... for me it's mostly the micro sized stuff. Very cool to fly a 9-cm quad in stereoscopic vision, not something everybody's doing or is capable of setting up regardless of available gear And of course just the fun of flying, but managing to make something that perfectly fits just that flying style you want to achieve at the moment. Even if it might seem easy, it isn't always. I currently have a mini H quad that's a real blast to fly around in just about any style you want while carrying an HD camera and being small enough to carry everywhere, but it's been since October that I'm still trying to change components and settings to get just the video steadiness that I want... Then the Oculus Rift are on the list, with better immersion than ever with an FOV that completely surrounds even your peripheral vision and will be interesting to try... still some fun things ahead
  11. Kilrah

    FPV in 3D vision

    Interesting, what are you using at the moment? single USB capture device, and software that transforms/duplicates the image for the Rift? You may have seen about this: http://emrlabs.ca/shop/index.php?main_page=product_info&cPath=3&products_id=5&zenid=uplon1ct5d0ifcdfd7729e97u6 Allows live 3D FPV with the Rift and other HDMI-based goggles or monitors. Been doing some tests lately (not with a Rift at this point) and getting great results. Advantage being no PC-related latency, "bluescreen" or reliability issues
  12. Kilrah

    Multirotors are dangerous?

    http://www.opale-paramodels.com/index.php/en/shop-opaleparamodels/18/rescue-systems/for-multirotors There are a few videos on there.
  13. Kilrah

    Multirotors are dangerous?

    There are parachute systems that work well on the market, pretty neat and lightweight too. Some use them. Prop guard is only good if you nicely come from the side where it is and not too hard not to break it. Won't be of much use anytime the machine is above you or when it falls out of the sky. I guess the take is why carry one and put up with it if it's only gonna potentially help in a minority of cases.
  14. Kilrah

    "FPV" is being trademarked!

    Pretty strange. I have had a few great exchanges with Val in the past - he's the one who got me hooked to FPV more than 10 years ago in the first place (even if it was already by the time his activity in the field had started to dwindle), and I'm still grateful for it. I still have the e-mails from 2003. I hope it's not the same as several other cases that I've witnessed of people who just go through bad days at some point, and instead of getting through them and/or talking to people to get some help become isolated and try to cling onto whatever they did in the past to try to value themselves. Val if you read this - don't waste your money on that (I did at some point, Denis did too), and rather come and have a chat, the field sure has changed a lot, but if you forget the few morons that eventually riddle any activity it's still as good as it gets, and a little bit of creativity like in the old days is still able to make things go forward in the good direction
  15. Voice announcements already are. Recognition... still not sure about that.
×