8flight EFB - what am I missing here?

midlifeflyer said:
dunno. i agree we’ll see a lot, but I see Starlink as more of a passenger convenience than as an in-flight tool. Closer to music in the airplane than ADS-B weather and traffic.
Click to expand...
I find utility with a Starlink internet connection with the ability of making brief flight phone calls or texts, filing/amending a flight plan in air for an upcoming leg after a lunch/fuel stop and pulling up regional radar that has improved latency compared to Nexrad.
 
robin ardoin said:
I find utility with a Starlink internet connection with the ability of making brief flight phone calls or texts, filing/amending a flight plan in air for an upcoming leg after a lunch/fuel stop and pulling up regional radar that has improved latency compared to Nexrad.
Click to expand...

As a T-Mobile user who has landed late and been caught without service (and resultant 3 mile walk to a hotel in rain), it's now essential travel gear for me. :D
 
TCABM said:
I thought you IT guys were smarter than that.
Click to expand...

Laziness is my reigning trait :D I have wanted to leave T since Legere stepped down, but I haven't figured out where to go -- replacing two phones and a 4G internet receiver that works inside my hangar. (and yeah, starlink doesn't work from inside the hangar, so this is quickly becoming joke-on-me) Strangely I live in a spot where Verizon is like 2 bars, so no idea where to head anymore. All carriers seem to suck, and I don't feel like weeding through fly-by-night MVNOs.
 
Is it still locked to the iCrap ecosystem? If not, I'll give it a shot. Otherwise, I'm sticking with iFly which is better than ForeFlight anyway (and cheaper).
 
TCABM said:
I would not characterize the proposed service as ‘move fast and break things’. What I will say is based on the public facing content says this service provides likely gives an incomplete weather picture from a strategic perspective - what is the weather doing over time and space - while also being marketed as a tactical decision making tool - where’s that hole I can punch thru?

That’s kind of a pandoras box in my opinion especially given the number of pilots that already do things in weather that an onboard radar would tell them not to do. If you convince these pilots that this is a better mousetrap, it may provide disastrous results.

Then again, that’s my lay opinion, which is why I really would like an actual weather expert’s thoughts on the tool.

That it’s reliant on a starlink connection hints at targeted a market user with the desire to put one of those things in their cockpit leading to watching cat videos on youtube instead of looking for traffic.
Click to expand...
It’s certainly not a replacement for onboard weather radar, and we would never position it that way. Particularly in the mountain west there are plenty of areas with poor or no NEXRAD coverage too.

But it certainly makes for much better decision making than a 10 or 15 minute old SiriusXM view of the weather. One of the most useful applications I’ve found though is before takeoff — I’ve seen storms, particularly in the Midwest, grow by 40 miles in the time other EFBs update radar data, so you could look and see that the weather is 50 miles away and no threat when now you’d see it’s actually 5 miles away; or has that storm really passed or is there another cell brewing behind the one that just went through.
GeorgeC said:
I also would be interested in @scottd 's opinion.

I thought the postprocessing that happens when the national radar mosaic is assembled does various normalization and clutter removal steps; seems that skipping those to trade for lower latency / incremental updates / finer granularity could introduce artifacts at the edges and/or mix images from different elevations.

I always thought it would be better to disseminate state vectors that could then be propagated from the valid time to the present, e.g. with a Kalman filter, but again, we're not there yet, and that would probably be more of academic than operational value anyway.
Click to expand...
Nationwide mosaics like MRMS or those from other weather providers like IBM/The Weather Company mainly focus on combining a variety of inputs to produce the most accurate 3D picture of the weather as it was several minutes ago (some filtering of non-precipitation returns, precip type, composite data if you’re into that, etc). We do the same thing, but our focus is on doing it in realtime and primarily in 2D since that’s what pilots are used to with datalink weather. So we use proprietary processing techniques to build and add to that picture with each radar sweep instead of taking all of the data from say the last 15 minutes and looking at it in one giant step. And there are some things in there, like eliminating returns from wind turbines that we seem to do much better than other providers. There are other things like returns at a given altitude that we don’t care about as much, since again it’s 2D. But the key is that we show you where the weather at altitudes you care about is 1-2 minutes ago at most, not 15 minutes ago.

In a way we do also have those state vectors — you can see the model data from the run say 30 minutes ago and what it predicts the weather is going to be doing now or in 15 minutes or an hour from now.
robin ardoin said:
Perhaps the 30 second period is the app data refresh rate and uploads the most current data transmitted from the NWS/NEXRAD. Akin to listening to the ATIS every 30 seconds to hear the same recording until it a new recording is issued. I also fail to see how it can refresh faster than a 360deg scan at multiple tilt angles can be acquired, processed and broadcast. I gleaned from BT that Matt claims to be using regional radar raw data which may be able to be delivered faster but it would only be one or two tilts in a 30 second period. ???
Click to expand...
New data comes in from each radar every 60 degrees of rotation, which is about every 3-4 seconds in precipitation modes. And we combine all that data individually in the app in realtime instead of waiting for some complete scan or doing the entire country at once (adding minutes of processing time instead of seconds, on top of using data that’s maybe 15 minutes old already). Unless you’re a meteorologist analyzing the 3D development of a storm (in which case you’re not using a composite anyway but each level individually), you don’t really care about anything but the lowest couple tilt angles, which is why radars in active precipitation modes also revisit the lowest tilt multiple times to better monitor the position of a storm. So we get the lowest tilt about every 2 minutes from a given radar site when it matters, and in most of the US there is enough overlap that coverage of a given location is even better than that. If the data is older than 2 minutes we’ll indicate that site as having potentially old data (a gray shading over the potentially impacted coverage area), so there are no “maybe”s.
 
robin ardoin said:
@mattguthmiller thanks for clarifying that, Matt.
Click to expand...
Of course!

And I guess I should also add that processing things in that manner is critical to that last bit too -- we can account for the time of each piece of data in the mosaic. So we'll show a timestamp indicating the most recent data received but also highlight any part of the mosaic that's older than 2 minutes (if that data's from an active precip VCP scan -- in clear air mode where sweeps of the lowest tilts might be 10 minutes apart, we don't show that until then). Whereas other mosaics will give you a time that that mosaic was created, even if some of the data in it was 30 minutes old at that time and other data was 30 seconds old. So we've even considered for example changing the timestamp as you zoom around the map, so that you know how old just the piece you're looking at right now is. We'll probably do something like that as we add radar sources from the rest of the world where they either aren't available in realtime or don't scan as frequently.
mattguthmiller said:
If the data is older than 2 minutes we’ll indicate that site as having potentially old data (a gray shading over the potentially impacted coverage area), so there are no “maybe”s.
Click to expand...
 
Is Starlink actually required? Just curious; in my open cockpit world I never fly so high that I lose cell service, and I've used Avare's live Nexrad data via cell internet to navigate around a line of storms (staying far enough away to always have an out and being aware of the data age, of course).
 
SkyHog said:
Is it still locked to the iCrap ecosystem? If not, I'll give it a shot. Otherwise, I'm sticking with iFly which is better than ForeFlight anyway (and cheaper).
Click to expand...
Matt mentioned if 3,000 android users subscribe, he’ll build 8Flight for android users. It would take roughly 6 months to develop and launch
 
BeechSundowner said:
Matt mentioned if 3,000 android users subscribe, he’ll build 8Flight for android users. It would take roughly 6 months to develop and launch
Click to expand...
Yep, subscribe when it's not compatible and if and when there are 3,000 who do...
 
FPK1 said:
8Flight has no Garmin panel interfaces so it's a deal breaker for me...
Click to expand...
Yep, that's the problem with ifly as well, and the reason I'm a GP subscriber. I don't know enough about it to understand why FF is the only other EFB that can talk to them. Maybe because FF pre-existed GP and they had an agreement in place, or maybe because FF is the 800lb gorilla. Either way, if garmin locked FF out of their system like everybody else... well, that would be interesting.
 
Jim K said:
Yep, that's the problem with ifly as well, and the reason I'm a GP subscriber. I don't know enough about it to understand why FF is the only other EFB that can talk to them. Maybe because FF pre-existed GP and they had an agreement in place, or maybe because FF is the 800lb gorilla. Either way, if garmin locked FF out of their system like everybody else... well, that would be interesting.
Click to expand...

Fltplan Go does as wella and is free.
 
Jim K said:
I don't know enough about it to understand why FF is the only other EFB that can talk to them.
Click to expand...
I’ve found no public developer resources that say how you can connect to Garmin avionics. They also have never responded to any emails I’ve sent asking if that documentation even exists.

My guess is it’s part liability, part market share.
 
RedundancyTanner said:
I’ve found no public developer resources that say how you can connect to Garmin avionics. They also have never responded to any emails I’ve sent asking if that documentation even exists.

My guess is it’s part liability, part market share.
Click to expand...
ifly said several years ago they had it figured out and would be releasing the feature "very soon". Waited a year. Never happened. Still hasn't...4...5? years on. I take that to mean it's not so simple as one would think.
 
Jim K said:
ifly said several years ago they had it figured out and would be releasing the feature "very soon". Waited a year. Never happened. Still hasn't...4...5? years on. I take that to mean it's not so simple as one would think.
Click to expand...
Technical ability isn't the same as legal liability.

I'd be inclined to be believe they did have it working, because it's not that hard to poke around a bluetooth connection and make a handshake that starts the flow of data. If something bad were to happen, Garmin would shift the responsibility to ifly accessing the device without authorized as the cause. But even before that, it's impossible to know if Garmin sent a cease and desist blocking them from releasing, or if their own legal counsel advised against it.

Same argument can be made for the sentry too.
 
mattguthmiller said:
So we use proprietary processing techniques to build and add to that picture with each radar sweep instead of taking all of the data from say the last 15 minutes and looking at it in one giant step.
Click to expand...
From a quick search I've found NEXRAD has real time data available through AWS. If that's what you're using, why not just say that?

NEXRAD on AWS - Registry of Open Data on AWS

registry.opendata.aws registry.opendata.aws
 
RedundancyTanner said:
From a quick search I've found NEXRAD has real time data available through AWS. If that's what you're using, why not just say that?

NEXRAD on AWS - Registry of Open Data on AWS

registry.opendata.aws registry.opendata.aws
Click to expand...
I’m not sure which part you’re referring to (that’s just the raw chunks, none of the processing etc), but we don’t. We have a dedicated connection. The AWS bucket is best effort only and occasionally delayed by several minutes or down completely.
 
Jim K said:
Yep, that's the problem with ifly as well, and the reason I'm a GP subscriber. I don't know enough about it to understand why FF is the only other EFB that can talk to them. Maybe because FF pre-existed GP and they had an agreement in place, or maybe because FF is the 800lb gorilla. Either way, if garmin locked FF out of their system like everybody else... well, that would be interesting.
Click to expand...
It only takes a simple “yes” from them to allow it like every other avionics manufacturer out there. For FF it seems to just be that enough customers demanded it, hence my request here.
 
Jim K said:
Yep, that's the problem with ifly as well, and the reason I'm a GP subscriber. I don't know enough about it to understand why FF is the only other EFB that can talk to them. Maybe because FF pre-existed GP and they had an agreement in place, or maybe because FF is the 800lb gorilla. Either way, if garmin locked FF out of their system like everybody else... well, that would be interesting.
Click to expand...
Garmin would have loved to force everyone to use Garmin Pilot by making their panel mount avionics incompatible with everything else. However, they quickly realized that they'd likely lose more money from lost avionics sales than they'd gain in new GP subscribers. And yes, they did try this very briefly at the beginning of the GTN series.

ForeFlight had enough market share even early on that Garmin was essentially forced to work with them. I would not expect Garmin hardware to work with anything other than Garmin owned products (GP, FPG) and ForeFlight, ever.
 
mattguthmiller said:
It only takes a simple “yes” from them to allow it like every other avionics manufacturer out there. For FF it seems to just be that enough customers demanded it, hence my request here.
Click to expand...
It's probably market share. GP will compete with FF and maybe Garmin even has some sort of license deal with FF, but Garmin likely doesn't want any more competition...?
 
I'm using Avare with NEXRAD. It's all free, but I sent Avare a nice bonus check in Dec. As for freshness, well - this is what I found on the noaa.gov about the NEXRAD data sweeps. I admit, I don't know how often the raw data is processed and then published. I don't even know if NEXRAD knows that. Whatever it is, I don't go close enough or fast enough that I would run into something that would affect me before getting an update, except of course in cases where the receiver has not gotten an update for a while. There is no flag or warning about the current dataset on the panel that I've found.

Volume Coverage Pattern 112 (New in 2020, with WSR-88D Build 19.0)​

This VCP implements the multi-pulse repetition frequency detection algorithm (MPDA), which helps mitigate range/velocity aliasing (the Doppler Dilemma). This VCP has the same elevation angles (14) as VCP 12, but completes 20 azimuthal scans in under eight minutes. It also employs the Sachidananda-Zrnic Algorithm on the lowest elevation angles.

Volume Coverage Pattern Supplemental Features​

Many of these VCPs may also include scanning features such as AVSET, Base Tilt, and SAILS or MRLE. Automated Volume Scan Evaluation and Termination (AVSET) concludes a VCP at angles 6.2 degrees or above based on weather coverage at high elevations. Base Tilt adds an additional elevation scan lower than 0.5 degrees at select radar sites. Supplemental Adaptive Intra-Volume Low-Level Scan (SAILS) repeats the lowest elevation angle up to 3 times during a VCP, while Mid-Volume Rescan of Low-Level Elevations (MRLE) repeats up to the lowest 4 elevation angles during a VCP.
 
mattguthmiller said:
It only takes a simple “yes” from them to allow it like every other avionics manufacturer out there. For FF it seems to just be that enough customers demanded it, hence my request here.
Click to expand...
I have a Garmin GTx 345 transponder, I've kept my Garmin Pilot subscription because I get ADS-b in directly from that to GP, I also like GP's logbook and currency tracking. My nav/com is an Avidyne IFD440, also use the IFD 100 app, I get ADS-b in on those, which will feed into my FlyQ app I still use because I have a "lifetime" subscription. I'm trying out 8 Flight, but not seeing any ADS-b coming in. Will it get ADS-b from my IFD440?
 
GBSoren said:
I have a Garmin GTx 345 transponder, I've kept my Garmin Pilot subscription because I get ADS-b in directly from that to GP, I also like GP's logbook and currency tracking. My nav/com is an Avidyne IFD440, also use the IFD 100 app, I get ADS-b in on those, which will feed into my FlyQ app I still use because I have a "lifetime" subscription. I'm trying out 8 Flight, but not seeing any ADS-b coming in. Will it get ADS-b from my IFD440?
Click to expand...
We have support for Avidyne and Dynon coming in a couple weeks.
 
You must log in or register to reply here.
Back
Top