Redge

Are there any specific settings I need to change to maximise the traffic in the simulator, or can I safely ignore the AI traffic settings tab?

IFR flight in SI has received a big upgrade in the past week, with approaches and vectoring having been the main focus. I've hardly had any time to use MSFS this past month, but I crammed in four IFR flights in the last week to try and make the most of my SI subscription. Two of them were flawless, one had a minor vectoring issue (which was easily resolved when I mentioned the mistake to the controller) and one of them was giving me completely incorrect vectors. However, since I last flew, a new bunch of updates have been released to attempt to fix a lot of the remaining issues with vectoring, so it might already be much better. One of the biggest things I noticed was that the ATC response times have really dropped. I don't think I experienced a single response that took over 10 seconds, and several of them were far quicker than that (maybe around 6-7?). The response delays that were a problem for many only a few weeks ago really seem to be getting close to being resolved.

@JYW - you're right, that was the incorrect snippet. Sorry! @steelrfan85 - I applied the fix a long time ago, so had forgotten what I actually did. But I've found the solution: I use vJoy to assign the following key commands (each to a different vJoy "button"): CONDITION LEVEL 1 CUT OFF - vJoy Button #2 CONDITION LEVER 1 LOW IDLE - vJoy Button #3 CONDITION LEVER 1 HIGH IDLE - vJoy Button #4 In Spad, I then have the following logic (in pseudo code): If MIXTURE_AXIS <= 25 (SET LVAR:CONDITION_LEVER_CUTOFF_1 to 0) && (SEND SET_FUEL_VALVE_ENG1 Event 0) && (Press & Release vJoy Button #2) If MIXTURE_AXIS >= 26 && MIXTURE_AXIS <=74 (SET LVAR:CONDITION_LEVER_CUTOFF_1 to 1) && (SEND SET_FUEL_VALVE_ENG1 Event 1) && (Press & Release vJoy Button #3) If MIXTURE_AXIS >=74 (Press & Release vJoy Button #4) I hope this helps!

I use Spad.next and had the same issue. I believe the solution I implemented is this one below (from a user called Ape42, who posted the solution on a forum somewhere, but I don't actually know where it was... it was a partial screenshot I'd saved to my phone many months ago): Sorry! @JYW was right... This is the incorrect snippet. See my solution a few posts down:

Because you have to start somewhere? Because a business might not have the funding to develop a complete product before they launch, and therefore needs to take willing guinea pigs along for the ride? Because AI generated responses and voice generation are expensive, but getting cheaper over time? Because some people want to fly, experiencing the challenge of interacting with voice ATC, on any route they please? Because the product is developing very quickly, and paying customers know that the features they want (such as injected AI aircraft) will arrive in the near future —Q2 release hinted — and are willing to provide feedback and support to make that happen?

FlightFX's VisionJet will become part of the base sim for MSFS 2024.

The demo accounts, while they say are for 24 hours, can actually be used multiple times. They will unlock them about once a month again, for an extra 24 hours so that you can test the improvements again later on. Might be a reason to give it a go sooner?

I've been using it for over a month for both IFR and VFR flight (mostly IFR) and I've never been cleared to higher altitudes when on a descent. Now in default MSFS ATC however.... 😅

It’s been a pain point for years. I guess the thinking goes that the customer’s already paid by that point, so why bother spending time on the hosting. I absolutely love their planes - was flying the 414 again today - but their website (websites? They’ve got several different versions running on the same host) are a bit of a train wreck.

Yes it is. A long time ago it wasn’t and I recall the developer being unconvinced how needed it was but I think the community changed his mind. I was just flying through some choppy weather and enjoyed seeing the yoke react while on AP. It’s the little things, right? 😄

One thing I hadn’t thought about before when comparing SayIntentions and BATC was the nature of the AI implementation. I was on the SI discord earlier today when it was mentioned that the controllers on BATC don’t appear to be true AI, whereas those in SI are. And by that, they meant that the responses you get from BATC will be scripted, but using AI generated voices, whereas those in SayIntentions are using an AI engine to generate the actual response, separate from the voice, so they are far more flexible. I don’t know if that’s true? (Or if anyone at this stage knows for sure what BATC’s implementation will be like). But it got me thinking about my experiences with SI over the weekend, and what it was that made it so great was the intelligent responses to my issues such as my missed approach, the diversion or the mechanical failure (feel free to see my massively-too-long post about it in this forum). I had another flight yesterday where I couldn’t clear the icing from my C414, so requested a lower altitude. The controller followed it all up after a couple of clearances to descend by asking me to report if the icing had cleared at the lower altitude. if BATC can’t do this kind of stuff, then I’ll very happily keep paying the SI subscription so that I can have ATC that does.

Yes, with only the free voices, it's a one-time payment, in which case you can apparently get unlimited AI ATC at that initial price. The carrot/stick is in the free credits. I think the idea is that once you try the basic/advanced voices, you won't want to go back. BATC I believe won't ship with VFR on day one (unless they've since changed their minds on that?). They've opted for a slightly different route (IFR and traffic injection from the start, and add VFR later), whereas SayIntentions went with VFR and IFR initially, with their own traffic injection coming later on. IFR departures via the SID / arrivals via the STAR are already a thing, and vectoring has also been added on departure (when appropriate) as well as when picking up IFR in the air, but there's still more work to do on that front. I believe SI's initial expectations were that IFR would be announced in the summer and rolled out then around Expo time, but its proved to be a lot quicker (and simpler than VFR) to implement, so it's been brought forward to right now instead. I guess traffic injection (for which they apparently already have a working prototype in-house) will then be worked on, once IFR is sorted, a few months ahead of schedule. I honestly love what both teams are doing, and I'm sure if the feedback is good for BATC I'll also give that a go (Brian is super supportive of BATC as well as the other online ATC services, and genuinely loves the technology, no matter who is implementing it, and what it enables us to do now in MSFS). In the meantime though, there's only one gig in town and my experience so far is so much more impressive than I had expected from just the release videos. The demo is there for anyone to sample it all before they have to spend a penny.

OK, well then I mean MSFS' interpretation of the data that they're getting — snow aside. But that's getting beyond the main point I was trying to make in this thread.

It’s definitely usable. It’s not up to PilotEdge standards yet (that’s the only comparison I have I’m afraid - I’ve never really used Vatsim) but it’ll handle a flight from A to B with a few quirks here and there. The IFR logic is really new - literally a week or two old - so there are definitely bugs, but the change log updates can be measured in hours, not days. Late one night last week I experienced a bug while trying the new mid-air IFR clearance pickup feature. I headed to the discord and mentioned it. Brian (the lead dev) immediately asked me to show him. The next morning I checked in and saw the bug fix had already been pushed out in the next release. I’d say go ahead and try it. Aside from the 24 hour limit, you get to sample the full product, and if you don’t think it’s ready you can try again after a month to sample the improvements. These guys (Brian and the whole team) just want people to experience it, so they’re pretty flexible about everything, and are really receptive to people, especially those who help test and try new features to help them make it a better product. The whole community there is really positive. I’ll stop there as I think I’m starting to sound a bit cult like 😂.

I hear you! When not flying in the US I spend much of my time flying over Norway. It’s not perfect, and you can definitely see where Meteoblue data shines (and where it doesn’t), but it is getting better as time goes by. The US visibility improvements (which might be universal, I just haven’t checked) are something that wasn’t announced afaik, but is a really welcome upgrade.

I really recommend it! I probably sound like a paid promoter, but I promise I'm not 😄 — I just love how much the product adds to the experience... One more anecdote from the flight that I forgot to mention (with a bit of context)... The service is single-player (mostly). But if you choose to, you can hear other pilots (and let other pilots hear you) so that it adds a lot of chatter to the experience, to make everything more immersive. Over the weekend the added a bunch more voices to the service (I think there's 50+ at the moment) and one of the new US controllers has a a pretty sleepy voice (someone described him as being the kind of voice you'd hear from a controller about to sign off his shift!). While I was flying, I heard another pilot interacting with him, and the (real) pilot added something a long the lines of, "...and don't feel so bad buddy" at the end of his transmission. The ATC responded with the correct ATC response followed by, "...Oh don't worry - I love my job buddy". 😆

This weekend I had probably the most immersive flight simulation experience I've ever had, with some surprises along the way. A combination of factors contributed to it — fantastic (i.e. terrible) real-world weather conditions, a magnificent add-on aircraft (The Blacksquare TBM 850) and an insanely immersive air traffic control experience (SayIntentions.ai) that pulled it all together. Over the weekend I found myself with a bit of spare time in the evening for a quick flight. Having paid close attention to the new IFR features on SayIntentions.ai ATC service (which I've been subscribed to for the last couple of weeks), I decided to try them out on a quick hop from KHVN (New Haven, CT) to KCDW (Essex County, NJ), which should have been a 25 minute flight. The weather on Saturday around New York was terrible. IFR or Low IFR conditions, ceilings at or below 200ft everywhere and strong easterly winds. Light rain was reported around KHVN, but the observations at KCDW were showing a higher ceiling of 400ft with heavy rain. The Foreflight radar indicated that the conditions were likely to improve within the hour at Essex County. I opened Simbrief and planned my IFR flight. I'd planned to fly via CMK V623 SAX. I'd be IMC the whole way, with moderate icing expected from 6000ft all the way to my cruise altitude of FL180. SayIntentions (SI) now has the ability to read your latest flight directly from Simbrief, so as soon as I'd generated the flight, it was 'filed' on SI. I powered up the radio in the TBM to get the ATIS, and then requested my IFR clearance: The controller had my flight plan and provided my CRAFT clearance as well as letting me know the expected departure runway — a nice added touch since I'd last tried the interim IFR experience. I got the plane started, requested taxi (and received correct taxi instructions!) and headed to the runway for takeoff. Tower dutifully cleared me for takeoff (after an instruction to line-up and wait). I was handed over to departure shortly afterwards and was into the soup at around 500ft — I wouldn't see the ground again (or anything else, for that matter) until just before my final approach. I reached my initial altitude of 2000ft and cleared to climb directly to my cruise altitude — a lucky break for the skies near New York (and something I believe they are fine-tuning on the SI side. I've had other IFR flights where I'm cleared to much higher altitudes in multiple steps). I was handed over to another controller (Boston Center), all the while monitored my wings for icing and the radar scope for precipitation on the way up to my cruise altitude. The radar returns were showing oranges and reds ahead of me (I've rarely seen such heavy precip in MSFS!), something I'd normally ignore, being a sim pilot, but I figured I'd try asking my SI controller if I could deviate around the weather. I expected the controller to ignore my request but the response I got actually surprised me! Brian, the lead developer of SI is currently implementing a huge number of features across the board. The software was launched very much as a VFR-Only service, and as soon as the pricing was announced, I think that's where most people left it in their minds (too expensive!). The truth is that since launch just a few weeks ago, he's already rolled out huge improvements, including global ICAO support—which has also added quite a few country-specific changes in places like Australia and Germany–FSS Flight Service, and of most interest to me, IFR coverage from clearance thru approach (and landing of course). There's a general rule in SI as well, which is, if they haven't specifically programmed it in (yet), it doesn't mean it won't work! This is based on GPT-4 after all. If you tell the controller what you want, they will most likely respond accordingly. And so it was on my flight. I'd requested to deviate to the left around the weather, and the controller came back to me to approve a deviation 20º to the left and to please let him know when I was back on course. I hadn't said anything about the weather ahead, and it just so happened that the only 'corridor' between me and the other side of the intense rain was exactly 20º to the left! I know this was all just a coincidence, but in that moment it didn't matter. It was absolutely perfect, and incredibly immersive! I avoided the weather, turned back on course, was handed to the next controller and then prepared for my descent into Essex County by checking the latest foreflight weather. Another recent improvement to SI is the increased ATIS ranges. MSFS ATC only gives you ATIS coverage within about 20-30 miles of your destination in my experience, but on SayIntentions the range depends on your altitude (and can be up to 100 miles around FL300 I believe). This is going to be further improved upon in the near future, when terrain obscuration will be added to the radio reception modelling mean that, like on Pilot Edge (and possibly others), if you're too low, or between mountains, you might not be able to pick up what ATC is saying to you. The ATIS information is taken directly from public sources, and matched my Foreflight weather for KCDW. Winds were still coming from the east, but the ceilings had lifted slightly (to 800') and no rain was mentioned any longer. It was looking like the RNAV 10 approach would be my best bet, with minimums of 428' AGL. I asked the controller which runway I could expect for the approach and he told me Runway 22. I'm not quite sure of the logic that is used to determine landing runways (I think it's based on sim weather, rather than real world ATIS) and I believe that might explain the discrepancy here, but I decided to ask him if I could get the RNAV RWY 10 approach anyway and he approved it. A couple of minutes later told me to begin my initial descent. The descent was a little less successful than the climb. I had to request to continue my descent each time, but this is all par for the course. This code is really new — like, initially released last Friday or something, so there are bound to be issues that need to be ironed out. Improvements with SI are coming in thick and fast multiple times a day! I emerged from the clouds, to a beautiful view at around 5000ft. Far to the west the weather looked bright and clear, and a thin layer of broken stratus lay below (since when did MSFS render stratus correctly!) while heavy rainfall blanketed the airport. Things were looking a little dicier for the approach than I'd hoped. I established myself on final approach, and was handed over to the Essex County tower. As I got closer, the rain got harder. Then suddenly, I hit a wall of rain that obscured by visibility to almost nothing. 1000ft, no contact. 800ft, no contact. 600ft, no contact. This was just meant to be a quick 25 minute flight, but... How does SI handle missed approaches I wonder? There's only one way to find out. So when I still couldn't see anything at 500ft I went missed and informed the tower. I'm a lazy sim pilot who doesn't really know what he's doing, so my missed approach procedures are rough at best. But I told the tower I was going missed. Tower responded, asking me to enter the pattern for a new approach — obviously the wrong response in these conditions. So I told her I was in IMC and couldn't comply, and that I would follow the missed approach procedure (stating that I was turning left and climbing to hold at WOKPU). She responded, confirming and repeating what I'd said, and once I got there I radioed again to say we were going to remain in the hold a little longer to figure out what to do next. She responded with an affirmative, and to take our time. Marvellous! Missed approaches aren't a thing yet in SI. But if you just talk to the controller as if they are a thing, the ATC won't break and will work around its own limitations to keep you immersed! SI was holding up fairly well up to this point, how much further could I push it? I decided to find out by trying to divert to another airport. I checked foreflight for the nearest MVFR or VFR airport. The list of options was short. The nearest one was KAVP (Wilkes-Barre/Scranton Intl), about 15 minutes away. I decided to see how SI handled diversions. I called the tower back, asking to divert to KAVP. She responded with a brief, "understood. Standby for your new clearance." What!? Was that an AI-ism, or was she really going to give me a new clearance? I waited a little bit to see what happened. Sure enough, about a minute later, she contacted me again, "Cleared to the Wilkes-Barr/Scranton International Airport via DIRECT. Climb and Maintain 4000". My jaw dropped a little at this instruction. I read back my new clearance and went on my way. The further west I flew, the clearer the skies were around me. A layer of mist was settling below and there was quite a lot of blue sky above by this point. I absolutely love how the mist and visibility in MSFS live weather has been fine-tuned lately. No more clouds of fog in a circle around each reporting weather station. Its blended in so well, and looks especially good around hills, with their peaks poking up above it. I started preparing for my arrival. Getting the weather, and figuring out what approach to expect. I was looking at my iPad for airport frequencies, when I suddenly heard the RPM drop on my TBM. Had I accidentally knocked the prop lever? I checked my Bravo, but it was still at max. I looked up, and saw a 150 RPM drop. I cycled the prop level, but I couldn't get it to increase beyond this level. Huh? I decided to check the Blacksquare Failures mode on the radar and Lo and Behold, I had a prop governor failure! My first ever random failure on the TBM 850. I opened the user guide PDF quickly to find the emergency checklists. The first line: "Land As Soon as Practical" and "Do Not Attempt Go Around" — Yikes! I'd already had one of those today, and the visiblity below me wasn't looking promising. I quickly entered all the arrival data on the panel and decided to inform ATC of my predicament. I told them I was having power issues, but that I was explicitly not declaring an emergency, I was just informing them at this point, but that I would like to land at KAVP as soon as possible. The controller responded calmly to say that he understood, and that I should keep him aware of the situation. Shortly afterwards the mist began to dissipate, and I was cleared for the ILS 4 approach (the approach I had been expecting). Luckily I didn't experience any further power issues aside from my lower RPM, and the winds were favourable making the landing as simple as possible. I taxied off the runway, and was handed over to ground, allowing me to taxi to parking and face a likely expensive (albeit virtual) phonecall to my mechanic about my broken plane. My 25 minute flight turned into 1.5 hour ordeal, with IMC, diversions and mechanical problems to add to the drama. Even without the icing on the cake of the random mechanical issue it would have been an incredible first experience of a full SayIntentions IFR flight. * * * I didn't really expect this write-up to turn into a 2000 word essay, so well done for sticking with me all the way to this point, but my three takeaways from this adventure are: If you haven't reverted to MSFS' own real-weather engine lately, then you really should give it another chance. It has been FANTASTIC the last few months! It was always better in Europe, but the US seems to have gotten a bit of love lately too. I can't say enough good things about the Blacksquare TBM 850. There are of course some extra features I wish it could have, but everything that's already included is so immersive. It's my goto plane in the simulator (and I secretly hope a new update is coming with even more features). SayIntentions is the ATC I always wanted in a flight simulator. It's mind blowing at times, and adds so much to my enjoyment. There's a global 24-hour demo where you can try whatever you want for 24 hours - please just go and install it and give it a whirl! Brian, the lead developer, mentioned yesterday that 75% of demo users end up subscribing (!!) That's huge! Give it a go and you'll see why.

I rarely fly in central Europe due to exactly this issue. As soon as I get east of EHAM (sounds like a movie title!) I get severe stuttering and poor FPS. The worst part is that the performance never really recovers after that, even if in overflying to somewhere empty like the Gulf (Although maybe SU15 solved this particular degradation - I haven’t checked yet) i recall flying a lot to EDDK early on without such big issues. I first noticed it after the Germany WU, and it still plagues me to this day even on a 7800X3D/4090.

Did you turn on the PA audio in the cockpit? From the Fenix KnowledgeBase, they show you how to do that, so that you get the audio from the announcements in the flight deck too: https://kb.fenixsim.com/cabin-announcements

How are you turning? I've heard that they've updated the flight model to be more representative of the actual aircraft - so that the wheels turn max ~10º and the rest of the turn has to be handled using differential braking. I've not experienced any tipping myself, but have found the aircraft to turn a lot less when using solely the rudder pedals. Perhaps your speed is a little high also?

Wow... I abandoned the PMS50 version after my first year, partly because I got GNS430 hardware so found myself using that much more, but also because the feature-set seemed to be drifting further and further away from the actual unit. Now that this has the navigation data (and is cheaper!) I think I'll have to give it a go!

I know what you mean. The other night I was enjoying a late afternoon flight on my new system (same as yours, except for a 7800X3D instead) and definitely got the feeling that the visuals were better than the equivalent I saw on my old system (again, same as yours except for the 10700k instead). Not quite sure exactly why it looks so much better with the equivalent settings — it could be some mechanism in the brain to help justify the huge expense of the upgrade! 😆

I'm still really eager to see what BATC has come up with, but I agree, for the equivalent immersion and voice quality etc, the cost of SI compared to BATC looks very competitive. In another thread I was disparaging the voice quality of the different BATC voice levels a bit, but having heard some extra samples from their premium voices, I now actually think they sound fantastic (on-par with, and perhaps even better than the quality of the voices in SI). But it would then cost multiple dollars per flight to get that with AI traffic. The (optional) inclusion of other pilots audio chatter in SI is also excellent, and not something I thought I wanted. It's cheap (because it doesn't require AI generation), and they don't really promote this much, but it can be used in a multiplayer context (if you're on the same servers - albeit with a bit of a delay currently), which is fantastic. I prefer my traffic in the FSLTL form though, so I have only used it a couple of times. I'll be interested to see how the traffic injection with SI goes (and the implications that it has for pricing, latency etc) but for now I'm definitely on team SI.

IFR support is already partially implemented (and supports Simbrief as well) and Brian (the lead dev) expects the bulk of IFR support to be added in the coming weeks. This really is an exciting project and, while I also balked initially at the monthly price, now that I’ve been using it for a couple of weeks, I can’t see myself flying without it in future - I’ll just have to buy fewer other things 😆 They’re adding new functionality with impressive speed. FSS support was added last week, worldwide ICAO support is available and being improved daily, they already have proof-of-concept support for AI traffic injection (just not released to us yet), and Oceanic HF radio support is currently being discussed. The demo gives you 24 hours free and is reset each month so that you can come back and sample the improvements. ATC response times and cost are the two things that they say they are working to reduce as quickly as possible (and I believe them).

The quality of the speech and sound is very good, and they sound like real people, but my beef with every BATC demo that I've seen so far, is that the ATC voices just don't sound like real air traffic controllers. The intonation is more like a conversation with someone in a cafe, than the short, sharp instruction over the radio from a controller trying to get a point across in as few words as possible (same for other aircraft around you) I think I must be in the minority with this viewpoint though, because I haven't seen many other people mention it since BATC was announced. Last week, Brian of SayIntentions was discussing voice quality on discord, and mentioned the fact above, that it's very hard to train the voices to sound like actual ATC. He's done an admirable job on his tool (they sound a lot more authentic to me), but when discussing other accents, he played an early prototype of one, which everyone responded very positively to. His reaction was along the lines of, "Wow! You guys have really low standards... This doesn't sound anything like ATC yet".