Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
I'm holding off once I found out about the character limit. I could be wrong, and I really hope I am, I sure hope that is for a day otherwise the software isn't worth purchasing.
Wow, I just went to their site. There is no reset you have to keep purchasing more characters. Here is something from their website.
Nope, this software is not for me. They even say the 50,000 character limit is about 1 hour of speech. After that you have to fill up that character limit by purchasing more.
You're better off using Pilot2ATC. Yes hefty price tag but once you buy it it's all done, and you can use it without restrictions.
I do like to think my wittier writing entertains, but if I somehow made it unclear that Basic = Basic, then it's back to the Ayn Rand Academy for me. On the other hand, if you're misunderstanding because you read something wrong and really can't grasp Basic, then I get a Coke.
We all know you're cheap, you've made that abundantly clear in other posts. I'm cheap, too. I wouldn't have pulled the trigger if I thought I needed the Premium stuff. But if I win the lottery, there it is.
As it is, AI experts speculate that the price of AI voices ought to go down over time the more popular they become. Servers will have more users, which could drive down the cost since it would be easier to generate profit. On the other hand, Azure has been pricey since its inception around ten years ago and remains pricey today.
I don't want to shill for BATC, but I do think what you get for your money is better than Pilot2ATC, which is becoming outdated. However, it's your choice. Please make it an informed choice, though.
I place Pilot2ATC squarely in the "cumbersome" camp. It's really fiddly, you have to spend quite a long time setting things up, and it is expensive enough - twice the price of BATC.
Oh well, maybe no one else does but rest assured, even purchasing the basic model at $30.00 or so dollars, and paying another $9.99 per week for more character points it will quickly add up to the point that you could sink a few hundred dollars after Six months of using the software.
No, only fools pay for something that will run hundreds to enjoy.
Basic voices, of which there are 170, don't cost anything to use beyond the initial purchase for $30. No limit at all. Infinite usage (or at least as long as BATC keeps operating - those voices are depending on server-access, after all).
And before you think I'm shilling for BATC: Nope, I took a look at it and decided it isn't for me (yet). The basic voices for non-english countries are often cringe as hell, ridiculously overblown thick accent and frequent mispronounciations. No VFR, no traffic injection handling yet (will come in the future), both of which are deal-killers to me. Too many bugs for me, too.
So I am totally not telling anyone to go buy this. But what you're yodeling here is just plain wrong.
I switched to off-line voices a while back which to me don't sound much different to the Azure voices, and it has at least been mainly error free in IFR flights for me. I can live with those until something MUCH better comes along.
It must be somewhat difficult to develop a good ATC system for ANY developers out there because in the case of BATC, although somewhat better than the default, it's a far cry from being the perfect alternative. It may improve as time goes by but it's a long way off at this point for me.
But it does work. Why are you spreading misinformation?
Yes, they should.
SayIntentions is the other big AI-based 3PD ATC right now. It's also more expensive, but the AI is closer to ChatGPT. It really understands context and you can say a lot of things that it will comprehend. I think it's exciting as well, but it has a tendency to tell you things and not follow through, including the dreaded direct-to into a mountain. I'm not sure if that's a bug or a current limitation to ChatGPT, more likely the latter, although of course the AI learns from its mistakes easily.
SI is maybe at least as buggy as BATC, maybe less so. It handles a lot of commands, and is very, very impressive in that regard. SI would certainly be at least equal to VATSIM in terms of being able to converse with ATC.
This is not all true for the basic voices. The premium voices have a character limit. However, 1 hr of speech is a lot, and this will likely last many sessions.
Modern ATC in the real world is in a state of transition a lot like the 1970's and early '80's - not good times. Pre-Pandemic was the highest volume of air traffic ever recorded, and we are at that point again: more aircraft in a limited airspace. There is immense pressure to get commercial aircraft in and out of airports. No wonder there are so many close calls, the past couple of years have seen some noteworthy changes to ATC procedures both in North America and Europe.
On the plus side of the ledger is GPS navigation. Computerized real-time geolocation down to a meter or two of resolution ensures that ATC can shave separations down to a minimum. GPS-style satellites will only fail in a Carrington Event or a war, so it's reliable over the majority of the globe. Not all of it, but much of it.
GPS transmission is almost immune to hacking as well. Unfortunately, that's not the case for GPS receivers. At DEF CON, the yearly White Hat hackers convention in Las Vegas, they proved conclusively that a single person can jam an aircraft or even an airport through an unguarded GPS receiver using parts you can get from Radio Shack (fortunately, Radio Shack is dead, so maybe it's not so easy after all?). While it's almost impossible to make an airplane disappear from GPS through jamming, it's easy to create a denial of service attack, simply to overload the receiver with false contacts.
So, it does look like human ATC might end up being replaced with AI, just like writers, illustrators, and John Lennon. An AI ATC might truly be efficient, but it might also go nuts and blow up Tesla-style. BATC and SI kind of give us a glimpse into that. Well, so does MSFS, at least from the perspective of a computerized ATC that fails to value human life.
Thanks TWELVEFIELD (sex & drugs for now then it is)
Programming IFR flights is easier than handling VFR because you're telling the ATC exactly where and when you are flying - your flight plan. So the computer doesn't have to generate data for Chicago if you are flying from Tokyo to Beijing, unless you're Tom Cruise in a Darkstar. Flying VFR is simpler for pilots but harder for an AI to rationalize - so many variables. A good IFR flight plan has as few variables as possible, which makes for safe flights for millions of people and also makes for truly boring flights. That's good: exciting, unpredictable flights are very bad for airlines.
The first 3PD ATC I recall (that was any good) was I think RealATC 2000 for FS2000. I don't think I remember the name right, but it has a single dev who was very British by name of Dave Barry or something like that. I am sure I am remembering it wrong, but I'm not far off. Anyone who remembers it right I promise not to insult them to their face for 24 hours. I get to pick which 24 hours, though.
By filing an IFR flight plan, the ATC generated pre-rendered callouts that were more detailed than what the ATC in MS2000 could provide (and with better voices), because they were triggered by logical IFR events. I think you could even plug in SIDs and STARs, which the early FS ATC has problems with. The downside was that you couldn't deviate from your flight plan, but you could declare go-arounds, diversions, and emergencies because those were things you had to include in your flight plan, kind of like SimBrief.
So the AI system is similar to that, the basic concept hasn't changed much. The AI is more conversant, that's all. In SayIntentions, the AI can receive commands for a wide variety of flights, but I'm not convinced it understands all of the human rules yet. It's getting there.
Air traffic, yes, that's going to be interesting to see what happens. BATC barely has functional traffic, and nothing that goes into the air yet. It's planned as the "second half" of the BATC experience. If it works, that would be great as long as it isn't pointy passenger jets. If it doesn't go well, BATC isn't all that expensive as long as you stay away from Premium, I could live with footing that bill.