Is AWS dying?

[JumboShrimp]

I disagree with getting "squashed" in LAX, ATL, LHR ect. If you get any slots in those airports you are nearly invisible as there isn't enough slots to force oversupply routes enough to cause bankruptcy.  They are by far the easiest places to play IMO.

Yup, and the direction has been to fewer and fewer slots, meaning more and more boring in slot deficient airports.

[MidWorld]

AWS is not dying, it is evolving.

====

I've been a member for several years and played the long games twice, both times dropping out not too long before the end. I come back every couple of months to check the forum and see what's up.

AWS is a blessing. It is smart, sophisticated, dynamic, fun, I love the community and appreciate Sami's moderation. As a business owner, I learned a lot from this game.

Here are my 2 cents on the directions of improvement.

1) The basic stuff. The interface requires a face-lift. One simple improvement that would make a world of difference would be a dynamic page of real-game-time airport arrival and departure timetables with all of the delays and cancellations. That should be fairly simple from the programming perspective. The reason this is important is that this is a good time killer to get through the end of the game day. It would pain-relief the major annoyance of waiting.

A replica of flightradar24 would be lovely, but probably too much from a programming perspective.

2) The big issues. There is just an unbearable amount of time wasted doing repetitive tasks like plane scheduling that AI can handle just as well, but significantly faster. Ultimately, both times this was the reason behind my decision to BK.

Micromanagement required to play narrows the player community down to people, who have time in abundance. Unfortunately, I am not one of them. Instead of playing the airline strategy, this becomes the game of operations with some strategic inputs.

I hope this will change with CBD in pax service, the introduction of the hub-and-spoke model, further developments to allow brand marketing, services in the air (seats, service classes beyond first/biz/economy, food, entertainment etc), and services on the ground (check-in, luggage, lounges, terminals). Naturally, the game should move to managing the rules, not operations.

From the economic perspective, this game has known inputs that are unknown in RL (like demand on the route) and oversimplifies the demand model. Business, leisure, VFR clients cannot be managed separately, airline product is narrowed down to seat type selection, turnaround time is the same for legacy and low-cost, alliances do not allow connecting traffic, there are no code-shares, interline agreements, and so on. Bringing the RL countries' GDP, wealth distribution, mobility, cultural ties etc into the game calculus would allow the game to be thrilling all the way til 2036 or even beyond with some new plane types.

====

I find this game fun and I would love to be able to support Sami's further effort not only through game credits purchase but through other means, like Patreon or direct donations to allow hiring assistance at Sami's discretion. I think, as a community we can crowdfund certain projects, thus bringing major improvements closer. I personally would love to volunteer to help with some of those issues.

[Continental Sky]

It is smart, sophisticated, dynamic, fun, I love the community and appreciate Sami's moderation.

[...]

Instead of playing the airline strategy, this becomes the game of operations with some strategic inputs.

I fully agree with everything you wrote, but I highlighted these two points above. I am relatively new to this game, but like it and plan to play long games, hopefully will manage to do it.

Waste of huge amount of time for repetitive micromanagement tasks is indeed big drawback of this game; the scheduling you mentioned is one issue, but there is, at least for me, another much more annoying issue that shows how some functions are made in a completely user-unfriendly way.

In interface for opening new route or route planning, as far as I can tell, airports can be sorted only by their size, name or code; I was not able to find any other option. Name and code are absolutely useless, who cares for the code if there is sufficient demand. Remains the size, which is usually not reliable - oftentimes smaller airports can have higher demand than airports of bigger size or traffic. As a result, if I see that demand drops below my desired target on, let's say, second page of results, I still cannot be sure that 7th or 11th page does not hide an odd airport with huge demand. Routes for first 20-30-40 or so aircraft in a base are relatively easily found, but after that, as good routes get exhausted, I have to open literally 100 or 200 pages to find suitable routes for one single aircraft. I'd say it is much worse for cargo than for pax.

It's beyond my understanding why there is no other option to sort airports in route search page, say by pax demand and supply, cargo and potential cargo demand and supply, number of pax, pax fee, landing fee, percentage of available slots, etc., those are all factors to decide when opening a new route. I am pretty sure it does not require PhD in nuclear physics from programming point of view, and would be enormous help in locating suitable routes without opening new page for every single airport pair.

[groundbum2]

I think this is fairly deliberate, if the big fat juicy routes were presented first then it would take out a lot of the "hunting" in a game. Since the demand is largely static, a lot of players record these numbers to spreadsheets etc, then can use these to find the juicy worms.

Simon

[Continental Sky]

OK, I see your point, but I don't agree.

Charm of this game should be in decision-making process, not in data mining. When looking for expansion, a dozen of questions arise:

- Should I go for this route where demand is barely enough to cover airplane capacity and there are no competitors, or that route where demand is huge, but so is the supply.
- Should I go for overnight route that departs at 23:55 and arrives at 04:15, or play it safe and go for a daily route.
- Should I leave default ticket prices, or increase, or decrease, and for what percent.
- Should I go for this route where pax demand is abundant but cargo is zero, or that route where pax demand is two thirds of plane capacity, but cargo is full.
- Should I go for turnaround time with 15% delay probability in order to fly every day, or go for 1% delay probability and skip one day for A-check.
- Should I assign this route two flights a day with ATR or one flight a day with MD-87.
- Should I go with stopover, to have full capacity, or direct flight with limited capacity.
- Airport fees there are three times more expensive than in my base airport, lemme calculate if it is profitable to fly there at all.
- That aircraft is ideal for these 10 route pairs, but that means another fleet for me; lemme check is it worth it.
- This aircraft is cheaper to purchase and cheaper for maintenance, but 30% slower than that aircraft; lemme calculate if that one extra route a day will cover the cost difference.

...and so on, and so on. Inevitably, each of the questions above has another alternative: "Lemme see used aircraft market, maybe there is an affordable plane so to cover both options!".

So, instead of focusing on those questions, one needs to open hundreds of tabs and harvest the data. If those data, the ones I mentioned in my previous post, were readily available, in a sorted table, it still does not mean too much for success or failure, the data themselves do not provide answers to above questions at all, one still has to make right decisions to succeed! It just takes out the cumbersome manual work.

If I don't have time to collect the data in Excel, or am not willing to select/copy/paste for hours, I have to do it again and again, every time I want to deploy new aircraft. I don't think it's a sensible concept. In addition, at least from what I read somewhere here on forums, cargo demand is not static, but shifts from potential to actual demand, based on supply.

[Crazybernie]

Regarding route data if those numbers would be easy to find or scrape because they are presented in an easy table, all of your next steps would be almost trivial to script.

[Mr.HP]

Regarding route data if those numbers would be easy to find or scrape because they are presented in an easy table, all of your next steps would be almost trivial to script.

Didn't know that script can be used to play the game and have advantages over common players
Where's the fun in that?

[knobbygb]

Didn't know that script can be used to play the game and have advantages over common players
Where's the fun in that?

It's not allowed to run scripts against the actual game server, but if you manually make a copy of some data, in a spreadsheet for example, then it's yours to do what you want with. How would you draw the line as to what 'advantages' you are allowed to use in your own  time on your own PC?  You could 'scrape' the data by capturing browser screens I guess.  I'm not sure exactly where the line would be.

[swiftus27]

Didn't know that script can be used to play the game and have advantages over common players
Where's the fun in that?

Does Hitler in the bunker need to come back out and explain it to others players?

[Mr.HP]

It's not allowed to run scripts against the actual game server, but if you manually make a copy of some data, in a spreadsheet for example, then it's yours to do what you want with. How would you draw the line as to what 'advantages' you are allowed to use in your own  time on your own PC?  You could 'scrape' the data by capturing browser screens I guess.  I'm not sure exactly where the line would be.

Oh i see. I was under the impression that the scripts run directly into the game. If its like you said, i think its should be fine

[Crazybernie]

Running scripts that control the game is not fine , i am just pointing out that presenting tables in easy tables would make it almost trivial to run scripts from your computer to control the basic mechanics of the game.

However due to the fact that the data changes slightly with every request to the servers and because you would need to run a pretty predictable amount of request to refresh the data , therefore Sami can easily spot people that would attempt that.

However having spreadsheets that require manual data input is fine. The important thing is that the game mechanics need to be run by real humans.

In my case due to Peanuts bk in London last friday I spent like almost 6 hours to rapidly schedule planes at one point i was tired and went to bed. Result next morning a competitor opened in LHR and had accumalated a respectable amount of slots.  If there was a script to run the scheduling that player wouldn't have stood a chance.

[gazzz0x2z]

in MT, I was absent 2 critical days in the beginning (father had covid while visiting us, takes time). Result? My powerstart was not as powerful as usual, and my opponent in GRU is still alive and thriving. That's life. And that's what is great : you have to actually play to achieve success. I'm under my objectives, and I know why. It leaves space to others.

[Mr.HP]

However having spreadsheets that require manual data input is fine. The important thing is that the game mechanics need to be run by real humans.

Totally agree. It takes time and effort to build the data, and players doing that should be rewarded

in MT, I was absent 2 critical days in the beginning (father had covid while visiting us, takes time)

Sorry to hear that, hope your father and family is fine

[gazzz0x2z]

Sorry to hear that, hope your father and family is fine

Thanks. Fortunately, vaccines do protect people good enough, so the harsh times IRL are behind me. Dominating BRazil will prove harder than forecasted, and that's fun, ultimately. Maybe even I'm gonna fail, which would be a great lesson as well.