+- FsPassengers Forums (http://www.fspassengers.com/forum)
+-- Forum: FsPassengers (http://www.fspassengers.com/forum/forumdisplay.php?fid=3)
+--- Forum: FsPassengers Customizing (http://www.fspassengers.com/forum/forumdisplay.php?fid=6)
+--- Thread: Proboards Pilot career website (/showthread.php?tid=10113)
Re: Proboards Pilot career website - jc301290 - 19-07-2006
my eyes hurt from reading all those words you wrote.damn.
so,after reading your first problem, why not do what most ppl do?
just let the player type in the ICAO or IATA code for the airport. make a subroutine to check the validity (whether the airports exist, whether they
are the same,whether the airline has a plane for handling the range).
EDIT:NVM.just read the previous post.
DAMN! your'e FANTASTIC! is only i coded as well as you. >_<.
Post Edited ( 07-19-06 14:01 )
Re: Proboards Pilot career website - Airliner - 19-07-2006
jc301290, what the heck are you talking about?
Oh, okay.
About the airports thing, that was just an example of all the junk that makes a program work. You see a combobox or textbox and either
select or enter something and bing - results. But, there's all sorts of "junk" that goes into making each object, that's all. I'll sit there with
a notebook and scribble out a design and say, "Okay, enter this here and click this and then this screen comes up and then you hear this
sound and..." BUT! When it comes time to put the ideas into a real program, ugh. Anyone can pile up the ideas but, to
actually "convince" the computer to do what you have in your head, well, that takes a little more effort, which is why I admire people like
Dan for actually sitting down and hammering out something that actually works! It's one thing to do it within a major business and with a
team of programmers. It's quite another for one person to sit down and type until their fingers bleed! Good job again Dan!
Okay, I tried to keep that one pithy! I sometimes forget that Bill has a patent on bloviating!
Re: Proboards Pilot career website - Airliner - 20-07-2006
So, I just completed a list of 9-10 factors that will be used to generate loads. I'm goning to work up a mini-program to simply spit out
some numbers and then setup an airline to crunch everything out and see how it works. Anyone interested it trying it too, let me know
and I'll give you a copy - when I get it finished. It'll be similar to Pilot Career - meaning, you'll have to manually enter the data and it will
then dump the numbers back to you. I want just the basics to start, in order to work out the calcs.
So, let's say you setup your first route: KBWI-KMCI, on a B737
Using the Great Circle Mapper page, this means:
967 miles and about 2 hours
Add 15 mintes to each end for class 4 airports
1 route out of destination + 1 route into origin = "hub" bonus of 2
Total cities served = 2
On-Time record = 0(no record yet)
Origin City size = 4
Destination City size = 4
Aircraft Class = 4
[(Origin + destination)/2] / Aircraft = 1 (or 100%)
Route Distance value = 100 as it's the only route so no comparisons
Route frequency = 100 as it is again the only route - no comparisons
Services/Comfort (need values for each service offered as well as catering value over time of flight)
Call this 85
Price check - comparing to sliding CPM scale to determine if above or below the average.
(extra services can outweigh higher fares)
Call this value 120 (lower than the average for this route size)
Flights within same time period:
Setup 4 flights - 2 each way.
WN10 BWI-MCI 0600
WN20 BWI-MCI 1500
If more than 1 flight within the same time period (0500-0900 for example), decrease by 20%.
No other flights here so each is 100%
So we have the following estimates:
2+2+0+100+100+100+85+120+100 = 609 / 9 = 67.67 or 0.68
Now, seating is set to 137 so load generated is 137 * 0.68 or 93 passengers.
93 * 250 = 23,250 lbs. (includes average baggage weight for two checked plus carryon each)
Crew = Captain, FO and 3 FAs so weight for them is 1025 (190 each + 15)
Total weight to add to payload is 24,275 lbs. Now, either load the plane manually or use FSP and fly the flight.
Upon completion, need to record:
total time in minutes
fuel burned
FSP satisfaction rating (if used)
Use these values to re-calc the pax load value and crunch out the other flights, including the FSP value if so desired.
Adjust income/expenses based on ticket sales, crew salaries, fuel burn, maintenance costs and lease payments, if any.
Spit back bottom line cash balance.
Also, will want some generic data to confirm the other calcs are coming out right.
Okay, I need to start coding!
Re: Proboards Pilot career website - clootie - 20-07-2006
Airliner,
Give me a shout if you need testing.
Cheers
Dave
Re: Proboards Pilot career website - jc301290 - 20-07-2006
hey!give me that shout too!
:D
Re: Proboards Pilot career website - Airliner - 21-07-2006
You got it! Here's a quick look at the basic interface I just threw together for the demo thing:
![[Image: untitled.jpg]](http://mysite.verizon.net/vzepecu6/sitebuildercontent/sitebuilderpictures/untitled.jpg)
Again remember, this is all manual for testing purposes. You have to enter all info - not everything is calc'd for you. So, you enter the
Origin and Destination AND the distance. There will be a limited aircraft DB so it will give you the estimated time with an average cushion
for ground handling based on cruise speed. Seats will be fixed; there's no catering, etc. After the flight, you will need to log the start time
and end time, like in Pilot Career (I have a routine to read this from FS but, skipping it for this).
The program will autogenerate a load before you fly and run the numbers for all flights after you log a flight. Crap! I forgot to add the FSP
pax satisfaction log box! Also, the fuel burned - okay, more work to do!
By the way, just to give an exmaple, it took about 15 minutes to make this interface. That's the EASY part. No comes the coding to make
it work. That should take a couple days.
Re: Proboards Pilot career website - Airliner - 22-07-2006
Okay, just finished writing the routines to run this thing - they took about an hour. Final step is to build the databases and make sure
everything works, so I should be giving "shouts" by next week.
For this test, the only factors that will influence loads are:
- FSP Satisfaction Rating (of course!)
- # of destinations served by each city
You will get a bonus on each route = the number of cities served by the origin and destination airports. This will sorta simluate
connecting traffic - traffic connecting to your departure and from your arrival.
- Total cities served - basic; your network size
- On-time record
- Route Distance Value - described earlier. Longer flights that you fly have a higher impact
- Route Frequency Value - If you neglect to fly routes, their loads will steadily drop
Leaving out of test for now:
- Services/Comfort Figure
- Price and Value Figure
- Mutiple flights in same time period Figure
- Market size versus aircraft size Figure
Re: Proboards Pilot career website - jc301290 - 22-07-2006
damn. I love you dude.
Re: Proboards Pilot career website - Airliner - 22-07-2006
Hey! None of that stuff! I may be from Massachusetts but I cerainly do not agree with Justice Marshall!
Re: Proboards Pilot career website - clootie - 22-07-2006
looking good airliner, what DB are you going to be using. FSP files, MSSQL or MySQL ?
Re: Proboards Pilot career website - Airliner - 22-07-2006
Well, right now for this test, the databases are going to be simple txt files, comma delimited. I'll probably use excel to make them and
then convert. As a matter of fact, this thing may go the way of the single player version - standalone. I could probably later adapt it for
online play but, the whole reason I had originally designed the online "universe" for was so that not only would airlines run routes
alongside others but also so that everyone could download the traffic files of all other airlines. This way, you would see all the various VA
liveries flying their schedules along with you, rather than stock, static files like we have now. The problem with that is obvious. The
schedule is no big deal - just run through the entire database and shove it into a flightplans.txt file. The problem is with all the aircraft and
liveries. That would be a major headache as it would be like PAI - download each airline one at a time, plus update everytime a new
model or paint is introduced. (I should mention that I DO currently use all their files. Hey, it's free and relatively painless to install!
Anyway, just had another idea on the way home. Rather than have a bonus multiplier built into the system (I was going to pull a Dan and
work in a x50 multiplier, which would gradually be reduced as your airline grows) but this has given way to something I think is far better.
Before each flight, you have to option to run a one day advance, one week advance or one month advance. Meaning, after you log your
flight, the new load averages and ratings are calculated for your airline and then, depending on your choice, the sim will run and generate
results for all other flights for that day only, the whole week or a full month. So, if you want to get things going, run monthly to start. This
will have the effect of basically a x30 multiplier on all flights in the schedule, but each flight be run individually.
Another quick adjustment is that, what the? I think my wife spilled something on the keyboard. Anyway, not only will your on-time
performance be the basis for generating other loads, it will also be used to generate arrival times for the AI flights as well. Let's say
you've run 20 flights on your own so far and have an on-time rating of 92% with an average delay of 15 minutes. The system will run the
AI flights in the schedule and use the 92% to determine if each flight is on-time or delayed. If delayed, the average delay time is used
with a random value to set the delay. These too will then effect further values. Bottom line is, the better you perform, the better your AI
pilots will perform. Same with fuel burn - each time you fly a flight, the fuel burn you log will be entered into that route and will be used for
all AI flights along the route. This means that whenever you start a new route, you MUST fly that route next. Until you fly a new route, it will
not be flown by the AI.
I was thinking about approaching Dan about incorporating all this into FSP in the future but, now I'm thinking this maybe should be
separate, as it's starting to get more complex. We'll see.
Re: Proboards Pilot career website - jc301290 - 23-07-2006
why not use PHP with a SQL DB?no need for downloading the traffic files for airlines...
Or,maybe your program could interface with a SQL DB instead?
Re: Proboards Pilot career website - Airliner - 23-07-2006
Quote:jc301290 wrote:
why not use PHP with a SQL DB?no need for downloading the traffic files for airlines...
Or,maybe your program could interface with a SQL DB instead?
You're not hearing me. Traffic files are not the problem - it's the liveries and models that would involve all kinds of headaches. I doubt
that anyone wanting to keep their traffic up to date would want to go through and check all airlines for updates everytime they fly. Of
course, there could be an automated way but, then everyone's aircraft folder would get huge.
Second, I don't know php - tried but never got the hang of it. Like I said before, for an online version, SQL would be the DB of choice along
with the VB2k5 language.
Re: Proboards Pilot career website - jc301290 - 24-07-2006
Oops. :$ My eyes glazed over as I was reading that.
Re: Proboards Pilot career website - Airliner - 25-07-2006
My eyes glaze over sometimes when I try to figure out what I scribbled in my notebook! "Huh? I want to put this there and why is this thing
over, oh!!!!!! Okay, I remember."
By the way, I got half the coding of the demo parts done last night.
Just for the record, here's a quick thing to illustrate what I meant before about the coding taking up a lot of space for silly little things,
here's a little piece of code. All this does is add a plane to your fleet. And, it's a simple group of things to, not at all the full data that will
be in the final.
SUB Button4Click (Sender AS QBUTTON)
'-- Purchase Aircraft
Button4.Enabled = 0
NewAcIndex = ListBox2.ItemIndex + 1
Check = 0
FOR X = 1 TO 500
IF AircraftGrid.Cell(1,X) = AircraftGrid.Cell(1,NewAcIndex) THEN
FOR Y = 1 TO 500
IF FleetGrid.Cell(2,Y) = "" AND Check = 0 THEN
Check = 1
FleetGrid.Cell(2,Y) = AircraftGrid.Cell(1,X)
Hours = VAL(AircraftGrid.Cell(9,X)) + 20
AircraftGrid.Cell(9,X) = STR$(Hours)
OldCash = VAL(AirlineGrid.Cell(1,1)):ACCost = VAL(AircraftGrid.Cell(11,X))
NewCash = OldCash - ACCost
AirlineGrid.Cell(1,1) = STR$(NewCash)
AirlineGrid.SaveToFile("Airline.txt",1,1,2)
CALL AirlineDisplay
FleetGrid.SaveToFile("Fleet.txt",1,1,501)
FleetGrid.LoadFromFile("Fleet.txt",1,1,501)
ListBox3.Clear
FOR Z = 1 TO 1000
IF FleetGrid.Cell(2,Z) <> "" THEN
ListBox3.AddItems FleetGrid.Cell(2,Z)
END IF
NEXT Z
END IF
NEXT Y
END IF
NEXT X
END SUB
This is what I'm talking about when I talk of the tedious stuff. But, I should be done in a couple days.