You can read the title of this week’s issue both figuratively and literally 
I made some really good progress on the performance editor and new formulas for aircraft takeoff performance…literally dealing with lift and drag (more below). But I also spent a bunch of time preparing the legal move of simulogics and the migration away from an annoying payment provider which I’d like to tackle as part of that move. Switching a payment provider is only marginally more effort than doing the paperwork required when the legal entity receiving the payments changes
But enough of the drag and on to the lift:
As mentioned several times before, our aircraft performance system is in pretty bad shape. The formulas are just laughably wrong and, consequently, so are the input data our volunteers have researched over the years. We’ve been planning to revise this for ages but there were several challenges along the way:
- as said, we don’t have a good system (in terms of formulas) to begin with, so researching “better” data can only do so much to alleviate the issues
- the actual act of researching the data is extremely cumbersome, with long feedback loops due to manual “data releases” to a test world
- in said test world, volunteers have no tools other than what any other player has at their disposal to evaluate aircraft performance
- the chicken-and-egg situation of “no data before new formulas” and vice versa sort of lead to a stand-still in this area
While I’d love to continue my work on the DS integration, especially the last point in the list above bothers me. And it does so to a point that I really want to get the aircraft performance problem solved first: There are volunteers willing and capable to help, but they are blocked by my (lack of) progress.
We already found a good (enough) approximation for payload/range performance. Last week, I finally moved on to takeoff performance. For this, I found a good set of approximations which I managed to implement in our performance editor. The results look extremely promising, especially because they take into account airport altitude and, in theory, temperature. Finally, Hot&High operations in AirlineSim!
Here’s a shot of a chart for the 747-8. The red lines are the output based on our formulas, the chart below shows the actual values from Boeing’s airport planning manual:
In terms of what we need for AS (which is, after all, a game…not dispatch software), this is almost perfect. There are few more improvements I’d like to look into, but won’t waste too much time on:
- Pretty sure I can find a better solution for the “minimum takeoff distance” on the left of the chart…great fit for low altitudes, but seems off for higher ones. Plus, there is a hard lower limit I still need to add.
- The curves become steeper for long takeoffs in high altitudes…probably a simple physical limit…just haven’t figured out what it is yet . If any pilot wants to enlighten me…please go ahead!
- A lot of charts show “brake energy limits” that I’d like to add. I figure it should be simple enough to do (famous last words), as brake energy should be some multiple of “mass * speed”, the latter of which is a byproduct of the formulas anyway.
At this point, I have to give kudos to Boeing for their charts. This is what an example for the A380 looks like:
The axis are flipped. There’s far less detail. And for some weird reason, the chart extends far beyond the MTOW of the aircraft.
Either way, the fit is still pretty good. Especially compared to what we’ve had so far. We won’t and can’t go overboard with this…I’ll likely ignore different flap settings, all runways in AS will always be dry and we’ll stick to cosy ISA standard temperatures of 15 degrees Celsius at sea level (for now :D). But I am really happy with this!
Only downside: The new formulas require a bunch of input variables that are impossible to find on the internet. So I’ll likely have to write some code to reverse-engineer them from the charts. As such…I got my major AS project for 2024 cut out for me, I guess.