I am following this thread since some time without saying anything, so I must warn youthat this will be a long, long post. And with spoilers, too
1) As far as I can see, one very hot point of the debate is whether penalties (including crashes) should be "cleaned up" in time. Hector's case in the preceding post is very well stated, and an excellent explanation why we should not do it, however I think the solution stands in this very phrase:
My question is: what do we want to see? Nice numbers or our real learning curve?
By following this thread it looks obvious that there is no single answer to this, so my suggestion here is that we may keep two scores per pilot, one that includes all and a rolling one. We can extend this to maintaining two "top tens", for instance a "best scores in the past 3 months" and a "best all-time scores".
Time for another warning
: being one of the developers, unless I get bashed for this idea this WILL be implemented this way. So do not hesitate to comment on this one...
2) Whatever we do to compute the pilot overall scores, one thing is for sure, we need to change the per-flight scoring method. There is still debate within the MT about the details, but the big idea is to grant bonuses for everything the pilot does right instead of subtracting penalties for what he does wrong.
To put it otherwise, a pilot gets nothing for an uncompleted flight (whatever the reason), some bonus for completing the flight in one piece and more bonuses for doing it right (like using the landing lights when AND ONLY WHEN needed).
This change will make necessary to restart FAC calculation at zero for everyone, because there will simply be no way of computing together "old" and "new" scores. I know this will be a blow for some of us (including myself
) but this is inevitable if we want to reform the score/penalties system.
3) One aspect still under debate is whether the overall pilot score should be like a "total" (as it is now) or an "average" (like getting a 1..100 score per flight, and compute the average for all flights). Feel free to comment on this, because we will probably implement the most popular option of the two.
4) I have seen, in this thread, very sensible comments about scoring long flights vs. short ones, and we will try to take into account the following principles:
- A 4-hour flight is not twice as difficult to perform as a 2-hour flight, because the most important parts (and those that require pilot workload) are take-off and landing, which occur one time in both cases;
- However this tends to be less true (in the favour of longer flights) when flying online, because in this case there is a continuous workload throughtout the flight (let's ignore the (in)famous "IVAO badge hunters" who do their online long hauls with "Do not disturb" remarks in their FP while going shopping).
Balancing the bonuses for online nd offline flights will definitely be a difficult and delicate thing, because online flights should be encouraged while the EHM policy remains NOT TO FORCE anyone to fly online.
5) Some penalties obviously need revisiting and for now, one method we think of is adding timers. For instance, turning off the lights by mistake will no longer be penalized IF the pilot realizes his mistake and corrects it within seconds.
The same should solve the overspeed problems. I know this is debatable too; Kivanc stated above, and I think he is right, that in real life passengers would not be very forgiving for their pilot going into overspeed.
However, this is one of the things that are more difficult in (MS)FS than real life, because the wind can change from 50kt tailwind to 50kt headwind in a split of a second. This hardly can happen in real life, while it is a common phenomenon for anyone flying online (and happens, in a lesser extent, also when flying offline wth real weather).
In this condition, we simply cannot require pilots to avoid overspeed (or even stall if the wind changes in the opposite direction), because there is no (known) method to do it. The best we can require (and apply penalties to those who don't) in this case is getting out of overspeed / stall within a short, given time.
6) As FL4 was mentioned at some point and Eric-Jean asked about it, here is a rough story of it:
I started the project 2 years ago but I must admit it did not always run at a fast pace. Also, it took me some time because I had lots of technical stuff to learn, both FS-related (like handling FSUIPC) and related to the broader .Net universe.
At this time the FL4 client can be used (and it is used, by me, as stated in some flights comments) to record non-PP flights. Along with massive technical improvements, which are not visible to pilots, let me point out a few new features (that are already developed, so I take no risk in naming them
):
- Possibility to save a flight in progress and resume it at a later time;
- Possibility to break a long flight as two connected flights, or land at the alternate and THEN resume to the main destination;
- Possibility to completely perform a flight (including PP) wthout a running Internet connection
- A more comprehensive flight selection mechanism.
So, at this point there is still to develop the whole PP bonus/penalty/scoring system and the user interface also needs a "facelift". This is why this thread about PP scoring happens at the best possible time, because in a few weeks we will have decided what to develop.
I hope I did not get anyone bored with such a long post
Andrei
