All… I gather this is a known issue and will probably be ignored, but I thought I would make a bug report anyway because it SHOULD BE considered a serious bug. If you are inclined to support it, use the link below.
Text of Report:
Bottom Line: the performance of the instructor is HIGHLY variable. This creates an unfair advantage for players who fly aircraft with better instructor performance versus their competition.
The Yak-3 features very good instructor performance near the limit of what a human with “real” controls can achieve. This should be the standard - the unaided player can access a little bit more performance, in exchange for greater skill/risk. The instructor pilots the P-51D series almost, but not quite, as well. However, the performance of the instructor is ABYSMAL in the F4U-4, the F6F-5, and the Bf-109 G6 to name three.
The logs and test data for all of the aforementioned are attached.
Also attached are graphs/plots showing the turn performance with/without the instructor and plots comparing the P-51D-10, F4U-4, and Bf 109 G-6 to the Yak-3 benchmark with simplified versus real controls. The performance balance of the aircraft changes SIGNIFICANTLY based on the instructor performance. This really shouldn’t be considered acceptable.
You cannot balance flight performance without also balancing instructor performance.
FWIW, I also tested the A7M1 and La-7… they are gimped as well, the La-7 arguably a bit worse than the A7M1. In fact, the La-7 instructor is Corsair levels of bad… 4+ deg/s of turn rate across the board.
I do agree that it looks like that a certain plane has an “optimized” instructor, and others not - i am not surprised by this 😎
What i do not agree is your a little bit too simplistic view on players which don’t use mouse aim and the instructor. Why?
Using mouse aim is basically flying with an auto pilot, the mouse indicates the intended flight path and the instructor controls the aircraft - and not the player.
Mouse aim and the instructor artificially stabilizes the aircraft in all relevant flight conditions except turn performance - and is unbeatable in head-ons or often overlooked when battle damages occur due to this artificial stabilization.
I recommend to test this in 1 vs 1 duels - i have done this years ago with my son and the accuracy gap and the flight behavior when damaged is far away from being a minor advantage.
So your claim that non-mouse aim pilots would have “a little bit more performance” is only relevant in pure 1 vs 1s - whilst in every other metric the mouse aim / instructor pilot plays basically with advanced training wheels and has a massive advantage.
The real solution for the issue described in your bug report is not to pull mouse aim / instructor pilots to the level of let’s say passionate pilots whilst leaving a small turning advantage for SFC or FR pilots - the solution would be to define a certain and fixed gap for the instructor limitations for all aircraft to eliminate the possibility that certain aircraft from certain nations were either intended or unintended buffed or nerfed whilst keeping a “healthy” gap between mouse aim and SFC/FR players.
Nevertheless i upvoted your bug report in the hope that gaijin kills the severe disadvantage for the planes you mentioned. 👍
My view on mouse aim/instructor vs SFC / FR controls:
I wrote some time ago that my total number of kills in head-ons is about 20-25, then 50, and now i would say in the range of 80-90 - out of more than 20.000 player kills. All of the other kills were either BnZ/BnR attacks or deflection shots whilst turnfighting in isolated 1 vs 1s i created on my own.
The increase of head-on kills is a result of hundreds of matches on Pacific maps in Hellcats and (mainly) P-47 variants in which i was forced to perform really long range head-ons at very high speeds as it is obvious that you can’t reliably energy trap stuff like A6M5s, A7M2s or Ki-84s as these bloody aircraft refuse to stall - and turn better😂- so my sole advantage of not using mouse-aim / instructor was not applicable.
If you watch vids from basically all CCs on yt it is obvious that a large chunk of their kills are rooted in long range head-ons followed by stalling out rookies. As a non mouse-aim / pilot you can’t afford to get hit in basically all aircraft as even a slight yellow wing tip ruins your turn advantage into a disadvantage as your advantage vanishes. You are forced to avoid the most effective tactic to get kills (=head-ons) due to the severe accuracy disadvantage.
Another major advantage of mouse aim / instructor is the massive increase of situational awareness by using keyboard for turning whilst looking around with mouse. I think a lot of players remember the Adam the Enginerd vid dancing with 4 enemies at the same time in his Ki-43 - you can’t do this whilst using a Hotas.
It’s not Statshark - this is data from WRTRI, from planes I flew in custom matches and logged/recorded.
A case can be made that instructor limits could certainly vary based on the skill/difficulty required to fly an air frame near its limits. I vehemently disagree that Instructor performance should be so variable as to change the character of the aircraft or it’s performance relative its peers.
Here is the F4U-4, Bf 109 G-6, and Yak-3 flown with realistic controls.
That’s pretty competitive performance that comes down to pilot skill and creating/using the areas where your aircraft is better to create an advantage and win.
Now here are the same aircraft with simplified controls.
Just like that, we’ve gone from a balanced match-up to one where the Yak-3 is clearly superior across the board. The difference is especially bad for the Corsair which loses its low speed rate advantage AND gets crushed at high speed.
Different instructor limits should not have that sort of effect on balance, relative flight performance.
I fully agree with you - players doing the actual flying SHOULD have a performance advantage.
Being able to hold the cursor on the screen and “click him out of the air” with as much difficulty as opening a document on your desktop does need to be balanced.
I don’t think I am qualified to say how much of an advantage the player actually flying the plane should get, but no argument that it should be there.
Ki-84 is probably the most held-back plane i’ve tested personally. 109s must be close since the wing slats don’t even deploy ever under Instructor. This also seems to affect how hard you can turn when your controls are compressing.
Damn, you mentioned the Ki-84 and I meant to test that as well. I think I’ve got AoA in my data, but not sure if I have critical AoA in there.
I didn’t do a “E” or “F” series 109 yet. Looking at the G-6, it’s about 4 deg/s from 550 kph to 250 kph (as low as I tested). That’s a 33% turn rate difference at 250 kph, a 25% difference at 300 kph, and 20% at 370 kph… so not insignificant.
Something worth adding and not shown in the graphs:
Coordinated Flight/Turns.
I recommend paying special attention to the rudder while turning with mouse aim, using keyboard with instructor in mouse aim, using autotrim + realistic and of course full-realistic controls (while actively managing rudder input to keep the ball centered).
Taking my Mustangs to RB I’ve noticed that Mouse Aim seems to get confused on whether you’re trying to actively turn or just make a minor adjustment with rudder to get a shot off. This leads to aircraft flying uncoordinated (the nose points out of the airstream introducing additional drag due to the fuselage being in the way and reducing the maximum turn performance by causing one wing’s AoA to rise faster than the other’s).
This has the consequence of more significant energy bleed and may contribute to highly different instructor limits. Given 109s are mentioned, I feel it’s worth noting that those planes slip in quite an extreme manner making rudder inputs all the more vital (they’re quite prone to assymetric wing stalls if you get lazy with the rudder especially from gyroscopic precession and changing throttle). I wonder if inability to make the rudder stay coordinated is why 109s are so limited with instructor.
Critical AoA is very important since in most planes, hitting 100% with Sim controls results in an accelerated stall (you turned too hard and one of your wingtips stalled, so you experience a sudden roll). A few aircraft are a little touchier, others less.
Also probably the easiest way to measure how much of a plane’s turning ability you’re using at any given point.
They do. 109 F-4 is UNABLE to pull hard enough in RB to rate at its best speed. In Mouse Aim it will lose a rate fight to a Yak-3, while in Sim it will win that fight.
In Sim you also have much greater instant turn ability, especially at low speeds where the Instructor struggles a lot to make the most of your plane’s abilities. The Zeros are very held back here whereas in Sim you can pull the nose around effortlessly.
That’s a solid point… the Mustangs bobble the nose around almost comically if you don’t use the keyboard to roll the aircraft first.
I wonder how much more performance I left on the table… since my uncoordinated, non-flying self used no (or no intentional) rudder input during testing.
@AlvisWisla - the instructor absolutely does change attainable/observed ITR/STR.
@PercussionCap
Yeah… I had heard “wing stalls” mentioned and really had no idea what those were/meant. Then I did this testing and realized pretty quickly that the two little brackets in the virtual cockpit shouldn’t touch… or I’d get (what I assume is) a wing stall.
I had to retest the Yak-3 because I stalled it so many times during my first test. I was still really, really struggling when I tested the F4U-4… but managed to keep that one in controlled flight. Testing was a little smoother there after… but a better pilot could definitely get a little more out of these planes. Tracking the Critical AoA would tell me how much I’m leaving on the table, so I will do that.
ETA: here’s a video for the Corsair. I re-hash some of the discussion here, so not necessary to read the description.
I can post the plots of turn performance (rate and radius) and critical AoA that track the video if desired. Sustained with instructor is about 19.3 deg/s @ 330-335 kph. Sustained without is 20.8 deg/s @ 300-305 kph. Average TR for the first 1620 deg was 20.8 deg/s with and 23.1 deg/s without. Max ITR w/ instructor was 23.6 deg/s @ ~540 kph. Max ITR without was 32.1 deg/s. @ ~582 kph. That’s a 36% difference.
This is extremely noticeable and even detrimental on some jets if you want to make minor corrections to the direction you want to go, and especially if you pitch downwards. There’s been several times where I move the mouse down and the plane just doesn’t move, or even rolls side to side trying to roll downwards.
I’ve also had very low speed vertical maneuvers be messed up because it can’t use the rudder to rotate the plane to go down again.
It’s also incapable of making a proper turn at all, it either goes into a very steep turn, or just uses rudder.
Apparently they are called “pitch indicator lines.” The harder I turn, the closer they get to one another. When they touch/cross, the aircraft leaves controlled flight. When I fly these tests, I try to bring them as close together as possible… but I am far from perfect in that regard.
It seems to vary slightly by aircraft… the harder your pull, the closer they get. If they touch you might be OK… if they pass one another, you’ll lose control.
Related: I did test the Ki-84 Ko… I really struggled to push the limits without departing controlled flight compared to some of the others. I seemed to lose control before the pitch indicators actually met.
