Forum Replies Created
-
Posts
-
Instead of a special option, I’d prefer a more generic solution, so we don’t run into the same problem with the future aircraft releases (and so we can “fix” all the existing aircraft that don’t really work out-of-the-box with CLS2Sim). Give us a way to control which sim variables are used and make this part of the aircraft profile — then only one of us needs to be smart enough to figure out which variables to use and everybody else can get the profile from the profile cloud (which would make the cloud so much more meaningful).
I don’t know what the best solution would look like, whether we should be able to pick variables from a list, or use RPN notation, or write LUA scripts, but I’m sure Brunner could take some inspiration from tools like FSUIPC, MobiFlight, Spad.next, or Axis and Ohs.
In case you don’t follow the discussion on the A2A forum, Scott Gentile posted this:
“It turns out we need to take another path for Force Feedback to work, through Asobo likely. We haven’t yet made this request as we’re deep into other things but we will and keep in mind they are busy too so it will likely take a while before, if, anything is done. We do want FF to work. It is important to us so we won’t be completely dropping this ball.”
As I wrote on the A2A forum, there are more people with VR headsets than people with force-feedback hardware. VR hasn’t seen any TLC by Asobo since SU5, when MSFS arrived on the Xbox. I find it hard to believe that they would divert developer resources to FF. Especially when there is no FF device officially supported by the sim (afaik). Especially at this point in time, when the deadline for MSFS2024 is inching closer day by day.
I think I understand Scott’s approach. He probably wants FF to feel right, and to get that, the sim has to supply some raw data like forces acting on control surfaces and what have you.
On the other hand, at this point, I would be happy if I could just use the Comanche incl. AP with CLS2Sim’s usual speed and rpm based approximations of forces. (Or the PC-12, or the DA-62 and 42, or the cool legacy AP in the Wilga.)
The number of MSFS planes that don’t work “out of the box” with Brunner hardware appears to be growing.
I believe, we either need Brunner to step up and make CLS2Sim more flexible, maybe by letting us configure which variables to read from and write to the sim, or we need some (open source) tool that replaces CLS2Sim or puts itself between CLS2Sim and MSFS.
tbh, I don’t have much hope. We are few: not many people use Brunner hardware, only a subset use MSFS, even fewer fly the Comanche (or the other aircraft I listed), and only some of them are not happy with workarounds like “use Vjoy and never touch the AP”. So the Comanche sits in my hangar as a reminder never to buy an aircraft unless other Brunner users confirm that it works as intended with my yoke.
Oops, when I wrote “Trim HW – elevator left/right”, I obviously meant “Trim HW – elevator up/down”. Brainfart.
Nice! I’m gonna try hardware trim then.
edit: For SOME reason, I can’t set up any buttons to control the hardware trim. I checked “Use hardware trim”, I applied the setting, I switch to the Buttons tab, but “Trim HW – elevator left/right” does not shop up in the list of options I can select. Only “Trim HW – Rudder left/right” and “Trim – elevator up/down”. I can’t figure out why.
Hey Jaybird, thanks for your response!
To work around issues with career add-ons (OnAir, FSE) which could not see the engine running because it’s simulated externally to MSFS (that’s what Accu-Sim 2 does), A2A added a tiny engine with a tiny propeller to the model. So tiny, that they don’t affect the flight model, but their mere presence is enough to get the simple data the career add-ons need (basically they just want to know whether an engine is running).
So currently, CLS2Sim gets some spurious data from the sim (like the thrust from the mini engine), mixed with real data (like airspeed), and the result of CLS2Sim’s calculations are probably correct for certain features, and out of bounds or wrong by orders of magnitude for others.
I have to test the AP issue again. I get the constant Trim Down warning (and if I act upon it, the AP will suddenly disconnect with the plane being trimmed perfectly for a death dive). “Use hardware trim” is unchecked in all my profiles, so there must be another setting that affects it. My main issue with the AP is, though, that CLS2Sim has no way of knowing if the AP is engaged, so it can’t behave differently when it is (“Don’t set pos in AP mode”), which can result in unwanted behavior in certain situations (when AP and CLS2Sim start fighting over control of an axis).
I suspect that forces in flight being “different” is largely caused by the missing prop wash effect.
I recently tried it, out of curiosity, and, no, currently XPForce doesn’t seem to be an option. It complains about not being able to create certain DX FFB effects. But even if it worked, my impression is that it’s not as comprehensive as CLS2Sim, which allows for A LOT of fine-tuning. XPForce uses the same method of getting data (SimConnect variables) and it’s also closed source.
So, yeah, even if you could use XPForce with MSFS, I doubt that you would want to.
The DX FFB support is probably most useful for older games that are not supported by CLS2Sim.
I’ll copy my response from earlier today on the SimWorks Studios Discord:
In the spirit of “teach a man to fish…”, let me tell you how I approach creating a profile.
(I hesitate to share my profiles in the cloud – I have no advanced knowledge, I don’t fly a Kodi IRL, so why would my profile be better or more realistic than anything you could create? I did upload some of my profiles before, and later found out that they’re buggy, but I can’t delete or change them now, because the “Brunner cloud” is… erm… peculiar.)
Effects / Yoke physics: I get the thrust settings by loading the aircraft in the sim on the runway, and copying the “Current engine thrust” for idle and max thrust. I let the yoke be pulled forward be Elevator weight, and eyeball the settings (for the Kodi: Pull force% -20, aerodynamic speeds 15/35), prop wash affects Elevator.
Motor vibrations, ground vibrations, turbulences: whatever feels good for you. For me, it’s very low motor vibrations (for the Kodi, I use 5% force), 5-15% ground vibrations on “even” ground. I don’t use the Turbulences effect. The Kodi doesn’t have a stick shaker, I think.
Buttons: most of them are “not bound” because I assign them in FSUIPC, 7 is Disable AP, 9 and 10 are Trim elevator up/down. I don’t use hardware trim, I guess that’s a matter of taste (another reason why I don’t like to share profiles – the profiles contain a mix of “hard facts” based on physics and personal preferences, which seems kinda stupid.)
I don’t touch anything on the “Analog Inputs” tab, everything is “Not bound”.
I use the “Axis Range settings” tab mostly to limit the movement along the roll axis, to match it to the yoke animations of the aircraft. For the Kodi, the limits are 100%, but other aircraft might only turn the yoke 80% in either direction. I have no deadzones set up, I leave everything else at the default settings.
Axis force settings: my approach here is very simple: since I have no way of finding out what forces are needed in the real aircraft, I base it simply on expected airspeed. I want the yoke to be “loose” at low speeds, “firm” at high speeds, and really hard to move at overspeed – it’s how I imagine the real yoke to behave.
I “open” the windows for Pitch and Roll, switch to the Force Scale Factor tab, leave everything as is, I just set the “Max Speed”. For the Kodi, I use 220 kts for both axes. For the 172 I would use 120 kts, for a personal jet I would use something like 450 kts.
If the result is “too firm” or “too loose”, I just change the “Max Speed” as required.
AP settings: Enabled on Pitch and Roll, Velocity 20, Force 1000, no Trim to zero, “Don’t set pos in AP mode” is enabled, “On overpower disable” is set to AP&FD (personal preference), Enhanced AP mode is disabled.
Hydraulics are off, Shutdown is set to my personal preference (it shuts down both axes), Trim is enabled with default settings (no hardware trim, personal preference, again). And my Yoke doesn’t have a display, it’s a cheap CLS E NG.
I hope there’s something in this wall of text you can use.
When I have an issue that requires feedback, I use the helpdesk, and the support has been excellent in every case.
But, yeah, the forum appears to be the red-headed stepchild that rarely gets attention. 😀
It’s probably meant as a “users help users” kinda thing.Brunner people, can you add a “yoke shakes on touchdown” effect, independent of Ground Vibrations, to the Effects tab, pretty-please?
Thanks for considering it.
I updated my Cloud profiles for the default C172 and DA62/DA62X for MFS, and I added a profile for the upcoming Twin Otter.
Use the release date or the CLS2Sim version number to identify the latest versions, the profile names are not unique (I tried to overwrite or remove the old versions, but couldn’t).(I haven’t tested the profiles with firmware v576 yet.)
Yes, the default curves were not linear, that’s why I used an “opposite” curve to correct for that, so the physical yoke movement matched the in-game yoke movement.
After SU7, I don’t need that “opposite” curve any more for the Asobo aircraft I fly (C172, TBM9, DA62).
For 3rd party aircraft, I still need the correction curve.I’m pretty positive that something changed, but I’m not sure whether it’s something in the sim or just how Asobo set up their own aircraft since SU7.
And, since you mentioned it, I haven’t launched P3D or XP in a year. My brain, spoiled by MFS, does no longer accept their environments as believable, which makes quite a difference in immersion when flying in VR. The vast majority of features a sim needs is present in MFS and working. Asobo is still struggling to find a good release rhythm, a compromise between quick development and not breaking shit with every release; they’re making progress though, e.g. by offering access to beta builds to all users for the first time now. I do hope either Lockheed-Martin or Laminar are able to provide a realistic (and pretty) world by the time Microsoft loses interest in flight simming once more, so I’m happy about anybody who’s sticking with those sims and keeping them viable.
I decided to do some more testing and suddenly it works. It took a restart of the sim. I guess either the sim or CLS2Sim was kinda confused before. Or maybe the AP detection becomes unreliable when re-connecting or modifying the profile several times in a row.
