Something weird going on with disappearing controls on CW

[imacken]

Haven't tried Spitfire missions for a while so I don't know when this behaviour started.

If I look at the Spitfire binds either from the main DCS UI or after loading a non-CW mission, I see the correct bindings - s1 attachment for an example - and they work in CW and non-CW missions OK as was always the norm. However, if I load a CW mission first after loading DCS, then all my binds disappear from the Spitfire - see c2 attachment - for all missions.

I don't have a CW option in the controls dropdown - see s3 attachment - and I have never seen this behaviour in all the years I've owned the Spit module.

Anyone able to throw some light on this?

[saburo_cz]

Be sure you have both folders with controls for both planes. (in C/user/saved games/DCS/Config/Imput/...)

If not, make missing one and copy there files from the first.

[imacken]

Thanks. 
Sure, that’s a workaround, but it’s not the answer to the issue. The point is that for years it worked, and now it also works if I select a non CW mission before a CW one. Then, all the binds work in CW and non-CW. If I select a CW mission first after loading DCS, then all binds in both CW and non-CW are completely gone!

Also, if I adopted the copy and paste approach, how could I make a change to CW binds without changing non-CW?

[Art-J]

5 hours ago, imacken said:

Sure, that’s a workaround, but it’s not the answer to the issue. The point is that for years it worked [...]

You're one lucky bastard then ;), 'cause I've NEVER seen the controls loading for CW at first mission properly for as long as CW is available in DCS. Granted, I've been using the two-folders-workaround ever since it was advised on the forum years ago, but every now and then I temporarily leave only one folder to see if maybe, juuuuuuust maybe, ED have finally fixed that bloody issue in Spits (and Mustangs for that matter - same problem).

And as always, turns out they haven't, and I see exactly the symptoms you described in your OP.

So, for the time being, two-folders workaround seems to be the only bulletproof solution for both Spit and Mustang. 

5 hours ago, imacken said:

Also, if I adopted the copy and paste approach, how could I make a change to CW binds without changing non-CW?

I might have misunderstood, but why do you want to have  different binds for non-CW and CW if they're functinally exactly the same aircraft? Honest question.

When both folders are there, I believe changes made in control options menu are being saved by the game in non-CW folder only. That's why they have to be manually copied to CW folder outside of the game.

Edited August 14, 2022 by Art-J

[imacken]

Ok, so good news for me, and maybe you guys as well.  I resolved my problem by loading the profile for each device WHEN IN A CW MISSION from the SpitfireLFMkIX folder.

Now, the CW version has all the bindings correct at all times, even on first mission load.

Even better, I can confirm that any binding change made in either non-CW or CW is applied to both successfully.

I hope this will work for you as well.

 

[Art-J]

Sounds promising, will give this method a try. If it solves the Mustang problem as well, that will be a breakthrough discovery.

When importing the non-CW profiles, did you do it while in the first-load-CW mission?

[imacken]

Yep, I did from the first mission where all the bindings were blank. 
 

Still working!

[imacken]

Just to let you know, it works for the Mustang as well!  When in a first mission with blank binds, I loaded the device profiles from P-51D-30-NA folder and all is working perfectly now.  Binds are all there, and any change go to all variants.

I hope this works for you guys.

[saburo_cz]

This "controls separation" was done at least year ago. For the Mustang and the Spitfire.

 

[imacken]

Thanks, but not sure how that explains the behaviour we are discussing., i.e. no binds on first CW mission, but CW binds there after loading a non-CW mission.

[Nealius]

Years ago the controls were split into two folders as pictured. Then a short time ago things were merged into a single folder. Same for the Mustang variants and (IIRC) recently the P-47 variants (which has led to a lot of confusion with bomb arming for me). I also had the same issue with the Spitfire CW and/or standard keybinds going AWOL, posted here but ED was unable to reproduce.

So, to recap, creating a separate folder to match the old, pre-merged status quo, resolves the problem?

[imacken]

I was interested in what is going on here, so I made some investigations.

I deleted (renamed) my Input folder so that DCS would create a new one on demand.
I loaded DCS and started in a non-CW mission, made a bind, and a SpitfireLFMkIV folder was created in a new Input folder
Quit that mission, loaded a CW one, then the bind above was there from the SpitfireLFMkIV folder
I then made a bind while still in CW mission, quit, then loaded a non-CW mission.  Bind was there and still only SpitfireLFMkIV folder.
However, I then quit out of DCS, reloaded, then started in a CW mission, there were no binds.
I made a bind, and then a SpitfireLFMkIVCW folder was created, and the binds applied to the non-CW missions as well.

So, conclusion 1) is that whatever variant you start in, the binds in the appropriate folder apply in that session to both variants.
This is awkward because, for example, if you start in a non-CW mission, make a bind change, then the next time you load DCS, if you start in a CW mission, the change will not be seen. This means that you have to make any bind changes to both variants by starting a mission for both on a DCS loading, or copy and paste the relevant folder.

The next question was, what does selecting from the main DCS UI menu ‘Options>Controls>SpitfireSim’ do?
1) If no mission has been loaded, or, if you loaded a non-CW mission, then changes made there are directed to the SpitfireLFMkIV folder and apply to both variants within that session’s missions.
2) If a CW mission has been loaded and quit from, then any changes in the main DCS menu are applied to the SpitfireLFMkIVCW folder and apply to both variants within that session’s missions.

In conclusion,
1) Binds apply to both variants and are dependent on what mission you have loaded first. If it was a non-CW mission, then the binds in SpitfireLFMkIV apply to both within that session. If it was a CW mission, then the binds in SpitfireLFMkIVCW apply to both variants within that session.
2) Changes made in the main UI Options>Controls menu are applied to SpitfireLFMkIV if no mission has been loaded, but if a mission has been loaded then changes are made to the variant folder of the mission that was first loaded.

All in all, this seems like a mess, and I’m pretty sure that this is not the way ED want this to work.

I feel we should raise (or update) a bug report on this after this. Unless someone else has already come to the same conclusion as me, of course!

Any comments appreciated.
 

[Art-J]

I recall there was a dedicated bug thread about it long ago (I think I might have posted in it), though I don't remember if in Mustang or Spit section of the forum. Can't seem to locate it at the moment.

There was no such thorough investigation presented in that thread, however - thanks a lot for checking and noting it all down. These observations should definitely be posted there as well, or in a new thread?

Makes me wonder how ED managed to avoid such a mess in Thunderbolt - we've got 3 variants to be used as separate units in the mission editor, and yet, they all load the same custom profiles from a single folder just fine, every time. So, it's doable in the source code, somehow, but why does ED struggle with converting Mustang and Spit to the same algorithm?

[imacken]

Thanks for that.

I'll post a new bug thread here unless you can find the old one.

[scoobie]

Thank you for taking time to investigate, @imacken, you're truly a hero, I'm not kidding.
There have been so many threads about it in the Spitfire forum and probably as many or more in the Mustang forum.

Some of them are general (and it is a general problem), like this one:

or this one:

 

Others are "incidentally specific", like this one:

 

(It's the same thing, but the OP simply stumbled upon it trying to bind bomb release button.)

Many of these reports are marked "cannot reproduce" or not marked at all.
Well, at least in case of Spitfire it's obvious the whole premise is "bugged", regardless of the code, because: 1) launch DCS, 2) open Controls menu, 3) change bindings for the Spit, 4) anwer this question: "Which variant of the Spit have you just modified?". This alone means the bindings CANNOT work as expected by the sheer power of Mother Logic. It's like 2+2=4. Now, if the bug reporting procedure is such that someone from ED must reproduce it or "it didn't happen", then, well - congratulations, sticking blindly to the procedure prevents ED from looking into and fixing an obvious bug bothering people every now and then for a few years now, if we count it collectively with the P-51. As a result, people are wasting their time, NineLine is wasting his time and every time the procedure gets stuck. I know that by and large procedures are neccessary or absolutely neccessary in many places, but when it happens that an otherwise sober procedure fails in 1 in every 200 cases, one simply needs to take a "detour" around the procedure. This is clearly one of such rare cases.

Unfortunately the issue is more complex/nasty than that because the Pony variants -25 and -30 do have seperate folders for bindings AND separate "caterogies" in Controls menu to define these bindings, so in theory the Pony should work just fine. Alas, it has the same issue - sometimes new bindings "disappear". In the Spitfire's case it's obvious it can't work every time, by design. And this alone should be enough to pass a bug report to a dev (procedural "emergency detour"). Hopefully if someone plunges into the code, he will discover why also the Pony is bugged the same way. Maybe it's a common code - writen for the Pony, then copied into the Spit? At least is smells like it.

Fingers crossed

EDIT: Sorry. It would actually be more productive if the "investigation" was started in the Pony as she was the first one to get this bug. Spitfire is a more complex case - it has (most likely) the Pony's bug, but on top of that she got the "variant selection problem". So it's better to start at the source, the Pony that is.

Edited August 18, 2022 by scoobie

[imacken]

I've just posted all this in the Bugs section here.