Haven't tried this yet, but hopefully when i've finished building my homebrew control unit (with steering/pedal box/shifter and the usual basics on for the 'land wheeled' module - i'll get around to it.
What will be interesting though, to discover, is how well you've executed and avoided the usual 'rookie' errors most attempts incorporate. As with supporting physical throttle controls or braking (where you override the usual control logic to inject new more progressive changes that are proportional but not necessarily linear in change vs effect) it'll be interesting to see how proportional the effect of control input will be, as that will ultimately play a big part in how it 'feels' on physical steering.
But in fairness, it appears you've covered or considered a fair few detailed aspects based on your description and itemised list of features, so there's certain hope of seeing some good execution.
Of course, that's assuming i ever stop trying to be the inherent and obsessive perfectionist and actually test the modules on a game...
@NotCrunchyTaco - well, at the end of the day, go with your instincts :)
After all, it's your pet project and it shows adequately what you were aiming for and isn't broken, so any fine tuning and refinement is your place to decide what needs it (if any), not us lot ...
After all, what to me feels 'synthetic' despite it fundamentally working and is usable - that may well be good enough for most and if they are happy with it.
I sort of figured it might be using a TP approach somewhere, but i guess (having not messed with using 'teleport' type functions in modding, although it's quite a basic fundamental method to drawn/erase/redraw offset in basic animation and motion simulation) that was the starting point that was either making the most sense to test in principle or was simply an easier jumping off point to start coding with. But that said, whatever the motive for how you did it, you had to start somewhere and at least you stuck your neck out to try a better than just-about-works attempt to get initial outside impressions.
@Cryon43 I can easily understand tha annoyance, and yeah - had the 'arrogant bst*rd' bit aimed my way too for trying to educate or at least bring rational thought to something overlooked. But that specific example of the keypress and filtering, i'm in total agreement over things like that, and no - it's not arrogance on your part for trying to educate on the importance of such things.
As for the limitations on available keys for controls and hotkey functions, that's precisely why i offload a lot of my 'keypress' functions to a dev board (Ardunio or Espruino or direct AVR/PIC setups) and use high speed serial compressed and encapsulated data over cable or wireless to feed control data back to C++ code at the PC (or whatever computer system in question). The fact you can go for broke with (in different arrangements and configs) pretty much whatever control pattern arrangement and tailor it to purpose according to means and resources and the most complex bit being deciding your encapsulated data protocol - for which i use something of personal spec which isn't dissimilar to that used by Firmata. It's extra effort and work, but in the long run, it works out better than some of the 'easy' quick wins that people employ to badly handle key press filtering etc.
As for GIT, yeah i'm aware of it and quite a few other versioning control setups - not a fan of them, but i do respect how valuable it is to have versioning control on bigger projects or simply where there's more than one person involved and working in a detached fashion. As for semantic versioning, yeah i used to follow that practice with my commercial developments in the telecoms and radio comms field - but rarely in my own pet projects - but for certain, if i was doing public releases of my projects or source code modules, i would indeed be using the framework for versioning. Again, i recognise the value of it and in the places where it's best used, i'm a supporter of such methods.
As for my skills and background in coding/development, it's all been on stuff that's long been under security restriction going back to the defrosted end of the 'Cold War' years. But amongst the unclassified stuff, has included development of digital trunking/transmission control and protocol design for radio and telecoms systems (usually for very long range private wire or secured radio links) - at the other extreme, building a very early self-guiding/self-landing drone (back in the 80s) running off an AP and INS based on 6502 and Z80 and 6809 processors - tested on a scratch build SST model that had to return home over a 30 mile glide path from around 10K ft altitude (getting permission was a nightmare), through desiging and building a very early example of a transportable 64-bit processor based multitrack studio that was (to all intents and purposes) fully laptop/desktop computing scale functional but had to be operated with virtual keyboards on a prototype touch screen and/or dedicated digital control panels.
So you could say, developer wise, i'm easily a newbie rating in game modding circles - but definitely not when it comes to the real world applications where bombproof and 'it just works' as perpetually as physically possible is concerned.
But it's definitely nice to see some rare animals called serious developers in these circles, even if their best stuff is undervalued or lacks recognition by many. But, in my experience, a lack of recognition of value or purpose never stopped the serious chasing their perfection or the desire to make it worth the effort.
After all, where would we be today if public perception (and often ignorance) defined technology and innovation?
Now this is yet another excellent example of what R* could easily have implemented, rather than their sorry excuse for 'what you are driving, or what you last drive' based ambient traffic spawn basis.
Would it have really taken them that much to employ this approach which effectively makes good their poor excuse - not really, but their effort speaks highly of how they were more interested in 'getting it out' than creating something that really shone.
Attention to details, making good what's bad or badly implemented by R*, or simply doing something sorely lacking on every level and is plainly a good useful idea - that's the essence of a good quality mod.
So the 'quality' of this one speaks for itself.
@Cryon43 - well, the mod (kinda feels like it's downrating it to call it that) itself speaks highly of you and what you could find (or maybe have found already) is what most mod developers overlook, that the attention to detail and efforts to make their pet project the best it can be technically as well as in execution is solid evidence of why the author should be given serious consideration for 'real world' projects, because the author who puts 'real world' effort to evolve their project this far is surely a candidate for bigger and better 'real world' professional level coding/design or at the very least worth having on a team to be consultant/quality control 'god' who's proactive as well as authoritive when it comes to ensuring that the project is bombproof and an excellent piece of work.
I mean, i considered the V version to be a benchmark (in it's different stages) to which my 'bot' projects for GTA V (notably the unreleased 'insurgentbot' hybrid mod) had to at least match in refinement and stability.
So even if the finer points of this mod and what it represents goes over the heads of many like an anonymous screaming warbird, you can bet safe money that some (including myself) realise how much time, effort and attention has gone it this and without that it wouldn't be what it is.
Having tested the current version of this mod, has inspired me to resurrect the buried/sealed away for eternity bot hybrid again and do some more with it with testing on 5RB - so it's more than simply a good release.
Given it's history, and being the original (as far as i known) AP/AD mod for V (so i'd assume maybe the case for IV since the author's working of it it goes back to then), it shows that what's already good from the off is worth the effort to maintain and refine/tweak where it could benefit from the effort.
And reading the updates list over time, and particularly the updates for this release, sounds like a case of refining/tweaking with purpose and getting nearer to the ideal the author wanted.
Certainly has been a worthwhile mod to develop for the author, if the execution demos anything beyond the fact it works well.
Notably, from the outset of the V version, unlike a lot of such AP/AD type mods - this one was always pretty resource friendly, and unlike some coding efforts, plays nicely on a resource level and doesn't interfere with other things by being a resource/thread hog. And when you think about it, just like any decent 'AP', it shouldn't be a resource hog that's intrusive beyond necessary evil levels - so that's evidence of the thought that went into it and (i'd assume) some good insight into how AP's work or some good info to work from, and managed to connect it well into what exists for ped/entity automation in the game.
It rates a lot of stars (if i was into doing star ratings) just for the combo of execution and persistence to keep it alive with improvements and fine-tuning/refinement.
You could be tempted to comment about the 'execution' - but it does what the author intended, and in my mind - if you think you can do a better execution, go code :P
Seriously, not my kind of feature add-on and maybe feels a little synthetic, but it works and if the author is happy, job done. But what it does highlight is how poorly implemented what rolls already exist in the R* code actually are. All this needs, at best, is a little tweaking to refine it (if the author feels it's worthwhile or needs it) but unless that's the case, pretty good result for something i normally take zero interest in.
If anything, combining the functional features of this with Jedijosh's Parkour mod would be a good combo for the overall effect, or better still, literally collaborate and create a combined mod incorporating both..
Nothing to criticise over, but still worthy of noting the effort and general execution as good/interesting as it was kinda a missing link (along with the parkour mod previously mentioned) in the original code.
Thinking back, the whole handling of climbing/tranversing and combat rolls/dodging did somewhat fall short in V compared to what existed in IV.
Well, a thanks is due regardless of whether anyone actually wants it - it's the thought that counts, and anyway, someone will use it to learn from or tweak and modify to suit their needs.
Like i say about the microcontroller and web-interface and android interfaces to mods and the vastly overlooked few examples we've had on here - it's not about whether it's literally a mod or inherently game-content/game-ready that decides what's worthy, it's as much about anything that can help someone in the modding arena (and in this case, way beyond that) is worth the effort of uploading and posting.
If you want a challenge - try recreating fully (and i do mean fully, as there's lots of odd features to it that often are overlooked) the venerable overkill piece of post-war hybrid construction :-
Convair B-36 Peacemaker or a B-47 or B-52.
If the gunner's positions work correctly, and guns appropriately, that would indeed make a visually good recreation complete - especially with the correct engine and exhaust note and drone (Merlin engines had a distinctive tone, as did the different prop/screws used on different Merlin equipped aircraft and boats and a very distinctive exhaust note with the obliglitory knack of shooting a jet of flame if the engine backfired).
Personally, though, i would have thought a B-17 or B-29 or a Liberator would have been more in keeping of an historic warbird bomber to be found in a virtual American airfield/airport.
In the same way where a Spitfire, Hurricane or Typhoon would more likely be replaced by A P51 variant (for Spit/Hurricane), or something like a ThunderBolt or Wildcat in the setting.
Just seems a bit at odds with the setting, a Lancaster.