Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[SAS~Storebror]

Thanks for the comprehensive explanation Herra.
I see two issues here:

• The IL-2 sound coding unfortunately is pretty complex and results quite often differ from what you'd expect. It took us literally months to get a more or less stable implementation of a lock tone for our guided missiles, without loosing it when you leave your cockpit and jump back in. Not for that we're all plain dumb but what has been coded wasn't what came out of the speakers. See UV3 sound mod: A very nice thing if it's working for you. If not you're lost. It's not that the coding contains specific errors, it's just that the right coding gives random results when it comes to sounds in IL-2.
• You're talking about flyby sounds only.
  I think another very important point would be the sound played when e.g. you're on ground in your aircraft, attacked by some transsonic aircraft, or quite similar when switching through external statics.
  The thing is that currently (irrealistically) you can hear jets approaching you regardless their speed.
  Transsonic low-level attacks are a matter of surviving or not and that's not modelled correctly yet. In a real world, the transsonic attack is inaudible unless you hear the boom, either the mach one or the attacker's bombs
  Now the problem is that in this case the "time to boom" is variable and we would (at least) need to implement some way to check whether you're in front of the aircraft in question or behind, leaving the mach cone aside for a moment. This will become extremely complex the more aircraft are around.

Just my $0.02

Best regards - Mike

[Herra Tohtori]

Yeah, unfortunately my level of coding skill and particularly understanding of deeper workings of IL-2 code is superficial only.

I have no way of knowing what can or can't be done with the engine (though in some cases, I have an idea). I can only help by offering my knowledge on "how things should work" (in ideal world)...

The things you mentioned about sound to static cameras and other actor point of views is completely valid, but probably a few orders of magnitudes more difficult to accomplish than what I described. I specifically limited my consideration to the fly-by view since it seemed the most simple method of implementing SOME kind of sonic boom that would have a bit better correspondence to real world sonic booms.

I don't know if the fly-by sound is also played when planes fly past any static camera, or if it's limited to F3 view specifically.

Actually implementing sound propagation would (in theory) take care of all the issues with audibility in front of supersonic aircraft as well as sonic booms. But it's a rather difficult thing to do even in simulations, much less so in a complex locked-code sound engine environment that sometimes doesn't do what you would expect it to do... like you said.



Btw, now that I think about it, perhaps there's a way to modify the Doppler shift coding to do something like - if velocity of closure is higher than speed of sound, instead of making the sound have even higher pitch it would cross-over to silence? Currently when flying supersonic or close to speed of sound you hear a very high-pitched sound when approaching, which transitions to low pitched sound when going away from fly-by camera.

No idea if that's even possible (hooking fly-by volume adjustments to the doppler shift effect), just brainstorming here...

[SAS~Anto]

Look its a good idea, but it falls in the too hard basket. For the amount of time we would spend working on it, we could have made JetWar twice over. And it would be for absolute minimal benefit.

[Typhoon Ib]

naahnaaah...
its not about "work vs effect/profit"
its about the challenge to pull it off.

and with a true sonic boom, you will have more jet wars supersonic "jet noise is for kids" players.
so the sonic boom is like a catalisator for jet war, making it so f****good, people will not fly and  the LOMAC series, but Jetwar.

...think about it... belieeeeeeve!
XD

[Herra Tohtori]

Look its a good idea, but it falls in the too hard basket. For the amount of time we would spend working on it, we could have made JetWar twice over. And it would be for absolute minimal benefit.

Fair enough, but then to be honest I'd rather forgo the - pardon me - crude imitation of sonic booms that we have at the moment.

It just feels fake to me, and if possible I'd prefer to have an option to turn it off...

[DeJotPe]

As you can see: F-14 can't be heard before passing by, those marks on the water has hyperbolic shape (as you know, intersection of cone and plane is hyperbola) so if the front of the sound wave is cone shaped, not flat, not spherical in front of plane, the plane must be supersonic.
And when it pasess you can hear the boom!
And before pass you can't hear him at all!

So please, please! Someone could really fix it, and make it default part of jet era?

[SAS~Anto]

DeJotPe:
I stand by my above response

If there is anyone who wishes to take this on, be my guest. But neither Storebror and I will be looking into it.

Fair enough, but then to be honest I'd rather forgo the - pardon me - crude imitation of sonic booms that we have at the moment.

It just feels fake to me, and if possible I'd prefer to have an option to turn it off...

Be my guest to fix it. Took us a bloody long time to get the current effect to work.

[DeJotPe]

Well, i thought it would be like this...
Calculations of mach cones, doppler effects etc. in my head seems to be very difficult to imagine, even without Il-2 random-sound-code-interpretation...

Right now we have recorded some sound of engines, pitch of them are dependent of RPM and speed, we also have some sounds of planes passing by with doppler effect audible.
If we want to have "speed of sound" simulated we would only need the sounds of engines at few diffrent RPMs, and then the engine sounds would change their pitch accordingly to it's (relative) speed.
But at speeds just below mach 1 the pitch of the engine sound should be raised (almost) to infinity, which would lead to strange and unrealistic effect of something like "dog-whistle". That's becase whitout distortions you can change the pitch of sound by one or couple of octaves, but not by several octaves... Even the keyboards/midi uses few samples for diffrent diffrent pithes of the same one tone....

[Herra Tohtori]

In fact the Doppler shift would be fine as it is. I know in theory it should be "overdriven" to infinity, but, this being a type of singularity in nature, I expect the math equations for regular doppler shift to simply go a bit berserk if you insert v=c in there (where c is speed of wave motion in medium). And, if you overdrive the signal over clipping point (to produce the "boom") you probably won't be able to distinct the pitch anyway.

There are other ridiculous effects - such as, when you go supersonic you should in theory be able to hear your own engine sound in reverse from the time when you weren't supersonic and the sound waves propagated ahead of you and you're now catching up with them.

As for the effect itself, I once researched what needs to be done in post-processing to make the fly-by sound like a sonic boom. Here's what I produced:

http://www.youtube.com/watch?v=qtDJUpgpL3Q&hd=1#t=5m30s

What you have here is "relatively simple" amplitude modulations:

-during approach, amplitude of fly-by sound is ramped to near zero
-during approach, amplitude of fly-by sound is increased to over clipping level but then quickly ramped back to 1 (nominal value)

It relies on the original fly-by sound (although I think at this point I used Tiger's fly-by sounds); however I didn't do anything to the frequency of the jet sound, only amplification.

This is a simplistic effect. It lacks the rumbling thunder-like characteristics of a sonic boom, because those actually come from the shock reflecting echoes from distance - just like the real thunder. Distance softens explosions and makes them lose most of their high pitch. A thunderclap happening very near you is a loud, powerful crack, whereas distant thunder gains a low, long-lasting rumble. As evidenced by multiple videos on youtube and elsewhere, a sonic boom happening very near to you would sound much like this to recording equipment.


Now based on what has been said by people who are actually involved in coding with the sound engine, I understand that doing this kind of amplitude modulation and getting it to correspond at the actual, accurate passing time would probably be a difficult task and low priority to accomplish.

I can't do it myself, I don't have nearly enough free time nor do I have the requisite understanding of the code, how to modify it, or the programming language in question. I'm not expecting you guys specifically to jump on it and fulfill my demands, that'd be childish and unappreciative of the work you've already done.

I'm posting this because if, at some point, someone with dedication, free time, and necessary skill set decides to appoint significant part of their life to improving sonic booms in IL-2 1946 modded game, I'm fairly confident this is probably the simplest way to get it at least somewhat right. This might not happen ever, or it might happen after a month. But as long as I'm an active member of the community I'll be available for physics related consulting on the matter, because I do care about physical verisimilitude a great deal.


Anto, I really do appreciate all the work everyone's been doing with the supersonic coding. I especially like the Prandtl–Glauert singularity - it behaves very much like what I see in videos of supersonic aircraft, and gives a good visual representation of crossing the sound barrier.