Echo/Delay of sounds in sectors

[Milian]

Today I had an idea. How about adding a echo/delay-feature to the sounds?

You only would need a "echo- thing", place it into a sector and give it three arguments:
- number of echos/repeats or length of echo
- distance (time) between each echo
- volume of the echo.

Or apply it to a line and add a sector tag (like Static_Init e.g.). This way you could connect the soundbehaviour to a whole area.

When the player is in one of the affected sectors, the delay will be applied to all environmental sounds (monster sounds, weapon sounds, ambient sounds).

With this feature, you could give every area its fitting sound appearance.
Imagine a Cyberdemon in a cathedral-like room with a powerful delay.
Or echos in mountain areas.

Would this be possible?

• Milian

[LaundriTorture]

Good idea but I'd like to see more variance and possibilites for this effect, imagine underwater sounds or things like that

[Caligari87]

Sounds simple and powerful to me, not to mention obviously atmospheric. This gets my vote!

Perhaps the second option (volume of echo) should be "echo decay", and tells by how much the sound is lowered per echo. That would nullify the need for the "Number of echos" option. Perhaps even a "Reverse Placement" bool variable so that it would seem like the echo is moving. So your thing options would say:

Code: Select all

Decay: 192
Delay Time (tics): 15
Reverse Stereo: 1

This would make an echo that would be delayed half-a-second, and played at 75% the volume of the original sound, so on and so forth, and flipping the speaker it plays from on every repeat.

Input?

[Killo Zapit]

What about just using ZDoom's EAX stuff? Though it is true that you can only have one EAX Effect active at a time, and the effect is applied based on what sector you are in and not what sector the sound was played from. Plus only a few soundcards support it. But you can use whatever echo effect you want, and even make your own with a special lump... In fact I am surprised more people haven't used it. Try typeing "eaxedit" at the console to try it out sometime.

[Grubber]

Sound effects without EAX would be great...

[Nash]

But Duke3D, Blood, Shadow Warrior etc pulled off environmental sounds in software mode without using any fancy soundcard hardware.

Even when ported to Windows (EDuke, JFDuke etc), the echos are still there.

Perhaps we can use that approach instead so that it isn't hardware-dependant?

[Belial]

Umm... What do you mean by echoes? I can't remember anything from the Build games that resembled true echoes.

If you're thinking of something like the starting area of Warp Factor (E2L03) or the spinning fan in E2L09, well, IMO the effect sucks.

[Medricel]

Yah, I'd like to be able to make custom environment EAX-esque effects as well. At the current moment, it seems as though I can only experience the default (and hardlocked) EAX effects. Granted, I can also apply EAX effects for my day to day Windows usage as well

[wildweasel]

I miss my SB Audigy card. *sniff* I have stupid onboard sound now while my beloved Audigy collects dust - all because my dad doesn't want to bother opening up this Dell box of mine.

But enough pathetic whining. I'd love a software solution for environments.

[Milian]

Good idea but I'd like to see more variance and possibilites for this effect, imagine underwater sounds or things like that

This would be hard to do.
Cause a echo/delay is easy to build, it's only repeating a sound with delay.
But to create underwater sounds or influence the frequency range or envelope (hope these are the right terms in english) of a sound, you'd need a filter with a variety of settings, and I'm sure Randy won't make ZDoom a soundediting program

Perhaps the second option (volume of echo) should be "echo decay", and tells by how much the sound is lowered per echo. That would nullify the need for the "Number of echos" option.

Every echo can be determined by 3 parameters.
- time between inital sound and echo
- degree of volume decrease (in %) with each loop
- number of echoes.

Imagine that in a mountain you won't hear an endless repetition of echoes, but only 1 or 2 which are quite loud, so you'll need both parameters.

What about just using ZDoom's EAX stuff?

What is EAX? I never heard of that.

[Grubber]

http://www.fmod.org/docs/HTML/FSOUND_FX_Enable.html

[Lexus Alyus]

Ya know, something like this might requre the usage of DSP or VSTi/DXi plug-ins... dunno how hard support for those might be and could be quite processor intensive... then again, I'm sure it's easier to code these effects directly into the engine .

What about incorperating convolution reverb technology in a game? . Heh, that would be cool... just load an impulse response of a mountain or church in and you can use a few preset environments in levels... that would make the game sound super realistic .

It intrigues me that so many developers concentrait on graphics and all these visuals but pay little attention to sound... sound technology is incredibly advanced now, but sound advancements are only being used for production software and to emulate analogue gear through digital means. I don't know how to program, but I know what is posible with sound these days... IMO, not enough is done to make an environment a real and beleivable place with sound.

for those that don't know, convolution technology is fairly new and all the rage in the digital production/sound scene. Basically put, it's reverb. Reverb is a kind of delay with many perameters and it's purpose is to emulate a real acoustic space. Up until recently this has been pritty basic and digital... but now we have convolution technology. Convolution reverbs (a little more processor intensive, but I beleive that it only has a few lines of code) basically take a sample, or impulse response (created from a sample... basically a recording of a real acoustic space) and then works out the acoustic charectoristics of a real space to calculate it's realistic reverb properties. As you can already see, this has loads of possibilities in the gaming world. Using delay to emulate mountains is very primitive considering how much progress the digital music industry has made... thats like using a sprite to create the illusion of depth rather than using a model... why is sound in computer games so primitive?

So, if you sample the reverb of a real forest and then use that for the whole of the sound in an area, you have a very ralistic forest environment... dead easy... but quite intensive on the processing front I suspect.

How does one sample an acoustic space if it has no solid form? Simple, you just blast a sound that comntains all frequencies in the audioble frequency range of the human ear. From that, the program can decifer the acoustic spaces properties by analysing the frequencies present in the recorded sound. The best sound that will have all frequencies is a starting pistol... the sound produced would also have the natural reverb decay of the room/area... which is also analysed to get a good reverb resault. Another sound is a sine wave sweep from 20Hz to 20K Hz (human ear frequency range). This is the best.

Of course, this idea can be extended. If you run a sine wave sweep through a guitar amp and then sample the sound from the guitar amp you can actually record the charectoristics of that guiatar amp... basically, if you load that impulse response you will in fact have the sound of the guitar amp, the sound of the microphone used to record that amp and the sound of any outboard pre-amp and equipment used... basically, you have an almost exact virtual representation of your amp... so, you have a virtual guitar amp! This method hasn't been used by virtual guitar amps yet (hence why a lot sound pritty shit) but, can easily be achieved with a convolution reverb.

Enough of my banter, here are some links .

Waves IR1

Emagics Space designer

Prosoniq Reyverb

Voxengo Pristine Space

The mightly Liquid channel outboard Mic Pre-amp and compressor modeller uses convolution technology to emulate the real analogue gear:

Sound on Sound article on the Liquid Channel

Why have I posted these links? Each article explains the concept of convolution reverbs... if you are an intelligent programmer then you can use this knowledge to makre games much better on the sound front .

I think you can see that I'm not stupid .

[dennisj1]

Does anyone have any documentation on the SNDEAX lump from ZDoom.Wad?

[Milian]

It intrigues me that so many developers concentrait on graphics and all these visuals but pay little attention to sound...

I agree 100%.
As a musician and sounddesigner it's really frustrating to see how the sounddesign is so extremely underrated. Music and sound is always the last thing, developpers (of games or software) add to their products.

I have another question:
In the listing of this forum, some threads are coloured. This thread is now green, others are red. What does that mean?
(Edit: Sorry, it's not green. The thread underneath is blue, so this appeared like green .. .anyway...)

[Nash]

Regarding convolution reverbs, anyone tried Soundforge's acoustic mirror yet? It's a cheap attempt but it's similar to that.

And just like Milian said, I too think that sound and music in computer games aren't getting the attention they deserve... but recently some games have been making really good use of sound. It's about damn time.

Milian - Red basically means "Randy says NO", green means "Good idea" (or something) and blue means the feature is already implemented.