Re: Theoretics and Pseudocode: Sound Cues.
@10I would exclude places like the network card where they aren't actually really ram as we know it. My list is only the not obscure stuff that matters for the sorts of optimizations being discussed here: l1, l2, l3 caches, ram itself
Re: Theoretics and Pseudocode: Sound Cues.
@9, Lets see The CPU cache, CPU registers, the disk, the RAM, the network card, and the graphics card? Those are at least some of the places I know pointer addresses can be stored and retrieved (though I'd generalize the disk, the networ
Re: Theoretics and Pseudocode: Sound Cues.
@9, Lets see The CPU cache, CPU registers, the disk, the RAM, the network card, and the graphics card? Those are at least some of the places I know pointer addresses can be stored (though I'd generalize the disk, the network and graphics
Re: Theoretics and Pseudocode: Sound Cues.
You will find that, more than being a way to waste time, that's a good approach to find out later that your efforts weren't worth it because you didn't understand. If you're still at the point that you seem to be at based off
Re: Theoretics and Pseudocode: Sound Cues.
7, alright thanks. That's always been one of my main issues, I try to make things as comprehensable and elaborate as I can so that I don't need to come back later to add more things, but I've been trying to break myself of it late
Re: Theoretics and Pseudocode: Sound Cues.
@5yes, you can do something like that too. But there's no real benefit here. You won't be able to change the speed of the footsteps for example. It's also *very* expensive to start concatenating decoded assets in that fashion.
Re: Theoretics and Pseudocode: Sound Cues.
5 reminded me. I'd like to start loading sounds into memory instead of streaming from disk, at least for things that don't have two many sounds. First of all, what's the recommended limit for when I should stop preloading and star
Re: Theoretics and Pseudocode: Sound Cues.
You can actually achieve gapless playback if you've got an audio backend which supports it. BASS for example supports preparing your samples in-memory beforehand, even merging multiple sounds into one with a bit knowledge of audio sample
Re: Theoretics and Pseudocode: Sound Cues.
Yeah, pretty much. You'll just have to make sure that when the queue refills, it doesn't start with the same sound that it ended with last time. But to be honest if you have 5 sounds and you assume uniform shuffling, the chanc
Re: Theoretics and Pseudocode: Sound Cues.
So grab the fully untouched list with a crawl of some kind up front on loading, and then clone it for the cue, and anytime a cue runs out, we can just grab that specific sound from the list that I grabbed up front and random.shuffle?We&#x
Re: Theoretics and Pseudocode: Sound Cues.
So grab the filly untouched list with a crawl of some kind up front on loading, and then clone it for the cue, and anytime a cue runs out, we can just grab that specific sound from the list that I grabbed up front and random.shuffle?We&#x
Re: Theoretics and Pseudocode: Sound Cues.
It's not expensive as long as you iterate the lists up front, not on every sound play (i.e. load the directories when the engine starts). You can either go through them in sequence, or use random.choice(mylist) to yank some out of a list
Theoretics and Pseudocode: Sound Cues.
Hello,I'm planning on porting your ideas into Python, but I'll accept answers in any language, even pseudocode. I've noticed that in a lot of audiogames where we pull a random sound for, say, steps or gunshots, we sometimes play the e
13 matches
Mail list logo
