I like what Boom does the most. Having a library made around a general theme then have one part that is really already pre designed sounds and one part that is just the "raw" recordings.

Then I use mostly the "raw" stuff but the designed stuff gives ideas.

And obviously they have a kickass quality although their files in 192kHz seems a bit over the top to me, I do like 96kHz the best, but whatever.

Ah thanks for the feedback guys!

I gotta tell ya, now that I've started recording various things, I am always looking around and thinking "hmmmm, I wonder what that would sound like," or "what could I do with that noise." I wish I had an awesome set up with whiz bang gear and a studio, but for the time being I have my portable recorder, will look at an audio interface soon also. But this is good as it will allow me not to just rely on top notch hardware, it will allow me to practice and use what I have to produce the best thing and learn from the bottom.

CCH - The variety aspect is very important I must admit. From what I have done, everytime I try and do as many takes in as many different ways that I can think of. I see the huge amount of different light switch takes, even though you think of it just being a light switch, the need to match the end users different contexts is important. Makes the collection more versatile. 60 SFX libraries? That's a huge amount, jealous! Are they used a lot? Yeah, I can imagine how important the metadata would be, don't want to waste time.

Valoon - I do like the idea of the "raw" recordings, as it would give the end user more control over it to match what they need to do. I have been doing some in 96kHz but also some in 48kHz as that's what a majority of people have asked for.

Another thing, at what dB level should the mastered/finished samples be at? What would be the norm?

Thanks.

I think it' the best to do one 96k version and one 48k. Boom does it but with 192k instead of 96k a least for some.

I don't really know about dB level. When I record sounds, I basically normalize then at the max possible and then I remove 1 dB that's what I am used to do but I only keep it for myself so it's not really a norm.

I am not sure how pros do it but I am pretty confident that for the raw stuff they just normalize it, it's raw so you don't want to touch it too much. For the designed stuff they probably add more effects.

It's true for music or stuff you don't touch much. But the bigger it is the more points of mesure there is on the sound wave, which means you can go crazy on the sound way more than a 44k when you have a 192k because the waveform will be reproduced from the samples way better.

At least that's what I've always been told, and I feel it too.

Ah I see. So pretty much physically there is no difference as we cannot perceive the difference but technically the extra resolution gives a much more clear control over the sample. How about downsampling then, would one like 96kHz be good as it can be halved to 48kHz?