...are not always the answer.
 
There seems to be a thing about bigger being better but I've often found finding the particular sounds you are looking for and making specific patches to produce that particular sound is actually more effective.
 
Relying on a library of every concieveable articulation, at every concieveable velocity from every concieveable mic position to get close to interpreting your midi performance doesn't always work as planned, and often sounds like a lack-lustre compromise.
 
So when for example your piece features a drummer making a dozen or so cymbal crashes of pretty much the same intensity just make a small patch for the exact sounds you are after which you can re-use when you want such crashes in your arrangement instead of loading a whole gargantuan sample library which will only approxiamate where you want to go.
 
A snare groove also will often only feature a limted range of dynamics so set up the perfect patch to suit, you'll soon build up a library of specific killer sounds that will often trump the use of a monster library.
 
Big libraries sure have their place but if it is specific sounds you are after then I recommend going the most direct route possible and get EXACTLY what you want rather than a near miss.

 I agree Jonbouy. Never could understand why when a company comes up with say a piano that is a computer emulation based on a real piano compared to a huge sampled streaming piano most people prefer the huge sampled piano when  they both can sound very similar depending.

 I think there are times when the streaming libraries beat anything else hands down but I have found that in my case there are times when working with certain sounds there are plenty of alternatives so that there is no reason to necessarily feel limited by or confined by not having the streamers. 

 

Yeah I don't know why I posted this even, other than I'll often listen to something I've made with a big library and gone, meh, and lately I've been more particular about getting exactly the thing I'm after in the first place.
 
I myself have been seduced by the idea that a multi-layered, multi-mic setup 'ought' to sound better when actually a few pre-mixed samples that are right on the money will often actually do the job right of the bat.

I thought size mattered .....

and bigger was always better....


http://www.youtube.com/watch?v=bS_ZcDaaduE

If the project is based on one album (private home studio); in the most simplistic and time-saving way I suggest narrowing the sound library down to only what is needed for example; once you figure out which sounds works best then save that as a template (Sonar).... to further explain.... 

For example; when I want to start on my next song I open an existing template which has the drum-kit loaded with separate outs and also several other plugins all with separate outs and usually with the same core sounds.  

So basically; everything is ready to go :) 

However; if running a commercial studio?... then I guess there will be mega-gig library at hand.... and whoa that's a lot of sounds to dig through {ARGHHHH}.   In that case I suggest the studio guys will need to have a clear understanding of the type of music the client does = get them to send a link/music long before booking so that way you'll be somewhat well prepared.  

SongCraft


In that case I suggest the studio guys will need to have a clear understanding of the type of music the client does = get them to send a link/music long before booking so that way you'll be somewhat well prepared.  

That's exactly what I do. My sampling library is remarkably huge. I've been collecting and creating samples for many years and have them all in order and placed in categories. Clients in need of specific sounds send me clips of what they are after and I either match them up or create them if possible.
 
The one thing I do like about having the larger sample libraries is, there are more hits per velocity which makes things more natural. Relying on a few samples without having a nice sample pool to draw from velocity wise can make things sound a little unrealisitic, but I do agree with Jonbouy's comments.

-Danny

Danny Danzi


Relying on a few samples without having a nice sample pool to draw from velocity wise can make things sound a little unrealisitic, but I do agree with Jonbouy's comments.

-Danny

That is key.
 
Retriggering the same same sample in succession sounds awful.  I will use at least 4 sample variations for each velocity layer.  I keep my samples organized though because sometime all you need is midi for a passage with a short range  of velocities say between 90-127.  I'd make a patch with a few velocity ranges to cover that specifically and include my random variation hits within that.
Things like crash hits for example where there may only be a dozen or so during a whole song, I'll likely pick out the EXACT ones I want rather than rely on the drum software to choose them for me.
 
So I end up with say a 1-200 kb patch that sounds killer and is tailored for a particular part rather than carry around say a 5Gb Kontakt multi in a project that may not sound as good.
 
I certainly use the big stuff when it is required though. 

I am someone that owns several large sample libraries but also makes patches too. It is also very important to get onto the more detailed aspect of the programming of a patch though and serious samplers like Emulators for example allow for very expressive playing. Many patches are not even programmed anywhere near where they can be in terms of touch sensitivity etc. I agree with Jonbouy in that they can be a turn off only at first though.

I think you have to be fluent with the samplers OS and be able to get in there and edit the patches in depth. For example in the Emu E5000 that I use you can send velocity to all manner of destinations. eg volume of course but also filter cutoff, attack, pitch, resonance, pan position and all of these in varying degrees. (And also most effect parameters!) EMU's can set up two very different filter types and morph between them under the control of velocity. Then the sound takes on a new dimension and feels and sounds incredible to play. BTW VST 3 really takes expressive control of synths into a very new level of capability.

Kurzweils can have many VAST processes going on over a patch and be under the control of touch sensitivity. The options there are almost limitless.

I have made patches by carefully sampling things and then doing all the pre production on a sound but then you need to get into the expression part of the patch to fully complete it. If you also add in the effects of after touch, pitch and mod wheels, foot control and breath control then some serious expression can be built into patches as well. 

These things apply to all VST's and synths as well as samplers. Samplers are OK and for me I never get bored with them. They seem infinite to me. There are so many ways you can edit the patch. Repeating sounds is very avoidable. I have got incredible conga patches for example that have like 7 or 9 different soft notes in succession. The idea is to also use them in such a way that for example you may only use 9 soft hits and rotate them 1 to 9 and restart etc. This will cycle over the time signature you are using and notes won't repeat themselves in the same spot rhythmically for tens or hundreds of bars etc..You can be clever like that and use maths to make things sound more natural. If you are a drummer you can sample cymbals crashes like 10 times and use all of the crashes in succession before repeating. Same with hand claps and finger clicks etc.. It is important for one crash to continue on with its volume envelope too if another crash comes in quick succession. 

All this does take time and require a lot of thought and input. But at least once you edit a patch for very expressive playing it can be stored as such and then you have it like that from then on.