I agree that the more variation we can get into sample libraries, the better.
The thing is, we are only now getting to use the kind of technology which gives us this choice.
In the old days of the Fairlight CMI with 2 seconds of 8 bit memory, there wasn\'t much point in putting a clunky sample into memory, because every time you played a new note you would hear that clunk. Sure, that single cello note sounded totally realistic on its own, but the moment you played a few notes in succession, anyone listening could easily tell that it was a sampled cello. In fact, it was often the quirky performance of the sample which was the giveaway. So the solution was to provide \'neutral\' samples which didn\'t give themselves away so quickly.
I think you need to have the \'basics\' accurately represented before you attempt to add all the human and instrumental quirks which trick the listener into thinking that he\'s hearing a real player.
Even with endless streaming and dimension switching, Gigastudio is only just barely able to do this, and will be better at it in V3. Most samplers which rely on ram can\'t even approach the kind of felexibility required.
Up until a couple of years ago, the sampling paradigm was all about managing to replicate an instrument with the minimum number of samples and ram allocation necessary. Quirky or alternate articulations were regarded as luxuries which could be included, but probably woudn\'t be taken advantage of. With disk streaming this paradigm has radically shifted. Gigastudio owners have beed a relatively small group of the sampler owning community, so it\'s not surprising that it\'s taken a while for library developers to take full advantage of the new possibilities Giga offers - but they\'re starting to now.
If developers can make a profit by creating libraries which not only offer all the standard articulations but also a wealth of alternatives, then we\'ll definitely see them. We\'ll also see more samples libraries of different models of the same instrument - just as we are seeing with pianos at the moment.
This is excerpt from another topic that I thought I would post here try and generate some discussion...It\'s concerning my opinion on \'clean\' samples...hope this generates some discussion...thanks:
Im glad there are some who understand and agree with my sentiments regarding \'dirty\' Rhodes samples. If someone wants to use a clean sterile Rhodes samples/modeling there are plenty of options available. But what isn\'t available (and what I endeavour to fill) is having that feeling of actually playing a Rhodes...those little imperfections that make such a wonderfully unique sound! If, after listening to my demo, you visit the logic audio page and check out the EVP88...the difference between the two is MIND BOGGLING(sic)! The response is the same with other available Sample CD\'s...don\'t misunderstand me...most of the available Rhodes samples sound great...but too clean...not authentic enough! The comparision(sic) is comparable to say...GigaPiano and the Malmsjo!!! Not that one is necessarily better than the other...just that, in my opinion, one provides a more accurate picture of a nice warm piano recording!
Seeing as the majority of the users here are interested in film/tv/game scoring...there seems to be too much of an emphasis on obtaining the absolute CLEANEST samples...I believe this is a misconception in the search for ultimate realism!!! Granted, poorly recorded samples are \'immature\'...but recordings with \'imperfections\' seem to be able to mask the great void between samples and real recordings. I shall once again consider the Malmsjo piano...to my \'young\' ears...some of the demo\'s on Hans\' site sound incredibly realistic! Now, this is not a result of poor recording...but the natural imperfections of the instrument being allowed to come through.
I guess my thesis here is that perhaps if, as I think we are all here help accomplish, going for ultimate realism via the use of samples...we may be wise to reconsider are posture on how \'clean\' our samples should be. It is my opinion that if developers continue to scrub the life out of the samples we use...then I doubt we\'ll ever get as close to that goal...(indiscerable realism through samples)...as we hope to be able to!
I thank you for taking the time to read my thoughts, and I hope this may be able to provide some additional discussion.
J.P., your samples sound darn good, i was just merely trying to point out how you could potentially do it better, and make more money selling a more Complete library since you allready have the expertise of sampling it once...
There will always be the need for Dirty versions of a certain instrument (sampled properly with the right gear), also having the clean version and being able to blend the two or use both would just makes a complete library you could sell at regular library prices, that was all i was trying to tell you, this has nothing and i mean nothing to do with clean samples versus dirty samples, it\'s all about offering the costumer options, it doesnt say one is better than the other....
Personally, and this is a personal personal taste, you can process samples , but once they\'re sampled with the processing, you can\'t unprocess them.
I can pretty easily take EVP88 and pass it through a Fender twin ang get a dirty sound, i can\'t however undirty a sampled version
of a library and clean it if i want a cleaner sound, that\'s why having both would make a great complete Rhodes library that nobody has done yet (specially that it could all be done in one pass, sampling the mic and input at the same time on 2 different tracks) .
Having said this, like Munsie said... what the hell , sell it for 10 bucks, people will buy it it sounds real good.