I have been thinking about this for some time now. How are smaple libraries created? I mean, I understand the basics about recording and editing performances, but what is the process that is used to create a library? What type of software/hardware is used? How is the crossfading of insturments accomplished? These are the kinds of things I\'d like to know.
The reason that I am asking this is because I am interested in maybe (with much emphasis on the \"maybe\") creating a few libraries of my own.
I can possible forsee a theremin library programed in the vein of GPO only adding one more element, a volume pedal.
The pitch could be produced using a midi controller and the volume by the added pedal.
That would then free up the Mod wheel for characteric slides between notes, (Or maybe vice-versa). The sustain pedal would be used to control legato and staccato playing.
The only aspect I haven\'t figured out yet is vibrato. Maybe two different samples is the way to go, one with vibrato, one without.
I once made a terrible theremin sample using the gigastudio map-wave thing. The clumsy way that I did it kind of disuaded me from trying it again. However, now I feel inspired to give it the old college try.
I know that this has absolutely nothing to do with GPO but I really admire the way that the library was created. So, naturally, the first people that I wanted to voice this idea to is Mr. Garritan and all of the GPO users.
I really feel at home on this forum and that is what sold me on GPO when I bought it.
Thanks for all of the great conversation. Please, any comments or concerns about my idea are welcome.
BTW, does anyone else think that there should be a thread for demo posting only? I think I have downloaded every one that has been posted thus far, well, besides the ones posted today, for I am currently at work.
Johnny I can\'t speak for the real developers, but in my own limited experience I have basically recorded the sound I want in Logic - done whatever processing I wanted (which is usually nothing) and then trim the sample where I want it to start and save it to disk. Then I open my sampler, load the samples, map them, and save the preset. But I don\'t have much experience and the little I have had has not made me want to do too much sample development [img]images/icons/smile.gif[/img]
You might try posting this to the main thread as there are many more developers and opinions over there.
Forgive my blasphemy, but why would you need to sample a theremin? its output is a simple sine wave, not a complex waveform (like a Stradivarius violin) that would need to be sampled every 1/2 note to make it convincing. You can get good results with any (virtual) analog synth with sine wave sound, portamento (tweak the porta time as you play to make it convincing) and vibrato. I came up with convincing patch on RolandJP8000 in under a minute. The only problem would be adding some warmth to it, but that\'s another story.
I just feel sampling it would be too much work, but don\'t let me stop you [img]images/icons/wink.gif[/img]
And I\'m not a keyboard freak detesting any non-keyboard instruments - I used a theremin in a song once (the Pocket Theremin to be precise [img]images/icons/smile.gif[/img] Took much practice to nail the notes. The song is at http://matejhrovat.go.to called Invisible Forest Creatures
I don\'t mean to sound like a downer, but as one having had some experience (BAD) with selling my own sample libraries I would say weigh it CAREFULLY. It is not something to be undertaken lightly!! (unless you aren\'t interested in making any money, which I don\'t suppose you would like!) (for fun.. that\'s a diferent story).
There winds up being A LOT of tedious work and then there is all the problems of duplication.
And dealing with people can be REALLY a pain sometimes.
I could go on and on about my experiences some years ago. Most of these experiences caused me grief.
Perhaps the biggest grief came from discovering why my sales evaporated shortly after coming out with my sample discs for sale.
I found out a large european soundware company that shall remain nameless, \'stole\' my samples (literally) and then turned around and \'sold\' them in their own library without even an acknowledgement of any kind.
Needless to say I was VERY p*ssed! You can\'t know what it feels like to be in that position until you ARE in that position.
I had no money for legal actions which is often insurmountable for even large soundware companies like EW.
If you are intent on doing it, I would advise much planning at the begining instead of the \"I should have done....\". This will insure less stress as you work along on the project.
And make no mistake, if you do it to make money, it IS a PROJECT!
It also cost much $$$$$ for advertising!! Even web advertising might not get you that many sales.
I had countless people contact me for endless details, sometimes they continued to call and bug me even after I had answered all their questions! (stupid questions a plenty, too!).
My email got clogged with this sort of drivel.
If you are SERIOUS about producing a quality product that will fetch a decent price you must be IMPECCABLE about the sound quality or customers will crucify you on the details and demand their money back!
I really did have all these experiences.
I made a nice demo CD because so many people demanded one before they would purchase. I\'m talking a $99.00 library here, not $990.00!!
Even then, many people said I should send it to them for free because they took the time to show their interest in my product. If I had done that I would have gone way in the hole on the entire project just paying shipping charges!
In the end I only made a few hundred dollars profit. NO WAY did this compensate me for my time and aggravation.
Printing costs too, are sooooo expensive.
I would suggest you get a good sound editor like Bias or Soundforge. Wavelab is nice for experiemntal sound design too.
I had some trouble with looping so I did several samples on a friends MAC running the \"Infinity\" looping software. Auto tune can be useful too.
Beyond that the recommendations for software get really deep. It just depends on what you are trying to accomplish. Again, remember people will hang you on your quality (or lack of it!).
All of the above is, of course for a COMMERCIAL library. If you just want to have fun and post your stuff on a BB somewhere on the web, go have a ball.....
I hope I didn\'t scare you off TOO much!!
At least you will learn as I did to really appreciate what Gary and Nick and Doug and all the rest have gone through to bring us these great sound libraries.
For ME now, it\'s \"leave the driving to us..\" all the way with those guys.
I\'m not really wanting to make money with the libraries. I just want to give it a try. I realize that the theremin is a simple sine wave, but I want to try to challenge myself instead of using a basic sine wave generator.
Nexus, thanks for the great reply. I can totally understand why you are bitter at some soundware developers. Like I said, for now, I\'m not in it to make alot of cash, just self gratification.
Nexus is right but here are some general things if you decide to plow ahead:
1. The initial recordings can be made using almost any medium from a mic/preamp/interface/laptop to a portable DAT or Minidisc recorder. I would recommend that you not skimp at this stage. Make the recordings the highest possible quality. This is where a thorough knowledge of standard audio engineering comes in handy – mic technique, equipment choices, recording environment, etc. It helps to have things planned in advance. Know your goals – this includes philosophy of recording (e.g. dry or ambient). Prepare.
2. Record the samples chromatically with as many techniques/articulations, alternate takes, and dynamic levels as your time and budget will allow. This can be very time-consuming – do not underestimate. Patience.
3. Use audio editing software like Sound Forge, Wavelab, or Cool Edit to trim, edit, and label the samples. Know your software. Read the manual from cover-to-cover; you WILL miss valuable features if you don’t. Label the samples systematically (VlnSusC5) and organize them into folders. Label the folders clearly in categories. This is tedious work. Use batch processing wherever possible to save time. Patience.
4. The next step depends upon the sampler format you choose to use. The advice is simple to give but hard to use: Learn everything you can about the architecture of the sampler. Once again, this means reading the manual cover-to-cover as a start. No exceptions. Depending upon the quality of the manual that will get you about 60% of the way there. After that you must push and prod and experiment until you have tried things that maybe even the designers of the software hadn’t noticed. You must know how everything works (intimately) to organize your thoughts on the features you wish to build into your sampled instruments. Diligence.
5. Analyze the instrument you are trying to emulate. You cannot successfully emulate an instrument you don’t understand. Your goal should be to get to the point where you can think musically like the player of the instrument. Break it down into characteristics. For each characteristic try to find a counterpart in the sampler’s features. Some will be obvious, some not so. Sometimes there will be more than one approach to a characteristic. Try them all. Evaluate the strengths and weaknesses. As you accumulate features see how they interact with one another. Balance analysis with intuition.
6. Bug hunt. The more complex the instrument, the more potential problems – some of your own making and some of the software’s. It helps to have other people’s eyes and ears. Accept the fact that you can never successfully \"proof\" your own work beyond a certain point. There will always be glitches. Chip away at them. Eventually, you will eliminate most of them. Patience.
7. Listen to criticism. You don’t have to follow it but listen to it anyway; it can give you a different perspective or crystallize the one you have. Caveat: Not all criticism is created equal.
While I have absolutely no intent on doing any sampling, your 8 comments were sure worth reading. So many of the things you mentioned like reading manuals, thinking like the artist who would be playing the instrument, paying attention to details, organization and keeping good notes, listening to criticism, and patience are so true in almost all creative endeavors like using GPO. Thanks for taking the time to post them so thoughtfully. I only wish I had more of these kind of virtues.