Just got back from a lesson with a wonderful music teacher who knows straight, non-electric acoustics like a PhD. I was telling him about my new expedition into using GigaStudio. His comment: \"A magnificent technology. Unfortunately, you have to watch out because many of the people who make the sample libraries use too much vibrato in making the samples in the first place, which gives the samples a subtle, unnatural and steady wah-wah-wah that a real acoustic instrument just playing in the open air doesnt have.\" I was interested in any comment from the experienced GS users or sample makers here - especially those oriented toward classical / jazz and traditional acoustic sounds?
eh, he might not understand the concept of gigastudio to its fullest.
with giga, you can record longer samples. Meaning less of that \"loop\" sound with vibrato. You get more of the natural vibrato from the actual players.
the only problem is you usually dont get control of the vibrato in this way. There are some exceptions.
the problem with vibrato is using it to emulate actual players, and \"emote\" with it.
depending on the instrument players use vibrato/tremolo differently.
anyhow. even with sampled vibrato, it can get stale. Since it will be the same every time with repeating notes. (again...here we go with auto alternating...then people complaining about sample bloat...and then....blah blah blah)
I think it depends on the sound and the application.
One of the things that made early string libraries sound so strident and dead was the tendency to record everyone with no vibrato. I think string players think that ensemble playing without vibrato is clever because it\'s hard to do and maintain great intonation. But it\'s rare that I want to call up an ensemble string sound and hear that funereal tone. I often ask our string players for more vibrato when they think we\'re about to ask for less...
For a lot of solo instruments, the vibrato is what gives you a large part of their character. Factors like when the vibrato comes in, how quickly it increases in intensity, whether it varies in depth or speed with the sustain of the note, whether it is just pitch, timbre or volume variation, or a combination, all contribute to a \'live\' sounding peformance.
Personally, I prefer solo instrument samples with the vibrato performed within the sample. I know that means you need a hell of a lot of articulations just to cover different tempos and styles of vibrato, but LFOs - even creatively programmed ones which use envelopes and randomisation, just don\'t cut it.
Your teacher may actually have been referring to the looping inadequacies of older libraries. When trying to fit long sounds into small amopunts of ram, everyone was forced to make the sustain portion of the sound loop - even cymbals. Take a close listen to the decay of a lot of drum machine crash cymbals (especially if you can open up the envelope release time). They definitely tend to display the \'wah-wah-wah\' that your teacher describes.
If the time stretching approach were used to alter vibrato depth/rate which was already recorded in a sample, perhaps in Giga there could be a simple UI function (where the sample developer had already \'marked\' the beginning of the vibrato, the way loops are marked\'), and the user simply had a few intuitive LFO-like attenuators to interact with - delay, fade, rate and depth - leaving the program to set the nitty gritty time stretching parameters, maybe even in real time...
There\'s an exciting new product coming up now from www.sonicimplants.com . It\'s called symphonic strings, and seems to feature the same kind of stuff that\'s in \"gigastrings\", only with a different sound.
Chadwick, you make a great point about LFO vibrato lacking character, I can\'t help but agree, but I also can\'t agree with the tendency to incorporate the vibrato within the sample-at least in solo samples. I can never find a speed, depth and expression that matches my intent. Sections with vibrato work OK, the massed instruments tend to mask detail anyway.
The best solution I\'m aware of is to record the original sample without vibrato, and then add in the desired result when performing the solo instrument by using a ribbon controller like on the Kurzweil 2500/2600 series. You can use pressure for volume, position for pitch, and control with one finger essentially the same way a violinist/cellist does. As detailed an expressive as you can be, using a keyboard as input device, and playable in real time, too!
Hmmm...Thanks for these helpful comments. I\'m a classical musician just getting into Giga. There\'s a distinction here that I\'m not clear about.
If a string sample intitially has \"too much vibrato\" for the taste and purposes of a particular Giga user is the ONLY possible origin of that:
A - The fact that the violinists simply played with a lot of vibrato, at the direction of the sample maker, as the sample was being made.
B - OR, does the sample maker (sorry, I\'m anything but a sound engineer) have the ability to insert vibrato at a microphone level or some further mixing level in the process of later making the finished CD for the user and HE, the sample maker, has chosen to produce samples with a fair bit of vibrato that were originally played quite flat?
The problem with artificial vibrato via Pitch bend/Ribbon controller is that its shifting the whole frequency range up and down.
The character of vibrato comes from more varied frequency deviations in different areas of the spectrum, as well as tremolo in certain areas of the spectrum
On way to adjust vibrato is to use a time stretch/compress process on your samples. This can be especially usefull if you have a time stretch/compress process that can be manipulated to change its stretch/compress rates over the length of the file. Allowing you to give more character to the vibrato.
Of course stretching can add artifacts.
Cool edit has this feature.
another way around it might be melodyne it allows you to stretch and compress files at specific points I belive, plus it does formant shifting and its time stretching algorithms are better that cool edit.
Anyhow, time/rate of vibrato is still an issue.
Even if real time control of depth and frequency of REAL vibrato were available, it would require careful playing and learning of the end programmer to get it to sound good.
And we all know that in the end that might be too much for the majority of people who want the \"out of the box\" experience.
real vibrato in the sample is still the choice for me, until we get some plugs that can work in both the frequency and time domain at the same time.... in real time...
even tho that is sort of possible with the NFX EQ...its still not powerful enough.
Thomas_J - Looking over Sonic Implants - is there any reason for a weekend amateur like me to be following ALL these other music technology developments in more than a casual way ? Sound fonts, plug-ins, VST, DLS, Halion, EWS24, Vsampler, Kontakt... on and on and on!
I\'ve got GS96 running nice and smooth (a small miracle in and of itself because I\'m basically lo-tech) on my Win98se P3-550 256 with 2 fast hard drives and the Terratec EWX 2496 and a simple MIDI controller. And I use good old $29 PowerTracks7 to sequence because it does just fine with GS and I dont have to look at a UI that looks like the control panel of an F-16 jet! My work is \"weekend\" classical and jazz composing so my only needs are good clean acoustic, traditional sounds - like pianos and oboes and sax and strings and basses and drums and flutes and trumpets.
So I still look into these other new software technologies on the net like VST and boom - I\'d have to make a whole sequencer change, or then I look into sound fonts a little and I see that bing - I\'d probably need to make a sound card change. The last capability hurdle for me is really just financially being able to afford (!) the good sample libraries for GS - and of course learning gradually how to work with them. Assuming I can clear that hurdle (with perhaps an upgrade to GS160 and more ram along the way) and gradually purchase some good libraries I\'m about ready to turn off my hour a night of reading up on the \"latest, greatest new thing\" in music technology and put that extra time where I really want it - into composing new music with GS. Sound reasonable?
The joy of just working with GS in this simple way is already immense (!) after years of just sequencing with pretty beepy, cheezy sounding acoustic instruments from synths. Or am I missing out on one other technology in this rather simple formula that would easily compliment what Im doing - given that I do not perform live, I am not a professional chasing deadlines, and I am not in a quest for far-out, new-age, or synth-like rock and roll sounds. Just call me \"old school\".
And good music to everybody at this helpful forum. -LifeForce