ben m
Joined: 15 Sep 2002
Posts: 337
Location: UK |
| Week 6 - Samplers 1 |
 |
|
Ok,
this week you need to cover each of the following subheadings that are on the syllabus as well as the additional questions - it is important for everyone that you manage to answer the questions.
I will post the address of the practical when I have finished compiling it.
Features
Voices - Aidan
Memory - Dani
Outputs - Mike
Storage - Nancy
Sampler Hierarchy
Sample - Aidan
Key Group - Dani
Key Group Zone - Mike
Program/Multiple Programs - Nancy
Sampling
Recording - Aidan
Original Pitch - Dani
Editing - Mike
Looping - Nancy
Recording in Stereo - Aidan
Creation of Jey Groups and programs
Sample Assignment - Dani
Note Span - Mike
Tuning - Nancy
Loop Type - Aidan
Keyboard Track/Constant Pitch - Dani
Velocity Zones - Mike
Filter - Nancy
Envelope - Aidan
LFO - Dani
Keyboard Parameters - Mike
Mute Groups - Nancy
Misc Questions
Aidan
 Aside from a MIDI lead, what other connections are (or can be) made from a hardware sampler?
How does a hardware sampler ensure that there is no latency?
How can sample looping be used to save storage space?
Dani
What are the main requirements of a glitch free loop?
Why are closed and open hi-hats often sent to the same mute group?
On hardware samplers, how much sampletime would you typically get out of 64MB storage?
Mike
Name 2 ways that a sample could be reduced in size to save storage space?
What are the benefits of multisamples? What are the drawbacks?
Mike
Why may some samples not want to be 'tracked'?
Nancy
What is a 'one shot' sample? What instruments may benefit from this kind of sample definition?
Name one pitch related problem that may occur when sampling from vinyl?
What is the importance of setting the 'root' note for a keyzone? |
Mon Apr 07, 2003 8:48 am |
|
|
MIKE_NOSTRADAMUS
Joined: 08 Jan 2003
Posts: 163
Location: Northfork Recording Studios |
| can't copy this session text only |
|
|
ben have you introduced an Icon graphic in this session?
I dont see one but one sure pops up when I go to copy all of the text and tranfer it to word processor.
i can't do any thing with the text!
Mike |
|
Tue Apr 08, 2003 3:05 pm |
|
|
|
|
resol69
Joined: 31 Dec 2002
Posts: 69
|
| Nancy's thoughts |
|
|
Some sites to check out:
http://www.harmony-central.com/MIDI/Doc/tutorial.html
http://www.tweakheadz.com/Sampling_Tips.html
http://tilt.largo.fl.us/faq/synthfull.html
http://debussy.music.ubc.ca/~courses/319/Notes/Kontakt.html
Features: Storage
You need memory to store samples. Hardware samplers use on-board memory for sample storage, and do not need to use the computer’s memory except as space to run the software. With 16 bit digital recording system, the rule of thumb of 5 Megabytes per mono minute applies. If you use up the storage memory, you can usually buy another memory card. You can also store samples on the computer’s hard drive if you want. You cannot use a sample memory card to store anything but samples.
Sample hierarchy: Programs/Multiple programs
Sampler hierarchy includes samples, key groups and programs. Samples are the basic sonic building blocks, with a name, sample rate and looping information; a voice is a complete sound that contains one or more samples with key mapping, velocity settings, filtering, envelopes and modulation. A program is one complete keyboard setting containing one or more voices, plus performance parameters such as effects settings. Samples, voices and programs. are organised into banks, and you can have 1000 Presets and 1000 associated samples per bank.
More on how it works:
Once you get the samples into your sampler (either by recording or loading from disk), the samples will not automatically appear in a program. You must specifically assign your samples to a program in order to play them back in the Select Prog mode, where normal playback is done. Programs are built from Keygroups. Keygroups are just what the name implies: a group of keys where samples are assigned. There are many parameters that can be adjusted within a keygroup, such as filter settings, envelope settings, MIDI channels, audio output assignments and so on.
Sampling - Looping
By using looping, as sample that takes up a certain amount of RAM storage memory can be sustained for a long period of time. This way, you can save memory space. To create a loop, you define the segment of sound you want to sustain and the sampler continually accesses this section from the RAM and keeps it playing, without using more RAM. The trick is to find a section of sound that does not change significantly in amplitude and composition. A poorly looped sample will click loudly and deeply. Some samplers can search for such sections for you.
Here are some looping methods I’ve read about:
Mountains and Valleys:
Look carefully at the waveform's overall mountains and valleys. Even with a complex orchestra ensemble, you should see a "pattern" that repeats. Like one giant mountain (call it Mount Everest) and several smaller hills and mountains interspersed with valleys and dips. When you see a repeating pattern trim and crop the sample to where there are 2 or three instances of this pattern. Set your loop points, at the same point in the different instances "after valley 3 and halfway up Mount Everest" for example. Now Zoom in X4. Look for the repeating pattern at that level. Fine tune the loop. Sometimes you will get lucky and it will loop perfect the 1st time using this method.
The Reversed Section Method:
Do the best you can by getting the body of the sound as close as you can. Copy this section and paste it at the end of the loop. While the section you just pasted is still highlighted, reverse the pasted section. This gives you the back and forth loop and often gives the seamless, popless ensemble loop. Lots of synths use this method. You have to watch out for phasing on stereo samples. Its often best to do this in mono, then copy the mono sample to both sides of a new stereo sample, adding a slight delay to one side, or altering the volume envelopes for each side to give the sound a subtle l/r movement--very good for ensembles. Or you can EQ each side differently, perhaps a +3 boost at 4k on one side and a +3 boost at 500hz on the other.
Crossfade Looping Method:
Find your best loopable segment--it should have a start time close to the beginning and an end time close to the end. Make the loop. Then, use the crossfade looping function on your editor. Then hit the trim command so your sample loops seamlessly from beginning to end. Crossfade Looping like this will often leave a "pop" at sample start, even though it loops perfectly. With acid loops the solution here is to go to the beginning and end of the loop and fade in and out from and to 0db very quickly, with the smallest part of the sample you can manipulate. Your ears will not detect it if you do it right. If you are doing Acid Loops, this is the only way to go and it gets rid of all those nasty pops at the sample start. If its a string ensemble, just leave the pop in and use the "sample start" parameter at the program level of your sampler to get rid of it. Often there is an option to compress the audio material inside the crossfade. If there is a volume difference at the start and end points you should use it. This will minimize the "wow and flutter" problem this method often introduces as you play a sample up the keyboard.
Creation of Key Groups and Programs
Tuning:
You can use the tuning function on to change the pitch of a sample.
Filters:
you can add filters to your programs to boost or cut certain frequencies.
Mute Groups:
Mute groups are used for assigning instruments that use the same voice-as, for instance, when you want open and closed hi-hats on different pads to share a single voice.
What is a ‘one-shot’ sample? What instruments may benefit from this kind of sample definition?
Some sampling divides a sampled sound into an attack section and a sustain section, and then uses looping techniques to minimize the storage requirements for the sustain portion. However, some sounds, particularly sounds of short duration or sounds whose characteristics change dynamically throughout their duration, are not suitable for looped playback techniques. These sounds are stored as a single sample segment which is played once through with no looping. Short sounds, such as a quick hit on a drum would benefit from this type of sampling.
Name one pitch related problem that may occur when sampling from vinyl.
If the vinyl is warped, it would change the pitch as it’s playing and being sampled.
What is the importance of setting the ‘root’ note for a keyzone?
Samples are distributed across the keyboard with each sample taking up a certain range of the keyboard (called a zone or keygroup). A series of zones (each containing a sample) is grouped together in an instrument. The root note is the pitch where the sample sounds untransposed. This provides a focus for editing a key zone. |
|
Fri Apr 11, 2003 4:40 am |
|
|
Dani
Joined: 28 Jan 2003
Posts: 35
|
|
|
|
Samplers
Memory.
Memory is probably one of the most important factors when thinking about sampling. It determines how much sample time you have got available and the quality of the samples.
The memory of a sampler is RAM Random Access Memory, the same stuff you find in your PC or MAC. As the technology of RAM moves forward so does that of the samplers. Speed increases and the reduction in price are also felt in the music technology community.
Early samplers had as little as 512kilobytes of RAM and sampled at a highest rate of 8bit. You could only sample up to about 15 seconds.
The Casio FZ1 had 1MB of RAM as standard but could be upgraded to a massive 2MB! This model sampled at 16Bit 36k (not quite CD standard). It could also process stereo samples. But obviousy that cut down the sample time considerably. 1 megabyte allows about 0.5 million sample points, giving a recording time of 15.625 seconds.
As RAM is volitile memory, it means that everytime the sampler is used, you have to load all your samples into RAM, if the only media you have available is a floppy disk, then this is very time consuming. Imagine loading in 8MB of samples from floppy disks, in the middle of a gig!
It wasn't until manufacturers started building samplers with hard drives and SCSI connections that sample RAM started to really increase.
It's ironic to think that when I had my lowley FZ1 back in the eighties, stugging to fit a drum kit, bass and piano into 1MB and all I wanted was more RAM and a hard drive, that the PC I'm using now has 479MB of sample RAM available, 160GB Hard Drive space and a great sampler built into hardware and I never ever use samples!
Keygroup
A keygroup is number of keys the sample will be present on. If you've sampled, say, a piano at middle C, you don't want that same sample to play at top C because it would sound nothing like a piano. (Which is ok if you are doing avant gard!) SO we tell the sampler "only play this sample on these keys". You then assign another sample to the next key group along to get your full multisampled piano.
Original Pitch
Origional pitch is the sound as it was first sampled. Taking a sound to far from it's origional pitch will result in either distortion and break up when moving down in pitch and a speeded up chipmunk like quality when moving up.
Sample Assignment
This is how you assign a sample to a key, think of a drum kit. You would assign a different drum sample to each key of the keyboard. You probably wouldn't assign a sample to a keygroup though, unless you want differently tuned snares or something.
Keyboard Track/Constant Pitch
Keyboard tracking is when the sampler slows down or speeds up the sample to match the change in pitch that accompanies a key press. Constant pitch is when the pitch remains the same regardless of what key is pressed.
LFO
Low Frequency Oscillator, uses a very low frequency waveform to create continuous (periodic) change to the volume, pitch, or timbre of a sound.
Modern samplers have amultitude of different waveforms to choose from. A sine wave, perhaps, to simulate vibrato or a random wave to create a unique sound.
What are the main requirements of a glitch free loop?
That the start and finish point are at the same level or crossfaded. and that the wave itself doesn't have any noticable artifacts that will present them selves when heard over and over again. A slight raise in volume in the middle of a string loop will really stand out when changing in pitch and will give a fluttering effect/
Why are closed and open hi-hats often sent to the same mute group?
Because in real life the two would never make a sound at the same time. You would want the closed high-hat to mute the open high-hat when played.
On hardware samplers, how much sampletime would you typically get out of 64MB storage
Sixteen minutes. |
|
Sun Apr 13, 2003 10:45 am |
|
|
|
|
Aidan
Joined: 08 Jan 2003
Posts: 38
|
|
|
|
WEEK 6 SAMPLERS 1
FEATURES
VOICES
A sampler is essentially a tapeless digital recording device much like a multi-channel synth except it uses samples rather than preset patches.Theses samples are the basic building blocks and will have names,sample rates etc.A voice is a block that contains two or more samples with details of keboard ,velocity settings,modulation etc.
SAMPLER HIERARCHY
SAMPLE
The sampler hierarchy is samples,key groups and programs.Programs are made up of keyboard settings which are made up of 1 or more voices plus effects,velocity settings etc.
SAMPLING
RECORDING
The original aim of samplers was to sample individual notes or sounds so that they could be played back on a keyboard much like any other keyboard.With advances in technology longer sampling rates are now available giving rise to recording of whole musical phrases,which can be played back from a single key.This proved very useful for the sampling of vocals which allows you to copy lines or choruses for insertion at appropriate points.You could also sample drum rhythms at same tempo, allocate them on keyboard in groups and effectively play the drums on the keyboard.
CREATION OF KEY GROUPS AND PROGRAMS
LOOP TYPE
Samplers have different ways of triggering sounds such as \one shot trigger mode and retrigger mode.One shot mode is where when you hit a key you want the original sample to carry on to its natural conclusion while a newly triggered one plays over the top.In retrigger mode you can get the original sound to stop and then trigger again from the beginning-the classic n-n-n-nineteen effect.
ENVELOPE
To recreate the effect of a sounds natural decay some samplers have envelope shapers to allow you modify the envelope of the sample sound.In most cases the attack of the originalsound can be left s it aws,but a new decay has to be created to stop the sound stopping abruptly when you release the key.In some cases such as an organ sound an abrupt stop is ok but most instruments have slower decay time which can be duplicated by using the release phase of the envelope shaper.You can also use the envelope shaper more creatively to set up any envelope you like just like a conventional synth.
MISC QUESTIONS.
1.) Aside from a MIDI lead you could use an SCSI connection which allows you to transfer samples over SMDI-a protocol that uses the SCSI connection.
3.)Because sample sounds are stored in RAM the maximum sampling time is always limited by the amount of memory you have.In order to maximise sample time and memory you can use sample looping.Lots of sounds have a distinctive attack portion after which the sound becomes more consistent.There is therefore no reason to sample the whole sound but just the initial attack phase and then loop this over and over.This saves on storage space. |
|
Sun Apr 13, 2003 5:56 pm |
|
|
|
|
All times are GMT. The time now is Sat May 17, 2008 3:57 am
|
|
|
|
| |
