How do I know the available room in my sampler?

• Regarding the definition we know that a sampler takes every second x sample of the sound ( or sample word where x is the sample rate in hz) with a multiplier of 1, 1,5, 2, depending on whether the resolution is 8, 12, 16 bits. The result is expressed with octets.
• Don't forget that due to the binary form of number in samplers 1 octet is 8 bits, but 1Ko =1024 octets, 1Mo= 1024x1024 octets.

• So, 1 second of sound occupies at 44,1 KHz 16 bits resolution, 88200 octets. Twice in stereo.
  
  
• At the opposite, with my sampler having 16 bit resolution,, at 44,1kHz, 8Mo of RAM memory how long a sample can I fit in RAM?
  • That's 8x1024x1024 divided by 44100x2 = 95 seconds of monophonic sample

How to calculate a drum loop's tempo?

• Tempo is expressed in bpm (beats per minute, that is the number of quarter notes in 60 seconds)
  
  
• All you have to do is first get the number of seconds between two quarter notes in the loop.To do so it's better to have a graphic display of the loop.If you have enough musical background you can use the location of, say the snare drum or divide by 4 ( if you're playing 4/4) the exact length of the repetitive motif.
  
  
  • Then when you have that y value between two quarter notes and you want x in bpm you apply the simple formula x = 60/y.

    Back to the top

How to change a drum loop's tempo?

• You have to use a time stretching algorithm which preserves the pitch of the sample using this simple rule
  • If x1 is the original tempo and x2 the desired tempo apply a factor of x2/x1.

    
    

    Back to the top

Do I need a computer to work with samples?

• It's true that some samplers have a large enough graphic display to overlook the shape of the sample, but I don't think you will be able to work precisely or professionaly on the looping side, even on some basic features ( optimisation, cut, time stretching).
• Furthermore due to the power of computer microprocessors, on compex operations such as time stretching or pitch shifting you will be able to work more quickly, no doubt about it.
• First thing you have to do is transfer the sample file from the sampler to the computer.To do so you have to work with SCSI because with MIDI the advantage on gaining time would be very questionable.You got to have a software that offers this possibility ( such as Alchemy on Mac, more about it at the adress: http://www.Passportdesigns.com/alch.html) and a sampler that has SCSI transfer possibilities (Smidi for example) (Emu,Ensoniq, Akai,Kurzweil).
• The second step is to pick the right software for the job.Today on the Mac platform the best beside Alchemy( limited in looping) is Infinity from Antares (URL:http://www.richarde.com/web/Infinity.html ).Actually version 2.5 this tool allows some SCSI transfer for Akai S samplers and Ensoniq ASR, but most of all gives you real time algorithmic possibility to loop samples that were up to now judged impossible to handle.
  
  
• In the real world, having both softwares will help you resolve almost every situation, if you're ready to deal with your own sounds or from audio version of sampling CDs.

  Back to the top

How to loop a sample?

• 1) First pick up a stable zone the longest the better, trying to preserve the sample attack and keeping away from the sample's end.

  
  
  2) When arriving at the loop end point ( where the sample jumps back to the start point), to avoid any discontinuity you have to find two points of same amplitude and use them as loop points.

• To be sure the levels are the same, the simplest solution is to take the value of 0 volt.This location is called 'Zero Crossing'.
• Most samplers and softwares are able to find these points automatically.

  Nevertheless, electrically correct points can be of no musical interest.
  
  

3) Another important criteria, especially for bidirectionnal loops is to pick a sample's part where there is no mismatch in the sample's direction and where the level stays fixed.

This points are called: Zero Slope Points.

4) At last and this is a general rule, you have to achieve a relative correlation between what's before and what's after the loop.

Back to the top

What is the detune of a loop ?

A very usual problem is the detune of short loops.
This annoying transposition is due to a relation between the sampling rate and the pitch of the sampled instrument.
For exemple, let's say you sample at 44kHz a flute playing a A 440. One cycle of sample will be 44000/400 = 100 sample word long.If the loop is shorter or longer, the cycle length is altered and its pitch different from the rest of the sample. To tune the loop you'll have to pick a loop that's a mulptiplier of 100 sample word in this particuliar case.
With the help of a software it's possible to detect the pitch of the sample and therefore jump to the required conclusion.



Back to the top

What is Crossfade Looping ?

• Even with perfect loops,it's possible to have audible flaws, such as click, pop, phases .

• As a matter of fact the sample has to be evolving in an optimal way when arriving at the loop points then go back to its original shape.It's important to reduce any discontinuities without altering the features of the sound.
• This optimisation is done by changing the information from both sides of the loop, hence the name 'crossfade looping'.It's a great help too, to make better a good loop.
• The crossfade needs some sample after the loop end, that's why in the first place you have to preserve that security zone.
• Crossfade can be uni or bidirectionnal.
• In order to create a new sample the wave begins by fading out before the start point just as the original sample fades out at the same distance of the end point.The result is a new sample that starts and ends as the original but is modified around the loop points.
  
  
• Two algoritrhms may be found in the crossfade technique: Linear Power crossfade ( follows a linear progression, to be used if the sample is very identical around the loop points) and the Equal Power crossfade ( follows a square progession, if the sample is different around the loop points).

  Back to the top

• Some looping advices.
• Try to find a musically satisfying place for the loop ( natural progression or use the autolooping function) and try to get rid of the flaws, staying very close to the points.
  
  
• Do not include the attack of the sample and keep a little samples after the end point.
  
  
• The longest loops are generally the better because less noticable by the human ear.
  
  
• Try to work on raw material, no LFO,chorus...
  
  
• Do not use crossfade looping to sole an inherent problem in the sample because as far as looping is concerned it's the rule " garbage in , garbage out".
  
  
• Generally the longest crossfade the better.The maximum crossfade is the shortest of these three elements:
  • The distance between the start of the sample and the loop start point.
  • The distance between the loop end point and the sample's end.
  • Half of the distance between the loop points.
    
    
    
• Some samples such as the piano-like sounds need a short loop located at the fading out volume portion of the sound.
  
  
• To get a short loop to be less noticable, you can either compress the loop or add a LFO effect ( vibrato, tremolo) or phasing to keep the ear busy.
  
  
• A sample with lots of change in its evolution is likely to be better looped in reverse mode.
  
  
• Some sounds mixing tuned elements and noise (cymbals, sanre drums) may be hard to loop in forward mode.
  
  
• Play back the sample at different pitch, some perfect loops may become very audible.
  
  
• Let your ear be the judge, arm yourself with patience and use the right tools for the job.
  
  
• As a conclusion, nobody can achieve perfect loop before having tried on lots of loops.Keep on trying and this insurmountable task will become possible, and why not easy.

Back to the top

Cross sampler compatibility.
There are that two or three years, the motivation to buy such or such sampler (outside its price, and its technical characteristics) was its sound library.
Well, things have changed since many maufacturers have "cracked" formats of each other, proposing conversion algorithms with cross-platform compatibility.
Thus in addition to the sampler you buy the right to read (by SCSI almost exclusively, that is Hard Disk, CD ROM and the like) the sound libarry of competitors, and that is for the better.
The AKAI S 1000 having been the first 16 bits, 44.1 kHz of the market, this format ( S1000 CD ROM) has remained standard in the world of the digital world and almost all marks know how to recognize it (with more or less happiness).
Let's emphasize nevertheless that the conversion can sometimes leave aside some parameters of the patch as envelopes, aftertouch adjustments , panoramic positions a.s.o.
Similarly, the conversion time can be short, for example a n EmuIV converting a S 1000 format , or long, for example a S 760 Roland reading the same format (10 times more).

Here is today the chart of compatibility with some precisions, for exemple transfer with a software , that can sometimes fulfill gaps between one manufacturer and another.

X=RECOGNIZED

	AKAI S2000-S3000	EMU 32SI	E64 EIV	ENSONIQ ASR 10	KURZWEIL K2000 K2500	ROLAND S760
Version OS	1.5 et 1.6	2.10	2.0	3.53	3.18	2.20
AKAI S1000	X	X	X	X	X	X
EMU III	X	X	X	.	X	.
ENSONIQ	.	.	.	.	X	X
ROLAND S700	.	.	.	X	X	X
AUTRE	S950-S900	EMAX II	EMAXII ESI32	.	S3000	.
SDS	X	X	X	.	X	X
SMDI	SCSI	X	X	SCSI	X	.

TECHNICAL SPECIFICATIONS

• SPECIFICATIONS DU SDS

The MIDI SDS was adopted in January 1986 by the MIDI Manufacturers Association and the Japanese MIDI Standards Committee. The SDS defines the standard method for transfer of sound sample data between MIDI-equipped devices. Sample dumps may be accomplished with either an 'open loop' or 'closed loop' system.
The open loop method simply involves the straight dump of all sample data from its source to the destination, with no timeouts,packet acknowledgements, or any other form of handshaking, much as in the manner of a sysex bulk dump, usually intiated at the source. The closed loop method allows the use of handshaking messages between the dump source and destination, and usually places the dump process under the control of theslave, to allow it time to process the incoming data as necessary.
As with any standard, it can not be assumed that a device adheres to it unless the accompanying documentation specifically indicates it. Even then, it is best to check its conformity with non-critical data.

SPEC: SAMPLE DUMP FORMATS
DUMP HEADER:

F0 7E cc 01 ss ss ee ff ff ff gg gg gg hh hh hh ii ii ii jj F7

where
cc = channel number
ss ss = sample number (LSB first)
ee = sample format (number of significant bits; 8->28)
ff ff ff = sample period (1/sample rate) in nanoseconds (LSB first)
gg gg gg = sample length, in words
hh hh hh = sustain loop start point (word number) (LSB first)
ii ii ii = sustain loop end point (word number) (LSB first)
jj = loop type (00:forwards only; 01:alternating)

DATA PACKET:

F0 7E cc 02 kk <120 bytes> mm F7

where
cc = channel number
kk = running packet count (00->7F)
mm = checksum (XOR of 7E, cc, 02, kk <120 bytes>)

The total size of a data packet is 127 bytes. This is to avoid overflow of the MIDI input buffer of a device that may want to receive an entire packet before processing it.
A data packet consists of its own header, a packet number, 120 bytes of
data, a checksum, and an EOX. The packet number begins at 00 and increments with each new packet. It resets to 00 after it reaches 7F, and continues counting.
The packet number is used by the receiver to distinguish between a new data packet, or a resend of a previous packet. The packet number is followed by 120 bytes of data, which form 60, 40, or 30 words (MSB first for multiword samples), depending on the length of a single data sample.
Each data byte hold seven bits, with the msb in each byte set to 0, in order to conform to the requirements of MIDI data transmission. Information is left justified within the 7-bit bytes, and unused bits are filled with 0.
Example: Assume a data point in the memory of a 16-bit sampler, with the value 87E5. In binary, that would be:

1000 0111 1110 0101
and would be encoded as the following MIDI data stream:
01000011 01111001 00100000

The checksum is the running XOR of all the data after the SYSEX byte, up to but not including the checksum itself.