Page 3 of 4

Re: Analog sampler?

Posted: Fri Nov 23, 2012 1:17 am
by Stab Frenzy
gmeredith wrote:I think this is the one I have here:

http://experimentalistsanonymous.com/di ... corder.pdf

It states in the beginning:
Invox’ proprietary analog/multi-leval storage technology
is implemented in advanced Flash non-volatile memory
cells, each of which can typically store more than 256
voltage levels. The APR9301 stores and reproduces voice
signals in their natural forms, eliminating the distortion
that is often introduced by encoding and compression.
So I guess that means analog storage as opposed to digital storage and AD/DA encoding/decoding and aliasing filters.

Cheers, graham
Non-binary digital, but digital nonetheless. IMO.

Re: Analog sampler?

Posted: Fri Nov 23, 2012 1:35 am
by gmeredith
But from my understanding of it the waveform is stored as an analog voltage, not a number. It stores it similar to a cassette tape - except it is in memory capacitors, which can store a variable range of voltages, like tape. Whereas digital recording has signals encoded as a series of on/off binary transistor switches representing a binary number (the definition of "digital") which have to be decoded to be heard. I know this is mostly academic, and I get your point though.

Gridsleep, the original poster, asked "Is there any sampler that stores actual analog waves and plays them back directly as analog signals", and when I read up on the specs of my sampler's chip it seemed to describe its process as being of this type of analog process.

My sampler circuit definitely has its own distinctive sound.

What do others interpret this as?? Analog? Digital? a very interesting topic :)

Cheers, Graham

Re: Analog sampler?

Posted: Fri Nov 23, 2012 3:18 am
by Stab Frenzy
gmeredith wrote:But from my understanding of it the waveform is stored as an analog voltage, not a number. It stores it similar to a cassette tape - except it is in memory capacitors, which can store a variable range of voltages, like tape. Whereas digital recording has signals encoded as a series of on/off binary transistor switches representing a binary number (the definition of "digital") which have to be decoded to be heard. I know this is mostly academic, and I get your point though.
Digital doesn't mean binary, binary digital is a subset of digital overall.

If you have 256 possible values and they're stored as 1-256 rather than 00000000 to 11111111 that doesn't make them any less digital. Being able to store 256 values in one slot as opposed to just on or off is a step forward, but it's not analogue.
Gridsleep, the original poster, asked "Is there any sampler that stores actual analog waves and plays them back directly as analog signals", and when I read up on the specs of my sampler's chip it seemed to describe its process as being of this type of analog process.

My sampler circuit definitely has its own distinctive sound.

What do others interpret this as?? Analog? Digital? a very interesting topic :)

Cheers, Graham
Plenty of digital devices have their own distinctive sound, distinctive doesn't mean analogue. :thumbright:

Re: Analog sampler?

Posted: Fri Nov 23, 2012 3:58 am
by gmeredith
Sure, I get what you mean - the "digital" is the quantisation factor of the amount of "slots" available for storage of voltages - and would be analogous to bit height of a digital binary sampler. In that same way, analog tape recorders are also "digital", since the tape coating isn't continuous, but actually has a distinct crystalline magnetic structure, where each iron oxide crystal on the tape is a tiny single magnet of definite size and shape - it also quantises the signal, though there are much more crystals on the tape than the memory chip capacitors referred to, so the quality is better. So in that line of thinking, even tape is "digital".

But having said that, do you think that the common term "digital" which is used in reference to musical instruments is actually specifically referring to BINARY devices which encode/decode binary numbers rather than what we've been talking about?

Re: Analog sampler?

Posted: Fri Nov 23, 2012 4:24 am
by Stab Frenzy
Actually no, tape doesn't work that way. There may be a finite number of particles on the tape but the magnetic influence of the particles is continuous as the head isn't detecting a single particle at a time, it's seeing a large number of them and responding to an average of what's effecting it. As the tape moves even the tiniest amount it is effect slightly more by some particles and slightly less by others. The change is continuous and not simply moving from one value to the next, and as you move from one point to the next you pass through all possible values in between.

Re: Analog sampler?

Posted: Sun Nov 25, 2012 12:11 am
by gridsleep
Back to laser storage, I refer to a Wired article:

http://www.wired.com/science/discoverie ... 5/10/69033

Scientists at Australian National University in Canberra were able to fire a laser bearing a discreet signal (information) into a crystal where the light speed was slowed dramatically. The light was able to be stored in the crystal for up to a second (a record in 2005) and when it was let out back into the active beam, the original information was intact. Data stored analogically in a crystal. It's already been done. It just needs to be refined. Imagine opening your sampler synth keyboard and seeing not electronic circuits and wires, but crystals and fiber optic cables. How cool would that be?

If a single photon can act as a transistor, as they have shown, imagine how much musical information can be stored in even a short pulse of light. Entire symphonies, perhaps.

Research continues.

Re: Analog sampler?

Posted: Mon Apr 22, 2013 3:30 am
by calaverasgrande
from what I recall about wire recording the wire does not hold up under heavy use. Heat is the enemy of magnetism, and the magnetic signal on the wire is very faint. Also, one of the reasons to move to tape is that it had a better signal to noise ratio than wire.

As far as what constitutes digital. From what I understand pretty much all analog delays could be considered digital in a way. They do slice up the voltage of the waveform into discrete voltage levels. Then shuffle those through a series of capicitors based on a clock. Finally spitting them out at the end and lowpassing the result to "smear" the graininess of the stepping.
The difference is that there are no stairsteps induced by quanitisation.

I think disc based playback is probably the best idea. Might be able to work something out with an older full height SCSI drive. A lot of those have multiple platters with magnetic material on both sides. A larger capacity 18 GB drive (from the 90s!) might have 5 discs. That gives you 10 heads, 10 surfaces to play back from simultaneously.
The drives are meant to turn at 7200rpm or 10,000 rpm, But you might be able to change the circuit to spin slower, then possibly convince the heads to read the samples at various speeds for the notes in a scale.

Re: Analog sampler?

Posted: Tue Oct 29, 2013 9:09 am
by Clearsong
This might not count as sampling, but when I was a kid, I used to record sequences of sounds from a cheap AM radio plugged into a couple of my dad's guitar pedals and then record that into a boombox (aka, ghetto blaster). The "sequencing" effect was achieved by pressing record/play every couple of seconds as I fiddled with the tuning. Then I went to school and put the tapes into 4 tape recorders my band teacher had and I would listen to the random polyphony. I was disappointed by the results nearly every time, but it yielded a few good things.

Truthfully, the first device ever to be called a sampler was digital as far as I know. It was the EMS Musys system and before that you just had pop/rock guys like the Beatles using keyboards that triggered tapes. And before them, you had brilliant geniuses making tape music, which used a lot of sound recordings as source material.

Re: Analog sampler?

Posted: Tue Oct 29, 2013 4:06 pm
by sequentialsoftshock
Clearsong wrote:This might not count as sampling, but when I was a kid, I used to record sequences of sounds from a cheap AM radio plugged into a couple of my dad's guitar pedals and then record that into a boombox (aka, ghetto blaster). The "sequencing" effect was achieved by pressing record/play every couple of seconds as I fiddled with the tuning. Then I went to school and put the tapes into 4 tape recorders my band teacher had and I would listen to the random polyphony. I was disappointed by the results nearly every time, but it yielded a few good things.

Truthfully, the first device ever to be called a sampler was digital as far as I know. It was the EMS Musys system and before that you just had pop/rock guys like the Beatles using keyboards that triggered tapes. And before them, you had brilliant geniuses making tape music, which used a lot of sound recordings as source material.

Huh ?

Wrong wrong wrong.

Re: Analog sampler?

Posted: Tue Oct 29, 2013 6:47 pm
by cornutt
Clearsong wrote: Truthfully, the first device ever to be called a sampler was digital as far as I know. It was the EMS Musys system and before that you just had pop/rock guys like the Beatles using keyboards that triggered tapes. And before them, you had brilliant geniuses making tape music, which used a lot of sound recordings as source material.
Well, the "keyboards that triggered tapes" were the Mellotron/Chamberlin, which have been much discussed. They were basically analog romplers, using tape as the storage medium. And then there was the Opitgan, which was also an analog rompler, using an optical disc as the storage medium.

But before those things, and before proper synthesizers, there was a whole school of electronic music known as the "tape studio". The BBC Radiophonic Workshop and the San Francisco Tape Music Center were two pioneering institutions in tape studio techniques. Tape studio basically involved taking a whole bunch of tapes, cutting them into bits of various lengths, speeding them up, slowing them down, making loops, and filtering them. The source material was sometimes musical intstruments, but more often it was field recordings and found sounds. The original version of the Dr. Who theme (circa 1962) was done using tape studio techniques.

Re: Analog sampler?

Posted: Tue Oct 29, 2013 8:06 pm
by commodorejohn
If you set up a bucket-brigade device with a 1:1 feedback from output to input, and a zero-feedback initial-setup mode, you'd have a self-sustaining analog sampler of n stages. Granted, you'd need a good number of typical 1024-stage BBDs to make a buffer of any significant length, but it could be done. It'd still be time-quantized, so it wouldn't be quite as free of aliasing as a Mellotron-type device, but if BBD delays count as "analog" I don't see why this wouldn't. Be interesting to hear what the subtle variations of analog circuitry would do to the re-buffered sound over time...

Re: Analog sampler?

Posted: Wed Oct 30, 2013 3:51 am
by Stab Frenzy
commodorejohn wrote:Be interesting to hear what the subtle variations of analog circuitry would do to the re-buffered sound over time...
I guess you've never used BBDs before. :lol: It would be squalling feedback within a couple of seconds.

Re: Analog sampler?

Posted: Wed Oct 30, 2013 5:36 am
by commodorejohn
Heh, no...only understand the basic concept, I've got a lot more brushing up to do before I'm ready to fool around with circuit-building...

Re: Analog sampler?

Posted: Wed Oct 30, 2013 8:58 am
by Stab Frenzy
BBDs can only hold the signal for a few fractions of a second, even the best ones like the MF-104SD have a really dark tone because the high frequency is lost just by sitting in the BBD caps for a second.

Re: Analog sampler?

Posted: Wed Oct 30, 2013 10:29 am
by pflosi
Isn't the "dark tone" mainly the filter filtering out the clock noise?