Register to reply

Analog Memory?

by Jyan
Tags: analog, memory
Share this thread:
Jyan
#1
Feb27-14, 09:11 PM
P: 33
Is anyone aware of physical or mathematical reasons why "analog" memory is not possible? That is, a system that takes in a continuous waveform, saves it, and then later releases the same continuous waveform without any sort of sampling or interpolation. I certainly can't think of a way to do this. But the point of my question is about the existance of a mathematical theorem or physical law that makes this type of system impossible. This seems like it may belong in the Engineering forum, but I think it is more of a question about the underlying physics rather than any sort of engineering issue.
Phys.Org News Partner Physics news on Phys.org
Engineers develop new sensor to detect tiny individual nanoparticles
Tiny particles have big potential in debate over nuclear proliferation
Ray tracing and beyond
ModusPwnd
#2
Feb27-14, 09:12 PM
P: 1,071
Sounds like a wax cylinder.
AlephZero
#3
Feb27-14, 09:44 PM
Engineering
Sci Advisor
HW Helper
Thanks
P: 7,177
Quote Quote by Jyan View Post
I certainly can't think of a way to do this.
http://en.wikipedia.org/wiki/Gramophone_record

Quote Quote by ModusPwnd View Post
Sounds like a wax cylinder.
I guess you are too young to have known when vinyl was the main option for music recording (good quality analog tape recorders and reel-to-reel tape were way too expensive and fiddly for most people, and cassette tapes were the equivalent of over-compressed MP3 files today).

Or, you assume that "newer technology" automatically means "better quality"

ModusPwnd
#4
Feb27-14, 09:45 PM
P: 1,071
Analog Memory?

Huh? Im just saying that what he is describing sounds like a wax cylinder (among other things). You think I know about wax cylinders but not vinyl?
256bits
#5
Feb27-14, 09:56 PM
P: 1,484
Even a mechanical systems have an 'analog' memory. On a a roller coaster ride, the car follows the curve of the track without any digital sampling or interpolation. The memory is the layout of the track and all cars will follow the same wave pattern.
Jyan
#6
Feb27-14, 09:58 PM
P: 33
This seems like a really stupid question now :/
Devils
#7
Feb27-14, 10:08 PM
P: 175
Analog delay lines are close to what you are referring to. They were superseded decades ago.

http://en.wikipedia.org/wiki/Analog_delay_line
berkeman
#8
Feb27-14, 10:18 PM
Mentor
berkeman's Avatar
P: 41,128
Quote Quote by Jyan View Post
This seems like a really stupid question now :/
Not stupid at all. Great responses
AlephZero
#9
Feb27-14, 11:50 PM
Engineering
Sci Advisor
HW Helper
Thanks
P: 7,177
Quote Quote by ModusPwnd View Post
You think I know about wax cylinders but not vinyl?
I've learned the hard way the second guessing what other people know is often a bad plan.

It would be quite possible for somebody to know about wax cylinders from a history lesson, but have never actually heard a vinyl record played on high-quality equipment (and broadcasting the music on FM radio with a high frequency limit of 16kHz doesn't count as "high quality.")
Drakkith
#10
Feb28-14, 03:36 PM
Mentor
Drakkith's Avatar
P: 11,898
Quote Quote by Jyan View Post
This seems like a really stupid question now :/
Not at all. I actually learned something.
sophiecentaur
#11
Feb28-14, 03:47 PM
Sci Advisor
Thanks
PF Gold
sophiecentaur's Avatar
P: 12,203
Quote Quote by Jyan View Post
Is anyone aware of physical or mathematical reasons why "analog" memory is not possible? That is, a system that takes in a continuous waveform, saves it, and then later releases the same continuous waveform without any sort of sampling or interpolation. I certainly can't think of a way to do this. But the point of my question is about the existance of a mathematical theorem or physical law that makes this type of system impossible. This seems like it may belong in the Engineering forum, but I think it is more of a question about the underlying physics rather than any sort of engineering issue.
Did you ever hear of the 'Bucket Brigade' analogue shift register / delay?
(Ref this wiki link)

That could qualify for what you're after and there are no 'moving parts'.
cjl
#12
Feb28-14, 03:54 PM
P: 1,018
Quote Quote by AlephZero View Post
Or, you assume that "newer technology" automatically means "better quality"
In the case of CDs vs vinyl, it does. Sadly, poor mastering has ruined the quality of a great many CDs, but in terms of actual audio fidelity, the CD beats the vinyl record hands down.
analogdesign
#13
Feb28-14, 06:00 PM
P: 534
Quote Quote by sophiecentaur View Post
Did you ever hear of the 'Bucket Brigade' analogue shift register / delay?
(Ref this wiki link)

That could qualify for what you're after and there are no 'moving parts'.
The bucket brigade is a sampled-data device so it strictly a continuous waveform processing device as the OP asked (although it obviously can be used on a continuous signal if Nyquist is respected and a reconstruction filter is used).

Another option is a Mercury Delay Line. This was used as an early analog memory in Radar systems during and after WWII. It was also used as one of the first digital computer memories.

http://www.computerhistory.org/revol...-storage/8/309
sophiecentaur
#14
Mar1-14, 01:42 PM
Sci Advisor
Thanks
PF Gold
sophiecentaur's Avatar
P: 12,203
The most impressive analogue memory / delay line I have come across was the 20ms, field store quartz delay lines that were used in the early analogue TV standards converters. They carried a video signal and the delay was achieved by multiple reflections on the (internal) faces of a quartz polygon, a couple of cm thick and the size of a dinner plate. Pretty impressive stuff and a lot more compact than the digital field stores that replaced them (at the time).
CWatters
#15
Mar1-14, 06:07 PM
P: 3,154
Multi head tape recorders and spring delay lines were a form of analogue memory used years ago. Analogue data put in one end came out the other a short time later (somewhat distorted usually).

I suppose it might be possible to bounce an analogue signal off the moon. I calculate you could store about 2.5 seconds worth in the round trip distance.


Register to reply

Related Discussions
In binary can we have a value with deci centi mili or more lower valued prefix? Computers 14
Meaning of Scaled and Inverted in context of picoammeter analog voltage output Electrical Engineering 3
New FQXI essay contest Is Reality Digital or Analog? Beyond the Standard Model 41
The Bender and Orszag analog for PDE Science & Math Textbooks 0
Memory span as the quantum of thought plus golden ratio the basis of everything? Medical Sciences 3