Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Bob Olhsson]

jazzius wrote on Tue, 19 October 2004 03:37

Is it possible that some future technology will incorporate the best of both worlds (analog and digital?)....is the human brain analog or digital?

The top supercomputer designers say the next step in computing will be digitally controlled analog computing. Why have each clock cycle represent one of two voltages when it could easily represent one of 20 using today's analog technology? We only used two because of the noise limitations of late 1950s analog technology. We only continue using two because of the huge amount of virtually free code available combined with the status-quo being in the financial interest of a number of large corporations.

[ted nightshade]

Mmmm, digital controlled analog. Repeatability possibilities, real tight calibration, definitely an interesting prospect.

[danlavry]

Bob Olhsson wrote on Wed, 20 October 2004 15:27

jazzius wrote on Tue, 19 October 2004 03:37 Is it possible that some future technology will incorporate the best of both worlds (analog and digital?)....is the human brain analog or digital? The top supercomputer designers say the next step in computing will be digitally controlled analog computing. Why have each clock cycle represent one of two voltages when it could easily represent one of 20 using today's analog technology? We only used two because of the noise limitations of late 1950s analog technology. We only continue using two because of the huge amount of virtually free code available combined with the status-quo being in the financial interest of a number of large corporations.

I believe it is so because it is easier to make a 2 state hardware. In the old days, we made memories out of magnetics, a 2 pole device, say N up is a "1" and S up is a "0". We then moved to resistor diode logic (diode is 2 direction device, conduct or block), and to transistor logic (transistor on or off) and so on.

A less easy to understand issue is power and heat. Being close to ground or to the rail, the devices run much cooler. Being at a midpoint between Ground and a rail is takes a lot of energy. One can design around it but it gets complex.

BR
Dan Lavry  

[Joe Bryan]

Bob Olhsson wrote on Wed, 20 October 2004 07:27

jazzius wrote on Tue, 19 October 2004 03:37 Is it possible that some future technology will incorporate the best of both worlds (analog and digital?)....is the human brain analog or digital? The top supercomputer designers say the next step in computing will be digitally controlled analog computing. Why have each clock cycle represent one of two voltages when it could easily represent one of 20 using today's analog technology? We only used two because of the noise limitations of late 1950s analog technology. We only continue using two because of the huge amount of virtually free code available combined with the status-quo being in the financial interest of a number of large corporations.

Multi-level signalling is used more and more often these days. Many high-speed digital interconnects use this to increase bandwidth without increasing clock rates.

As for whether it's important that digital audio designers have analog backgrounds, I believe it's critical. Digital design is trivial compared to analog design. I don't condider someone an engineer unless they have the proper analog background, and anyone designing audio gear MUST have good ears.

-Joe

Joe Bryan
VP Engineering
Universal Audio

[Rebecca Robinson]

Dan,
can you reccomend any books/resources for those of us wishing to learn more about DSP?
Thanks

[danlavry]

Rebecca Robinson wrote on Fri, 29 October 2004 09:20

Dan, can you reccomend any books/resources for those of us wishing to learn more about DSP? Thanks

The first book I got was
"Discrete-Time Signal Processing"
Alan V. Oppenhieim and Ronald W. Schafer
Prentice Hall Signal Processing Series. 1989 ISBN 0-13-216292-X

But there are a number of books by Rabbiner, Gold, Oppenhime and Schafer, where 2 or 3 of them got together and wrote a DSP book, with some overlap between the books. They are good textbooks for EE students, but not "easy reading".

If you want a good book, less theoretical "hands on", I liked
"Digital Signal Processing, A Practical Approach"
Emmanuel C. Ifeachor and Barrie W. Jervis
Addison Wesly publishing, 1993 IBSN 0 201 54413 X

This last one is a good introduction. There is no one book that gives you everything, and you do need some math background. My tutorials on the www.lavryengineering.com (under support) are for people that want basic common sense understanding of filters, with zero math.

 

[Level]

Dan...you old fucker you....you did come in at a good time. 10 years..and you have a good "foothold" of the goings on.

I do design but I also assimilate..which is to take a set of existing ideas and modify them VS original ideas and IDEALS and try to fit them into the scope of diversity of all the crap we are forced to work with. Don't get me wrong..I am not the master designer of circuitry or am I someone that will "rock the World" so to speak..but since my first purchase of a digital recording (on vinyl mind you) of Ry Cooder, Circa 1976 and I was Actively recording at that time, developing loudspeakers and just pulling it all in.....


Set standards of the industry got screwed up big time. I blame monitors and rooms first..then non conformity of knowns...... second.


(0.....) is phrasing of my speaking so use them accordingly.......


To see what the largest issues of the recording arts are..one must see where the foul ups occur.

opportunity best oppurtunity as a "fix" is to provide some "sane standards" as to level, balance and translation.

audiophilia take audiophilia for example.

"they" are in the business of "remastering" the DAC's through DAC's, Formats, resistors, capacitors, power cords, schemes, demons and loudspeaker design. It grew "out of hand" around 1985.
Quotes:
"Our common denominator that we work toward is our weakest link"

"Our excellence dictates we are non conformists"

"Our Art dictates advancement of emotion universally"

Now, put these three together..and speak to me.

[JGreenslade]

Bob O wrote:

Quote:

I'm told that analog actually has a much brighter future than most people assume.

Something I find hilarious is that when you venture into consumer emporiums, certain gear has "digital ready" invariably printed on it in gold writing. With some modern analogue amps / processors having bandwidths up to 300K, shouldn't they have bright letters decreeing "analogue ready"? :-)

An associate of mine has been designing gear since the late '50s, and worked at firms such as Leak / Vox / many others. One day I asked him how he became an analogue specialist rather than digital - "I couldn't see digital taking off, it always struck me analogue would eventually prevail". He went on to design numerous analogue computers, and was particularly keen on base-5 logic.

Maybe one day Ron will have the last laugh :-)

Cheers,
Justin