avatar

reynaud

Silverado

Posts: 196 Member Since:13/07/2011

#21 [url]

Mar 4 14 2:18 AM

dcollins wrote:


Thank you for the link Dave. Good thing I understand Dutch as the majority of the interesting notes aren't in English. 

Glad they've made this information publicly available, good weekend reading. 

cheers,
Reynaud

Quote    Reply   
avatar

dcollins

Platinum Blonde

Posts: 2,343 Member Since:27/01/2011

#22 [url]

Mar 4 14 12:48 PM

I think most of them are also translated. His papers for Burroughs are all in English. I don’t pretend to understand all of what EWD says there, but many moons ago I think he taught me to look at programming in a different, pardon the expression, holistic way.

That doesn’t mean I like, or even can code very well, but there is an enthusiasm and obvious genius to the writing that is appealing even if I’m only getting 1% of it.


DC

Quote    Reply   
avatar

reynaud

Silverado

Posts: 196 Member Since:13/07/2011

#23 [url]

Mar 4 14 1:11 PM

I am pleased I don't code for survival but the one aspect that ensures it is somewhat pleasant is that it is essentially problem solving (which I thoroughly enjoy). Which makes it no different from most other engineering fields. Still, Matlab can be a lot fun and tends to keep me out of trouble (sometimes). It also ensures I truly appreciate the multitude more gifted than I.

cheers,
Reynaud

Quote    Reply   
avatar

cgc

Gold Finger

Posts: 944 Member Since:06/02/2011

#24 [url]

Mar 4 14 1:44 PM

bob olhsson wrote:
My experience with the old stuff is that it often sounds really good until you apply any kind of digital signal processing. I think fragile is probably the best term for it.

A lot the issues with older digital gear is the fixed point processing which requires far, far more care in algorithm design compared to floating point.  The really primitive stuff could turn the quantization noise and low headroom into a strength, the Lexicon 224 being an example (only 20 bit accumulator and 6 bit multiplier).  

In the case of a DSP effects unit, a design goal in the 80s might be to just exceed the THD+N of all 24 tracks in playback (-60dB or so) which was possible with 12bit SAR ADC and 20-24 bit fixed point processing.  Today in a multibit Sigma-Delta ADC and 32 bit float DSP system that THD+N figure can be -100dB for the processed signal.  It is possible and even subjectively pleasing to recreate some of the distortion of the old methods particularly with spatial/ambience processors like reverb and delay.  The signal decay from the added distortion actually makes for more 'realistic' echoes mimicking the deformation of the sound wave reflecting off a surface.  

Quote    Reply   
avatar

dcollins

Platinum Blonde

Posts: 2,343 Member Since:27/01/2011

#25 [url]

Mar 4 14 1:49 PM

reynaud wrote:
I am pleased I don't code for survival but the one aspect that ensures it is somewhat pleasant is that it is essentially problem solving (which I thoroughly enjoy). Which makes it no different from most other engineering fields. Still, Matlab can be a lot fun and tends to keep me out of trouble (sometimes). It also ensures I truly appreciate the multitude more gifted than I.

cheers,
Reynaud

I probably use Wolfram Alpha every day for _something_, now he has a language that should be amazing.

http://www.wimp.com/programminglanguage/

Quote    Reply   
avatar

cgc

Gold Finger

Posts: 944 Member Since:06/02/2011

#27 [url]

Mar 4 14 5:36 PM

dcollins wrote:

Alpha is great but that language looks next level.  Is it real?

Maybe we could do this forum in Wolfram?

post(message, embed(video)), or plotFFT(aliasing(squarewave(15kHz), samplerate(44.1khz))

Quote    Reply   
avatar

jj09

Tin Man

Posts: 22 Member Since:03/10/2015

#31 [url]

Oct 9 15 9:00 PM

cgc wrote:
I wish he had made it very clear that the mathematical equation for the analog LCR filter and the digital version are identical and only the precision of the solution differs in the two methods of implementation. The biggest misconception about analog is that it somehow is not math based.

Well, actually, a filter that works on sampled data does in fact have a different response, because it's created using a bilinear Z transform from the analog filter design.

This, of course, does not refute the point that you have to have the ***mathematical*** analog transfer function first.  The mathematics for analog filters is actually more complex and tricky. Seriously.

James D. Johnston

Last Edited By: jj09 Oct 9 15 9:03 PM. Edited 1 time.

Quote    Reply   
avatar

chance

Aqua Marine

Posts: 2,664 Member Since:30/01/2011

#32 [url]

Aug 15 16 2:01 PM

dcollins wrote:
In my continuing work avoidance efforts (Facebook) I’ve found a number of people, some very experienced in technical areas, that don’t really understand digital sampling.  
“It’s all stair steps!” “No information can exist between the samples” “LP’s have infinite resolution!” “Shannon and Nyqvist are why CD’s sound bad"

Et cetera.

Now comes a pair of videos so well-presented, so compelling, they are sure to not change the minds of any True Believers.

But for the rest of us:

Video Episode 1: A Digital Media Primer for Geeks

Video Episode 2: Digital Show & Tell


DC

 

I just watched the first video. At the end they showed all the open source tools used to make this video and saw "Audacity" I remember using Audacity years ago (Can't remember how or what I used it for tho)
Very educational video. I enjoyed it and learned a lot. Thanks for posting

Quote    Reply   
Add Reply

Quick Reply

bbcode help