On Friday, January 5, 2018 at 9:35:21 AM UTC+13, Steve Pope wrote:
> Steve Underwood  <steveu@dis.org> wrote:
> 
> >On 01/04/2018 01:01 AM, Steve Pope wrote:
> 
> >> I'm sort-of retired, as are probably others, but I also think that DSP
> >> has become a less-appreciated niche than it once was.
> 
> >Interesting viewpoint. DSP isn't exactly going away. Quite the opposite. 
> >It continues its march deeper and deeper into systems, displacing more 
> >analogue elements of them.
> 
> Yes, and this is why I think it's become less of a niche.  A design
> group is more likely to already have DSP design talent.  This is 
> especially true if you look at it at the algorithm level.  You don't need
> a specific DSP expert anymore to design an FFT or a Viterbi decoder.
> (Or, so program managers now believe.)
> 
> Steve

even home hobbyists and amateurs are using FFTs. Processors are so fast you don't need to program them in assembler any more! Other areas are still in our domain though such as adaptive filters and algorithms and the hardware is a mystery to most so designing special interfacing , FPGAs etc remains in our grasp. However, LabView Rio and myRio is beginning to take over that area too for many applications - at least in prototype. You no longer need to know much about FPGAs to program them since you use the data flow language LabView. If you want an expensive yet quick solution that's the way to go

Steve Underwood  <steveu@dis.org> wrote:

>On 01/04/2018 01:01 AM, Steve Pope wrote:

>> I'm sort-of retired, as are probably others, but I also think that DSP
>> has become a less-appreciated niche than it once was.

>Interesting viewpoint. DSP isn't exactly going away. Quite the opposite. 
>It continues its march deeper and deeper into systems, displacing more 
>analogue elements of them.

Yes, and this is why I think it's become less of a niche.  A design
group is more likely to already have DSP design talent.  This is 
especially true if you look at it at the algorithm level.  You don't need
a specific DSP expert anymore to design an FFT or a Viterbi decoder.
(Or, so program managers now believe.)

Steve

On Thu, 4 Jan 2018 05:39:53 +0000, Steve Underwood <steveu@dis.org>
wrote:

>On 01/04/2018 01:01 AM, Steve Pope wrote:
>> I'm sort-of retired, as are probably others, but I also think that DSP
>> has become a less-appreciated niche than it once was.

>Interesting viewpoint. DSP isn't exactly going away. Quite the opposite. 
>It continues its march deeper and deeper into systems, displacing more 
>analogue elements of them.
>
>Steve

I think it is a maturing field, though, so the number of researchers
needed is less.   The automated tools available allow practitioners to
get a lot done without having the depth of knowledge that was
typically needed in the past.

And those people who do dive in and want and/or need to know a lot
have google and the rest of the intarwebs to get their questions
answered, ironically probably sometimes by searching comp.dsp history.

So, we're dinosaurs.  Embrace it.   ;)

On 01/04/2018 01:01 AM, Steve Pope wrote:
> I'm sort-of retired, as are probably others, but I also think that DSP
> has become a less-appreciated niche than it once was.
Interesting viewpoint. DSP isn't exactly going away. Quite the opposite. 
It continues its march deeper and deeper into systems, displacing more 
analogue elements of them.

Steve

Eric Jacobsen <eric.jacobsen@ieee.org> wrote:

>On Wed, 03 Jan 2018 04:54:16 GMT, no-one@notreal.invalid (Robert

>>Well, if someone finds a really good replacement forum this same group
>>of people, please post back here where you are going so the rest of us
>>can follow you.

>Retirement?  

>Actually I think some people moved to dsprelated.com and some to
>dsp.stackexchange.com.   Neither are quite the same as usenet and
>naturally all have different cultures and groups of people.

I'm sort-of retired, as are probably others, but I also think that DSP 
has become a less-appreciated niche than it once was.

S.

On Wed, 03 Jan 2018 04:54:16 GMT, no-one@notreal.invalid (Robert
Scott) wrote:

>Well, if someone finds a really good replacement forum this same group
>of people, please post back here where you are going so the rest of us
>can follow you.

Retirement?  

;)

Actually I think some people moved to dsprelated.com and some to
dsp.stackexchange.com.   Neither are quite the same as usenet and
naturally all have different cultures and groups of people.

On Tuesday, January 2, 2018 at 11:54:19 PM UTC-5, Robert Scott wrote:
> Well, if someone finds a really good replacement forum this same group
> of people, please post back here where you are going so the rest of us
> can follow you.

agreed
mark

Well, if someone finds a really good replacement forum this same group
of people, please post back here where you are going so the rest of us
can follow you.

>On Thursday, November 2, 2017 at 3:25:44 AM UTC-5, robert
bristow-johnson
>wrote:
>> this used to be a very active newsgroup.
>
>We've solved all the world's signal processing problems, and the
solutions
>have been amalgamated in a Matlab toolbox. Look for it in R2018a.

It's kind of true!  I learned multirate DSP with polyphase FIRs a few
decades ago from this group.  Combined with RBJ's Filter cookbook, I've
been able to solve 1,001 problems!  

---------------------------------------
Posted through http://www.DSPRelated.com