Rune Allnor wrote:> On 11 Jul, 18:12, Steve Underwood <ste...@dis.org> wrote: >> Rune Allnor wrote: >>> On 10 Jul, 00:31, steve <bungalow_st...@yahoo.com> wrote: > >>>> But if you talk with an experience analog engineer, someone who >>>> excelled in linear algebra class, he still doesn�t want to hear about >>>> dot products or linear transforms or optimum least square >>>> approximations, he wants to hear about a �bank of sinc filters� >>>> analogy. You have to find an analogy that is consistent with a persons >>>> background and build on it. >>> The key phrase you use is 'wants to hear.' Not 'needs to hear' or >>> 'benefits from hearing'. I know this is considered by many to be >>> controversial, but my view is, and has always been, that analog >>> electronics and DSP are separate diciplines that should be >>> approached separately, each on its own terms. >>> Rune >> The first fast practical Fourier transform machines were purely >> analogue, working directly in terms of sinc functions. 'Wants to hear' >> is a perfectly reasonable thing to say here, because its just a >> different perspective on exactly the same thing. > > The first self-propelled vehicles were steam locomotives. > While that fact certainly holds some historical interest, > do you think it is useful to explain the workings of > present-day cars or trains in terms of steam engines?Your arguments used to make sense, but lately they are getting pretty silly. These vehicles differ in numerous basic ways. Their only commonality is "ground transport". A Fourier Transform implemented by optical or other analogue means is the same thing as one implemented on silicon. Regards, Steve
Can you recommend a good explanation of the proof of the Fourier Transform?
Started by ●July 1, 2008
Reply by ●July 12, 20082008-07-12
Reply by ●July 12, 20082008-07-12
Rune Allnor wrote:> On 11 Jul, 18:12, Steve Underwood <ste...@dis.org> wrote: >> Rune Allnor wrote: >>> On 10 Jul, 00:31, steve <bungalow_st...@yahoo.com> wrote: > >>>> But if you talk with an experience analog engineer, someone who >>>> excelled in linear algebra class, he still doesn�t want to hear about >>>> dot products or linear transforms or optimum least square >>>> approximations, he wants to hear about a �bank of sinc filters� >>>> analogy. You have to find an analogy that is consistent with a persons >>>> background and build on it. >>> The key phrase you use is 'wants to hear.' Not 'needs to hear' or >>> 'benefits from hearing'. I know this is considered by many to be >>> controversial, but my view is, and has always been, that analog >>> electronics and DSP are separate diciplines that should be >>> approached separately, each on its own terms. >>> Rune >> The first fast practical Fourier transform machines were purely >> analogue, working directly in terms of sinc functions. 'Wants to hear' >> is a perfectly reasonable thing to say here, because its just a >> different perspective on exactly the same thing. > > The first self-propelled vehicles were steam locomotives. > While that fact certainly holds some historical interest, > do you think it is useful to explain the workings of > present-day cars or trains in terms of steam engines?Your arguments used to make sense, but lately they are getting pretty silly. These vehicles differ in numerous basic ways. Their only commonality is "ground transport". A Fourier Transform implemented by optical or other analogue means is the same thing as one implemented on silicon. Regards, Steve
Reply by ●July 13, 20082008-07-13
On 13 Jul, 04:39, Steve Underwood <ste...@dis.org> wrote:> Your arguments used to make sense, but lately they are getting pretty > silly.No they aren't. They are as 'silly' now as they always were. Mind you, I tend to write with people in mind who are able to cope with challenges to 'obvious' or 'standard' expositions and explanations. If you can't handle to be provoked to contemplate once more stuff you thought you knew from decades ago, don't read my posts.> These vehicles differ in numerous basic ways. Their only > commonality is "ground transport".My point exactly. Which is why it makes no sense to use one as an analogy to learn about the other. As far as I am concerned, that's the case with DSP and analog electronics as well. They do the same sorts of tasks, but that's where the similarity ends. The technicalities and constraints are so different that there is more damage than help in trying to use one to understand the other. I've mentioned this before, but I once developed an acoustic detector for passive sonar which seemed to be able to detect targets some 8-12 dB quieter than the then known detectors. The only problem with my detector was that it was a pure DSP detector -- it could not be implemented in an analog world -- and was based on linear algebra. The people who needed to know about it were unable to discuss anything DSP that could not be expressed in terms of RLC analogies. So my detector was lost.> A Fourier Transform implemented by > optical or other analogue means is the same thing as one implemented on > silicon.It's described by the same very *fundamental* maths concepts, but there the similarity stops. The *day-to-day* technicalities as well as overall constraints, be it theoretical or practical, are very different. I tried to read some optics books a few years ago, but couldn't get much out of them. While I knew what the FT is about and how it worked, the maths involved in optics is technically very different from that used in DSP (opticians seems to use Bessel functions and cylindrical geometry all over the place). And the technology used to obtain and manipulate the spectra is completely different in optics. All in all, my DSP knowledge was of little help to me when reading those books. It might actually have been what prevented me from learning optics, since I already knew some of the concepts and parts of the terminology from a completely differnt setting, and thus tried (and failed) to make what I read about optics fit into the technical framework I know from DSP. Rune
Reply by ●July 13, 20082008-07-13
Rune Allnor wrote:> On 13 Jul, 04:39, Steve Underwood <ste...@dis.org> wrote:...>> These vehicles differ in numerous basic ways. Their only >> commonality is "ground transport". > > My point exactly. Which is why it makes no sense to use one > as an analogy to learn about the other.Not so. Practical ways to steer these vehicles don't depend on the prime mover, nor does the need to allow for differential motion of wheels on the same axle. Considerations of stability are similar. There's a lot that once learned about steam cars that informs the construction of gasoline engines. (Do you know about quasi-differential action of railroad cars wheels? they are conical, larger near the flange, and when the car presses to the outside of a curve, the outside circumference is larger than the circumference bearing in the inside track. A different kind of differential, but a differential nonetheless.> As far as I am concerned, > that's the case with DSP and analog electronics as well. They > do the same sorts of tasks, but that's where the similarity ends. > The technicalities and constraints are so different that there > is more damage than help in trying to use one to understand the > other.I think that depends on on one's customary mode of approaching problems. Carry-over and the regognition of potential parallels -- always with a mental search for counterexamples -- works for me.> I've mentioned this before, but I once developed an acoustic > detector for passive sonar which seemed to be able to detect > targets some 8-12 dB quieter than the then known detectors. > The only problem with my detector was that it was a pure DSP > detector -- it could not be implemented in an analog world -- > and was based on linear algebra. > > The people who needed to know about it were unable to discuss > anything DSP that could not be expressed in terms of RLC > analogies. So my detector was lost.Perhaps if you had been able to describe it in terms of analogies they understood it wouldn't have been stillborn? ... Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●July 13, 20082008-07-13
On 13 Jul, 16:30, Jerry Avins <j...@ieee.org> wrote:> Rune Allnor wrote:> > � � � � � � � � � � � � � � � As far as I am concerned, > > that's the case with DSP and analog electronics as well. They > > do the same sorts of tasks, but that's where the similarity ends. > > The technicalities and constraints are so different that there > > is more damage than help in trying to use one to understand the > > other. > > I think that depends on on one's customary mode of approaching problems. > Carry-over and the regognition of potential parallels -- always with a > mental search for counterexamples -- works for me.I used to use analogies to learn new stuff in the past. I stopped doing that when I realized that I only gained relevant insights that way when I discovered where, why and how the analogy became *invalid* in the context of the primary problem. After that I started looking for abstractions of the primary problem. So in the context of the DFT, where this thread started, I skipped 'DFT is a filter' because proper-context use of the term 'filter' involves stuff like 'gain', 'passband', 'stopband' etc, which only obfuscate attempts to understand the concept of the DFT. I don't use the statement 'DFT is a correlation' because proper-context use of the term 'correlation' involves stuff like 'mean', 'variance', 'bias', 'stationary' etc, which only obfuscate attempts to understand the concept of the DFT. Once you start distilling the concept of the DFT into its core parts, the main concept you are left with is the inner product. [And I use 'inner product' because I know the term extends to functions, whereas the term 'dot product' may or may not.] To quote the introduction of Stephen C. Dewhurst's 'C++ Common Knowledge': "I respect [the] readers and try to communicate with them as I would in person to any of my colleagues. Writing at an eight-grade level to a professional isn't writing. It's pandering." (ISBN 0-321-32192-8, p. xiii.) I couldn't agree more.> > I've mentioned this before, but I once developed an acoustic > > detector for passive sonar which seemed to be able to detect > > targets some 8-12 dB quieter than the then known detectors. > > The only problem with my detector was that it was a pure DSP > > detector -- it could not be implemented in an analog world -- > > and was based on linear algebra. > > > The people who needed to know about it were unable to discuss > > anything DSP that could not be expressed in terms of RLC > > analogies. So my detector was lost. > > Perhaps if you had been able to describe it in terms of analogies they > understood it wouldn't have been stillborn?The only reason why I was able to develop the idea at all was that I approached the problem at its own terms, in the proper context, using the theoretical and practical tools I had at my disposal. Trying to explain it in any other terms would amount to explain a problem with a modern car using steam engine terminology. Talking about the 'boiler' or 'coal store' just wouldn't make sense. Rune
