User talk:Bob K

Welcome!

Hello, Bob K, and welcome to Wikipedia! Thank you for your contributions. I hope you like the place and decide to stay. Here are a few good links for newcomers:

I hope you enjoy editing here and being a Wikipedian! Please sign your name on talk pages using four tildes (~~~~); this will automatically produce your name and the date. If you need help, check out Wikipedia:Where to ask a question, ask me on my talk page, or place {{helpme}} on your talk page and someone will show up shortly to answer your questions. Again, welcome! DV8 2XL 11:38, 19 November 2005 (UTC)[reply]

Subheterodyne

Thanks for the pointer, but I've never seen "subhetrodyne" used anywhere else in any textbook or other reference material. It gets 0 Google hits (or did the last time I checked, yesterday). I'd prefer to squash this ugly and incorrect coinage as quickly as possible. Wikipedia articles unfortunately gets copied hundreds of times over on the Web and I'd prefer to have them accurate if at all possible. --Wtshymanski 14:16, 17 November 2005 (UTC)[reply]

subhetrodyne [sic] is not the right spelling, which probably accounts for the 0 hits. But I agree that it is unusual, and I understand your concern, particularly since super does indeed appear to historically derive from supersonic. Or maybe in this century it is time to start putting that archaic modifier to more practical use. Just a thought.

FYI, here is another example of subheterodyne : (excerpt below)

3.22 A 4.02 GHz satellite television signal enters an Earth station receiver with IF frequency 70 MHz. What is the LO frequency and the image frequency for the cases of:

(a) High-side injection (superheterodyne)?

(b) Low-side injection (“subheterodyne”)?

--Bob K 21:59, 17 November 2005 (UTC)[reply]

non-breaking spaces

i was just editing negative frequency and saw you added a lot of non-breaking spaces. are you doing that on purpose? — Omegatron 14:50, 29 November 2005 (UTC)[reply]

Yes. I am used to separating sentences by 2 spaces, as I was taught long ago in high school. I have no clue what they are teaching these days. Without the nbsp, wikipedia displays only one of the spaces. It looks better with 2, IMHO. --Bob K 23:48, 30 November 2005 (UTC)[reply]

Ahhh. Style varies widely. Lots of people type two spaces in the markup and there's nothing wrong with that, but don't use nbsp to make it display as a wider space, please. It clutters up the markup for little gain, and is very atypical.

"There are no guidelines on whether to use one or two spaces after the end of a sentence but it is not important as the difference shows up only in the edit box. See Wikipedia talk:Manual of Style archive (spaces after the end of a sentence) for a discussion on this."

From the Manual of Style — Omegatron 01:58, 1 December 2005 (UTC)[reply]

Frequency spectrum

Can you look at frequency spectrum? I think it needs some work. — Omegatron 00:52, 6 December 2005 (UTC)[reply]

Yikes, you're right. It's in bad shape. Not an easy task though. No doubt I'll be thinking about it now, somewhere in the back of my mind.

Hilbert transform

Hi,

can you please explain the significance of the brackets in $[{\hat {s}}(t)]$ which I notices in the "Hilbert transform" article? --KYN 22:30, 10 December 2005 (UTC)[reply]

Sorry for the confusion. I tried putting them in parens (), but then I had parens inside of parens. The brackets just looked better, but I have to admit that it gives the appearance of a deeper meaning. Please fix as you see fit. Thanks. --Bob K 17:48, 11 December 2005 (UTC)[reply]

And do you happen to know if there is a wikipedia convention regarding use of

\star

or

*\,

to represent convolution? Either one seems to be acceptable, as long as the text also specifies what it means. --Bob K 18:00, 11 December 2005 (UTC)[reply]

OK, that explain things! About the sign for convolution, I noticed that the convolution article uses asterisk (

*

) to represent convolution and the cross-correlation article uses star (

\star

) to represent cross-correlation and I believe that these notations are the conventional also outside wikipedia. --KYN 21:54, 11 December 2005 (UTC)[reply]

fractions

I like the stacked version better than the unstacked version. IMO, that's the whole point of having a cool math editor. --Bob K 21:11, 15 December 2005 (UTC)[reply]

A major point of having a cool math editor is versatility: you can make it do what you want it to. "Stacked" is appropriate in some cases, but seldom within superscripts. Bob K, did you think I was saying "stacked" should never be used? If so, go back and read what I wrote, since you missed it entirely. Michael Hardy 21:53, 15 December 2005 (UTC)[reply]

PS: I'm betting you two will get outvoted on this one if it comes to that. Michael Hardy 21:53, 15 December 2005 (UTC)[reply]

I'm willing to risk it. How does that work? Of course I do not think you were saying that stacked should never be used. If you were that extreme, I probably wouldn't have botherd to ask for your reasons. --Bob K 22:52, 15 December 2005 (UTC)[reply]

Looking over some of your edits, it seems to me your use of TeX is often clumsy: you don't use \, between f(x) and dx; God only knows why you always use the matrix environment for "stacked" fractions \frac{a}{b}; you actually use "stacked" fraction inline rather than in "displayed" TeX, and that's uncouth ... several other problems. Take a look at some of my recent edits of articles you've worked on. Also, are you aware that the title phrase should be bolded at its first appearance in an article (see [[Wikipedia:Manual of Style)? Michael Hardy 22:29, 15 December 2005 (UTC)[reply]

Sure I'm clumsy with Tex. Maybe I will improve. Maybe not. Why in the world would I care if there is a \, between f(x) and dx? Somewhere I read that the \, can go anywhere. When it comes to stacked fractions, I know where I learned that... check out row 2. --Bob K 22:52, 15 December 2005 (UTC)[reply]

thanx for your support on DFT.

i'm just curious, Bob, a little bit about your story. how long has it been since you've been in school? (i'll date myself more directly, i enter my 6^th decade with the New Year.) if you want, you can find my email at my home page and send it. whatever. L8r, r b-j 06:59, 16 December 2005 (UTC)[reply]

thanx for your support on that RfC

would you consider putting it in the project page rather than the talk page? it will more "official" there. as you probably know, i usually don't back down from internet fights. it doesn't mean i'm in full agreement with you about the periodic extension inherent to the DFT, but i wanna slug that out on comp.dsp rather than here at WP. r b-j 21:30, 11 January 2006 (UTC)[reply]

Sure but need some help. See your e-mail. (Let me know if you didn't get it.) --Bob K 22:45, 11 January 2006 (UTC)[reply]

Dirac delta function

Hi Bob K, I wanted to tell you that the notation(http://en.wikipedia.org/w/index.php?title=Dirac_delta_function&diff=50641735&oldid=50639580) is conventional among physicists. They use it almost exclusively. They seem to think it more clear when using very very long integrants. I'm not so sure about that; I've always preferred the maths convention. Thanks for changing that and the sift/shift thing. Although I'm not sure sift is the right word, shift is definately not correct. Select would be better I think. --MarSch 18:11, 29 April 2006 (UTC)[reply]

sampling theorem, sinc.

hi Bob,

i'm still at a loss as to why "disambiguating" Sinc function helped. now we'll have multiple occurances of $\mathrm {sinc} (x)\$ meaning two different things. i would prefer if it was the normalized sinc all around, but losing that battle, adding the π to get $\mathrm {sinc} (\pi x)\$ is better than to have two inconsistent definitions.

also, i am still not clear what the "heuristic" basis for the sampling theorem does for anything. it doesn't prove anything and the explanation relying on that part of the DTFT article is circular. first we have to establish, independently, that sampling a signal causes its spectrum to be repeatedly shifted by multiples of the sampling frequency and overlap added. then, after that, to motivate the definition of the DTFT, you ask "what is the F.T. of that ideally sampled signal?" and the result, with a couple of substitutions x[k] for x(kT) and ω for 2πfT, is the DTFT. but to use that as a mathematical basis for the sampling theorem is circular. it serves no purpose.

lastly, you confuse the terms "aliases" with "images". e.g. an "anti-aliasing filter" is what goes before an A/D converter and an "anti-imaging filter" is what comes after the D/A. those copies of the original baseband spectrum shifted to multiples of the sampling frequency are images not aliases. if the signal is undersampled and a piece of the first image overlaps the baseband spectrum, that overlapping portion are aliases.

sorry to be so terse. i'm tired of typing too much at User:LutzL.r b-j 05:49, 17 May 2006 (UTC)[reply]

Yeah, you guys have been working overtime. I hope it was worth it. Regarding the multiple defs of sinc(), I think the best that Wikipedia can do is to accurately describe the complexity of the outside world, not just the world of mathematicians or EEs. A textbook writer has the advantage of tunnel vision. But sinc() has two common definitions (and always will in the foreseeable future). And almost everyone who uses it is careful to specify which one they are using. Those who don't are either naive or arrogant. So people are accustomed to seeing sinc defined differently for math articles than for EE articles. They expect it. Unfortunately that is the standard that has evolved, and it is too pervasive for Wikipedia to fix, not to mention that neither side wants to give up their preferred definition. The

{\rm {{sinc}_{\pi }\,}}

notation is rarely used and not self-evident; i.e., the casual reader will still have to look for the definition. And of course any given article is still free to locally define and use

{\rm {{sinc}_{\pi }\,}}

. Another advantage of the disamb page is that some articles mention the sinc() function in a non-specific way; i.e., they don't really care which definition you use. So those articles have the additional option of referencing the disamb page. --Bob K 11:59, 17 May 2006 (UTC)[reply]

i agree with you about the uselessness of the

{\rm {{sinc}_{\pi }\,}}

notation. to me the simplest thing to do is to say

{\rm {{sinc}(\pi x)={\frac {\sin(\pi x)}{\pi x}}\,}}

when that is what you want and to say

{\rm {{sinc}(x)={\frac {\sin(x)}{x}}\,}}

when that is what is needed. i wrote that thing on Talk:Sinc function because i thought it would be nice to change the default notation (forcing those who wanted

{\frac {\sin(x)}{x}}\,

to say

{\rm {{sinc}(x/\pi )\,}}

but that met with enough resistance that i dropped the idea (since the default continuous Fourier transform is not to my liking either. i just think that using

{\rm {{sinc}(x)\,}}

in all cases and then having to explain or reference which sinc() you're using is less concise than having one definition and tossing in a π factor when needed. r b-j 00:55, 18 May 2006 (UTC)[reply]

As I said 9-May, I do understand your quibble with the DTFT, and I can replace it with the Fourier transform. In fact the [19-March version] (before I changed it) actually did that, except that it happened to use the label

X_{dtft}(f)\,

for the transform. I merely moved the derivation to the DTFT article, and linked to it. --Bob K 00:22, 18 May 2006 (UTC)[reply]

since the DTFT thing appealed to the PSF, why not use that proof instead?. i really do not like the

X_{dtft}(f)\,

notation. it's another notational convention introduced that is neither necessary nor used anywhere else.

best, r b-j 00:55, 18 May 2006 (UTC)[reply]

I was sort of expecting some resistance to the

X_{dtft}(f)\,

notation, and it is unnecessary. I will do something about that soon (but not tonight). My reason for not embracing the PSF is that the PSF is not how most newbies get their first glimpse of insight into the sampling theorem. I was trying to make it easy to understand, and therefore fun. I don't care if it not rigorous, and I don't care if it is not labelled "proof". All I hope to do is convince some reluctant reader that sampling really works and that they can understand why and maybe even explain it to their friends. --Bob K 01:53, 18 May 2006 (UTC)[reply]

well, that so-called "rigorous proof" was meant to be that for the newbie. it was explicit and could be called "bloated" (i step-by-step evaluate some expressions), but it isn't deep or complicated. it is about the conceptually simplest proof i can think of. most EE textbooks that do it are almost the same except they don't represent the dirac comb as a Fourier series and apply the frequency shifting operation, instead they identify the F.T. of the dirac comb as another dirac comb in the frequency domain and then apply convolution in the frequency domain. i have always considered that to be a bit harder for the newbie to grasp. r b-j 03:29, 18 May 2006 (UTC)[reply]

If you look past my unfortunate labelling, you will see that

X_{dtft}(f)\,

is just what you have described... the convolution of

X(f)\,

with "another dirac comb in the frequency domain". So I think your issues are:

labelling
Fourier series vs. comb

I agree to fix the labelling. And as you said, the comb is the method preferred by EE texts. (So I am in good company.) So apparently I have a "proof" afterall. --Bob K 11:14, 18 May 2006 (UTC)[reply]

I also understand your distinction between alias and image, and in certain contexts (such as anti-aliasing filter) I can see that it is useful. But the word aliasing (in the discrete-time sense) is also used more broadly than that, and I did not have to look very far for an example: Aliasing#Aliasing. Maybe that article should explain the narrower definition as well. --Bob K 01:53, 18 May 2006 (UTC)[reply]

Phase image

Can you comment on Talk:Phase (waves)#Misleading image?? please? — Omegatron 03:54, 24 June 2006 (UTC)[reply]

Nyquist–Shannon

BobK, do you have an opinion on the dispute I provoked with Rbj on Nyquist–Shannon sampling theorem? Dicklyon 05:20, 5 August 2006 (UTC)[reply]

Single-sideband modulation/Proofs

I have added a "{{prod}}" template to the article Single-sideband modulation/Proofs, suggesting that it be deleted according to the proposed deletion process. All contributions are appreciated, but I don't believe it satisfies Wikipedia's criteria for inclusion, and I've explained why in the deletion notice (see also "What Wikipedia is not" and Wikipedia's deletion policy). You may contest the proposed deletion by removing the {{dated prod}} notice, but please explain why you disagree with the proposed deletion in your edit summary or on its talk page. Also, please consider improving the article to address the issues raised. Even though removing the deletion notice will prevent deletion through the proposed deletion process, the article may still be deleted if it matches any of the speedy deletion criteria or it can be sent to Articles for Deletion, where it may be deleted if consensus to delete is reached. --Zvika 19:23, 30 May 2007 (UTC)[reply]

The above is a rather silly template. You created this page a year and a half ago, so I don't expect you to remember why you did, but I decided to let you know of the proposed deletion just in case you have a reason for keeping it. If not, just forget about it. Have a nice day! --Zvika 19:24, 30 May 2007 (UTC)[reply]

Thanks for the explanation. I agree with your proposal. The history of how this came about is recorded at Talk:Single-sideband_modulation#Needs_work, in case you're interested. I had originally placed the information in the main article, but it was suppressed by Wtshymanski, who objected to the use of the Hilbert transform. I was too new to argue that point, but I knew the proof was useful, so I created the subpage. The next day, Splash objected to that and supported the original concept, which gave me the conviction to reinstate it. I would have removed the subpage myself, but I don't have that power. --Bob K 08:39, 31 May 2007 (UTC)[reply]

\align

Hello. Please see this edit. We've had \align for some time now. Michael Hardy 00:31, 22 August 2007 (UTC)[reply]

LSSA

Bob, can you look at the ongoing mess at Least-squares spectral analysis please, and see if this is something you can offer an opinion on? See also Wikipedia:Requests for comment/Geoeg. Dicklyon 05:10, 18 October 2007 (UTC)[reply]

Blackman Window

Hi Bob,

I noticed that you immediately undid my change to the formula for the Blackman window. Having a single parameter more clearly shows the intuition behind the formula. In fact, you can generate a whole family of Blackman windows by varying the single parameter 'a'. It is also possible to write the Blackman and Hann windows in such a way that one can easily see how Hann is a special case of Blackman for a particular value of 'a'. I personally think structure that intuitively demonstrates concepts and relationships has value. I know I personally was quite annoyed when first learning this subject that formulas were typically flattened and that I had to reverse engineer things like how the Blackman Window was derived and the fact that it does not have three degrees of variation.

Regards, A Uio —Preceding unsigned comment added by Auiow (talk • contribs) 03:49, 7 March 2008 (UTC)[reply]

Thanks for the explanation, and I agree. In fact I think we should add your insights to the article as unobtrusively as possible. I'll probably make an attempt later today.

--Bob K (talk) 14:12, 7 March 2008 (UTC)[reply]

File:Sinc functions.png listed for deletion

An image or media file that you uploaded or altered, File:Sinc functions.png, has been listed at Wikipedia:Images and media for deletion. Please see the discussion to see why this is (you may have to search for the title of the image to find its entry), if you are interested in it not being deleted. Thank you. Skier Dude (talk) 00:28, 11 January 2009 (UTC)[reply]

Convolution

Hi, your edit summary said that some characters do not appear correctly in Internet Explorer. They seem fine on my copy, which has no special fonts installed. But if this is generally a problem for people running internet explorer, then something should be done to make changes across several articles. Is it possible to get a screen shot to see what it looks like on your copy? Thenub314 (talk) 11:04, 19 January 2009 (UTC)[reply]

When User:Bdmy changed ƒ*g_T to ƒ∗g_T, the asterisk became an empty square (aka "box") on my IE screen. Upon further reading, such as Talk:Set_(mathematics)#Box-fixing and Template_talk:SpecialChars#Internet_Explorer, I learned that the problem is associated with IE. I have Firefox, and it works, but I use IE for everything else, as do many other readers.

--Bob K (talk) 14:28, 19 January 2009 (UTC)[reply]

This is good to know. I when I tried IE on my vista machine it worked fine, when I tried on my XP machine it displayed a box. I wonder why they only fixed the bug on Vista? Oddly, as an a side, on my XP system the PNG graphics also looked very pixelated. I am surprised the issue about boxes has never come up on the WT:WPM. I will try to keep this in mind when editing articles, and make sure to use * instead of ∗. Thenub314 (talk) 15:05, 19 January 2009 (UTC)[reply]

Ah Ha! I discovered an easy fix. I changed Times New Roman to Lucida Sans Unicode in a pull down menu at Tools -> Internet Options -> Fonts -> Webpage Font:. That fixes the problem for me. But how do we disseminate that information to the readership?

--Bob K (talk) 14:55, 19 January 2009 (UTC)[reply]

Well, one could include a link on the special characters template that discusses possible fixes. Thenub314 (talk) 15:05, 19 January 2009 (UTC)[reply]

Yes, I was just looking into that. But I haven't figured out where the "fix" page should exist. Any suggestions?

--Bob K (talk) 15:20, 19 January 2009 (UTC)[reply]

More progress... I just created section Changing Internet Explorer's (IE) default font

--Bob K (talk) 17:03, 19 January 2009 (UTC)[reply]

Nice work! It is looking much better. Thenub314 (talk) 19:07, 19 January 2009 (UTC)[reply]

Sorry

Sorry on the omega, brain was working so good last night. Dicklyon (talk) 14:56, 28 March 2009 (UTC)[reply]

filter banks

No comment on whether this deletion was right or wrong (I have no idea) but you might want to look at the WOLA dab page which has the same link with the same claim. SpinningSpark 18:23, 30 September 2009 (UTC)[reply]

Aliasing folding image?

Hi Bob,

In this edit to Aliasing, rewriting the folding section to make it briefer, you removed the image at right. (I’d added it earlier, this edit being my last revision.)

Did you mean to do this? It seems a useful image, showing quickly how folding occurs graphically, which is a useful supplement or alternative to the text. If so, I’d be happy to re-add it. Thanks!

—Nils von Barth (nbarth) (talk) 21:10, 1 July 2010 (UTC)[reply]

Hi Nils,

That was too many half-lifes of my memory decay time to actually remember. But in all likelihood, yes, it was intentional, because when I look at your picture now... and try to put myself in the shoes of a novice, I find that it doesn't really help. Maybe, with a verbal description specific to the picture, it would be more helpful. And I'm not so sure about that.

- Bob

Looking back, the blurb was pretty terse (obviously “folding” isn’t esp. helpful, but original blurb wasn’t much longer). I’ll see if I can write a better blurb. I drew this picture because when I saw such an image (in the context of superheterodyne), it was like a light bulb went off because it clarified the back-and-forth nature of aliasing, and visually demonstrated the many-to-one aliasing map.

I’ll try writing a blurb that relates to the text (or, better, relates the text and image to each other) and check with you; thanks for your feedback.

—Nils von Barth (nbarth) (talk) 00:52, 3 July 2010 (UTC)[reply]

Hi Bob,

I’ve had a shot at a somewhat more detailed blurb and better integration with the text in this revision; hope it’s a bit better!

BTW, in researching this, I found a similar image diagram (Spectral imaging of the atmosphere, by G. G. Shepherd, Figure 2.9(a), p. 43), which might be a useful reference.

—Nils von Barth (nbarth) (talk) 08:25, 3 July 2010 (UTC)[reply]

Yes, I think that is helpful, and I apologize for yesterday's lackluster response. I was tired from a long dayhike, and my Wikipedia "skills" are rusty. When I awoke this morning with a fresher mind, I tried to analyze what I thought might confuse people. I think it's the fact that the figure depicts both folding and periodicity, which are two different (and often confused) things. You can look at just one "period" of the spectrum (of a real-valued sampled signal) and see symmetry, which is not generally true for complex-valued signals. That's the real point of "folding". So I would prefer to see a figure that tells that story, if possible.

--Bob K (talk) 10:43, 3 July 2010 (UTC)[reply]

I guess what I'm trying to say is that your figure might be even better for this purpose if it just depicted the [0, f_s] region.

--Bob K (talk) 10:56, 3 July 2010 (UTC)[reply]

File:Aliasing-folding-1-period.svg

Only one period needed to show folding.

OIC – that’s a really good point. You’re right, to just show folding the figure is overkill and confusing.

I’ve re-drawn the figure so it only shows 1 period (hence folding), and hopefully tells the folding story better.

The larger figure (showing both folding and periodicity) seems useful to show all aliases (the global picture), though I’m not sure where best to put it – suggestions?

…and no worries regarding earlier response – thanks for elaborating on your intuition.

—Nils von Barth (nbarth) (talk) 17:59, 3 July 2010 (UTC)[reply]

Well I think I might have fixed it with my explanation in the article. I rather like what we have now.

--Bob K (talk) 18:50, 3 July 2010 (UTC)[reply]

I was too slow, so I changed the text again. But there seems to be a spurious f_s/2 at the upper left corner of the figure.

--Bob K (talk) 19:05, 3 July 2010 (UTC)[reply]

Oops – hope current is ok; feel free to use whichever image you think best.

Regarding the f_s/2 in the upper left – that’s to give vertical scale: point being that detected frequency increases from 0 to f_s/2, then back down from f_s/2 to 0, as it says in the text. Hope it’s not too confusing.

—Nils von Barth (nbarth) (talk) 20:36, 3 July 2010 (UTC)[reply]

Sorry to be picky, but the vertical scale is completely arbitrary and irrelevant. A peak amplitude of 1 would do just fine with the current text. So would an unspecified (blank) amplitude. But f_s/2 makes it appear that the peak value has a meaning relevant to folding.

--Bob K (talk) 12:22, 4 July 2010 (UTC)[reply]

OIC why the diagram is confusing (and why the diagram in the book is flipped): this isn’t a waveform, but rather a function of “input wavelenth to detected/computed wavelength” – if one takes all computed wavelengths to be in the interval [0,f_s/2], then as actual input frequencies vary in frequency, this is the graph of the output (computed, detected) wavelength in that interval. The vertical scale is “output Hertz of detected signal”, not “amplitude”.

If you check this diagram, you’ll note that it also uses Hertz scale on both axes, just as I have, but the graph is flipped. I drew the graph without a flip so that it’s an actual function (x = input Hz, y = output Hz), but given your comments, it seems that this graph is easily misinterpreted as a waveform; perhaps it would be better to flip it?

…and no worries about being picky – I appreciate care and attention – thanks!

—Nils von Barth (nbarth) (talk) 18:58, 5 July 2010 (UTC)[reply]

The diagram in your reference may be correct, but it is not helpful (IMO). It would not help me to understand folding as much as the one we now have in place (along with the verbal explanations). As you can see, I took my interpretation of your diagram and added more labels, including one to clarify the term "folding", which is what this section is about.

--Bob K (talk) 19:31, 5 July 2010 (UTC)[reply]

Oh, I see that you’ve made another diagram (more detailed and labeled) – thanks, looks great and much better!

—Nils von Barth (nbarth) (talk) 21:32, 5 July 2010 (UTC)[reply]

Bob’s image – aliasing & folding of a sinusoid.

Also, to clarify for myself – your diagram is showing a spectrum (amplitudes for various frequencies), while my diagram (and ref) were showing the (input frequency to output frequency) function, which is more abstract and hence potentially confusing. The mathematician in me think that drawing the function is useful, but I agree that particularly for an introduction (which is, in fact, the intended audience), showing a spectrum is fine and probably rather clearer.

—Nils von Barth (nbarth) (talk) 21:41, 5 July 2010 (UTC)[reply]

I'm glad you like my diagram. But if you read carefully, it is not an actual spectrum, because that concept is not a prerequisite for the article. Instead, it just depicts the aliases of a simple sinusoid when its amplitude is varied as an arbitrary function of its frequency between ƒ_s/2 and ƒ_s.

--Bob K (talk) 22:43, 5 July 2010 (UTC)[reply]

Oh, good point (re-reading) – just varying sinusoid, not spectrum/sum. Admittedly a bit subtle (I’m used to seeing such diagrams with spectra), but I don’t think it should carry any risk of confusing people as to what folding is, since the effect is the same.

—Nils von Barth (nbarth) (talk) 23:18, 5 July 2010 (UTC)[reply]