Seems like good, important work. Also seems irrelevant to the discussion, as near as I can tell. My assertion (still unsupported with citations -sorry) is that the auditory system is incapable of detecting or encoding information about the phase of the incoming signal in the neural spike train. If nothing else, the mechanical resonance of the hair cells prevents them from detecting the incoming signal's phase.
I'm not disagreeing AT ALL that the phase of the spike trains contains information, or that phase shifts IN THE SPIKE TRAIN are irrelevant. I'm talking about the ability of the cochlea and related neurons to detect and encode phase information for incoming audio signals above a few kHz; above the speech region, basically.
And, moreover, the whole discussion is irrelevant to whether a 50 ns phase shift in a speaker cable, at 25 kHz, is audible. That's my fault; I brought it up to demonstrate WHY such short phase differences are inaudible. But we've gotten off on a bit of a tangent...:-)
Hi, welcome to the party. If you look about 5 levels up, you'll see someone asserting that skin effect is audible in audio cables. I point out that in the example he cites, the total power loss is 0.020 dB at 25 kHz. I contend that this is completely and totally, under any imaginable condition, inaudible by humans and thus irrelevant. So yes, I am ignoring it. I also pointed out that the phase shift described in his citation was about 50 ns at 25 kHz, again too short (not significant enough) to be audible under any conceivable conditions. I also pointed out that neuron firing rates in the auditory path are limited to a few kHz, and that those neuronal spikes are synchronous with the incoming audio signal up to some few kHz, which (and this is the point of contention) may limit the auditory system's ability to report phase to the brain to signals less than some few kHz. Consequently, a phase shift of 50 ns is totally indiscernable by the human auditory system.
I can show, as does the grandparent poster's citation, that the signal DOES arrive largely intact, with neither amplitude NOR phase significantly affected. In this case, when I say significantly, I mean "to a level which is in any way perceptible to the human auditory system".
In other words, skin effect in audio interconnects is totally irrelevant - either from a phase or amplitude point of view.
I think we're talking at cross-purposes here. I asserted, based on the information I've soaked up in the last couple of weeks of discussion, that auditory path neurons do not present phase information to the brain for frequencies above a few kilohertz.
You've replied with links to researchers who work in neural spike train coding. I saw references to Berger's work, but could not find the work itself. None of the titles suggested anything about auditory path neural coding, but not being able to read the papers I can't be sure of that. Liaw's work is available from the web, but in the few minutes I have I haven't been able to find any reference to auditory path coding there, either. There's a lot of work with the hippocampus, true; but I don't see the connection between the hippocampus and the encoding of phase information by the auditory neurons.
I have no intention of turning this into a discussion of neural coding, because quite frankly I only know enough to know that I don't know enough. I will ask the nice gentlemen and ladies who've given these presentations to point toward some of their references, and happily share them with you. Note, however, that these will be works dealing with the auditory path only, and probably with some clinical studies of phase comprehension in hearing tests.
I mentioned spike rate as a limiting factor because the work that was referenced in these presentations, and which I have yet to cite for you, showed the phase alignment of auditory path neural spikes with the incoming signal, at signal frequencies up to the maximum spike train frequency and occasionally a bit beyond. If the phase of the input signal is being encoded somehow into the spike train from the auditory neurons, then that coding process is not well understood - at least not by any of my sources.
Are you really telling me that you are aware of work that specifically shows the transmission of phase information, in the auditory path, at frequencies significantly higher than the maximum spiking rate?
How neurons encode information is NOT - repeat, NOT - a known process. There are several theories regarding whether coding is purely based on spike frequency, spike synchrony, or some other coding systems we haven't discovered yet. I've spent the last two weeks listening to some of the world's brightest and most knowledgeable experts on neural coding, auditory system, visual system, etc at the Telluride Neuromorphic Engineering Workshop and the best thing that can be told is that there is more going on than spike rate, but there is no consensus on exactly what that code is. If you know more and better, then you should be here next year. Contact the organizers.
What I DO know, and what actual experiments have shown, is that auditory neurons fire in synchrony with the incoming audio signal up to a few kHz. Beyond that, there is no correlation between the phase of the incoming audio signal and the spike train from the neuron. The neurons continue to fire, but not in phase with the audio signal. Of course higher frequencies are being detected and passed on to the brain; but absolute phase information, as far as anyone can tell, is not preserved beyond that few kHz limit. Again, your evidence to the contrary is welcome. Try to be specific and cite research wherever possible.
Hmm... which argument do you wish to present? That there are no differences between a tube system and a DSP-emulated system? That different people have different ideas about what is "better"? Or that psychoacoustics play a part in what is considered "better"?
If you wish to argue the latter two, I won't disagree a bit. The first argument I can partially agree with.
If you wish to tell me that you can accurately reproduce the spectral response of a clean tube amplifier using a clean SS amplifier and a DSP, I will not argue a bit. I notice that you haven't mentioned anything about temporal response or phase response, neither of which is particular easy to measure with a spectrum analyzer. However, I don't know of any system which claims to actually do any of these things. If anyone sells a DSP-based SS amplifier which claims to emulate tube amplifiers for hi-fi, I'm not aware of it. Can be done, perhaps. Probably. But I don't think it has.
However, if you slide over into the realm of music production applications, the story is different. The distortion characteristics of a real tube amplifier are significantly different from anything simulated by DSP to date, and the difference is not subtle to those of us who have earned our living through the use and repair of tube-based guitar amps. Bottom line, there isn't a DSP-based emulator that really sounds like a small-box Plexi Marshall on a 4x12 [shivers in ecstasy - that thing was AWESOME], or like a Vox AC30 Top Boost, or even really like a Blackface Fender Champ - three tubes, one speaker! There are dozens which try, and a few which come close, but the state of the art simply is not yet up to snuff. It's not simply a matter of frequency response; there are temporal effects, resonance effects, mechanical effects - it's very complicated, and people really can tell the difference. I haven't even gotten into room emulation or microphone and placement emulation - moving the mic 1/4" laterally in front of the speaker makes an amazing difference. As a repair tech, I've played though at least one or two different examples of just about every guitar amp out there. I've also played through most of the DSP simulators, and have owned several. If you can make a DSP emulator which really sounds like those amps, call me. I'll buy one, and I've got the contacts in the music equipment biz to make us both filthy rich selling them.
Oh, and I've also got multiple EE degrees, including several courses in DSP, ASP, and human auditory perception. Looks nice in the business plan, ya know?
Went. That is, in fact, a pretty clear and accurate explanation of skin effect. And it shows me exactly what I already knew. A power difference of less than 0020dB at 25 kHz is, in fact, totally imperceptible to the human ear - at least, according to rigorously scientific studies. Many audiophiles will claim differently, I'm sure.
Additionally, the 50 ns phase shift given in the example is also indiscernible to the human auditory system. In fact, the human auditory system loses the ability to phase lock at frequencies above a few kilohertz. I don't want to go too far afield here, but the study of the physiology of the human auditory system can be interesting. Bottom line is that the spiking rate of the neurons of the auditory system is on the order of a few kHz, and for signals above those frequencies the only information reported is amplitude, not phase. Below, 2-3 kHz, the neurons spike in synchrony with the incoming audio signal. 50 ns worth of phase shift at 25 kHz is inaudible, and I challenge anyone in the world to prove differently. I will gladly eat my own words if anyone can do so - but I'm not particularly worried.
There are, of course, phenomenae that science does not understand or has not yet discovered. However, I personally hold any audiophile results which claim to be able to discern either 0.02 dB of power loss or 50 ns of phase shift, at 25 kHz, to be psychologically biased and inaccurate.
I would hate to refer to you as either being a liar or as being deluded, so I hope that you don't really believe that 0.02 dB is an audible difference. Thank you, though, for a most interesting and polite discussion.
Hmm... except that when shouting in a pipe, the wavelength of the sounds produced by your voice are of the same order of magnitude as some dimension of the pipe. Morever, the shouting in a pipe causes distortion through the Helmholz effect, wherein the pipe resonates at certain frequencies and 'smears' sounds at and near those frequencies. In an audio cable, the wavelength of even the highest frequencies you can possibly imagine hearing (say, 40 kHz per Pioneer's high-speed DAT tests from the '90's) are so much longer than any possible path length difference that the smear... well, lessee. Got a calculator around here somewhere...
The wavelength of a 40 kHz sine wave, in a waveguide (coax cable) whose propagation velocity is 1.5e8 m/s (half the speed of light; a pretty slow waveguide) is approximately 37.5 KILOMETERS. Even if the skin effect presented two paths whose length differed by a factor of four (fairly extreme case), you'd have to have a 25 meter length of cable to achieve 1 degree of phase shift. Summing two signals with phase shift of 1 degree results in a net amplitude loss of 0.00066 dB. One degree of phase shift at 40 kHz results in a time 'smear' of about 70 ns.
Perhaps you have ears capable of discerning a 0.00066 dB amplitude fluctuation at 40 kHz - I know I don't. Perhaps you can hear time smear of 70 ns - I know I can't. Perhaps you also have 25 meter - that's 80 some-odd feet for those of us in the US - interconnects; my whole apartment isn't 80 feet long. However, I think you have none of those things. I think I've presented a pretty extreme case, and shown that electrically, what you claim happens is either not there or is so insignificant that the difference is truly academic.
However, it may be that I've misunderstood your usage of the term "skin effect". It's very common in the audiophile world for salesman and such to misappropriate scientific terminology, banking on the "baffle them with bullshit" principle. If I have misinterpreted your statements, kindly set me aright so that I may further understand exactly what you mean.
There are several manufacturers of very expensive cables who claim that they are directional, and that one specific end of the interconnect must be the source end. They claim that the only way to tell which way the cable is aligned coming off of the roll from the manufacturer is to listen to it, very carefully. They also claim that the cables must be "broken in" through constant listening, and that if you don't use the cables for a period of some weeks, they lose their conditioning and must be "re-broken-in". This is for line-level interconnects, not even speaker-level. They claim that these effects are due to "microdiodes" in the crystalline structure of the copper (or silver) they use for conductors. They cleverly ignore the fact that the signals are AC, and thus current flows equally in both directions. They also totally gloss over the fact that solid-state diodes (at least all solid-state diodes known to real science) have a voltage drop on the order of hundreds of millivolts, which for an audio line-level signal is usually on the order of half of the overall voltage amplitude. It's simply amazing, and sounds perfectly plausible to those who don't know any better.
Perhaps you could explain to me, in small words that a musician, studio engineer, and holder of multiple EE degrees (that's me) can understand, exactly what skin effect you are referring to that is perceptible at audio frequencies, and under what conditions.
Do you work in the snake oil - I mean, audiophile business? I've never heard a bullshit spiel that elegant before. What are you selling?
None of that shit - skin effect, switching time - has any relevance at audio frequencies.
I take it you've already tried the one about the "microdiodes" in wire that make it unidirectional, never mind the fact that the signal IS AC, ferchrissakes?
So true, so true. I read a post on this very site a few days ago wherein some idiot was taking back an iRiver because it 'only' claimed 30 dB worth of bass boost, or some such nonsense.
30 dB?!?! That's ridiculous! How can you even consider using that much bass? You can't possibly hear the music at that point, just some insane thumping and a little human voice off in the background!
I realize that you are trolling, but I'll bite. Bob Carver's claims notwithstanding, nobody has been able to accurately simulate the sound of a tube amplifier using analog or digital filtering. Nor is there any significant amount of hum in a decent tube amp.
There are dozens of companies willing to sell you simulators of overloaded tube amps. You, personally, may even think that they sound just like overloaded tube amps. However, they don't. I have played guitar, repaired guitar amplifiers, engineered and mixed studio albums, and I know from experience that the state of DSP for tube emulation simply isn't there yet. You may disagree; that's fine. But you're wrong.:-)
Parent poster is correct. Several products exist which attempt to use DSP to mimic the sound of tube amplifiers, almost all of them in the guitar effects realm. None of them are yet there. They sound a lot LIKE a tube amp, but they don't capture it all. CAN they? Probably someday. But not yet.
I base this opinion (yes, opinion) on: several years as a professional audio engineer; several years as a guitar amplifier repair technician; several years as a semi-professional guitarist; and two degrees (working on a third) in Electrical Engineering.
Sibling posters who believe that DSP can do anything are correct up to a point: DSP can achieve any given transfer function, up to a desired level of accuracy. You need more accuracy? Increase the bit depth and sample rate, tweak the processing. However, the bug stumbling block is this: you gotta know what transfer function you want to emulate first.
Might I suggest you work on your cause-and-effect a bit more?
I have three rocks in my garden. All of them are red. None have been stolen. Ergo, red rocks cannot be stolen.
See how stupid it sounds? Dude - the top five cars on that list are also the TOP FIVE most common cars in the U.S. If you can establish a relationship between the theft rates of similarly-equipped cars, where the only variable is RFID or not, then you've got a case and you are welcome to tell us all about it. If you just wanna spout uneducated shit...... well, OK, welcome to Slashdot!
Beats me. Apparently the metallurgists say that amorphous steel is stronger. [ shrug ] Sounds good to me.
However, the claim about the amorphous steel being "twice as strong and hard" smells funny to me... twice as hard? On what sort of scale? Isn't steel already about 2/3 as hard as diamond anyway? Does that mean that amorphous steel is 4/3 as hard as diamond? How would they measure that?
Re:Thank you ...
on
Amorphous Steel
·
· Score: 3, Informative
"glassy", as in "amorphous", as in "non-crystalline". Does NOT mean transparent like window glass. Think obsidian - it's black (or green), opaque, and shatters in totally random directions. That's because it has no crystalline structure and thus no lines or planes of fracture. This is non-crystalline steel. It's not transparent aluminum - but then, nothing is.
Yes, but the earth's field is static. The magnetic pulse from the railgun will be dynamic, meaning that it changes very rapidly over time, and thus one can use really high AC gain and low DC gain on your front-end. If you put up a big loop antenna, you WILL see the signature from this thing from much further than 100 m. Lightning strokes are detected this way; the NLDN uses only about 100 stations to detect lightning all over the US.
You can think of the strokes as being a series of irregularly-spaced delta functions. They aren't perfect delta functions, of course, because there isn't any such thing. Now, a perfect delta function has power at all frequencies (i'm simplifying), but for a very short period of time. The lightning flash (collection of leaders and strokes) often has several strokes, but since the temporal spacing is uneven there is no fundamental frequency you can measure.
It goes up and down, but it's not oscillating - at least not harmonically.
The use of a large, really large fixed object such as a really tall tower or a skyhook would allow for the capture of a significantly increased amount of lightning, true. That is a very good point. That doesn't solve the storage problem, though.
Our rockets themselves aren't that expensive, truthfully. And we can usually reuse the rocket proper, we just load a new motor in it. Those are cheap. What's expensive is the spool of wire attached to the rocket. You'd think that getting 700 m of 32 ga wire wrapped on a spool would be cheap, but to get a spool wrapped with 32 ga Kevlar-reinforced wire which will spool off cleanly every time is, in fact, a non-trivial task. That's where most of the $500 figure comes from. That DOESN'T include the costs of site maintenance, personnel, range safety, etc.
The typical successful trigger occurs at something between 300 m and 700 m here in Florida. It may be different elsewhere, where the clouds are higher.
Lightning flashes HAVE been triggered with large lasers. But we don't have one. We could probably get one, if we could find someone to fund the purchase, but then the amortized cost per flash would probably (yeah, I'm just guessing) exceed the cost of the rockets we use now.
Most of my class notes are pretty well encapsulated in the book my profs wrote, Lightning: Physics and Effects by Rakov and Uman. I imagine Amazon has it, but it's real spendy - I don't even own a copy. However, it's pretty much a definitive and comprehensive treatment of the subject. The bibliography alone is worth the price.
There is a great big electric field associated with a nearby channel, and field gradients can result in some really interestingly large voltages to appear across things like the ground.
However, to call these fields radio waves implies that they are oscillatory in nature, which is simply not accurate. I'm a lightning researcher, and in the course of my work I've studied lightning electric fields recorded during close lightning strikes. It's not my personal favorite area of interest, but I know enough to say that "radio waves" is a poor description.
The reason that you don't stand near trees during an electrical storm is because 1) the flash is likely to initiate a side channel which passes from the tree-trunk (radio tower, light pole, bus stop, etc) through you, making you very unhappy; and 2) because the HUGE injection of current into the ground causes the ground itself to "rise" from a nominal 0 volts to several kilo- or even mega-volts, and that voltage falls off as the square of the distance... so that if your two legs are 10 feet and 12 feet respectively from the channel termination point, you might experience a voltage of several kV (or more) between them. This causes a current to flow up one leg and down the other, and makes you (and your goodies, don't ya know) very unhappy. This is worse if you happen to have four legs which CAN'T be placed together, like if you're a cow or horse. Zap!
Slashdot Mirror
Seems like good, important work. Also seems irrelevant to the discussion, as near as I can tell. My assertion (still unsupported with citations -sorry) is that the auditory system is incapable of detecting or encoding information about the phase of the incoming signal in the neural spike train. If nothing else, the mechanical resonance of the hair cells prevents them from detecting the incoming signal's phase.
... :-)
I'm not disagreeing AT ALL that the phase of the spike trains contains information, or that phase shifts IN THE SPIKE TRAIN are irrelevant. I'm talking about the ability of the cochlea and related neurons to detect and encode phase information for incoming audio signals above a few kHz; above the speech region, basically.
And, moreover, the whole discussion is irrelevant to whether a 50 ns phase shift in a speaker cable, at 25 kHz, is audible. That's my fault; I brought it up to demonstrate WHY such short phase differences are inaudible. But we've gotten off on a bit of a tangent
Hi, welcome to the party. If you look about 5 levels up, you'll see someone asserting that skin effect is audible in audio cables. I point out that in the example he cites, the total power loss is 0.020 dB at 25 kHz. I contend that this is completely and totally, under any imaginable condition, inaudible by humans and thus irrelevant. So yes, I am ignoring it. I also pointed out that the phase shift described in his citation was about 50 ns at 25 kHz, again too short (not significant enough) to be audible under any conceivable conditions. I also pointed out that neuron firing rates in the auditory path are limited to a few kHz, and that those neuronal spikes are synchronous with the incoming audio signal up to some few kHz, which (and this is the point of contention) may limit the auditory system's ability to report phase to the brain to signals less than some few kHz. Consequently, a phase shift of 50 ns is totally indiscernable by the human auditory system.
I can show, as does the grandparent poster's citation, that the signal DOES arrive largely intact, with neither amplitude NOR phase significantly affected. In this case, when I say significantly, I mean "to a level which is in any way perceptible to the human auditory system".
In other words, skin effect in audio interconnects is totally irrelevant - either from a phase or amplitude point of view.
I think we're talking at cross-purposes here. I asserted, based on the information I've soaked up in the last couple of weeks of discussion, that auditory path neurons do not present phase information to the brain for frequencies above a few kilohertz.
You've replied with links to researchers who work in neural spike train coding. I saw references to Berger's work, but could not find the work itself. None of the titles suggested anything about auditory path neural coding, but not being able to read the papers I can't be sure of that. Liaw's work is available from the web, but in the few minutes I have I haven't been able to find any reference to auditory path coding there, either. There's a lot of work with the hippocampus, true; but I don't see the connection between the hippocampus and the encoding of phase information by the auditory neurons.
I have no intention of turning this into a discussion of neural coding, because quite frankly I only know enough to know that I don't know enough. I will ask the nice gentlemen and ladies who've given these presentations to point toward some of their references, and happily share them with you. Note, however, that these will be works dealing with the auditory path only, and probably with some clinical studies of phase comprehension in hearing tests.
I mentioned spike rate as a limiting factor because the work that was referenced in these presentations, and which I have yet to cite for you, showed the phase alignment of auditory path neural spikes with the incoming signal, at signal frequencies up to the maximum spike train frequency and occasionally a bit beyond. If the phase of the input signal is being encoded somehow into the spike train from the auditory neurons, then that coding process is not well understood - at least not by any of my sources.
Are you really telling me that you are aware of work that specifically shows the transmission of phase information, in the auditory path, at frequencies significantly higher than the maximum spiking rate?
Happened to me this morning. So, today's the day. What are you going to do?
How neurons encode information is NOT - repeat, NOT - a known process. There are several theories regarding whether coding is purely based on spike frequency, spike synchrony, or some other coding systems we haven't discovered yet. I've spent the last two weeks listening to some of the world's brightest and most knowledgeable experts on neural coding, auditory system, visual system, etc at the Telluride Neuromorphic Engineering Workshop and the best thing that can be told is that there is more going on than spike rate, but there is no consensus on exactly what that code is. If you know more and better, then you should be here next year. Contact the organizers.
What I DO know, and what actual experiments have shown, is that auditory neurons fire in synchrony with the incoming audio signal up to a few kHz. Beyond that, there is no correlation between the phase of the incoming audio signal and the spike train from the neuron. The neurons continue to fire, but not in phase with the audio signal. Of course higher frequencies are being detected and passed on to the brain; but absolute phase information, as far as anyone can tell, is not preserved beyond that few kHz limit. Again, your evidence to the contrary is welcome. Try to be specific and cite research wherever possible.
Hmm... which argument do you wish to present? That there are no differences between a tube system and a DSP-emulated system? That different people have different ideas about what is "better"? Or that psychoacoustics play a part in what is considered "better"?
If you wish to argue the latter two, I won't disagree a bit. The first argument I can partially agree with.
If you wish to tell me that you can accurately reproduce the spectral response of a clean tube amplifier using a clean SS amplifier and a DSP, I will not argue a bit. I notice that you haven't mentioned anything about temporal response or phase response, neither of which is particular easy to measure with a spectrum analyzer. However, I don't know of any system which claims to actually do any of these things. If anyone sells a DSP-based SS amplifier which claims to emulate tube amplifiers for hi-fi, I'm not aware of it. Can be done, perhaps. Probably. But I don't think it has.
However, if you slide over into the realm of music production applications, the story is different. The distortion characteristics of a real tube amplifier are significantly different from anything simulated by DSP to date, and the difference is not subtle to those of us who have earned our living through the use and repair of tube-based guitar amps. Bottom line, there isn't a DSP-based emulator that really sounds like a small-box Plexi Marshall on a 4x12 [shivers in ecstasy - that thing was AWESOME], or like a Vox AC30 Top Boost, or even really like a Blackface Fender Champ - three tubes, one speaker! There are dozens which try, and a few which come close, but the state of the art simply is not yet up to snuff. It's not simply a matter of frequency response; there are temporal effects, resonance effects, mechanical effects - it's very complicated, and people really can tell the difference. I haven't even gotten into room emulation or microphone and placement emulation - moving the mic 1/4" laterally in front of the speaker makes an amazing difference.
As a repair tech, I've played though at least one or two different examples of just about every guitar amp out there. I've also played through most of the DSP simulators, and have owned several. If you can make a DSP emulator which really sounds like those amps, call me. I'll buy one, and I've got the contacts in the music equipment biz to make us both filthy rich selling them.
Oh, and I've also got multiple EE degrees, including several courses in DSP, ASP, and human auditory perception. Looks nice in the business plan, ya know?
Went. That is, in fact, a pretty clear and accurate explanation of skin effect. And it shows me exactly what I already knew. A power difference of less than 0020dB at 25 kHz is, in fact, totally imperceptible to the human ear - at least, according to rigorously scientific studies. Many audiophiles will claim differently, I'm sure.
Additionally, the 50 ns phase shift given in the example is also indiscernible to the human auditory system. In fact, the human auditory system loses the ability to phase lock at frequencies above a few kilohertz. I don't want to go too far afield here, but the study of the physiology of the human auditory system can be interesting. Bottom line is that the spiking rate of the neurons of the auditory system is on the order of a few kHz, and for signals above those frequencies the only information reported is amplitude, not phase. Below, 2-3 kHz, the neurons spike in synchrony with the incoming audio signal. 50 ns worth of phase shift at 25 kHz is inaudible, and I challenge anyone in the world to prove differently. I will gladly eat my own words if anyone can do so - but I'm not particularly worried.
There are, of course, phenomenae that science does not understand or has not yet discovered. However, I personally hold any audiophile results which claim to be able to discern either 0.02 dB of power loss or 50 ns of phase shift, at 25 kHz, to be psychologically biased and inaccurate.
I would hate to refer to you as either being a liar or as being deluded, so I hope that you don't really believe that 0.02 dB is an audible difference. Thank you, though, for a most interesting and polite discussion.
Hmm... except that when shouting in a pipe, the wavelength of the sounds produced by your voice are of the same order of magnitude as some dimension of the pipe. Morever, the shouting in a pipe causes distortion through the Helmholz effect, wherein the pipe resonates at certain frequencies and 'smears' sounds at and near those frequencies. In an audio cable, the wavelength of even the highest frequencies you can possibly imagine hearing (say, 40 kHz per Pioneer's high-speed DAT tests from the '90's) are so much longer than any possible path length difference that the smear... well, lessee. Got a calculator around here somewhere...
The wavelength of a 40 kHz sine wave, in a waveguide (coax cable) whose propagation velocity is 1.5e8 m/s (half the speed of light; a pretty slow waveguide) is approximately 37.5 KILOMETERS. Even if the skin effect presented two paths whose length differed by a factor of four (fairly extreme case), you'd have to have a 25 meter length of cable to achieve 1 degree of phase shift. Summing two signals with phase shift of 1 degree results in a net amplitude loss of 0.00066 dB. One degree of phase shift at 40 kHz results in a time 'smear' of about 70 ns.
Perhaps you have ears capable of discerning a 0.00066 dB amplitude fluctuation at 40 kHz - I know I don't. Perhaps you can hear time smear of 70 ns - I know I can't. Perhaps you also have 25 meter - that's 80 some-odd feet for those of us in the US - interconnects; my whole apartment isn't 80 feet long. However, I think you have none of those things. I think I've presented a pretty extreme case, and shown that electrically, what you claim happens is either not there or is so insignificant that the difference is truly academic.
However, it may be that I've misunderstood your usage of the term "skin effect". It's very common in the audiophile world for salesman and such to misappropriate scientific terminology, banking on the "baffle them with bullshit" principle. If I have misinterpreted your statements, kindly set me aright so that I may further understand exactly what you mean.
There are several manufacturers of very expensive cables who claim that they are directional, and that one specific end of the interconnect must be the source end. They claim that the only way to tell which way the cable is aligned coming off of the roll from the manufacturer is to listen to it, very carefully. They also claim that the cables must be "broken in" through constant listening, and that if you don't use the cables for a period of some weeks, they lose their conditioning and must be "re-broken-in". This is for line-level interconnects, not even speaker-level. They claim that these effects are due to "microdiodes" in the crystalline structure of the copper (or silver) they use for conductors. They cleverly ignore the fact that the signals are AC, and thus current flows equally in both directions. They also totally gloss over the fact that solid-state diodes (at least all solid-state diodes known to real science) have a voltage drop on the order of hundreds of millivolts, which for an audio line-level signal is usually on the order of half of the overall voltage amplitude. It's simply amazing, and sounds perfectly plausible to those who don't know any better.
Perhaps you could explain to me, in small words that a musician, studio engineer, and holder of multiple EE degrees (that's me) can understand, exactly what skin effect you are referring to that is perceptible at audio frequencies, and under what conditions.
I'm waiting patiently.
Do you work in the snake oil - I mean, audiophile business? I've never heard a bullshit spiel that elegant before. What are you selling?
None of that shit - skin effect, switching time - has any relevance at audio frequencies.
I take it you've already tried the one about the "microdiodes" in wire that make it unidirectional, never mind the fact that the signal IS AC, ferchrissakes?
So true, so true. I read a post on this very site a few days ago wherein some idiot was taking back an iRiver because it 'only' claimed 30 dB worth of bass boost, or some such nonsense.
30 dB?!?! That's ridiculous! How can you even consider using that much bass? You can't possibly hear the music at that point, just some insane thumping and a little human voice off in the background!
What an idiot.
I realize that you are trolling, but I'll bite. Bob Carver's claims notwithstanding, nobody has been able to accurately simulate the sound of a tube amplifier using analog or digital filtering. Nor is there any significant amount of hum in a decent tube amp.
:-)
There are dozens of companies willing to sell you simulators of overloaded tube amps. You, personally, may even think that they sound just like overloaded tube amps. However, they don't. I have played guitar, repaired guitar amplifiers, engineered and mixed studio albums, and I know from experience that the state of DSP for tube emulation simply isn't there yet. You may disagree; that's fine. But you're wrong.
Parent poster is correct. Several products exist which attempt to use DSP to mimic the sound of tube amplifiers, almost all of them in the guitar effects realm. None of them are yet there. They sound a lot LIKE a tube amp, but they don't capture it all. CAN they? Probably someday. But not yet.
I base this opinion (yes, opinion) on: several years as a professional audio engineer; several years as a guitar amplifier repair technician; several years as a semi-professional guitarist; and two degrees (working on a third) in Electrical Engineering.
Sibling posters who believe that DSP can do anything are correct up to a point: DSP can achieve any given transfer function, up to a desired level of accuracy. You need more accuracy? Increase the bit depth and sample rate, tweak the processing. However, the bug stumbling block is this: you gotta know what transfer function you want to emulate first.
Sheesh, do you also use 2 cars in tandem because one is always broken?
Never driven an MGB, have you?
Might I suggest you work on your cause-and-effect a bit more?
... well, OK, welcome to Slashdot!
I have three rocks in my garden. All of them are red. None have been stolen. Ergo, red rocks cannot be stolen.
See how stupid it sounds? Dude - the top five cars on that list are also the TOP FIVE most common cars in the U.S. If you can establish a relationship between the theft rates of similarly-equipped cars, where the only variable is RFID or not, then you've got a case and you are welcome to tell us all about it. If you just wanna spout uneducated shit...
So this is what passes for a troll post these days? Nice.
I also like the fact that the very first mod was "-1, Overrated". How does that work?
Aha! Thank you.
Oh, BTW, first post.
Beats me. Apparently the metallurgists say that amorphous steel is stronger. [ shrug ] Sounds good to me.
However, the claim about the amorphous steel being "twice as strong and hard" smells funny to me... twice as hard? On what sort of scale? Isn't steel already about 2/3 as hard as diamond anyway? Does that mean that amorphous steel is 4/3 as hard as diamond? How would they measure that?
Enquiring minds want to know!
Right. And what's a BPL again?
Yet Another Annoying Acronym.
"glassy", as in "amorphous", as in "non-crystalline". Does NOT mean transparent like window glass. Think obsidian - it's black (or green), opaque, and shatters in totally random directions. That's because it has no crystalline structure and thus no lines or planes of fracture.
This is non-crystalline steel. It's not transparent aluminum - but then, nothing is.
Yes, but the earth's field is static. The magnetic pulse from the railgun will be dynamic, meaning that it changes very rapidly over time, and thus one can use really high AC gain and low DC gain on your front-end. If you put up a big loop antenna, you WILL see the signature from this thing from much further than 100 m. Lightning strokes are detected this way; the NLDN uses only about 100 stations to detect lightning all over the US.
Err... what?
20 amps is 64 times more than 25 kiloamps?
I'm sorry, I obviously wasn't prepared to debate you in your own forum.
You can think of the strokes as being a series of irregularly-spaced delta functions. They aren't perfect delta functions, of course, because there isn't any such thing. Now, a perfect delta function has power at all frequencies (i'm simplifying), but for a very short period of time. The lightning flash (collection of leaders and strokes) often has several strokes, but since the temporal spacing is uneven there is no fundamental frequency you can measure.
It goes up and down, but it's not oscillating - at least not harmonically.
The use of a large, really large fixed object such as a really tall tower or a skyhook would allow for the capture of a significantly increased amount of lightning, true. That is a very good point. That doesn't solve the storage problem, though.
Our rockets themselves aren't that expensive, truthfully. And we can usually reuse the rocket proper, we just load a new motor in it. Those are cheap. What's expensive is the spool of wire attached to the rocket. You'd think that getting 700 m of 32 ga wire wrapped on a spool would be cheap, but to get a spool wrapped with 32 ga Kevlar-reinforced wire which will spool off cleanly every time is, in fact, a non-trivial task. That's where most of the $500 figure comes from. That DOESN'T include the costs of site maintenance, personnel, range safety, etc.
The typical successful trigger occurs at something between 300 m and 700 m here in Florida. It may be different elsewhere, where the clouds are higher.
Lightning flashes HAVE been triggered with large lasers. But we don't have one. We could probably get one, if we could find someone to fund the purchase, but then the amortized cost per flash would probably (yeah, I'm just guessing) exceed the cost of the rockets we use now.
Most of my class notes are pretty well encapsulated in the book my profs wrote, Lightning: Physics and Effects by Rakov and Uman. I imagine Amazon has it, but it's real spendy - I don't even own a copy. However, it's pretty much a definitive and comprehensive treatment of the subject. The bibliography alone is worth the price.
Err... no.
There is a great big electric field associated with a nearby channel, and field gradients can result in some really interestingly large voltages to appear across things like the ground.
However, to call these fields radio waves implies that they are oscillatory in nature, which is simply not accurate. I'm a lightning researcher, and in the course of my work I've studied lightning electric fields recorded during close lightning strikes. It's not my personal favorite area of interest, but I know enough to say that "radio waves" is a poor description.
The reason that you don't stand near trees during an electrical storm is because 1) the flash is likely to initiate a side channel which passes from the tree-trunk (radio tower, light pole, bus stop, etc) through you, making you very unhappy; and 2) because the HUGE injection of current into the ground causes the ground itself to "rise" from a nominal 0 volts to several kilo- or even mega-volts, and that voltage falls off as the square of the distance... so that if your two legs are 10 feet and 12 feet respectively from the channel termination point, you might experience a voltage of several kV (or more) between them. This causes a current to flow up one leg and down the other, and makes you (and your goodies, don't ya know) very unhappy. This is worse if you happen to have four legs which CAN'T be placed together, like if you're a cow or horse. Zap!