First off, I wasn't implying that high-quality headphones aren't valuable. I have $80 Sony headphones that have good frequency response. As to your question about balanced headphones...
Most high-performance analog signal processing these days is balanced. For example, the analog data path in a communications transceiver is almost certainly balanced, as are the data converters. There are a couple of key benefits of balanced (called differential in the industry) signal processing. The key one is rejection of interference that appears the same on both wires (since the signal is the difference of current or voltage on the wires). Also important lately is an increase of 3dB in SNR by using a differential signal path. This is simply because the signal on the two wires is perfectly correlated, while the noise on the two wires is uncorrelated. That said, differential signal processing sounds like a good idea for headphones, right? Well... it COULD be.
The problem is for a signal to accrue the benefits of balance it has to balanced everywhere there could be interference. Remember the point here is to have the absolutely cleanest signal possible (this is for audiophiles after all). The problem is that the signal IS NOT REALLY BALANCED. Look at the FAQ I posted the link to, refer to Art. III (Balanced Sources). If you look at the handsome diagrams you will see some problems. Now, to be a differential or balanced signal you need to have a signal that is equal and opposite. In the case of a vinyl source they get a single-ended source from the Phono and put it through two op-amp circuits, one inverting and one non-inverting, and they are depending on the outputs of the two circuits to have exactly the same phase relationship. True, they will be close because the audio is much lower in frequency that the bandwidths of the amplifiers, but it isn't truly balanced here. And the mismatch between the two halves is most likely MORE than the distortion/interference you would expect from a good quality single-ended headphone. Ouch!
For the digital source, it is a train wreck! That is NOT the way DACs are supposed to be used! I have designed quite a few data converters and they in no-way-shape-or-form match each other well. (In digital audio we are talking about supreme precision, so the matching isn't even close) If they could match that well, it would be possible to put a bunch in parallel and create SUPER FAST data converters. You can't do that easily in practice due to all kinds of DISTORTION due to mismatches between channels. There is no way that the overall signal path would be limited in performance by anything here than the mismatch of the DACs themselves. I would guess if you looked at the spectrum of the "balanced" signal it would be full of tones due to the DACs. OUCH!
That said, it is quite possible that subjectively this sounds good, because the ear finds certain kinds of distortion pleasing. For example, overdriven vacuum tubes sound good to a lot of people. However, from a technical standpoint, this is a supreme waste of money, and probably sounds worse than a good quality $100 - $200 set of single-ended headphones.
I would say luck plays a larger role in the success of technology startups than most people would like to think. You can make the argument that if you were unlucky in market, competition, etc. then you really didn't have the business acumen you needed, but I think that is kind of a circular argument: successful people have business acumen, and understanding business makes you a success.
PS I was the first engineer at a startup, so I've seen it in all its glory and pain.
That is a good statement, but unfortunately people have been doing exactly what you suggest since at least the late 1970s. The problem is, there is no good way to predict or even fathom what innovations will drastically change whatever assumptions have to be made to enable the forecast in the first place.
There were grave technological barriers to making transistors with gate lengths smaller than 1 micron. It was not clear in the early 80s if they could be solved. In fact, they have been solved and at a cost much, much, lower than really anyone predicted. This has happened several times in the semiconductor business.
Now, the semiconductor roadmap is assuming that some technological innovation will enable us to make deep UV lithography an econonmic reality. It is behind schedule, and direct e-beam lithography is far to expensive to use to pattern wafers (it is used in mask fabrication at present). Maybe the end is nigh. Based on history, I think there will be a work around. There is so much money at stake.
This whole Moore's law obsession is not germane. This is a business. When it makes sense to continue to pour billions into semiconductor manufacturing R&D we will. When it doesn't, we will stop. It's quite simple. It is common knowledge that the improvements in software have mainly come from faster computers (Boyle's law of software and all that). When it is too expensive to continue to improve semiconductors, we will do something else, and it will become cost effective to do more research into software.
I'm with you Packetmon. I got through a demanding program in Electrical Engineering in the 90s without owning a computer. I used the UNIX terminals in the engineering building for homework, and used the Mac lab to write my papers. Saved me a lot of money, and got me out of the dorm. Nowadays everyone has a personal laptop. I wonder if the students work alone more.
In case you didn't know, SiC was actually the first commercially successful semiconductor. It is also known as Carborundum and was the basis of the original low cost "crystal radio sets" that were used by hobbyists and cheapskates to demodulate AM radio. Basically you had a wire called a "cat whisker" and you moved it around the SiC crystal until you made a good rectifying contact. Then, you would adjust the channel with a variable condenser (now called of course a capacitor) and it drove high impedance headphones. It was a "free energy" radio that had terrible performance but it was pretty cool because it used the EM radiation in the signal itself as the power source.
The more things change, the more old ideas/materials/circuits get dusted off.
I have to pay (in theory) for my own articles all the time. Of course I never do. I save my own copies... duh. Even if I didn't I could always have a coworker email me a copy. Of course at work I have unlimited access to the papers, but if I were to quit I could be in that situation.
Makes sense to me, its my work but the "paper" was published by someone else. If any researcher emails me, I'd be happy to send them a paper if they don't want to pay.
I remember POP-CORN quite fondly. It was the only way to reset your clocks after a power outage if you didn't have a wind-up or battery powered watch/clock you trusted. My friends and I used to play the lamest game when we were in elementary school. We would dial POP-CORN and try to time it so the voice would say "exactly". It was a timing thing. Instead of saying "three forty-two and twelve seconds", to hear it say "three forty-three exactly" was a real score! Ah, the early 80s.
Science may be somewhat underpaid, but engineering (which is in one definition the application of scientific principles) is not in the UK. We have design teams (I work for an American semiconductor manufacturer) in the UK that are quite talented and well paid. Of course they need to be well paid since they live in Kent.
That said, science is hard. It kicked my ass to get through engineering school. Physics, Maths, and Chemistry are what they are. You can't make them easier. If you want to be competent, you have to accept the pain associated with acquiring the needed skills and understanding (not so much "knowledge").
I suppose the fact that science positions in the UK are unpaid explains why I worked with so many good English and Scottish physicists in my first job at an American government laboratory.
Another hard game: Project X on the Amiga. It took two of us playing co-operatively to beat it. And this is a side-scrolling shoot-em-up!
I thought Shadow of the Beast was incredibly hard. What pissed me off about it was you had to get through the entire level again after you died. So, if you had a hard time at one spot near the end of a level, too bad, you had to work for 5 min to get there again. That f*in game made me want to throw my joystick through my 1084 monitor! Another tough one was WINGS. I guess it was hard since actually surviving a WWI dogfight was hard, so it at least was realistic. I killed so many poor aviators.
How sad. I designed some of the ICs for the drift chamber trigger for PEP-II. I am proud to have been a part of that project, but I haven't worked on it since 1997. Thanks for the update.
There are others besides the RHIC. The Beavatron at Lawrence Berkeley National Laboratory, and of course the Stanford Linear Accelerator are still under operation. The Stanford Linear Accelerator has the B-Factory as one of its targets. The B-factory is an extremely advanced detector (in high energy Meson regime particles) and is the center of the BaBar project.
I think we should be friends too... we have a lot in common. I downloaded London and Footfalls. REALLY GOOD STUFF. I think you guys are in touch with your inner Waldorf, no? I made a small donation to thank you guys for putting up that synthpop for free. Please email me at carl_r_grace@yahoo.com
Wouldn't JAVA make more sense as Starbucks' stock symbol? I liked SUNW. I have fond memories about learning to program on SUN and HP workstations. HP has already mostly phased out their UNIX workstation line, and this seems to be (potentially) a first tentative step for SUN to become more like IBM and move away from hardware as their bread and butter.
I write this from a SUN Linux box, so I certainly hope this isn't the case.
I attended UCSB for a while as well, but did not get a degree there. I went somewhere else.
I would say that the majority of wild Isla Vista residents are in fact UCSB students. There are many Santa Barbara City College students there, but all the people in my apt building were UCSB. Also, there are a lot of Mexican laborers in the town.
The Engineering and Materials science departments are outstanding. But it is a large university and there is a lot of party people. It is a fact. UCSB has the largest percentage of white students in the UC system. That is also a fact.
The ECE department is universally regarded as excellent and myopic... all they care about is compound semiconductors. We can thank Nobel Laureate Herbert Kroemer for that.
I agree with you that UCSB is a wonderful school. But the fact is there are a lot of 24 hour party people as undergraduates. Denying that does not improve the reputation of the science there.
I'm sure engineered bands are somewhat to blame for this, and it does seem that people tend to prefer the music they grew up with during their teens and early twenties.
People have thought the music they listened to in their teens and early twenties was better than current music since the 1902. Seriously, I think Depeche Mode's best album was "Violator", the Cure's "Disintegration" is better than anything out now, and Morrissey's "Your Arsenal" is one of the masterpieces of modern rock. Of course I'm biased as hell and young people today wouldn't even know some of these artists. They are probably nostalgic for pop punk.... gee, I miss when music was good, like "Simple Plan" and "Sum 41".
And no, I don't have "exotic cables", just quality speakers and a hefty power amp with plenty of headroom to spare.
Come on now, don't you know that Monster Cable makes everything sound better? hehe. When I was in college I had a buddy with a hi-fi amp that oscillated when connected to speakers with a long run of Monster Cable. It was just too damn capacitive. I replaced it with Radio Shack speaker wire and it sounded great. What a rip off.
stuttering vocoders and dropped calls as "tolerable"
I'm with you, Mr. Transistor. We have a joke around the office, that Cingular's (now AT&T) cell-phone slogan which is "fewest dropped calls" actually translates to "We Suck Less". We have the technology to make things work SO much better, the problem is entirely economic. The large companies need to get a return on investment, so they basically just try to see what they can get away with. For the cell-phone example, if Cingular built more base stations, and limited the number of simultaneous phones on the network at a given time, they would have virtually zero dropped calls... but substantially less revenue.
PS.. what kind of xtr are you? MOS, BJT or HBT? Perhaps an exotic HEMT?
It's illegal to crank commercial volumes, but every local station does it anyway - advertisers love it. I have to turn down the volume every time a stupid loud commercial comes on.
You don't seem to understand it, but that is the crux of the loudness war. The local stations do not in fact crank the volume on commercials. That would be illegal. In fact what they do is compress the dynamic range of the audio, so the "apparent loudness" is increased. The peaks (which is how the FCC defines volume) are the same, but the RMS volume (essentially the average sound level and what our ear perceives as volume) is increased. Think about it, a CD is 16 bit, so the max volume is obviously 2^16=65536 for any particular data sample. So, they can't make the volume 2^17. What they can do, however, is compress the dynamic range, so instead of the average volume level to be at 4096, say, it is now 16483.
Commercials on TV suck, don't they. The audio is compressed to hell and back.
It most certainly required new ideas and new technology to get to the end result. However, my standpoint is that while the processes to accomplish this task may very well be innovative, the idea of 40nm threading itself is not
I think I finally understand your point. I may even agree now. Me saying "screw 45nm, let's get down to 32nm" is hardly innovative. What is innovative are the ideas that scientists and engineers are going to have that enable us to circumvent the technological hurdles involved in scaling down to 32nm. I think we have found common ground. Thank you for the comment.
Yes, but the wafers themselves cost much much much more. Also the cost of the tooling equipment hasn't yet been depreciated. Generally a given chip will be cheaper in a process one step behind the leading edge, because of the tooling cost. For analog circuits (not the kind we're talking about) the oldest process possible is the cheapest, because the whole of the equipment needed to manufacture is free from a relative standpoint.
I guess the issue is we have different definitions of the word "innovation". I should note, that your strict definition of innovation is non-standard and is not generally agreed on. I would argue that it is more that complicated. You state that innovation is taking a route that is completely unlike anything we're seeing today. Does that mean the only innovation we've ever seen in electronics is the invention of the vacuum tube? It was the first amplifying device. Everything since then is basically a refined amplifying device. Integrated circuits are simply the same old circuit but now built on the same substrate. How is that innovation? Has there been no innovation in electronics since the turn of the 20th century?
Slashdot Mirror
Fross,
First off, I wasn't implying that high-quality headphones aren't valuable. I have $80 Sony headphones that have good frequency response. As to your question about balanced headphones...
Most high-performance analog signal processing these days is balanced. For example, the analog data path in a communications transceiver is almost certainly balanced, as are the data converters. There are a couple of key benefits of balanced (called differential in the industry) signal processing. The key one is rejection of interference that appears the same on both wires (since the signal is the difference of current or voltage on the wires). Also important lately is an increase of 3dB in SNR by using a differential signal path. This is simply because the signal on the two wires is perfectly correlated, while the noise on the two wires is uncorrelated. That said, differential signal processing sounds like a good idea for headphones, right? Well... it COULD be.
The problem is for a signal to accrue the benefits of balance it has to balanced everywhere there could be interference. Remember the point here is to have the absolutely cleanest signal possible (this is for audiophiles after all). The problem is that the signal IS NOT REALLY BALANCED. Look at the FAQ I posted the link to, refer to Art. III (Balanced Sources). If you look at the handsome diagrams you will see some problems. Now, to be a differential or balanced signal you need to have a signal that is equal and opposite. In the case of a vinyl source they get a single-ended source from the Phono and put it through two op-amp circuits, one inverting and one non-inverting, and they are depending on the outputs of the two circuits to have exactly the same phase relationship. True, they will be close because the audio is much lower in frequency that the bandwidths of the amplifiers, but it isn't truly balanced here. And the mismatch between the two halves is most likely MORE than the distortion/interference you would expect from a good quality single-ended headphone. Ouch!
For the digital source, it is a train wreck! That is NOT the way DACs are supposed to be used! I have designed quite a few data converters and they in no-way-shape-or-form match each other well. (In digital audio we are talking about supreme precision, so the matching isn't even close) If they could match that well, it would be possible to put a bunch in parallel and create SUPER FAST data converters. You can't do that easily in practice due to all kinds of DISTORTION due to mismatches between channels. There is no way that the overall signal path would be limited in performance by anything here than the mismatch of the DACs themselves. I would guess if you looked at the spectrum of the "balanced" signal it would be full of tones due to the DACs. OUCH!
That said, it is quite possible that subjectively this sounds good, because the ear finds certain kinds of distortion pleasing. For example, overdriven vacuum tubes sound good to a lot of people. However, from a technical standpoint, this is a supreme waste of money, and probably sounds worse than a good quality $100 - $200 set of single-ended headphones.
Carl
I forgot to add... here is a FAQ that explains why you NEED this headphone amp....
http://www.headphone.com/products/faqs/balanced-headphones/balanced-vs-unbalanced/
If you know anything about analog circuits and audio you're going to have to try not to laugh out loud.
We had a good laugh at work about this balanced headphone amp: http://www.headphone.com/products/headphone-amps/the-max-line/headroom-balanced-max-amp.php
We were thinking if there are really people paying $4k for this stuff, we're in the wrong business (Analog Integrated Circuits)
Audiophiles are idiots. The issue is they have more pretension than technical acumen... so they are easily taken.
Carl
I would say luck plays a larger role in the success of technology startups than most people would like to think. You can make the argument that if you were unlucky in market, competition, etc. then you really didn't have the business acumen you needed, but I think that is kind of a circular argument: successful people have business acumen, and understanding business makes you a success.
PS I was the first engineer at a startup, so I've seen it in all its glory and pain.
That is a good statement, but unfortunately people have been doing exactly what you suggest since at least the late 1970s. The problem is, there is no good way to predict or even fathom what innovations will drastically change whatever assumptions have to be made to enable the forecast in the first place.
There were grave technological barriers to making transistors with gate lengths smaller than 1 micron. It was not clear in the early 80s if they could be solved. In fact, they have been solved and at a cost much, much, lower than really anyone predicted. This has happened several times in the semiconductor business.
Now, the semiconductor roadmap is assuming that some technological innovation will enable us to make deep UV lithography an econonmic reality. It is behind schedule, and direct e-beam lithography is far to expensive to use to pattern wafers (it is used in mask fabrication at present). Maybe the end is nigh. Based on history, I think there will be a work around. There is so much money at stake.
This whole Moore's law obsession is not germane. This is a business. When it makes sense to continue to pour billions into semiconductor manufacturing R&D we will. When it doesn't, we will stop. It's quite simple. It is common knowledge that the improvements in software have mainly come from faster computers (Boyle's law of software and all that). When it is too expensive to continue to improve semiconductors, we will do something else, and it will become cost effective to do more research into software.
I'm with you Packetmon. I got through a demanding program in Electrical Engineering in the 90s without owning a computer. I used the UNIX terminals in the engineering building for homework, and used the Mac lab to write my papers. Saved me a lot of money, and got me out of the dorm. Nowadays everyone has a personal laptop. I wonder if the students work alone more.
Carl
In case you didn't know, SiC was actually the first commercially successful semiconductor. It is also known as Carborundum and was the basis of the original low cost "crystal radio sets" that were used by hobbyists and cheapskates to demodulate AM radio. Basically you had a wire called a "cat whisker" and you moved it around the SiC crystal until you made a good rectifying contact. Then, you would adjust the channel with a variable condenser (now called of course a capacitor) and it drove high impedance headphones. It was a "free energy" radio that had terrible performance but it was pretty cool because it used the EM radiation in the signal itself as the power source.
The more things change, the more old ideas/materials/circuits get dusted off.
I have to pay (in theory) for my own articles all the time. Of course I never do. I save my own copies... duh. Even if I didn't I could always have a coworker email me a copy. Of course at work I have unlimited access to the papers, but if I were to quit I could be in that situation.
Makes sense to me, its my work but the "paper" was published by someone else. If any researcher emails me, I'd be happy to send them a paper if they don't want to pay.
This is a non-issue.
I remember POP-CORN quite fondly. It was the only way to reset your clocks after a power outage if you didn't have a wind-up or battery powered watch/clock you trusted. My friends and I used to play the lamest game when we were in elementary school. We would dial POP-CORN and try to time it so the voice would say "exactly". It was a timing thing. Instead of saying "three forty-two and twelve seconds", to hear it say "three forty-three exactly" was a real score! Ah, the early 80s.
Science may be somewhat underpaid, but engineering (which is in one definition the application of scientific principles) is not in the UK. We have design teams (I work for an American semiconductor manufacturer) in the UK that are quite talented and well paid. Of course they need to be well paid since they live in Kent.
That said, science is hard. It kicked my ass to get through engineering school. Physics, Maths, and Chemistry are what they are. You can't make them easier. If you want to be competent, you have to accept the pain associated with acquiring the needed skills and understanding (not so much "knowledge").
I suppose the fact that science positions in the UK are unpaid explains why I worked with so many good English and Scottish physicists in my first job at an American government laboratory.
Another hard game: Project X on the Amiga. It took two of us playing co-operatively to beat it. And this is a side-scrolling shoot-em-up!
I thought Shadow of the Beast was incredibly hard. What pissed me off about it was you had to get through the entire level again after you died. So, if you had a hard time at one spot near the end of a level, too bad, you had to work for 5 min to get there again. That f*in game made me want to throw my joystick through my 1084 monitor! Another tough one was WINGS. I guess it was hard since actually surviving a WWI dogfight was hard, so it at least was realistic. I killed so many poor aviators.
How sad. I designed some of the ICs for the drift chamber trigger for PEP-II. I am proud to have been a part of that project, but I haven't worked on it since 1997. Thanks for the update.
There are others besides the RHIC. The Beavatron at Lawrence Berkeley National Laboratory, and of course the Stanford Linear Accelerator are still under operation. The Stanford Linear Accelerator has the B-Factory as one of its targets. The B-factory is an extremely advanced detector (in high energy Meson regime particles) and is the center of the BaBar project.
Hi Null Device,
I think we should be friends too... we have a lot in common. I downloaded London and Footfalls. REALLY GOOD STUFF. I think you guys are in touch with your inner Waldorf, no? I made a small donation to thank you guys for putting up that synthpop for free. Please email me at carl_r_grace@yahoo.com
Carl
Wouldn't JAVA make more sense as Starbucks' stock symbol? I liked SUNW. I have fond memories about learning to program on SUN and HP workstations. HP has already mostly phased out their UNIX workstation line, and this seems to be (potentially) a first tentative step for SUN to become more like IBM and move away from hardware as their bread and butter.
I write this from a SUN Linux box, so I certainly hope this isn't the case.
Thanks for the clarification.
How nice to see another electronics geek on here... I'm a 65nm MOSFET.
I attended UCSB for a while as well, but did not get a degree there. I went somewhere else.
I would say that the majority of wild Isla Vista residents are in fact UCSB students. There are many Santa Barbara City College students there, but all the people in my apt building were UCSB. Also, there are a lot of Mexican laborers in the town.
The Engineering and Materials science departments are outstanding. But it is a large university and there is a lot of party people. It is a fact. UCSB has the largest percentage of white students in the UC system. That is also a fact.
The ECE department is universally regarded as excellent and myopic... all they care about is compound semiconductors. We can thank Nobel Laureate Herbert Kroemer for that.
I agree with you that UCSB is a wonderful school. But the fact is there are a lot of 24 hour party people as undergraduates. Denying that does not improve the reputation of the science there.
I'm sure engineered bands are somewhat to blame for this, and it does seem that people tend to prefer the music they grew up with during their teens and early twenties.
People have thought the music they listened to in their teens and early twenties was better than current music since the 1902. Seriously, I think Depeche Mode's best album was "Violator", the Cure's "Disintegration" is better than anything out now, and Morrissey's "Your Arsenal" is one of the masterpieces of modern rock. Of course I'm biased as hell and young people today wouldn't even know some of these artists. They are probably nostalgic for pop punk.... gee, I miss when music was good, like "Simple Plan" and "Sum 41".
And no, I don't have "exotic cables", just quality speakers and a hefty power amp with plenty of headroom to spare.
Come on now, don't you know that Monster Cable makes everything sound better? hehe. When I was in college I had a buddy with a hi-fi amp that oscillated when connected to speakers with a long run of Monster Cable. It was just too damn capacitive. I replaced it with Radio Shack speaker wire and it sounded great. What a rip off.
stuttering vocoders and dropped calls as "tolerable"
I'm with you, Mr. Transistor. We have a joke around the office, that Cingular's (now AT&T) cell-phone slogan which is "fewest dropped calls" actually translates to "We Suck Less". We have the technology to make things work SO much better, the problem is entirely economic. The large companies need to get a return on investment, so they basically just try to see what they can get away with. For the cell-phone example, if Cingular built more base stations, and limited the number of simultaneous phones on the network at a given time, they would have virtually zero dropped calls... but substantially less revenue.
PS.. what kind of xtr are you? MOS, BJT or HBT? Perhaps an exotic HEMT?
It's illegal to crank commercial volumes, but every local station does it anyway - advertisers love it. I have to turn down the volume every time a stupid loud commercial comes on.
You don't seem to understand it, but that is the crux of the loudness war. The local stations do not in fact crank the volume on commercials. That would be illegal. In fact what they do is compress the dynamic range of the audio, so the "apparent loudness" is increased. The peaks (which is how the FCC defines volume) are the same, but the RMS volume (essentially the average sound level and what our ear perceives as volume) is increased. Think about it, a CD is 16 bit, so the max volume is obviously 2^16=65536 for any particular data sample. So, they can't make the volume 2^17. What they can do, however, is compress the dynamic range, so instead of the average volume level to be at 4096, say, it is now 16483.
Commercials on TV suck, don't they. The audio is compressed to hell and back.
It most certainly required new ideas and new technology to get to the end result. However, my standpoint is that while the processes to accomplish this task may very well be innovative, the idea of 40nm threading itself is not
I think I finally understand your point. I may even agree now. Me saying "screw 45nm, let's get down to 32nm" is hardly innovative. What is innovative are the ideas that scientists and engineers are going to have that enable us to circumvent the technological hurdles involved in scaling down to 32nm. I think we have found common ground. Thank you for the comment.
Yes, but the wafers themselves cost much much much more. Also the cost of the tooling equipment hasn't yet been depreciated. Generally a given chip will be cheaper in a process one step behind the leading edge, because of the tooling cost. For analog circuits (not the kind we're talking about) the oldest process possible is the cheapest, because the whole of the equipment needed to manufacture is free from a relative standpoint.
I guess the issue is we have different definitions of the word "innovation". I should note, that your strict definition of innovation is non-standard and is not generally agreed on. I would argue that it is more that complicated. You state that innovation is taking a route that is completely unlike anything we're seeing today. Does that mean the only innovation we've ever seen in electronics is the invention of the vacuum tube? It was the first amplifying device. Everything since then is basically a refined amplifying device. Integrated circuits are simply the same old circuit but now built on the same substrate. How is that innovation? Has there been no innovation in electronics since the turn of the 20th century?