I started with a WAV file of full scale Gaussian white noise. Continue reading
I just tested the audio performance of the Korg MR-2, so figured I would run similar tests of the Canon 5D-MkIII to see if there’s any need to use an external recorder. Continue reading
The Korg MR-2 does not live up to its premium/professional price. When a field recorder costs as much as a smartphone, you expect it to be reliable and high performance. I’ll list the flaws first, state my recommendations, and then delve into the testing details. Continue reading
When I heard Jay-Z and Kanye’s “Murder to Excellence” on the radio and internet, the song sounded muffled. The cymbals aren’t crisp. The entire song, including the vocals, sounds like it’s coming out of bad speakers. I figured maybe the spectral range of the song was too broad for mp3 compression and it was losing the high frequency detail. I wanted to hear the real thing, so I bought the CD to get a good copy of the song. Here’s what I got. It’s the WAV file of Murder to Excellence from the Watch the Throne CD.
This time, I ran the oil 10 months and 7 days for 20,824 miles. As we can see, the wear levels are still good and the oil properties held up. I’m going to go for 25,000 miles next time. Also, I cleaned dirt that had accumulated near the oil fill hole on the engine. Not sure if some of it got inside and somehow dissolved in the oil.
It’s also time to change the air filter. I’ll change the filter and check the intake hoses. The whole intake system may get reworked if I get around to installing a carb-heat style hot air intake system to increase efficiency. Hot air ought to reduce pumping losses and may improve combustion a bit too. I got the idea while thinking through all the reasons fuel economy drops in the winter. Carb heat systems on aircraft are simple, like dryer vents that suck air from a metal scoop near the exhaust manifold, so it should be easy enough to add the same feature to my car. A more complex design could include a bypass that opens up whenever the throttle moves beyond a certain point. It would have to use its own spring and pull-only actuation so there’s less chance of a jam holding the throttle open.
I just found a printed handout from this Norman Augustine lecture and there’s no trace of it on the internet. So, I’ve typed it in for you. — Charvak
Excerpts from the Dec. 5 inaugural Brunel Lecture Series in Complex Systems
The fact is there’s no such thing as a simple system. And, complex systems are made particularly challenging because the interactions in those systems are so easily overlooked or misunderstood.
Today, I’d like to share with you some lessons from my own experience in systems engineering.
1. The first lesson that I would like to cite is the importance of taking a broad view of what constitutes a system. Remember Andrew–a particularly devastating hurricane that hit Florida a few years ago? When the hurricane hit, the telephone companies were having difficulty getting the telephone system back in operation for the first couple of days. The reason was not lack of wire or trucks or switching centers. The recovery was stopped by the lac of child care centers. Most telephone employees come from two-wage earner families and when the hurricane knocked out all the child care centers, one parent had to stay home with the children. Only about half the work force showed the day they so badly needed all the work force and more. The telephone company quickly called in their retirees who set up day care centers so the regular work force could work. The problem, of course, was that they had too narrowly defined the system.
2. The next major lesson is bound the problem. This may sound contradictory to the first lesson. It is, but no one ever said that systems engineering was easy! To true transportation engineers, the air transport system is really only a part of the transportation infrastructure. They would be thinking of highways and ships and perhaps even how to move information. The challenge for systems engineers is to determine what could be the boundaries of the system for practical purposes. If you included too little in the system, it makes the system untenable; if you include too much, it makes it unanalyzable.
3. The next lesson is to watch for unintended consequences, sometimes these can be extremely subtle connections or interactions. The first day I worked when I was out of college, a wise, old engineer said, “No change is a small change.” If I’d paid attention, I could have saved my employer billions of dollars with mistakes I’ve made by not realizing there’s no such thing as a small change. Let me illustrate by referring to the Standard ARM, an anti-radar missile used during the Vietnam War. It had been tested extensively and performed beautifully. In Vietnam, it was flown against radars in North Vietnam and, at 16 seconds time of flight, they would all blow up. Back at the test range, they worked fine. Investigations found nothing. The only difference between the missiles that went to the fleet and the test ones was that one had an actual explosive in the nose, whereas the other had an inert warhead. On the one that went to the fleet, someone put a sticker on the side of the missile that said, “Live round.” Investigators discovered that the glue that held the sticker on would debond at the temperature the missile reached at about 16 seconds in flight. In wind tunnel models, the stickers would blow up through a strange airflow right through the guard beams and detonate the warhead.
4. The next one, question everything, is a lesson that Warren Buffett told me before I was to teach my first class. He said the most important lesson that I could teach my students was to always have someone around who could tell the Emperor he has no clothes. It’s very good advice to always have someone around, preferably yourself, but others too, who can challenge what you’re doing. Prior experience and inborn biases can cause a person to have very fuzzy vision. And if prior experience can cause fuzzy vision, arrogance can cause absolute blindness.
What’s the danger here? When the Hubble Space Telescope was first launched, the media called it the Near-Sighted Mr. Magoo. The subcontractor who built the optics for this telescope, one of the finest optical manufacturers in the world, built the flight article and then tested it. When they tested it, they got a pattern that indicated a huge error. They said to themselves there’s no way we could have made an error like that. Something’s wrong with the test. So the telescope went in orbit and it turned out they had made a huge error. Fortunately, NASA was able to put a set of corrective eyeglasses on the telescope and eventually we got sensational pictures.
5. Another lesson is the things you worry about usually aren’t the things that do you in. That’s because you pay attention to them and you can usually solve them. Rather you tend to have your lunch eaten by things that were overlooked or that you thought were under control but were not. This past year, I served on a commission to review the Osprey, a tilt-rotor aircraft. Our question was whether the concept of having an aircraft that’s both helicopter and fixed-wing in performance might be fundamentally flawed. Aircraft of this type had a tragic record: five crashes, 22 Marines killed the prior year alone. Everybody was very focused on this complex rotor system. We dug into the five crashes. If my memory serves me correctly, there were three unrelated mechanical problems, one maintenance error, and one pilot error. None had much to do with the very complex new concept.
6. Then, watch for details that will get you if you don’t watch out. There are some painful examples of not getting the details right. Remember the Mariner spacecraft that we built for Jet Propulsion Laboratory, one of the finest technical organizations in the world? To our great embarrassment, NASA and we were working in different units–English and metric–and we lost that spacecraft.
7. The next lesson is treasure your anomalies. While the details can hurt, they can also help. In that Mariner mission, there were a number of earlier corrections on the way to Mars that all had the same directional bias, which is peculiar. Had we questioned that, we might possibly have discovered that we were working in different units, but nobody challenged it. They just put in a correction.
8. Then, a lot of redundant systems aren’t redundant. An L1011 with three engines was flying from Miami to Nassau. It had an oil loss warning indicator on one engine, so the pilot turned around to fly back to Miami. On the way back, the second engine gave a no-oil indication, then the third engine. The pilot said there’s no way that you can have three separate engines with totally separate oil systems all lose their oil on the same flight. Well, it turns out there is a way. Just before they had left Miami, the maintenance people had put a new chip detector in each engine. The chip detectors had all come from the same supplier, who had left off an O-ring in the assembly process. Happily, the engines didn’t fail.
9. Finally, complex systems often involve human components. When I was assistant secretary of the Army for research and development, we were producing the Pershing Missile. The engineers realized that somebody could accidentally get two huge cables reversed and that would be very bad. So, they designed one cable bundle with a 16-pin connector and the other with an 18-pin connector. The only problem was the strongest soldier in the United States Army forced a 16-pin connector into an 18-pin connection. A fire followed.
Today, I shared with you nine lessons from my own experience in systems engineering. You might say, why not ten? The reason is because I’m sure that everyone–particularly the old hands in the audience–have one of their own to add.
I recently arrived in Switzerland and faced the problem of notifying my friend when to pick me up from the train station. Had I known Switzerland still has payphones, I would have dropped 50 cents in one and sent an SMS. But imagine landing in a country without payphones, without internet, and with no working SIM card. How could you inform a friend what time you may arrive? A little signal sent on a 2-way FM radio could work. I came up with an even cooler idea. Payment networks are the most prevalent means of communication. You could set up a program that keeps checking your bank account for a transaction. Then, make a cash withdrawal or purchase just before boarding the train. The program could detect the transaction and send an email to indicate which train you’re on.
Back in the nineties, we could send prearranged messages free of charge by using payphones to ring other phones. Or we could make collect calls and encode the phone number to call back using dictionaries of names. Of course, then you could end up placing a collect call as someone unlikely such as “Tanya Rosenberg van Kent”
Here’s a good article on digital audio.
The article explains a lot of things people who like music should know. First, we’ve all seen the “THD” or “total harmonic distortion” specs on all audio equipment. It turns out that harmonic distortion isn’t really an issue and it might even make music sound better. Violins have so many harmonics already, a little harmonic distortion is no big deal. Also, it’s a poor metric because digital devices are more prone to other types of distortion. It’s no surprise that they achieve <0.01% THD or something small like that, without much expense. It reminds me of $50 digital cameras with 14 MP that take awful pictures. Anyway, don’t buy audio gear based on the THD numbers. Now I wonder if maybe the expensive tube amps with poor THD numbers actually do sound better because they avoid the truly bad types of distortion that digital systems can create.
The article also does a great job of explaining quantization distortion and how dither turns it into quantization noise. Also, read about dither or at least look at the pictures so you’re clear on exactly how dither works and how amazing an idea it is:http://en.wikipedia.org/wiki/Dither
The article goes on to discuss 1-bit DSD, which I spoke highly of from a technical standpoint in this previous post. However, I agree with the authors that 24-bit or higher PCM is the best choice for audio processing and 1-bit DSD only makes sense as a final distribution medium. Any mathematical operations on 1-bit DSD are very complex and I don’t really know how the distortion and noise introduced will sound, but PCM is easy to understand.
I recently bought a couple pairs of high-end headphones. As the audio forums suggested, the headphones didn’t sound very good plugged into my computer’s soundcard. My ears felt fatigued after listening, despite low volumes. People on the forums say the poor quality is because sound cards or most digital players don’t have the power to drive large headphones. But they sounded bad even with the volume turned way down. Now, I think it’s more of an issue of quantization noise at low volumes. A D/A converter should receive the raw original bitstream, not scaled down, and then the volume adjustment should be an analog operation. I got a fairly inexpensive HeadRoom BitHead portable headphone amplifier / USB soundcard which makes the headphones sound great!
Also, the more I’ve learned and listened, the more I think that mp3 audio may in fact be of sufficient quality or even indistinguishable from CD, SACD or DVD-Audio. The inadequate sounding mp3′s I’ve heard are probably that way because the original recording or mastering was poor, not a fault of the compression. It is possible to get a crisp, clear, precise sound from an mp3. A bad D/A conversion system probably does more audible harm than the mp3 compression.
I recently interviewed a candidate who listed having worked in an ADC group at Analog Devices. Wanting to ask a technical question on a topic he would/should know, I noticed that I wasn’t sure exactly how an ADC works. So, I thought up three solutions before asking the question, but still wondered how ADC’s work fast enough for applications like audio. I’m familiar with the ramp and counter method used in microcontrollers from having used them. The successive approximation method is also a straightforward and fairly good solution, but even that takes 24 cycles for audio quality ADC. Next, I reinvented multistage subranging ADC as a way to expand flash ADC. While trying to figure out what my invention was called, I stumbled across another really cool ADC that I would never have thought of. The operation of most ADC’s can be explained to anyone because they don’t require special knowledge to understand, but delta-sigma ADC’s are completely based on frequency-domain thinking. Learning about them helped improve my understanding of DSP and control theory.
Here’s a good discussion of various ADC’s:
Read all 3 parts of this post. Part 1 helps understand what the modulator does. Part 2 makes Nyquist simpler than ever before. Part 3 gives insight into how the noise shaping really works.
I like this one because it explains how you extract 16 bits of width out of just 64 samples.
And once you understand all that, you’ll appreciate this piece on upsampling in CD players
but don’t buy one because it’s all overpriced. Advertising the $1.25 DAC http://www.ti.com/product/