Oh, and don't forget the wig, a la Andre Rieu. Sesame Street has a great Rieu-esque puppet during the "Number of the Day" segment (you understand if you have a toddler).
Brian May's (of Queen) guitar was made by him and his dad. The neck is oak, I believe, and it was from an old mantelpiece. IIRC, May has said that it was worm-eaten, so there are channels in the wood.
No one yet has duplicated that Red Special sound. Most folks in rock music either have a Fender (Tele/Strat) thing going or a Les Paul sound. Paul Reed Smith (Santana, Mark Tremonti, others) sort of splits the difference.
Nope. Typical vocoders for VoIP (like G.729 and G.723.1) are really pretty bad at encoding music. They're not nearly as bad as the first LPC vocoders, which assum e a monophonic (but harmonically complex) signal, but they still are optimized for speech, not music.
Want to know how it sounds? Just listen to background music on your cell. Yuck.
I'm assuming that VBR implies some sort of compressed vocoder other than G.711 or G.722, which are not variable packet size.
I may be wrong, though. One could do G.711 with silence detection and during silence periods, send noise description packets that are much smaller than the voiced packets. However, that would mean just two packet sizes, really.
Isn't this how digital watermarking is done, a la Digimarc? I believe it is. They call them "covert digital watermarks." It is for copyright protection, in their application, but any data could be embedded as noise in the same way.
The drawback with digital watermarking for copyright protection is that if someone alters the file's size or resolution or JPG compression, the watermark is at risk for being corrupted.
With secret data in one's own pictures that are on one's HDD, then that's not an issue.
Hmmm. JPGs have EXIF metadata attached. What if there were a slick way to have data encrypted and spread out among the EXIF headers of multiple JPG files? There would have to be some sort of key and map for the data, but it wouldn't show up as normal plaintext.
There is a function that I use a lot in my work, namely, the Fast Fourier Transform. Back before digital computers, someone came up with a method for computing the Fourier Series for a set number of harmonics (frequency and amplitude of the waveform were normalized). One cycle of the waveform was drawn at a given scale, and a stylus was used to trace the waveform. While the stylus moved, various gears in the machine turned at various rates, giving the harmonic content in the waveform by how much each gear moved. Very ingenious for the day.
It is quite possible that something like this could be implemented in nanotechnology, and would be able to generate spectral content pretty rapidly. I'd be interested in seeing how specialized mechanical processors of the past might be reimplemented in micro for current computing problems.
... but not for engineering.
C may be dying for the computing world at large, what with so much OOP stuff out there. But for embedded systems, there are TONS of legacy and even new pieces of code running in C. It seems that for a lot of small-run processors (which I typically work on-- DSPs, mostly), there aren't a) a lot of people writing C++ compilers for the silicon and b) there aren't a lot of developers crying out for object-oriented solutions.
We're quite content with banging out good old C, and of course, for the occasional tight, oft-used loop, it's optimized via assembly. We embedded programmers don't typically want a lot of overhead and data abstraction and such in our code; I've done some C++, and I haven't minded it, but we need visibility into the nuts and bolts of what's going on down on the metal, and C++ just adds a lot of stuff that gets in the way. And yes, I know that C++ can be made to be very C-like, but why bother?
Every processor that I've used that is using a high-level compiler at all is using C. I'll have good job security for the forseeable future knowing how to program DSPs using C and assembly.
Good comments, DG. I agree with a lot of what you have said. Yes, there is a legacy business model that is based on reprints. However, no one wants to take a pay cut.:) There must be a price paid for lost revenue if I were to regard my business as a service, and not a content provider.
There is a good deal of discussion of this issue among photogaphers (I know we're getting a little OT, but there are some analogies with the recording/film industry). Many disdain photogs who give away negatives or files. It does cheapen the industry as a whole. Why, you may ask? Even though you mention that printing with a negative was/is an art form that needs skill, just because a digital file is copiable without degradation doesn't mean that anyone can make a beautiful print from every out-of-camera image.
I spend about 20-40 hours (yes, hours) working with the images from a wedding. The last wedding that I did started with well over 1000 images and I ended up with about 900 final images. I massage exposure, color, apply soft effects, etc. It is a lot of work. If I were to just give images away, I would have NO motivation to spend that much time working with the images-- I'd just give them raw, and then, they'd be of lower quality. Plus, I don't want a client posting (online or in a print for others to see) an image that they say that I took, when I haven't had control over the printing process. What if it's an awful, low saturation, green-cast print from Walgreens? I have a reputation to uphold. There are some considerations that go with my business that make it a disadvantage to merely see it as a service and not control the final product-- word of mouth and repeat business depend on great PAPER prints. Brides still buy albums, and families still buy 8x10s and 11x14s for their stairways.
In short, only those photogs who take PERFECT images out of the camera are going to be the ones giving out unprocessed images, and they are going to be the ones charging a premium for their lost revenue. Either that, or the unskilled will simply give out their junk, lowering the quality. Higher prices, lower quality-- you pick.
As a semi-pro photographer, I do see the value in protecting one's IP (Intellectual Property, not Internet Protocol, in case any were confused). The images, media, whatever, are the keys to the bank. My clients frequently want me to give them the digital files, as if they're "free," when, if I do that, I make $0 on prints, which is where much of the money is to be made in photography (in the present biz model).
Analogizing to HD-DVD (or whatever digital media there is), the companies who produce such media have the right to protect their content. The rub lies in them protecting their content without interfering with our fair use. That's a really hard thing to implement/enforce. I don't really have any ideas on how it should be done, but I'm simply emphasizing that they have created the content, and thus, they have a LEGAL right to protect it. Just because it's in a digital format, that doesn't mean it's now free (either as in "free beer" or "free speech"). The content still belongs to the original creators, though one owns a copy.
The folks bitching about "free speech" and publishing encryption keys are way off, IMHO. If I had watermarked thumbnail images on my website, and full res images accessible by password, then someone cracked that password and published on the web, I'd be upset. They have stolen the keys to the bank, then made copies of the keys for anyone to find. It's stealing, immoral, etc. There's nothing admirable about stealing and aiding and abetting others to as well. I know it's in our nature to want to get around the system (esp. as geeks), but it just ain't right.
Moore's Law is a law like the Law of Gravity. Newton's formulation of the LoG didn't cause gravity to start operating. Newton simply made a mathematical expression of observed reality.
Designers aren't sitting around making sure their gate counts conform to Moore's Law. They design the fastest, highest-performance silicon they can.
I get really PO'd at references to Moore's Law being a determiner of things as opposed to an observer.
There are three things that I would add (I am a semi-pro photographer who uses a dSLR).
First, the megapixel race has really distorted the idea of image quality in the pop mindset. Many think that the more sensor sites a camera has, the better images it will produce. WRONG. With the dSLRs, the sensors are generally quite a bit larger than the P&S and camera phones. Larger sensors means more photons per sensor, and thus, less measurement noise. A 6-megapixel dSLR vs. a 6-megapixel P&S competition would almost certainly show the dSLR blows away the compact in noise performance (assuming a larger sensor on the dSLR). Camera phones probably have much smaller sensors even than dedicated P&S cams -> higher noise.
Second, the response time of many camera phones is horrific. I have an LG vx9900 (enV) with a 2MP camera, and while it takes OK pictures, I can't get a good shot of my daughter to save my life. I really need Nicholas Cage to see about 2 seconds into the future in order to press the shutter at the right moment. My Nikon D70s is well nigh instantaneous. I rely on extremely fast
response in my portrait and wedding business. Those split seconds make the difference between a great expression and a so-so one.
Third, one of the things many people love about high-quality portraits is selective focus, or narrow depth-of-field. I use this outdoors to get my subject in focus and the background blurry. With my 80-200mm f/2.8 lens (cost twice as much as my camera body), I get incredibly soft background blur that is very pleasing. There is no way that a camera phone with a focal length of a few millimeters is ever going to achieve that, since such a short focal length lens will have huge depth of field. One needs a telephoto lens, and the faster (lower f number), the better. This means $$, generally. The fast telephotos-- the extreme example being the sports sideline shooters with their huge glass on monopods, their cameras hanging off their lenses-- are pretty pricey. But for many portraits, a fast telephoto is invaluable.
For me, I can't imagine a camera phone ever approaching the capabilities I need from a camera. It's the physics of the thing. The orig article is obviously written by someone who is not really a serious photographer. Like someone mentioned-- white balance. Duh. Digital photography 101.
Slashdot Mirror
As Jack Handey rightly said, "If you drop your keys into a river of molten lava, forget 'em, 'cause man, they're gone."
Forgot the Soviet Russia meme! Nice pile-on.
Imagine a Beowulf Cluster of /. memes, you insensitive clod!
Perhaps, after he decoded the digitised data, he accidentally crashed his red-coloured lorry into the rear garden?
I'm so gruntled, that I would never do such a thing.
If you're at a restaurant and your breakfast order is announced to the entire establishment, you might be a redneck.
Long live the Waffle House. Praise the Lord for that Eternal Yellow Glow.
If vegetarians eat vegetables, what do humanitarians eat?
Oh, and don't forget the wig, a la Andre Rieu. Sesame Street has a great Rieu-esque puppet during the "Number of the Day" segment (you understand if you have a toddler).
Brian May's (of Queen) guitar was made by him and his dad. The neck is oak, I believe, and it was from an old mantelpiece. IIRC, May has said that it was worm-eaten, so there are channels in the wood.
No one yet has duplicated that Red Special sound. Most folks in rock music either have a Fender (Tele/Strat) thing going or a Les Paul sound. Paul Reed Smith (Santana, Mark Tremonti, others) sort of splits the difference.
But Brian May is in a class all by himself.
Imagine a Beowulf cluster of... Nevermind.
Nope. Typical vocoders for VoIP (like G.729 and G.723.1) are really pretty bad at encoding music. They're not nearly as bad as the first LPC vocoders, which assum e a monophonic (but harmonically complex) signal, but they still are optimized for speech, not music.
Want to know how it sounds? Just listen to background music on your cell. Yuck.
I'm assuming that VBR implies some sort of compressed vocoder other than G.711 or G.722, which are not variable packet size.
I may be wrong, though. One could do G.711 with silence detection and during silence periods, send noise description packets that are much smaller than the voiced packets. However, that would mean just two packet sizes, really.
...my brain returns more and more 404 pages.
Mmmmmmm. Bacon. *AGLLLL*....
Grranimals. That's what the average Joe needs: "Hmm. Crysis has a picture of a hippo on it...this video card also has a hippo. Best Buy, meet Visa."
Isn't this how digital watermarking is done, a la Digimarc? I believe it is. They call them "covert digital watermarks." It is for copyright protection, in their application, but any data could be embedded as noise in the same way. The drawback with digital watermarking for copyright protection is that if someone alters the file's size or resolution or JPG compression, the watermark is at risk for being corrupted. With secret data in one's own pictures that are on one's HDD, then that's not an issue.
Hmmm. JPGs have EXIF metadata attached. What if there were a slick way to have data encrypted and spread out among the EXIF headers of multiple JPG files? There would have to be some sort of key and map for the data, but it wouldn't show up as normal plaintext.
Kenny Rogers put his order in for one.
..but all its sphincter muscles relax and contents of its bowels and bladder spew forth...
You mean, if Windows dies, they'll release the source?Did you try Mountain Dew instead of Coke?
There is a function that I use a lot in my work, namely, the Fast Fourier Transform. Back before digital computers, someone came up with a method for computing the Fourier Series for a set number of harmonics (frequency and amplitude of the waveform were normalized). One cycle of the waveform was drawn at a given scale, and a stylus was used to trace the waveform. While the stylus moved, various gears in the machine turned at various rates, giving the harmonic content in the waveform by how much each gear moved. Very ingenious for the day.
It is quite possible that something like this could be implemented in nanotechnology, and would be able to generate spectral content pretty rapidly. I'd be interested in seeing how specialized mechanical processors of the past might be reimplemented in micro for current computing problems.
... but not for engineering. C may be dying for the computing world at large, what with so much OOP stuff out there. But for embedded systems, there are TONS of legacy and even new pieces of code running in C. It seems that for a lot of small-run processors (which I typically work on-- DSPs, mostly), there aren't a) a lot of people writing C++ compilers for the silicon and b) there aren't a lot of developers crying out for object-oriented solutions.
We're quite content with banging out good old C, and of course, for the occasional tight, oft-used loop, it's optimized via assembly. We embedded programmers don't typically want a lot of overhead and data abstraction and such in our code; I've done some C++, and I haven't minded it, but we need visibility into the nuts and bolts of what's going on down on the metal, and C++ just adds a lot of stuff that gets in the way. And yes, I know that C++ can be made to be very C-like, but why bother?
Every processor that I've used that is using a high-level compiler at all is using C. I'll have good job security for the forseeable future knowing how to program DSPs using C and assembly.
Good comments, DG. I agree with a lot of what you have said. Yes, there is a legacy business model that is based on reprints. However, no one wants to take a pay cut. :) There must be a price paid for lost revenue if I were to regard my business as a service, and not a content provider.
There is a good deal of discussion of this issue among photogaphers (I know we're getting a little OT, but there are some analogies with the recording/film industry). Many disdain photogs who give away negatives or files. It does cheapen the industry as a whole. Why, you may ask? Even though you mention that printing with a negative was/is an art form that needs skill, just because a digital file is copiable without degradation doesn't mean that anyone can make a beautiful print from every out-of-camera image.
I spend about 20-40 hours (yes, hours) working with the images from a wedding. The last wedding that I did started with well over 1000 images and I ended up with about 900 final images. I massage exposure, color, apply soft effects, etc. It is a lot of work. If I were to just give images away, I would have NO motivation to spend that much time working with the images-- I'd just give them raw, and then, they'd be of lower quality. Plus, I don't want a client posting (online or in a print for others to see) an image that they say that I took, when I haven't had control over the printing process. What if it's an awful, low saturation, green-cast print from Walgreens? I have a reputation to uphold. There are some considerations that go with my business that make it a disadvantage to merely see it as a service and not control the final product-- word of mouth and repeat business depend on great PAPER prints. Brides still buy albums, and families still buy 8x10s and 11x14s for their stairways.
In short, only those photogs who take PERFECT images out of the camera are going to be the ones giving out unprocessed images, and they are going to be the ones charging a premium for their lost revenue. Either that, or the unskilled will simply give out their junk, lowering the quality. Higher prices, lower quality-- you pick.
As a semi-pro photographer, I do see the value in protecting one's IP (Intellectual Property, not Internet Protocol, in case any were confused). The images, media, whatever, are the keys to the bank. My clients frequently want me to give them the digital files, as if they're "free," when, if I do that, I make $0 on prints, which is where much of the money is to be made in photography (in the present biz model).
Analogizing to HD-DVD (or whatever digital media there is), the companies who produce such media have the right to protect their content. The rub lies in them protecting their content without interfering with our fair use. That's a really hard thing to implement/enforce. I don't really have any ideas on how it should be done, but I'm simply emphasizing that they have created the content, and thus, they have a LEGAL right to protect it. Just because it's in a digital format, that doesn't mean it's now free (either as in "free beer" or "free speech"). The content still belongs to the original creators, though one owns a copy.
The folks bitching about "free speech" and publishing encryption keys are way off, IMHO. If I had watermarked thumbnail images on my website, and full res images accessible by password, then someone cracked that password and published on the web, I'd be upset. They have stolen the keys to the bank, then made copies of the keys for anyone to find. It's stealing, immoral, etc. There's nothing admirable about stealing and aiding and abetting others to as well. I know it's in our nature to want to get around the system (esp. as geeks), but it just ain't right.
Moore's Law is a law like the Law of Gravity. Newton's formulation of the LoG didn't cause gravity to start operating. Newton simply made a mathematical expression of observed reality.
Designers aren't sitting around making sure their gate counts conform to Moore's Law. They design the fastest, highest-performance silicon they can.
I get really PO'd at references to Moore's Law being a determiner of things as opposed to an observer.
There are three things that I would add (I am a semi-pro photographer who uses a dSLR).
First, the megapixel race has really distorted the idea of image quality in the pop mindset. Many think that the more sensor sites a camera has, the better images it will produce. WRONG. With the dSLRs, the sensors are generally quite a bit larger than the P&S and camera phones. Larger sensors means more photons per sensor, and thus, less measurement noise. A 6-megapixel dSLR vs. a 6-megapixel P&S competition would almost certainly show the dSLR blows away the compact in noise performance (assuming a larger sensor on the dSLR). Camera phones probably have much smaller sensors even than dedicated P&S cams -> higher noise.
Second, the response time of many camera phones is horrific. I have an LG vx9900 (enV) with a 2MP camera, and while it takes OK pictures, I can't get a good shot of my daughter to save my life. I really need Nicholas Cage to see about 2 seconds into the future in order to press the shutter at the right moment. My Nikon D70s is well nigh instantaneous. I rely on extremely fast response in my portrait and wedding business. Those split seconds make the difference between a great expression and a so-so one.
Third, one of the things many people love about high-quality portraits is selective focus, or narrow depth-of-field. I use this outdoors to get my subject in focus and the background blurry. With my 80-200mm f/2.8 lens (cost twice as much as my camera body), I get incredibly soft background blur that is very pleasing. There is no way that a camera phone with a focal length of a few millimeters is ever going to achieve that, since such a short focal length lens will have huge depth of field. One needs a telephoto lens, and the faster (lower f number), the better. This means $$, generally. The fast telephotos-- the extreme example being the sports sideline shooters with their huge glass on monopods, their cameras hanging off their lenses-- are pretty pricey. But for many portraits, a fast telephoto is invaluable.
For me, I can't imagine a camera phone ever approaching the capabilities I need from a camera. It's the physics of the thing. The orig article is obviously written by someone who is not really a serious photographer. Like someone mentioned-- white balance. Duh. Digital photography 101.