Until PNG is more widely accepted in browsers, gif will have its place.
PNG is fully supported in every major browser (Mozilla, Konqueror, Opera) except IE, which supports PNGs that have all the features of GIfs (ie, no alpha).
PNG is widely accepted. PNG is technically superior to GIF, and a good way to persuade Microsoft to implement alpha transparency in IE is to use PNGs exclusively.
Re:An honest question - who cares?
on
AAC Put To The Test
·
· Score: 3, Informative
There *are* lossless codecs like FLAC and SHN, but they generally achieve between 10 - 30% compression.
Actually, the compression ratio for SHN is much better. As much as 74% compression can be achieved on techno and pop. I would call 55% typical for live shows from etree.org.
FLAC has similiar compression rates. FLAC's strengths lie in its ability to compress 24bit audio and built-in checksums.
Actually, the policy at Vanderbilt is 1GB down,.5GB up per day. If you exceed those limits, you get capped to 64kbps each way. This policy seems to be effective in limiting abuse of bandwidth, but still allowing legitimate uses.
I've only heard stories of service being revoked in cases of copyright infringement.
I found plans for a power supply here. For +5V, it uses the MAX787 and for +3.3V, the MAX788, two efficient switching power supply ics. That should achieve 80% effeciency or so, and supply 5A of power for each.
For +12 and -12, it uses LM2587, another switching regulator, then steps the -12 down to -5. This seems like overkill to me; the efficiency gained isn't worth the extra parts.
For +12, I would use an LM1084 (powered from the car battery), which doesn't have the effeciency of a switching regulator, but doesn't require any inductors.
For -5 and -12, I would use 2 ICL662s (powered from +5 and +12, respectively), which will only supply 300mW each, but that's all we need for negative supplies.
Most MP3 compression utilities filter out frequencies above 20kHz or below 20Hz; CD-quality sound can produce frequencies at 44kHz and the inverse of the length of the sound. Clearly, the data filtered out is not tiny.
It is true that most low-grade speakers and headphones cannot produce frequencies above 20kHz or below 20Hz (check the frequency response), but higher grade equipment can. This is often the reason somebody can't tell the difference between CD sound and MP3.
Psychoaucustics models say that few people can hear frequencies, by themselves, above 20kHz. Suppose that you can't. However, multiple frequencies above 20kHz can produce frequencies below 20kHz. Try playing a root and a fourth on a very well-tuned string instrument. It should resonate and octave below the root. Playing a root and an octave above it also produces and octave below the root. Various other combinations produce different tones.
The frequencies above 20kHz do matter, whether you can hear them or not. These high frequencies often contribute to tone; differences in ambouchere, reed, mouthpieces, instruments for reed players, differences in strings, where the string is played, plucking/bowing style for string players.
Another way to think about it: try to produce a square wave with only sine waves. To reproduce it exactly, you need the wave that has the same period, then period/3, then period/5, then period/7, etc. When you take away the high frequencies, the wave isn't the same.
Higher bit rate MP3s are not CD quality sound. MP3 still distorts the original, and many find that objectionable when hearing it from quality equipment.
For actual CD-quality sound, I suggest a lossless algorithm such as shorten (.shn).
There should be a geographical notation TLD - sort of a business.city.state/province.country
People have pointed out that this exists (.us) and gets little use, because it is largely irrelevant. Who cares where amazon.com or yahoo.com is located? And with the.mtl tld, why wouldn't we have problems the same as with.com?
We need to have a business type TLD system - so that Bob's Car Market and Bob's soul kitchen and Bob's spandex emporium all can have a reasonable web address. So businessname.businesstype.country might work for them.
How do you determine businesstype? Coca-cola could be considered to be in the soft drink business, the refreshment business, pop, soda, soda-pop, etc. A very large number of business types could be thought of that describe this company. This puts a large burden on those registering the synonyms, those verifying the truthfulness, and the name servers. 20 different addresses instead of 1 in the current system would require beefier hardware for the master servers.
Growth isn't provided for in this model either. Suppose coke decided to can foods. Many other synonyms would have to be registered. This is also a flaw in the current system. Slashdot started out as more of an.org; nonprofit, no employees. It then morphed into a.com; for profit, a few employees. Should it be exclusively a.com? It would be a huge hassle to have everyone switch, but it shouldn't be a.org.
Machines on "always-on" connections and not behind firewalls are of course the most vulnerable...
This is misleading. Machines on always-on connections are no more vulnerable than machines on dial-up. However, these machines are more likely to be attacked because they always have a network connection.
This confuses some people. My grandmother thought if she got cable Internet access, her computer could suddenly be compromised easily because the cable Internet made her computer insecure. It wouldn't be any less secure; it would just have a higher chance of being attacked.
This is not really what we need, this is really what spammers need. What would prevent a company from picking up a few dozen of these at their local Walmart with the intent to spam? They don't have anything with which to identify the spammers.
Or is this a neccessary evil for complete anonymous access? Would you rather get a few dozen spams daily or have non-anonymous Internet access?
Slashdot Mirror
They might even learn how to spell!
repost from October 3 2003, by LesPaul75.
Except that climbing a mountain doesn't cost an additional $8 billion of the public's money over the next 5 years.
Try the Pragma: no-cache and/or Cache-control HTTP headers.
PNG is fully supported in every major browser (Mozilla, Konqueror, Opera) except IE, which supports PNGs that have all the features of GIfs (ie, no alpha).
PNG is widely accepted. PNG is technically superior to GIF, and a good way to persuade Microsoft to implement alpha transparency in IE is to use PNGs exclusively.
Actually, the compression ratio for SHN is much better. As much as 74% compression can be achieved on techno and pop. I would call 55% typical for live shows from etree.org.
FLAC has similiar compression rates. FLAC's strengths lie in its ability to compress 24bit audio and built-in checksums.
Actually, the policy at Vanderbilt is 1GB down, .5GB up per day. If you exceed those limits, you get capped to 64kbps each way. This policy seems to be effective in limiting abuse of bandwidth, but still allowing legitimate uses.
I've only heard stories of service being revoked in cases of copyright infringement.
For linux, check out lm_sensors. It monitors fan speed, temperature, and voltages via the i2c bus or isa.
It works pretty well. I've noticed that my pus's +/-12 goes out of bounds often, but neither seems to affcect stability.
Have fun, but I suspect that the builtin sensors don't update often enough to detect a dirty supply.
For +12 and -12, it uses LM2587, another switching regulator, then steps the -12 down to -5. This seems like overkill to me; the efficiency gained isn't worth the extra parts.
For +12, I would use an LM1084 (powered from the car battery), which doesn't have the effeciency of a switching regulator, but doesn't require any inductors.
For -5 and -12, I would use 2 ICL662s (powered from +5 and +12, respectively), which will only supply 300mW each, but that's all we need for negative supplies.
The "typical application"s of the ICL662 and LM1084 should work fine; check out the datasheets.
National Semiconductor (LM ics)
Maxim-IC (MAX and ICL ics)
(BTW, last time I checked, coilcraft doesn't give away samples for the inductors that the website references)
Most MP3 compression utilities filter out frequencies above 20kHz or below 20Hz; CD-quality sound can produce frequencies at 44kHz and the inverse of the length of the sound. Clearly, the data filtered out is not tiny.
It is true that most low-grade speakers and headphones cannot produce frequencies above 20kHz or below 20Hz (check the frequency response), but higher grade equipment can. This is often the reason somebody can't tell the difference between CD sound and MP3.
Psychoaucustics models say that few people can hear frequencies, by themselves, above 20kHz. Suppose that you can't. However, multiple frequencies above 20kHz can produce frequencies below 20kHz. Try playing a root and a fourth on a very well-tuned string instrument. It should resonate and octave below the root. Playing a root and an octave above it also produces and octave below the root. Various other combinations produce different tones.
The frequencies above 20kHz do matter, whether you can hear them or not. These high frequencies often contribute to tone; differences in ambouchere, reed, mouthpieces, instruments for reed players, differences in strings, where the string is played, plucking/bowing style for string players.
Another way to think about it: try to produce a square wave with only sine waves. To reproduce it exactly, you need the wave that has the same period, then period/3, then period/5, then period/7, etc. When you take away the high frequencies, the wave isn't the same.
Higher bit rate MP3s are not CD quality sound. MP3 still distorts the original, and many find that objectionable when hearing it from quality equipment.
For actual CD-quality sound, I suggest a lossless algorithm such as shorten (.shn).
People have pointed out that this exists (.us) and gets little use, because it is largely irrelevant. Who cares where amazon.com or yahoo.com is located? And with the .mtl tld, why wouldn't we have problems the same as with .com?
We need to have a business type TLD system - so that Bob's Car Market and Bob's soul kitchen and Bob's spandex emporium all can have a reasonable web address. So businessname.businesstype.country might work for them.
How do you determine businesstype? Coca-cola could be considered to be in the soft drink business, the refreshment business, pop, soda, soda-pop, etc. A very large number of business types could be thought of that describe this company. This puts a large burden on those registering the synonyms, those verifying the truthfulness, and the name servers. 20 different addresses instead of 1 in the current system would require beefier hardware for the master servers.
Growth isn't provided for in this model either. Suppose coke decided to can foods. Many other synonyms would have to be registered. This is also a flaw in the current system. Slashdot started out as more of an .org; nonprofit, no employees. It then morphed into a .com; for profit, a few employees. Should it be exclusively a .com? It would be a huge hassle to have everyone switch, but it shouldn't be a .org.
This is misleading. Machines on always-on connections are no more vulnerable than machines on dial-up. However, these machines are more likely to be attacked because they always have a network connection.
This confuses some people. My grandmother thought if she got cable Internet access, her computer could suddenly be compromised easily because the cable Internet made her computer insecure. It wouldn't be any less secure; it would just have a higher chance of being attacked.
This is not really what we need, this is really what spammers need. What would prevent a company from picking up a few dozen of these at their local Walmart with the intent to spam? They don't have anything with which to identify the spammers.
Or is this a neccessary evil for complete anonymous access? Would you rather get a few dozen spams daily or have non-anonymous Internet access?