All you never wanted to know about cable modems
on
Modem Accelerators?
·
· Score: 2
There are cable modem standards called DOCSIS. Not all modems
follow these standards. For the modulation demodulation part have a look at the
RF specification starting at section 4 on page 23. Upstream uses QPSK or 16QAM, downstream uses 64QAM or 256QAM. I believe that for a given number of homes there will be one downstream signal, but multiple upstream signals. (the number of upstreams would probably not equal the number of homes.)
For the hardcore RF geeks these specs are a great read. You can see how these specs were designed as opposed to "happened."
The European CE mark is quite a bit more strict then any of the FCC's guidelines. Look for CE the next time you buy anything electrical. It is an indication of a better product.
Rogers has been doing some stupid stuff with their dhcp servers. I had been using dhcpclient (2.2.x) and for a while was restarting the client every hour. Now I am on DSL. The last straw for me was the email forwarding problem. Other rogers's users could not send mail to my email address which forwarded back to my rogers account.
Bike Friday: bikes that fold.
on
Personal Transport?
·
· Score: 3, Informative
Just to add to the previous poster: Bike Friday bikes are not folding bikes. They are bikes that fold. My roommate of a few years ago had one. He loved it. He was a serious bicyclist and would do 150+ km on a sunday ride on his Bike Friday. Not to mention his trips all over the world. Bike Friday bikes are the best for the international traveller. The bike folds down to fit into one suitcase which becomes a bike trailer.
This dosen't help the SF commuter though. Bike Friday bikes don't exactly fold down to fit in your back pocket, and you can't take bikes on BART.
Yes. What you are suggesting here is the correct way to do it. (or atleast one of the correct ways) You won't find the MAX chips in your run of the mill power supply though. What I thought the top poster was suggesting was to cannibalize a regular AC to DC power supply and hook the 12v from the car straight into it.
PSK31 is not better then packet. It is different. It was designed to get 50wpm in 31Hz. The authors don't consider PSK31 a digital mode. Think of PSK31 as an alternative to morse.
The real kicker is that the 8259 is long gone, but modern chipsets emulate cascaded 8259s. The other joke is the A20 address line. All because of two decade old legacy software.
I belive that a powersupply in exsiting pc can be modifyed to take 12vDC by bypassing the 1st stage of the powersupply
You probably don't want to do that. First off you probably can't as the transformer probably supplies 12v, 5v, 0v, -5v, -12v not just 12v.
Also the transformer plus rectifiers also filter the incomming power. In theory you could use voltage regulator to take 12v and drop it down to 5v. But, you now have a whole lot of power to dissiplate through the voltage regulator.
From the official psk31 web page there is a link to psk31 software. At the bottom of that page it mentions a couple of harward implementations of PSK31. I would think that you could hook one of these up to any type of serial terminal (say Dos + telix) and you would be good to go. As mentioned before some old laptops make good serial terminals. Be carefull though. Some of those old laptops have awfull battery times. And just about any laptop that is over 5 years old will need a new battery. There is probably a battery shop in your yellow pages that will rebuild batteries for you.
It depends on the NIC and its driver. Some NIC's have always allowed the MAC to be changed on the card. On others it couldn't be done and required a kernel hack.
two words: Ford Pinto. Ford was harmed for years after the public lost trust in Ford due to the Pinto.
For those who might not know. The Ford Pinto was a cheap poorly designed car of the 70s that had a nasty tendancy to burst into flames when struck from behind in minor collisions. Ford execs knew of this problem but decided against a recall as the cost was greater then the cost of a few lawsuits. see the internal Frod memo and more information
In her novel, Tea from an Empty Cup, Pat Cadigan predicted a world with 2 Internets. One was 100% accountable. It was the main network used for real bussiness. There was no annonymity. The second network was designed to allow for anonymity. It was an "any thing goes" network where spoofing was the rule not the exception. I would like to see these networks. When I need to get work done I would use the accountable network. When I want to view pr0n I would use the other network. I think having two distinct networks like this would be a good compromise for the privacy advocats, and those tired of DOS attacks.
Ofcourse there are a *few* (as in many) technical difficulties to resolve first.
It is possible that you could tell the difference between 16bit 44.1ks/s vs 24bit 48ks/s. With 16bit sound the absolute best signal to noise ratio (SNR) you can get is about 96dB. With 24bit you can get a theoretical SNR of 144dB. The source SNR will dominate the SNR of the ouput, so assumeing good shielding the 24bit sound should be quite a bit better then 16bit.
Now there are some limitations. Most stereo gear is designed for CD sound (16bit) and won't do better then 96dB. The max theoretical SNR of an amplifier is about 130-140dB (can't remember exact value). I believe that is a limitation imposed by physics and probably has something to do with the charge of an electron. So while 24bit may seem like overkill it is probably a standard that will survive longer then any of us.
Regarding 44.1ks/s vs 48ks/s. Higher is always better. Nyquist's Theorem states something along the lines of: You can reproduce any signal if your sampleing rate is atleast twice as high as the highest frequency in the signal. So 44.1ks/s should be able to reproduce 22.1kHz, but it can't. Nyquist assumes an infinate sequence of samples which is clearly impossible. With more samples you can also do better digital signal processing. There are just more samples to work with.
To tie this altogether we have to consider one other reality: Electronic gear produces 3rd order harmonic noise. 3rd order harmonics suck. They sound bad, sometimes are even painfull. It results in a lot of higher frequency noise. So we have to compensate for this with better then required sample rates and sample bits. This is why the next CD standard will probably be 24bit and 96ks/s.
In general you can do this under *nix with #!at the top of the file. So if you had a *nix command/usr/bin/rpn your example would translate into:
#!/usr/bin/rpn
b %1
b %2
a
s
Just about every *nix command that accepts a file containing commands as an argument will work this way.
I never did like REXX back when I was using OS/2 and I don't know why. It looks like REXX is an easy language to program with. And really I should have jumped at the chance to use a "real" scripting language versus basic and those hideous dos batch files.
In my post I was not refering to NCs. I was refering to *nix boxes without a disk running programs locally and storeing all data and programs on a server. Running *nix and dos+netware like this was quite common circa 1990. It also worked quite well, except when the network went down. All too common with thinnet (coax).
Search Google for "linux diskless howto" and you will get several links. Most *nixes can net boot. Sun hardware supports netboot out of the box to the point that netboot is the default if no hard disk is installed. Even lowly dos can netboot with the help of netware. It is old school. In the early nineties we figured out that netbooting sounded a lot cooler then it really is.
Your OS/2 comment is bang on. Technically OS/2 beat win95 in every way (back in 92/93). But IBM charged money for the dev kits. MS gave their's away (atleast for the first little bit). IBM failed to get hardware vendors onboard. MS paid hardware vendors to write drivers for win95. IBM ingnored the home market. MS put the home market square in their marketing sights. IBM's OS division couldn't even convince IBM's pc division to ship OS/2.
OS/2 2.0 had the jump on win94 (ninety four) by a good year. When win95 finally shipped OS/2 had had two years to catch the market but did nothing.
Slashdot Mirror
There are cable modem standards called DOCSIS. Not all modems follow these standards. For the modulation demodulation part have a look at the RF specification starting at section 4 on page 23. Upstream uses QPSK or 16QAM, downstream uses 64QAM or 256QAM. I believe that for a given number of homes there will be one downstream signal, but multiple upstream signals. (the number of upstreams would probably not equal the number of homes.)
For the hardcore RF geeks these specs are a great read. You can see how these specs were designed as opposed to "happened."
The European CE mark is quite a bit more strict then any of the FCC's guidelines. Look for CE the next time you buy anything electrical. It is an indication of a better product.
Rogers has been doing some stupid stuff with their dhcp servers. I had been using dhcpclient (2.2.x) and for a while was restarting the client every hour. Now I am on DSL. The last straw for me was the email forwarding problem. Other rogers's users could not send mail to my email address which forwarded back to my rogers account.
Just to add to the previous poster:
Bike Friday bikes are not folding bikes. They are bikes that fold. My roommate of a few years ago had one. He loved it. He was a serious bicyclist and would do 150+ km on a sunday ride on his Bike Friday. Not to mention his trips all over the world. Bike Friday bikes are the best for the international traveller. The bike folds down to fit into one suitcase which becomes a bike trailer.
This dosen't help the SF commuter though. Bike Friday bikes don't exactly fold down to fit in your back pocket, and you can't take bikes on BART.
Yes. What you are suggesting here is the correct way to do it. (or atleast one of the correct ways) You won't find the MAX chips in your run of the mill power supply though. What I thought the top poster was suggesting was to cannibalize a regular AC to DC power supply and hook the 12v from the car straight into it.
Thanks for the tip on the Maxim parts.
PSK31 is not better then packet. It is different. It was designed to get 50wpm in 31Hz. The authors don't consider PSK31 a digital mode. Think of PSK31 as an alternative to morse.
The real kicker is that the 8259 is long gone, but modern chipsets emulate cascaded 8259s. The other joke is the A20 address line. All because of two decade old legacy software.
I belive that a powersupply in exsiting pc can be modifyed to take 12vDC by bypassing the 1st stage of the powersupply
You probably don't want to do that. First off you probably can't as the transformer probably supplies 12v, 5v, 0v, -5v, -12v not just 12v. Also the transformer plus rectifiers also filter the incomming power. In theory you could use voltage regulator to take 12v and drop it down to 5v. But, you now have a whole lot of power to dissiplate through the voltage regulator.
From the official psk31 web page there is a link to psk31 software. At the bottom of that page it mentions a couple of harward implementations of PSK31. I would think that you could hook one of these up to any type of serial terminal (say Dos + telix) and you would be good to go. As mentioned before some old laptops make good serial terminals. Be carefull though. Some of those old laptops have awfull battery times. And just about any laptop that is over 5 years old will need a new battery. There is probably a battery shop in your yellow pages that will rebuild batteries for you.
It depends on the NIC and its driver. Some NIC's have always allowed the MAC to be changed on the card. On others it couldn't be done and required a kernel hack.
two words: Ford Pinto. Ford was harmed for years after the public lost trust in Ford due to the Pinto.
For those who might not know. The Ford Pinto was a cheap poorly designed car of the 70s that had a nasty tendancy to burst into flames when struck from behind in minor collisions. Ford execs knew of this problem but decided against a recall as the cost was greater then the cost of a few lawsuits. see the internal Frod memo and more information
In her novel, Tea from an Empty Cup, Pat Cadigan predicted a world with 2 Internets. One was 100% accountable. It was the main network used for real bussiness. There was no annonymity. The second network was designed to allow for anonymity. It was an "any thing goes" network where spoofing was the rule not the exception. I would like to see these networks. When I need to get work done I would use the accountable network. When I want to view pr0n I would use the other network. I think having two distinct networks like this would be a good compromise for the privacy advocats, and those tired of DOS attacks.
Ofcourse there are a *few* (as in many) technical difficulties to resolve first.
It's frightening to find a government where having sociopathological impulses helps you achieve political power.
I hate federal politics too.
It is possible that you could tell the difference between 16bit 44.1ks/s vs 24bit 48ks/s. With 16bit sound the absolute best signal to noise ratio (SNR) you can get is about 96dB. With 24bit you can get a theoretical SNR of 144dB. The source SNR will dominate the SNR of the ouput, so assumeing good shielding the 24bit sound should be quite a bit better then 16bit.
Now there are some limitations. Most stereo gear is designed for CD sound (16bit) and won't do better then 96dB. The max theoretical SNR of an amplifier is about 130-140dB (can't remember exact value). I believe that is a limitation imposed by physics and probably has something to do with the charge of an electron. So while 24bit may seem like overkill it is probably a standard that will survive longer then any of us.
Regarding 44.1ks/s vs 48ks/s. Higher is always better. Nyquist's Theorem states something along the lines of: You can reproduce any signal if your sampleing rate is atleast twice as high as the highest frequency in the signal. So 44.1ks/s should be able to reproduce 22.1kHz, but it can't. Nyquist assumes an infinate sequence of samples which is clearly impossible. With more samples you can also do better digital signal processing. There are just more samples to work with.
To tie this altogether we have to consider one other reality: Electronic gear produces 3rd order harmonic noise. 3rd order harmonics suck. They sound bad, sometimes are even painfull. It results in a lot of higher frequency noise. So we have to compensate for this with better then required sample rates and sample bits. This is why the next CD standard will probably be 24bit and 96ks/s.
Maybe his opinions are worth more? :-)
Parent article should not have been modded as flamebait.
Ha! That "SkyCam" is obviously a hoax. It isen't foggy.
In general you can do this under *nix with #!at the top of the file. So if you had a *nix command /usr/bin/rpn your example would translate into:
#!/usr/bin/rpn
b %1
b %2
a
s
Just about every *nix command that accepts a file containing commands as an argument will work this way.
I never did like REXX back when I was using OS/2 and I don't know why. It looks like REXX is an easy language to program with. And really I should have jumped at the chance to use a "real" scripting language versus basic and those hideous dos batch files.
Of course you can do that under linux. IBM Object REXX for Linux or other REXX for Unix.
Actually I think the Net Howto was written for 0.99.
hmm. good idea.
hmm, I guess I was hopeing it was 92/93. 1990 makes me feel old. :-)
In my post I was not refering to NCs. I was refering to *nix boxes without a disk running programs locally and storeing all data and programs on a server. Running *nix and dos+netware like this was quite common circa 1990. It also worked quite well, except when the network went down. All too common with thinnet (coax).
Search Google for "linux diskless howto" and you will get several links. Most *nixes can net boot. Sun hardware supports netboot out of the box to the point that netboot is the default if no hard disk is installed. Even lowly dos can netboot with the help of netware. It is old school. In the early nineties we figured out that netbooting sounded a lot cooler then it really is.
Your OS/2 comment is bang on. Technically OS/2 beat win95 in every way (back in 92/93). But IBM charged money for the dev kits. MS gave their's away (atleast for the first little bit). IBM failed to get hardware vendors onboard. MS paid hardware vendors to write drivers for win95. IBM ingnored the home market. MS put the home market square in their marketing sights. IBM's OS division couldn't even convince IBM's pc division to ship OS/2.
OS/2 2.0 had the jump on win94 (ninety four) by a good year. When win95 finally shipped OS/2 had had two years to catch the market but did nothing.