The USSR used 220V mains. I only built one amplifier, the two bad tubes were in a radio (one was bad when I got the radio (heater failure) the other went bad some time later (heater failure again).
As I said, my experience is that the heaters fail first (or the tube starts leaking, which makes the getter white or transparent). Maybe the tubes made in the USSR have weak heaters, or very rigid other parts?
I have replaced a total of two tubes in my devices, both of them were dark. One was a tuning indicator (I got the radio with it bad) and the other was a Soviet version of EL84 - that worked for a year or so and then burned out. In all of those devices I have not seen a tube that does not work but lights up.
Also, the tubes in my headphone amp are on at least 12 hours per day (sometimes I do not turn it off for weeks) and in use 6 years already and they still work. One tube (it was used when I got it, it has a black spot (burn?) in the getter) sometimes starts to act up (becomes noisy) but a slight tap gets it back in line. Granted, those are low power tubes (two 6N23P/6DJ8 and one 6N2P (kinda similar to 12AX7)) but still, they are quite long lived. The 6N23Ps are not even military grade.
So, why not combine dozens (or hundreds) of transistor output stages to get the equivalent of a single valve-based amplifier? That way, you get the same output power, but you never need to replace any burned-out tubes.
Because then the circuit would be much more complex (the need for matching all the small transmitters so they all work well in parallel) and a failed transistor could result in a lot of failed transistors. Tube circuits are simpler and tubes can tolerate overloads better.
Oh, and another thing. HVDC substations replaced their mercury valve rectifiers many years ago because new silicon-based technology could do the same job, at the same power level, with much less hassle. That's a higher level of power than broadcasting.
As the result of a rectifier is DC, it is simpler to combine a lot of smaller components in parallel and they dissipate less power than a transmitter would (since the output devices have to operate in linear mode in a transmitter, while they are on or off in a rectifier)
I have a few devices that use vacuum tubes and I have not encountered a tube that heats up but does not work due to low emissions. They all either work acceptably or not light up at all (a tube full of air also does not light up, at least from the nominal heater voltage). Maybe in the USSR made tubes the heater is the first to go.
Then again, I do not have a tube tester (always planning to build one, but always find something better to do), so maybe most of the tubes in my devices have really low emission, but the devices work OK, so I don't bother replacing the tubes with unused ones to see if the performance changes.
The device that produces the microwaves is called a magnetron and is a vacuum tube (vacuum tubes do not have to be made of glass, in fact, a lot of early vacuum tubes used in radios were metal).
Yes, but they are cheaper than the subscription. AFAIK NetFlix costs $17/month, so $204/year. Hard drives usually last longer than a year.
I have a VHS collection. I don't have a VCR. Even if I did, the quality of the media is not what it used to be.
I have a VHS collection and three working VCRs. The quality is not that much worse than broadcast analog TV, even for relatively old tapes. While VHS tapes degrade over time, they do not need to be replaced often (a tape will last longer than a hard drive stored in some box).
Basically, I dislike paying money continuously to keep something static (like the access to some movie), I'd rather pay money once (even if it's more expensive up front) so I don't need to pay again later. Just like with audio tapes or records - I bought it it's mine forever unless I damage or destroy it by being not careful.
This is why I archive things to data or audio tapes. I know that audio tapes last long (I have one recorded in 1951, still works) and I read somewhere that data tapes also are good for archiving (but have not used them long enough to say for certain).
I want bigger though. I wish the disk manufacturers were making 5.25" full height drives (even if they were spinning at 5400 or 3600 RPM), we could have 20TB or even bigger drives now. Add a 5-10GB SSD as cache and it would probably be quite fast too, though I would just use a 2TB HDD drive for "fast" needs and the big drive for archiving etc.
While I download less and less, it is because I archive everything and so I have less stuff to download. I like 1080p rips, but do not hunt for them, 720p or even SD can be good enough (though I will download the higher resolution version if I get the option to).
Streaming depends on other people, the data I have only depends on me. I may choose to stream instead of grabbing the tape, but I want the tape to be there, in case I cannot stream (internet connection goes down, the streaming provider goes down or something else).
Don't you need to pay a monthly fee for NetFlix? Also, are you certain, that NetFlix (or a similar service) will be available 10, 20, 30 years later? I know that my record and audio tape collection, VHS tapes, DVDs and data tapes will still be there, they do not depend on some company staying in business and I do not need to pay a monthly fee to keep them.
In an isolated environment the delay does not matter. So what if you have to wait a week to see the first episode, then the next episodes will come out once a week, just like they do in the USA. You can even be one year behind and have no problems. Until you want to discuss the TC show with other people. If you go to an international forum, there will be people who saw the show when it aired in the US (Americans and those who downloaded the torrents), they will be discussing an episode you have not seen yet. If you wait the week to see, say, episode 3 the guys on the forum will be discussing episode 4 and so on. The only way for you to discuss that show in that forum is to wait until the show takes a break and you can catch up.
Explain how the books aren't convenient compared to the show.
TV shows or movies have much higher "bitrate". That is, if you made a movie from a book, it would be faster (for most people) to watch the show than to read the book.
Also, at least for me, reading a paper book is less convenient physically than watching a movie. The books usually are made so they close by themselves unless I put a heavy object on them (I use an old battery from a UPS), but then I cannot read the text that is under that object, so I have to move it to read that portion of the text. Then lift it, turn the page and put it back down. On the other hand, I can just start the movie/TV show on my PC or VCR and watch it while doing nothing (or eating, or putting some device back together after repairing it or something else).
It seems like Microsoft wants the same look on a tablet and desktop so that it would be easier to use them (no need to learn two UIs). The problem is that the UI will suck on at least one type of device on which they are planning to use it. It is like making a car that is controlled like an airplane. Or an airplane that is controlled like a car. Both would suck.
My printer is inkjet, so any leak (there were a few) can be remedied with paper towels and (in some cases) water. There is a HP 2500 laser printer, but mine is ink (for whatever eason HP assigned the same number to very different devices).
Too bad it is not possible in the real world. Windows is already quite good at removing frozen apps from memory, but what if the app was accessing a device? If the driver or the device freezes only a reboot will restore it, if the app is using a device and freezes with the device locked - restart.
In Windows 8, the start screen has large buttons that are easier to hit with a mouse (better for Fitts' law).
Yes, because that's why I bought a 24" monitor with 1920x1200 resolution - I want bigger buttons. No, I can click on small buttons just fine (and I play FPS games with mouse sensitivity turned way up), I bought a bigger monitor with a big resolution so that more icons (and windows) can fit on the screen. If I only wanted to use one app at a time, I could have stayed with the 15" monitor that can only do 1152x864.
I got that printer for free, though it needed some repair (which I did myself) and, as I said, it is quite cheap to print (newer inkjets are more difficult to refill and the ink costs more for less (the black ink cartridge in my printer is 69ml)).
The printer supports postscript, but for some reason it is extremely slow if I use PS, I managed once to make it print fast using postscript under Linux, but only once. Usually it is "processing job" for half an hour before printing. PCL works fine (printing starts almost immediately). Then again, the printer was made in 1999, it probably just is that slow.
Unifying the look between a tablet and desktop is a bad idea because the devices use very different screens and input hardware.
A tablet has a relatively small, but high-dpi screen, the only input by default is the touchscreen which may not even support a stylus and fingers are quite big compared to the pixels on the screen. The result: 1. Icons and text have to be big relative to the screen so the user can see them. 2. Buttons and other active areas have to be big and far apart, so it is possible to accurately select them using a finger.
On the other hand, a desktop has a big, but low-dpi screen (my monitor is 24", but the resolution is only 1920x1200), however, the mouse can be pixel-accurate, which means: 1. Icons have to be smaller relative to the screen, so more of them fits on the screen (nobody with normal eyesight needs a 4cm icon (yes, the smallest button in the Metro interface is about 4cm by 4cm if I enlarge the screenshot so it fills the screen)). 2. Buttons and other active areas can be smaller and closer together, so the user does not need to move the mouse as far.
These requirements are essentially mutually exclusive, which means that whatever interface you come up with will suck when used in tablets or desktops or both.
One fun thing with HP driver I experienced is this:
I have a HP Professional Series Color 2500CM printer (quite old, but can print on A3 pages and the cartridges are very easy to refill, making the printing very cheap). The printer is old enough that Windows XP has a built-in driver for it, and it works quite well. Once I decided to download a driver from HP and try that out. It showed error messages saying that the ink cartridges and print heads have expired. As it turns out, HP has burned expiry dates to the printheads and cartridges, but the Windows XP driver does not check, which allows me to use the parts until they wear out (the yellow ink cartridge was supposed to expire in 2002 and I'm still using it).
All wired ISPs offer unlimited data service (actually, the majority of them do not offer a capped service, even if you wanted to) and have no problem with people using their service (the response to an overloaded network is an upgrade to the network).
Cellphone providers only offer capped services though, which sucks. Other wireless providers (WIMAX etc) offer unlimited service (IIRC).
This is the result of having a lot of wired ISPs, the competition is quite fierce - one ISP started to offer a faster connection and another (my ISP) upgraded everyone's speed (my connection previously was 200mbps local/80mbps international now it is 300mbps to everywhere) without asking for more money (I actually pay a bit less than I did for the 200/80 connection).
Lithuania. I pay ~23EUR/month, the connection is fiber (as I understand the network itself is PON) and the fiber goes to the room (the fiber-to-gigabit converter is in the house (I live in an individual house)).
In my country the ISP contracts specify minimum bandwidth or "ratio", for example, 30:1, which means that my 300mbps connection may sometimes be 10mbps (thankfully it doesn't happen often, if at all, usually at most the speed drops to 60-70mbps). Once the speed did not rise above ~20mbps for a week or so, I called the ISP and it was fixed (turned out to be a configuration problem on their end).
But it could be useful if the safer option is chosen. For example, in your case car 1 should have slowed down. On the other hand, if, say, car in front detects a pothole it can announce its intention to slow down before applying the brakes.
The range depends on the power used, antennas and the data rate. As cars most likely do not need 50mbps connection, the range would be longer. Also, in a mesh network, a node can forward packets between two nodes that are too far from each other to communicate directly.
Slashdot Mirror
The USSR used 220V mains. I only built one amplifier, the two bad tubes were in a radio (one was bad when I got the radio (heater failure) the other went bad some time later (heater failure again).
As I said, my experience is that the heaters fail first (or the tube starts leaking, which makes the getter white or transparent). Maybe the tubes made in the USSR have weak heaters, or very rigid other parts?
I have replaced a total of two tubes in my devices, both of them were dark. One was a tuning indicator (I got the radio with it bad) and the other was a Soviet version of EL84 - that worked for a year or so and then burned out. In all of those devices I have not seen a tube that does not work but lights up.
Also, the tubes in my headphone amp are on at least 12 hours per day (sometimes I do not turn it off for weeks) and in use 6 years already and they still work. One tube (it was used when I got it, it has a black spot (burn?) in the getter) sometimes starts to act up (becomes noisy) but a slight tap gets it back in line. Granted, those are low power tubes (two 6N23P/6DJ8 and one 6N2P (kinda similar to 12AX7)) but still, they are quite long lived. The 6N23Ps are not even military grade.
So, why not combine dozens (or hundreds) of transistor output stages to get the equivalent of a single valve-based amplifier? That way, you get the same output power, but you never need to replace any burned-out tubes.
Because then the circuit would be much more complex (the need for matching all the small transmitters so they all work well in parallel) and a failed transistor could result in a lot of failed transistors. Tube circuits are simpler and tubes can tolerate overloads better.
Oh, and another thing. HVDC substations replaced their mercury valve rectifiers many years ago because new silicon-based technology could do the same job, at the same power level, with much less hassle. That's a higher level of power than broadcasting.
As the result of a rectifier is DC, it is simpler to combine a lot of smaller components in parallel and they dissipate less power than a transmitter would (since the output devices have to operate in linear mode in a transmitter, while they are on or off in a rectifier)
I have a few devices that use vacuum tubes and I have not encountered a tube that heats up but does not work due to low emissions. They all either work acceptably or not light up at all (a tube full of air also does not light up, at least from the nominal heater voltage). Maybe in the USSR made tubes the heater is the first to go.
Then again, I do not have a tube tester (always planning to build one, but always find something better to do), so maybe most of the tubes in my devices have really low emission, but the devices work OK, so I don't bother replacing the tubes with unused ones to see if the performance changes.
The device that produces the microwaves is called a magnetron and is a vacuum tube (vacuum tubes do not have to be made of glass, in fact, a lot of early vacuum tubes used in radios were metal).
Don't drives need to be replaced?
Yes, but they are cheaper than the subscription. AFAIK NetFlix costs $17/month, so $204/year. Hard drives usually last longer than a year.
I have a VHS collection. I don't have a VCR. Even if I did, the quality of the media is not what it used to be.
I have a VHS collection and three working VCRs. The quality is not that much worse than broadcast analog TV, even for relatively old tapes. While VHS tapes degrade over time, they do not need to be replaced often (a tape will last longer than a hard drive stored in some box).
Basically, I dislike paying money continuously to keep something static (like the access to some movie), I'd rather pay money once (even if it's more expensive up front) so I don't need to pay again later. Just like with audio tapes or records - I bought it it's mine forever unless I damage or destroy it by being not careful.
This is why I archive things to data or audio tapes. I know that audio tapes last long (I have one recorded in 1951, still works) and I read somewhere that data tapes also are good for archiving (but have not used them long enough to say for certain).
I want bigger though. I wish the disk manufacturers were making 5.25" full height drives (even if they were spinning at 5400 or 3600 RPM), we could have 20TB or even bigger drives now. Add a 5-10GB SSD as cache and it would probably be quite fast too, though I would just use a 2TB HDD drive for "fast" needs and the big drive for archiving etc.
While I download less and less, it is because I archive everything and so I have less stuff to download. I like 1080p rips, but do not hunt for them, 720p or even SD can be good enough (though I will download the higher resolution version if I get the option to).
Streaming depends on other people, the data I have only depends on me. I may choose to stream instead of grabbing the tape, but I want the tape to be there, in case I cannot stream (internet connection goes down, the streaming provider goes down or something else).
Don't you need to pay a monthly fee for NetFlix? Also, are you certain, that NetFlix (or a similar service) will be available 10, 20, 30 years later? I know that my record and audio tape collection, VHS tapes, DVDs and data tapes will still be there, they do not depend on some company staying in business and I do not need to pay a monthly fee to keep them.
In an isolated environment the delay does not matter. So what if you have to wait a week to see the first episode, then the next episodes will come out once a week, just like they do in the USA. You can even be one year behind and have no problems. Until you want to discuss the TC show with other people. If you go to an international forum, there will be people who saw the show when it aired in the US (Americans and those who downloaded the torrents), they will be discussing an episode you have not seen yet. If you wait the week to see, say, episode 3 the guys on the forum will be discussing episode 4 and so on. The only way for you to discuss that show in that forum is to wait until the show takes a break and you can catch up.
Explain how the books aren't convenient compared to the show.
TV shows or movies have much higher "bitrate". That is, if you made a movie from a book, it would be faster (for most people) to watch the show than to read the book.
Also, at least for me, reading a paper book is less convenient physically than watching a movie. The books usually are made so they close by themselves unless I put a heavy object on them (I use an old battery from a UPS), but then I cannot read the text that is under that object, so I have to move it to read that portion of the text. Then lift it, turn the page and put it back down. On the other hand, I can just start the movie/TV show on my PC or VCR and watch it while doing nothing (or eating, or putting some device back together after repairing it or something else).
It seems like Microsoft wants the same look on a tablet and desktop so that it would be easier to use them (no need to learn two UIs). The problem is that the UI will suck on at least one type of device on which they are planning to use it. It is like making a car that is controlled like an airplane. Or an airplane that is controlled like a car. Both would suck.
My printer has 76MB RAM, it probably is the maximum it can handle.
My printer is inkjet, so any leak (there were a few) can be remedied with paper towels and (in some cases) water. There is a HP 2500 laser printer, but mine is ink (for whatever eason HP assigned the same number to very different devices).
Too bad it is not possible in the real world. Windows is already quite good at removing frozen apps from memory, but what if the app was accessing a device? If the driver or the device freezes only a reboot will restore it, if the app is using a device and freezes with the device locked - restart.
In Windows 8, the start screen has large buttons that are easier to hit with a mouse (better for Fitts' law).
Yes, because that's why I bought a 24" monitor with 1920x1200 resolution - I want bigger buttons. No, I can click on small buttons just fine (and I play FPS games with mouse sensitivity turned way up), I bought a bigger monitor with a big resolution so that more icons (and windows) can fit on the screen. If I only wanted to use one app at a time, I could have stayed with the 15" monitor that can only do 1152x864.
I got that printer for free, though it needed some repair (which I did myself) and, as I said, it is quite cheap to print (newer inkjets are more difficult to refill and the ink costs more for less (the black ink cartridge in my printer is 69ml)).
The printer supports postscript, but for some reason it is extremely slow if I use PS, I managed once to make it print fast using postscript under Linux, but only once. Usually it is "processing job" for half an hour before printing. PCL works fine (printing starts almost immediately). Then again, the printer was made in 1999, it probably just is that slow.
Unifying the look between a tablet and desktop is a bad idea because the devices use very different screens and input hardware.
A tablet has a relatively small, but high-dpi screen, the only input by default is the touchscreen which may not even support a stylus and fingers are quite big compared to the pixels on the screen. The result:
1. Icons and text have to be big relative to the screen so the user can see them.
2. Buttons and other active areas have to be big and far apart, so it is possible to accurately select them using a finger.
On the other hand, a desktop has a big, but low-dpi screen (my monitor is 24", but the resolution is only 1920x1200), however, the mouse can be pixel-accurate, which means:
1. Icons have to be smaller relative to the screen, so more of them fits on the screen (nobody with normal eyesight needs a 4cm icon (yes, the smallest button in the Metro interface is about 4cm by 4cm if I enlarge the screenshot so it fills the screen)).
2. Buttons and other active areas can be smaller and closer together, so the user does not need to move the mouse as far.
These requirements are essentially mutually exclusive, which means that whatever interface you come up with will suck when used in tablets or desktops or both.
One fun thing with HP driver I experienced is this:
I have a HP Professional Series Color 2500CM printer (quite old, but can print on A3 pages and the cartridges are very easy to refill, making the printing very cheap). The printer is old enough that Windows XP has a built-in driver for it, and it works quite well. Once I decided to download a driver from HP and try that out. It showed error messages saying that the ink cartridges and print heads have expired. As it turns out, HP has burned expiry dates to the printheads and cartridges, but the Windows XP driver does not check, which allows me to use the parts until they wear out (the yellow ink cartridge was supposed to expire in 2002 and I'm still using it).
All wired ISPs offer unlimited data service (actually, the majority of them do not offer a capped service, even if you wanted to) and have no problem with people using their service (the response to an overloaded network is an upgrade to the network).
Cellphone providers only offer capped services though, which sucks. Other wireless providers (WIMAX etc) offer unlimited service (IIRC).
This is the result of having a lot of wired ISPs, the competition is quite fierce - one ISP started to offer a faster connection and another (my ISP) upgraded everyone's speed (my connection previously was 200mbps local/80mbps international now it is 300mbps to everywhere) without asking for more money (I actually pay a bit less than I did for the 200/80 connection).
Lithuania. I pay ~23EUR/month, the connection is fiber (as I understand the network itself is PON) and the fiber goes to the room (the fiber-to-gigabit converter is in the house (I live in an individual house)).
In my country the ISP contracts specify minimum bandwidth or "ratio", for example, 30:1, which means that my 300mbps connection may sometimes be 10mbps (thankfully it doesn't happen often, if at all, usually at most the speed drops to 60-70mbps). Once the speed did not rise above ~20mbps for a week or so, I called the ISP and it was fixed (turned out to be a configuration problem on their end).
But it could be useful if the safer option is chosen. For example, in your case car 1 should have slowed down. On the other hand, if, say, car in front detects a pothole it can announce its intention to slow down before applying the brakes.
The range depends on the power used, antennas and the data rate. As cars most likely do not need 50mbps connection, the range would be longer. Also, in a mesh network, a node can forward packets between two nodes that are too far from each other to communicate directly.