The big issue most people had with IBM's 75GXP line was the suddenness with which the quality fell.
Usually your average geek can say "Oh yeah, don't buy E-machines, don't buy Packard Hell," etc. But IBM had come through damnnation, and were on their way to sainthood before the 75GXP drive release.
In the early 90s, IBM had a terrible reputation for hard drive reliability, there are even clandestine tales of them dumping entire wharehouses of the things. But I digress...
By the mid-ninties, IBM had redeemed themselves with a couple generations of decent drives. Then they released the series of Deskstar drives, and suddenly IBM could do no wrong. There were two high-performing and incredibly reliable generations of the Deskstar before the doomed 75GXP was released. The drives were so outstanding that even WD cashed in on the technology, re-branding some 14GXP drives as their own.
When IBM released the 75GXP, everyone and their dog was the first to proclaim it the king of the hill. Nobody questioned reliability, despite having the highest density ever for an IDE drive, untested glass platters and fluid bearings. It was only after significant amounts of the drives had reached circulation that the reliability issues surfaced.
ONE GENERATION FOLKS - a whole decade of trust gone up in smoke. Nobody could have called that, and it doesn't usually happen - until recently that is.
It hurts when you think you're safe, and then suddenly one day *POOF*, you're among the downtrodden.
My 75GXP gave me 14 months, and managed to leave me convinced I was the exception...before it blew up on me.
To be perfectly honest, this trend is going to continue so long as hard disk makers keep undercutting profits for marketshare. I remember a time not too long ago when hard drives under $100 new were non-existent, and high-end IDE drives topped $300. Now you can pick up a perfectly-servicable 40GB drive for around $60, and performance IDE tops out at ~$200.
The users cry "WHERE DID MY RELIABILITY/HAPPINESS GO?"
The companies cry "WHERE DID MY PROFITS GO?"
Reminds me of the video game market in 1984.
It's a sick cycle folks, but unless the hard drive industry takes a serious crash this cycle of slashing profits for marketshare will never end. Companies like IBM and Fujitsu won't liquidate their HD operations when they run into problems, they'll just keep handing them off to someone else for a loss and keep the quality bar at ground level.
Until then, we'll all have to be wary, both of anecdotal evidence, as well as the absolute BS MTBF and marketing crap these manufacturers are handing us. It's anecdotal evidence that got SO MANY into this whole 75GXP hole in the first place.
Open Windows Explorer and type in your internet address on the bar. Boom, you're browsing the net with Windows Explorer, and when it's minimized it looks inoccous enough. Warning though, earlier versions of IE-integrated explorer don't display the internet address on the minimized icon, but the versions that come with Win2k and XP do.
EVEN BETTER: SURF THE WEB WITH YOUR MSDN BROWSER!
Most of you folks who code have MS Visual Studio on your machine. Open up MSDN and click on Go, URL. Type in the URL ( and I mean a COMPLETE URL, don't leave out the http://, this is a primitive browser )
The good thing about this is it NEVER shows the internet address in the title bar / minimized bar, and unlike Windows Explorer it REALLY looks like you're doing something.
In fact, there are LOTS of products out there with HTML browsers built in, you just have to look for them.
Acually, it would make sense to package the "multiplayer" joystick with all multiplayer games.
So you've got the base unit with 10 single-player classics, and you can hook up the multiplayer joystick which only contains a rom with multiplayer games.
Is it just me, or is that game count pitifully small? I cannot imagine the actual game royalties topping $2-3 for all 10, and I know that most Atari carts were around 4K so ROM space certainly isn't an issue.
I believe people would be willing to plunk down the dosh if it say, cost $25 and came with 30 or so games + the ability to expand to multiplayer for like $10-15. To me, 10 games just seems too constrictive, and is likely to encourage folks like us to hack around a bit and find a way to reprogram the unit.
Actually, RCA Color, that to which you are referring, took many years to develop, and was originally not chosen by the NTSC.
Orginally, the NTSC chose CBS Color ( a mechanically-timed color wheel system ) because RCA's "no moving parts" system was late and looked terrible.
But RCA had an established Black and White user base, and the CBS color sets were incompatible, so CBS color sets didn't sell. A few years later, NTSC formally retracted their endorsement of CBS Color and endorsed RCA color, which hasn't changed for 48 years.
I only wonder if the same shiznit is going to happen to HDTV, we'll be stuck in 480i land forever!:)
Well, at 24k after taxes, you're taking in roughly double the US national mandated minimum wage ( state minimums can be higher ).
I imagine since you are military, you get free housing, or you are supplied a housing credit if you live off-base. Consider how much housing in Japan costs, and suddenly you're "bringing in" 40k+
But even if you do pay for housing, you still qualify for EIC, and that can add a little to the pile come tax time.
The target fighter has one point of reference. How are you suppsed to discern a line in space from one point?
IF the figher knew the power of the beam incident on itself, it could be possible to calculate the elevation angle of incidence by how much power is absorbed by the plane's skin.
The fighter could also calculate the angle of incidence from the longest length of elliptical-shaped burn created by the incident beam. But to measure this requires that the beam not move at all, and to USE it requires that the target fighter know both the attacking beam's aperture and distance from source.
IF the fighter knew all the above, it could analyze the elliptical-shaped burn created by the incident beam to find the azimuth.
With elevation and azimuth we have a target.
But let's be honest; distance, water vapor and smart pilots will guarantee you never know the exact power or configuration of the originating beam. So at best, you get an intelligent guess at the direction of origin.
This is coherent light we're talking about playing along a convex surface ( for example, a missle warhead ).
A small portion of the light inherent on the missle will be reflected directly back at the aircraft shooting the laser. However, thanks to 2x the distance plus the spreading of the light caused by the convex reflective surface, the power density inflicted upon the aircraft would be very low, even near the laser.
Basically, to harm the shooter with his own beam, you'd need to capture the beam perpendicular to a flat reflective surface. Not only is that damn near impossible in flight, even for a fraction of a second, but it would screw your aerodynamics:)
So then they just shoot your slow ass down with an AMRAAM.
By keeping the average bitrate to a range DSL can stram over, it's not so bad. You've got a movie at your fingertips.
But on the other hand, I know what 650MB movies look like, and no matter what format you use they still look like crap. Worse than VHS, let alone DVD.
Sounds like they'd be better suited to cap the movies at around a gig and have the users have like a 2 minute streaming buffer or so. At around a gig, you can get much better than VHS quality.
Me too. Jesus, theses aren't even "next generation" fans, they're just the same old crap ( Temperature controlled high-speed fans, and low noise high-output 120mm fans are old news ), and a couple of regular fans dressed up iMac style.
Big deal.
Next on Slashdot: "Cowboy Neal Discoveres a New Brand of Lay's Potato Chips"
Live a little. There's more to life than the details. The fact that this post is modded to hell and back as "funny" only shows how sick most people are of arguing over nothing.
LCDs still aren't quite there yet.
on
LCD Round-up
·
· Score: 4, Informative
Look, let me clear up some common misconceptions.
LCDs do not refresh at a certain rate per second like CRTs. In fact, once a pixel is set on an LCD, that pixel remains set to that color until it is changed.
THERE ARE NO UPDATES ON AN LCD. Each pixel is wired, and stays the same color until it's signal changes.
This is why there is noticable blur on an LCD. On a CRT, we would just see the whole screen getting updated at an incredibly low rate and call it insane flicker. But LCDs simply have a certain delay between when you change the pixel's color signal, and when the pixel gets updated. It looks blurry because it's not uniform rederawing of the screen like a CRT refresh.
There are three problems presently with LCDs that manufacturers will have to address before they overthrow CRTs:
*Even highend LCDs do not have the response time to even deliver 60fps video without blurring, and by far games are the worst thing to view on an LCD with a slow response time. As you look around and maneuver, the whole scene is blurred. The best LCDs on the market right now have around 25-30ms response rate, which is barely above 30fps. I believe we had this debate years ago ( 60 vs 30 ), and if the horsepower in today's video cards is anything to judge by, I'd say 60fps minimum won. I know personally I can't live with anything less. Sure, not everyone needs this kind of response time, but making it avaliable for the performance player is still a necessity.
*Most lowend mass-market LCDs have even worse response times (~45ms), and end up looking terrible when you view a video, or even when you're just scrolling through Explorer. People have come to expect a certain responsiveness and capability after paying for a multigigahertz toaster.
*Very few LCD screens have addressed the fact that their contrast ratios are terrible, even compared to cheap CRTs. I know a lot of you are proponents of LCDs because of their lack of flicker, but the truth is low contrast can cause just as much strain on the eyes, especially when reading.
The heaviest things in a powersupply are the transformer and the massive heatsink. Of course, you will also notice a bit more heft if they don't use the cheaper thinner sheet metal for the case too.
Very simple. A larger transformer can handle more current.
A large heatsink can mean ONLY ONE THING: the powersupply needs to dissappate large amounts of power lost to inefficiency. But this can have two causes - either the PS is efficient, but truely capable of high power, or the PS is inefficient, and they're trying to palm it off to you.
Let me tell you, Enlight has always been a good price/performance ratio. Myself and friends of mine have had a good history with Enlight PS.
I still have the 250W PS that came with my old 7237 case back in the day. It's 5 years old now, and has seen 3 upgrades, and has run 24/7 now in my Celermine server for 8 months to boot. Keep in mind, this is a PS I have accidentally shorted out and tripped the internal breaker at least 3 times, it's one tough mofo:)
I bought the 340W Enlight to replace it so I could upgrade to an Athlon system, and I'm fairly impressed. It comes with 2 fans, one of which is plugged into a 3-pin motherboard jack, and can be BIOS controlled.
I've had it loaded down with 2 hard drives, 2 opticals, an XP 1600+, and a Radeon 8500 and have had zero stability problems. The PS fans are actually quieter than the fan on my Coolermaster, which is great because I only paid $65 for the case and PS.
As I explained above, RS-232 ( which uses a UART ), is an asynchronous communications standard. The chip itself is clocked.
There is no clock shared between the two points of communication. Each end "agrees" on a clock speed, but there is no guarantee how accurately each end produces said clock speed.
A receiver detects a new packet when it receives the start bit, and it samples the incoming serial stream using it's own local clock. This is the asynchronous part of the communications, the receiver really has no idea if it's sampling right, and clock skew between the sender and receiver can produce errors.
Actually, most popular communications formats are "asynchronous".
Don't confuse yourself. Synchronous communications involve a real-time shared clock between points.
Then you have asynchronous communications standards like RS-232. The sender and receiver choose a baud rate, and the receiver waits for a start bit, then starts sampling the stream using it's local clock. So long as the clocks are close enough, and the packets are short enough, you'll never get an error.
Then you have standards like Fast Ethernet, which are also asynchronous. AFAIK, the clock used to decode the Ethernet packet is contained somewhere in the preamble, and a PLL is tuned to the packet's clock rate. This is to avoid the obvious problems of the simple async communications of RS-232.
A SAMPLE OF THE ACTUAL CLOCK used to encode the packet is avaliable to the receiver, but the receiver can only use this to tune it's local clock. It has to do the decoding asynch.
Actually, cable providers have gotten away with a very common picture quality issue from the very beginnings of cable.
Analog cable shares the same spectrum as broadcast, it's just closed-circuit and shielded. But even this didn't stop interference between the two.
Many idiodic local stations insisted that their channel be carried over cable on the same number as their broadcast transmission, guaranteeing that the two unsynched signals would display a nice pretty ghost on your TV.
Some cable providers were smart, and initially wouldn't let ANY channel into cable spectrum used by local broadcasters. But as channels filled up, they gave in and used the spectrum, ensuring that each and every show viewed on the brand new channels would have a nice ghosed image of the local CBS/NBC/ABC broadcast affiliate using the same spectrum.
Digital cable claims to fix this problem, although in reality it has simply allowed cable providers to go back to not using spectrum already used by local broadcasters. Everything in the normal VHF - UHF range of channels is still analog, but now they have hundreds of channels beyond that in which to toss the overflow.
You bitch, you moan, but believe me, it's a LOT better than some of the incredibly crappy cable services I've witnessed in my lifetime, and nobody every successfully sued them. And it's just a stepping stone to the real deal - high bandwidth digital cable.
Re:woooaa
on
Lego Segway
·
· Score: 2, Informative
Actually, the problem is it's only good at one thing:
Balancing while navigating flat terrain.
Anything less than flat, and the distance sensor is going to be confused. It may overcorrect and send the thing tumbling. Really broken terrain, or areas with grass and plants, OR DENSE TRAFFIC would send this simple sensor crying home to momma.
Ginger, on the other hand, uses a complex internal gyro system to determine which way is up, which means that no external interference is possible ( short of rigging gravity, and even then GINGER will always be right! ). The only way to make it tip over is to give it an external impulse that it's motors cannot respond to fast enough.
Most IDE hard drives either go out within one month or last more than 5 years anyway.
Not true anymore, although I would have agreed with this statement 2 years ago. It took months for the 75GXP reliability issues to surface, and most folks got a good 3-6 months or even a year out of them before encountering issues. The general consensus was that the REPLACEMENT DRIVES were failing within a month, but that's just because most of the replacements were probably just reflashed and low-level formatted RMAs.
My personal IBM 75GXP experince lasted 14 months, and I was right at the edge of calling it safe when the drive started grinding and produced bad clusters like rabbits.
New and unproven technoligies like glass platters, fluid bearings, and extremely high density disks coupled with shrinking profit margins are pushing performance IDE drives to the limits of reliability.
But don't worry. If IDE hard drives continue to tank, PC standards bodies will jump the gate and start setting minimum reliability ratings. After all, these HD manufacturers couldn't give a damn about consumers, but big OEMs can move mountains.
It's not the tracks holding Amtrack back...
on
Jet Turbine Locomotives
·
· Score: 2, Insightful
...It's the switching and signaling.
Amtrack's older fleet of Diesel trains have an operating speed of 103mph. The more recent trains introduced in the mid 90s can do 110mph in service.
Most of the switching and signaling on Amtrack's thousands of miles of non-electrified track require a human to see and react to the signal. So, with the exception of high-speed electric routes, most Amtrack railways are limited to 79mph for safety reasons. Municipalities also have the power to limit train speeds at a local level, if 79mph is considered too dangerous.
Amtrack introduced a computer-based switching system to counter this, but I imagine just like everthing else Amtrack has attempted, it was implemented half-assed at best.
http://www.cnn.com/TECH/9610/12/high.speed.trains/
Until a switching system SOMEWHERE NEAR the complexity of the FAA's air-traffic control system is introduced, it doesn't matter how fast you can make the engine go.
Slashdot Mirror
The big issue most people had with IBM's 75GXP line was the suddenness with which the quality fell.
Usually your average geek can say "Oh yeah, don't buy E-machines, don't buy Packard Hell," etc. But IBM had come through damnnation, and were on their way to sainthood before the 75GXP drive release.
In the early 90s, IBM had a terrible reputation for hard drive reliability, there are even clandestine tales of them dumping entire wharehouses of the things. But I digress...
By the mid-ninties, IBM had redeemed themselves with a couple generations of decent drives. Then they released the series of Deskstar drives, and suddenly IBM could do no wrong. There were two high-performing and incredibly reliable generations of the Deskstar before the doomed 75GXP was released. The drives were so outstanding that even WD cashed in on the technology, re-branding some 14GXP drives as their own.
When IBM released the 75GXP, everyone and their dog was the first to proclaim it the king of the hill. Nobody questioned reliability, despite having the highest density ever for an IDE drive, untested glass platters and fluid bearings. It was only after significant amounts of the drives had reached circulation that the reliability issues surfaced.
ONE GENERATION FOLKS - a whole decade of trust gone up in smoke. Nobody could have called that, and it doesn't usually happen - until recently that is.
It hurts when you think you're safe, and then suddenly one day *POOF*, you're among the downtrodden.
My 75GXP gave me 14 months, and managed to leave me convinced I was the exception...before it blew up on me.
To be perfectly honest, this trend is going to continue so long as hard disk makers keep undercutting profits for marketshare. I remember a time not too long ago when hard drives under $100 new were non-existent, and high-end IDE drives topped $300. Now you can pick up a perfectly-servicable 40GB drive for around $60, and performance IDE tops out at ~$200.
The users cry "WHERE DID MY RELIABILITY/HAPPINESS GO?"
The companies cry "WHERE DID MY PROFITS GO?"
Reminds me of the video game market in 1984.
It's a sick cycle folks, but unless the hard drive industry takes a serious crash this cycle of slashing profits for marketshare will never end. Companies like IBM and Fujitsu won't liquidate their HD operations when they run into problems, they'll just keep handing them off to someone else for a loss and keep the quality bar at ground level.
Until then, we'll all have to be wary, both of anecdotal evidence, as well as the absolute BS MTBF and marketing crap these manufacturers are handing us. It's anecdotal evidence that got SO MANY into this whole 75GXP hole in the first place.
Assuming you have Windows 98 or later:
Open Windows Explorer and type in your internet address on the bar. Boom, you're browsing the net with Windows Explorer, and when it's minimized it looks inoccous enough. Warning though, earlier versions of IE-integrated explorer don't display the internet address on the minimized icon, but the versions that come with Win2k and XP do.
EVEN BETTER: SURF THE WEB WITH YOUR MSDN BROWSER!
Most of you folks who code have MS Visual Studio on your machine. Open up MSDN and click on Go, URL. Type in the URL ( and I mean a COMPLETE URL, don't leave out the http://, this is a primitive browser )
The good thing about this is it NEVER shows the internet address in the title bar / minimized bar, and unlike Windows Explorer it REALLY looks like you're doing something.
In fact, there are LOTS of products out there with HTML browsers built in, you just have to look for them.
Acually, it would make sense to package the "multiplayer" joystick with all multiplayer games.
So you've got the base unit with 10 single-player classics, and you can hook up the multiplayer joystick which only contains a rom with multiplayer games.
Is it just me, or is that game count pitifully small? I cannot imagine the actual game royalties topping $2-3 for all 10, and I know that most Atari carts were around 4K so ROM space certainly isn't an issue.
I believe people would be willing to plunk down the dosh if it say, cost $25 and came with 30 or so games + the ability to expand to multiplayer for like $10-15. To me, 10 games just seems too constrictive, and is likely to encourage folks like us to hack around a bit and find a way to reprogram the unit.
Actually, RCA Color, that to which you are referring, took many years to develop, and was originally not chosen by the NTSC.
:)
Orginally, the NTSC chose CBS Color ( a mechanically-timed color wheel system ) because RCA's "no moving parts" system was late and looked terrible.
But RCA had an established Black and White user base, and the CBS color sets were incompatible, so CBS color sets didn't sell. A few years later, NTSC formally retracted their endorsement of CBS Color and endorsed RCA color, which hasn't changed for 48 years.
I only wonder if the same shiznit is going to happen to HDTV, we'll be stuck in 480i land forever!
Well, at 24k after taxes, you're taking in roughly double the US national mandated minimum wage ( state minimums can be higher ).
I imagine since you are military, you get free housing, or you are supplied a housing credit if you live off-base. Consider how much housing in Japan costs, and suddenly you're "bringing in" 40k+
But even if you do pay for housing, you still qualify for EIC, and that can add a little to the pile come tax time.
Lightning is 1,000,000,000 to 10,000,000,000 joules.
Funny, I always thought lightning had 1.21 jigawatts.
Doc! ...what the hell is a jigawatt?!?
Impossible.
The target fighter has one point of reference. How are you suppsed to discern a line in space from one point?
IF the figher knew the power of the beam incident on itself, it could be possible to calculate the elevation angle of incidence by how much power is absorbed by the plane's skin.
The fighter could also calculate the angle of incidence from the longest length of elliptical-shaped burn created by the incident beam. But to measure this requires that the beam not move at all, and to USE it requires that the target fighter know both the attacking beam's aperture and distance from source.
IF the fighter knew all the above, it could analyze the elliptical-shaped burn created by the incident beam to find the azimuth.
With elevation and azimuth we have a target.
But let's be honest; distance, water vapor and smart pilots will guarantee you never know the exact power or configuration of the originating beam. So at best, you get an intelligent guess at the direction of origin.
This is coherent light we're talking about playing along a convex surface ( for example, a missle warhead ).
:)
A small portion of the light inherent on the missle will be reflected directly back at the aircraft shooting the laser. However, thanks to 2x the distance plus the spreading of the light caused by the convex reflective surface, the power density inflicted upon the aircraft would be very low, even near the laser.
Basically, to harm the shooter with his own beam, you'd need to capture the beam perpendicular to a flat reflective surface. Not only is that damn near impossible in flight, even for a fraction of a second, but it would screw your aerodynamics
So then they just shoot your slow ass down with an AMRAAM.
Dy-4 with VxWorks on VME?
Hmmmmmm....
Actually, I believe they were walls. She actually complained to him because he wouldn't spring to have all 4 walls covered, she was stuck with 3.
But really, the concept of overconnectedness can truely be traced back to 1984, and the telescreens.
Lets see.
By keeping the average bitrate to a range DSL can stram over, it's not so bad. You've got a movie at your fingertips.
But on the other hand, I know what 650MB movies look like, and no matter what format you use they still look like crap. Worse than VHS, let alone DVD.
Sounds like they'd be better suited to cap the movies at around a gig and have the users have like a 2 minute streaming buffer or so. At around a gig, you can get much better than VHS quality.
...Is finally a reality!
Anthem for everybody!
Now all we need are Electric Monks to do the discussion and believing for us!
Me too. Jesus, theses aren't even "next generation" fans, they're just the same old crap ( Temperature controlled high-speed fans, and low noise high-output 120mm fans are old news ), and a couple of regular fans dressed up iMac style.
Big deal.
Next on Slashdot: "Cowboy Neal Discoveres a New Brand of Lay's Potato Chips"
No, the man we need is...
ULTRAMAN!
Woot, he can grow to insane sizes and shoot fireballs, only problem is he looks like a Power Ranger...hmmm
Did someone say Power Rangers!?!?
Lets make "A fighting force of Extra-ordinary Magnitude!!!!!"
Big Jim Slade, Former Tight-End for the Kansas City Cheifs, will save us!
No, the quote is correct.
The quote was muttered while watching a video with a repetitive "Cigarette Dangles..." lyric.
Of course, it's not as good as Beavis trying to "speak French" and sing along with a video ( really Japanese, but they think it's French ).
Pssst. It's humor.
AKA:
LOLOLOLOL
TEH FUNEY
HAHAHAHAHAHAHHHAAAHAH!
Live a little. There's more to life than the details. The fact that this post is modded to hell and back as "funny" only shows how sick most people are of arguing over nothing.
Look, let me clear up some common misconceptions.
LCDs do not refresh at a certain rate per second like CRTs. In fact, once a pixel is set on an LCD, that pixel remains set to that color until it is changed.
THERE ARE NO UPDATES ON AN LCD. Each pixel is wired, and stays the same color until it's signal changes.
This is why there is noticable blur on an LCD. On a CRT, we would just see the whole screen getting updated at an incredibly low rate and call it insane flicker. But LCDs simply have a certain delay between when you change the pixel's color signal, and when the pixel gets updated. It looks blurry because it's not uniform rederawing of the screen like a CRT refresh.
There are three problems presently with LCDs that manufacturers will have to address before they overthrow CRTs:
*Even highend LCDs do not have the response time to even deliver 60fps video without blurring, and by far games are the worst thing to view on an LCD with a slow response time. As you look around and maneuver, the whole scene is blurred. The best LCDs on the market right now have around 25-30ms response rate, which is barely above 30fps. I believe we had this debate years ago ( 60 vs 30 ), and if the horsepower in today's video cards is anything to judge by, I'd say 60fps minimum won. I know personally I can't live with anything less. Sure, not everyone needs this kind of response time, but making it avaliable for the performance player is still a necessity.
*Most lowend mass-market LCDs have even worse response times (~45ms), and end up looking terrible when you view a video, or even when you're just scrolling through Explorer. People have come to expect a certain responsiveness and capability after paying for a multigigahertz toaster.
*Very few LCD screens have addressed the fact that their contrast ratios are terrible, even compared to cheap CRTs. I know a lot of you are proponents of LCDs because of their lack of flicker, but the truth is low contrast can cause just as much strain on the eyes, especially when reading.
The heaviest things in a powersupply are the transformer and the massive heatsink. Of course, you will also notice a bit more heft if they don't use the cheaper thinner sheet metal for the case too.
Very simple. A larger transformer can handle more current.
A large heatsink can mean ONLY ONE THING: the powersupply needs to dissappate large amounts of power lost to inefficiency. But this can have two causes - either the PS is efficient, but truely capable of high power, or the PS is inefficient, and they're trying to palm it off to you.
Let me tell you, Enlight has always been a good price/performance ratio. Myself and friends of mine have had a good history with Enlight PS.
:)
I still have the 250W PS that came with my old 7237 case back in the day. It's 5 years old now, and has seen 3 upgrades, and has run 24/7 now in my Celermine server for 8 months to boot. Keep in mind, this is a PS I have accidentally shorted out and tripped the internal breaker at least 3 times, it's one tough mofo
I bought the 340W Enlight to replace it so I could upgrade to an Athlon system, and I'm fairly impressed. It comes with 2 fans, one of which is plugged into a 3-pin motherboard jack, and can be BIOS controlled.
I've had it loaded down with 2 hard drives, 2 opticals, an XP 1600+, and a Radeon 8500 and have had zero stability problems. The PS fans are actually quieter than the fan on my Coolermaster, which is great because I only paid $65 for the case and PS.
As I explained above, RS-232 ( which uses a UART ), is an asynchronous communications standard. The chip itself is clocked.
There is no clock shared between the two points of communication. Each end "agrees" on a clock speed, but there is no guarantee how accurately each end produces said clock speed.
A receiver detects a new packet when it receives the start bit, and it samples the incoming serial stream using it's own local clock. This is the asynchronous part of the communications, the receiver really has no idea if it's sampling right, and clock skew between the sender and receiver can produce errors.
Actually, most popular communications formats are "asynchronous".
Don't confuse yourself. Synchronous communications involve a real-time shared clock between points.
Then you have asynchronous communications standards like RS-232. The sender and receiver choose a baud rate, and the receiver waits for a start bit, then starts sampling the stream using it's local clock. So long as the clocks are close enough, and the packets are short enough, you'll never get an error.
Then you have standards like Fast Ethernet, which are also asynchronous. AFAIK, the clock used to decode the Ethernet packet is contained somewhere in the preamble, and a PLL is tuned to the packet's clock rate. This is to avoid the obvious problems of the simple async communications of RS-232.
A SAMPLE OF THE ACTUAL CLOCK used to encode the packet is avaliable to the receiver, but the receiver can only use this to tune it's local clock. It has to do the decoding asynch.
Actually, cable providers have gotten away with a very common picture quality issue from the very beginnings of cable.
Analog cable shares the same spectrum as broadcast, it's just closed-circuit and shielded. But even this didn't stop interference between the two.
Many idiodic local stations insisted that their channel be carried over cable on the same number as their broadcast transmission, guaranteeing that the two unsynched signals would display a nice pretty ghost on your TV.
Some cable providers were smart, and initially wouldn't let ANY channel into cable spectrum used by local broadcasters. But as channels filled up, they gave in and used the spectrum, ensuring that each and every show viewed on the brand new channels would have a nice ghosed image of the local CBS/NBC/ABC broadcast affiliate using the same spectrum.
Digital cable claims to fix this problem, although in reality it has simply allowed cable providers to go back to not using spectrum already used by local broadcasters. Everything in the normal VHF - UHF range of channels is still analog, but now they have hundreds of channels beyond that in which to toss the overflow.
You bitch, you moan, but believe me, it's a LOT better than some of the incredibly crappy cable services I've witnessed in my lifetime, and nobody every successfully sued them. And it's just a stepping stone to the real deal - high bandwidth digital cable.
Actually, the problem is it's only good at one thing:
Balancing while navigating flat terrain.
Anything less than flat, and the distance sensor is going to be confused. It may overcorrect and send the thing tumbling. Really broken terrain, or areas with grass and plants, OR DENSE TRAFFIC would send this simple sensor crying home to momma.
Ginger, on the other hand, uses a complex internal gyro system to determine which way is up, which means that no external interference is possible ( short of rigging gravity, and even then GINGER will always be right! ). The only way to make it tip over is to give it an external impulse that it's motors cannot respond to fast enough.
Most IDE hard drives either go out within one month or last more than 5 years anyway.
Not true anymore, although I would have agreed with this statement 2 years ago. It took months for the 75GXP reliability issues to surface, and most folks got a good 3-6 months or even a year out of them before encountering issues. The general consensus was that the REPLACEMENT DRIVES were failing within a month, but that's just because most of the replacements were probably just reflashed and low-level formatted RMAs.
My personal IBM 75GXP experince lasted 14 months, and I was right at the edge of calling it safe when the drive started grinding and produced bad clusters like rabbits.
New and unproven technoligies like glass platters, fluid bearings, and extremely high density disks coupled with shrinking profit margins are pushing performance IDE drives to the limits of reliability.
But don't worry. If IDE hard drives continue to tank, PC standards bodies will jump the gate and start setting minimum reliability ratings. After all, these HD manufacturers couldn't give a damn about consumers, but big OEMs can move mountains.
...It's the switching and signaling. Amtrack's older fleet of Diesel trains have an operating speed of 103mph. The more recent trains introduced in the mid 90s can do 110mph in service. Most of the switching and signaling on Amtrack's thousands of miles of non-electrified track require a human to see and react to the signal. So, with the exception of high-speed electric routes, most Amtrack railways are limited to 79mph for safety reasons. Municipalities also have the power to limit train speeds at a local level, if 79mph is considered too dangerous. Amtrack introduced a computer-based switching system to counter this, but I imagine just like everthing else Amtrack has attempted, it was implemented half-assed at best. http://www.cnn.com/TECH/9610/12/high.speed.trains/
Until a switching system SOMEWHERE NEAR the complexity of the FAA's air-traffic control system is introduced, it doesn't matter how fast you can make the engine go.