that do not require the registration process. It's meant for when you install on alot of machines, but works just like any other XP install. Minus the registration.
SCSI operates at a few megahertz. its a very slow and rapidly aging bus. Its generally thought of as 'fast' because its fast when you're dealing with mechanical drives and accessories.
If this new ram were to be added on to current systems, it would likely be in the form of a PCI add-in card. That's still a bottleneck, but nowhere near as slow as using a bus spec for mechanical subsystems.
Most people on slashdot talk out of their ass constantly. I mean, look how many people on this thread suggest turning it into an IDE or SCSI drive, which is absolute nonsense. It has half-nanosecond write times. A bus spec built for millisecond mechanical drives is woefully inadequate for such storage technology.
Of course if they knew anything at all about basic bus architecture (aside from obscure acronyms), it would be obvious.
The only conceivable 'drop-in' on current machines this might be useful for is on some sort of PCI setup. PCI's bandwidth is far higher than SCSI or IDE, but is still a considerable bottleneck. I can see this being used as a reasonable large caching scheme.
Really though, for effective use of such technology requires a redesign of the whole system.
Why would you waste the speed of such a thing with a SCSI connector? SCSI is *really, really, really* slow compared to this. I mean, SCSI is meant for mechanical drive speeds with millisecond seek times. Not nanosecond. If you were to see this adapted for existing machines, it *might* use PCI for the transfer, but even then PCI is really slow too.
Using this technology requires rebuilding entire system busses to use it effectively.
I posted something else along the lines of this, but how would you do it under heavy load? The disk is so enormously slow compared to RAM, you'd overwhelm whatever buffer you're using to do the write-back. You'd have to throttle back requests on the RAM, thus negating the performance increases.
I thought one of the key points of ACIDity was to maintain data integrity in the event of catastrophic system failure (ie, power goes out)?
With a dynamic RAM system (DRAM also isn't all that reliable...SRAM is better, and SRAM is very expensive) you are highly vulnerable to this.
I suppose you could implement a kind of write-back system to the disk where you pile up things in some kind of buffer, but under heavy load, you're going to overwhelm it. Or at the very least cause the thrashing that this supposedly helps avoid.
Really though, at that point, you're just using the RAM as a cache. While this sounds all nice and fanciful, it doesn't sound to me like he's thought it all the way through. Perhaps some people who know more about database design can point out any simple mistakes ive made...
All military hardware today is shielded against EMP. I would think sentient robots 200 years from now would also be aware of their own weaknesses and be shielded as well.
it should be the other way around. An LCD 'turns on' its pixel so it becomes opaque - that is, black.
A pixel is 'turned off' (goes clear) and lets the white backlight through it to display white.
This is why LCD's are not as good at displaying black as CRT's. The backlight is always on, and the pixels can have varying degrees of opaqueness when turned on. This makes black more of an uneven very dark gray (well, uneven on my LCD anyway..newer ones might be better at it)
Also, they won't necessarily be black or white. I have some green dead pixels.
the read/write head of a harddrive is not unlike that of an airplane wing. It floats/flies micrometers above the surface of the platter on a cushion of air as the platters spin. This is necessary because the platters have microscopic hills and trenches as polishing something perfectly smooth isnt quite possible, and to avoid crashing the head into the platters.
This is why most harddrives on their labels say "DO NOT COVER EXHAUST HOLES"
Im surprised it worked for 2 days. Maybe thats how long it took for the air already in the drive to get pushed out the complex baffle filters that are behind the exhaust holes.
kazaa cannot locate 90% of the rare music i look for.
If you want popular or semi-popular things, kazaa works well. For rare things, you might, if lucky, find one person somewhere who has it and it almost always returns 'Needs more sources'.
one of the best places for columbia news...
on
More on Columbia
·
· Score: 1
is the Houston Chronicle. Since nasa has such a huge houston presence, theres usually a front page story every day that catches angles regarding the shuttle that larger news organizations ignore.
Today for example had interviews with some engineers at USA regarding the Cult of Safety, and a bunch of other things.
They've got a whole ongoing section dedicated to the investigation and how its going.
Every commercial jet built within the past 40 years has been fly by wire.
Only poorly programmed general purpose computers crash. The brake-by-wire system not only uses redundant wires and chips, all software within is proven correct through mathematical analysis. And the system is task specific - it only brakes.
Further, your mechanical brakes are far more likely to fail catastrophically. What do you do when a hydraulic brake line gets cut? All fluid will seep out from the entire system, rendering all your brakes useless.
With brake by wire, not only is it easy and feasible to run 'back up' wires to each wheel, a cut will only affect that specific wheel rather than all of them.
How such a flaw - and it was a very obvious one, that testing would've uncovered - ever made it out of the lab is beyond me.
Avionics code is bug free because its proven mathematically for correctness and undergoes tens of thousands of hours of rigorous lab testing. At least that's how its supposed to work.
For example, the code flying the Boeing 777 flew literally millions of flights in every possible variable and condition before a single prototype of the plane ever left the ground. This was to catch any possible faults that the rigorous mathematical analysis didn't involve.
How or why ariane failed to do apparently either of these is shameful and shocking. But perhaps it is incidents like the ariane that led honeywell - and all modern avonics designers - to go the lengths they do now to validate the code.
You'll probably appreciate this game I wrote. It's web-based, about secret Agent spy stuff. Point is to level up, overthrow a government. Lots of wacky fun and twists to it. Its easy to play...sort of like LORD, but by no means a clone.
i've seen this on a number of machines. I believe as soon as you logoff it reverts to the 'default' state.
While this might be annoying to some, in general, its a good thing on public computers. Besides undoing any software installs a user might've tried, it also removes old cookies and temp files that might contain someone's personal info.
Most public uni. computers ive seen all have zip drives. If you want to download and save something, I suggest putting this to use. It is, after all, a public computer.
they dont ignite exactly on contact, its a highly exothermic reaction, but the needed component is powdered aluminum. This is probably what the long white trail following the orbiter was, as its altitude was much too high for contrails.
i never had the headache problem, but i do remember the instruction book recommending a gaming pause every 15 minutes. In fact, I recall at least one game that automatically paused after 20 minutes and told you to take a break.
All in all, I dont think it was THAT bad. I kinda liked the little bugger. Sure the red on black was odd, but the effect was neat and it worked. The sound was pretty good too, since your ears were right by the speakers and it used true stereo sound with a fairly high sampling rate for the time.
while hydrogen is fairly common throughout space (obviously), theres really not that much free hydrogen floating around. Most of the kind that would be accessible with such a method is found in gaseous clouds, of which are nowhere near earth.
i emailed this to my father, at his company they have one of these on nearly every workstation. He emailed back that 2 weeks ago one of his employee's homes burned down, and the UPS is the suspected culprit.
A few months ago in that very same office one of the UPS's begin emitting a high pitched whistle from within (described as a 'tea kettle' sound), likely very hot gases escaping.
Slashdot Mirror
that do not require the registration process. It's meant for when you install on alot of machines, but works just like any other XP install. Minus the registration.
If this new ram were to be added on to current systems, it would likely be in the form of a PCI add-in card. That's still a bottleneck, but nowhere near as slow as using a bus spec for mechanical subsystems.
Of course if they knew anything at all about basic bus architecture (aside from obscure acronyms), it would be obvious.
The only conceivable 'drop-in' on current machines this might be useful for is on some sort of PCI setup. PCI's bandwidth is far higher than SCSI or IDE, but is still a considerable bottleneck. I can see this being used as a reasonable large caching scheme.
Really though, for effective use of such technology requires a redesign of the whole system.
Using this technology requires rebuilding entire system busses to use it effectively.
and if your power outage or other complete system failure occurs before the disk operations are complete?
I posted something else along the lines of this, but how would you do it under heavy load? The disk is so enormously slow compared to RAM, you'd overwhelm whatever buffer you're using to do the write-back. You'd have to throttle back requests on the RAM, thus negating the performance increases.
With a dynamic RAM system (DRAM also isn't all that reliable...SRAM is better, and SRAM is very expensive) you are highly vulnerable to this.
I suppose you could implement a kind of write-back system to the disk where you pile up things in some kind of buffer, but under heavy load, you're going to overwhelm it. Or at the very least cause the thrashing that this supposedly helps avoid.
Really though, at that point, you're just using the RAM as a cache. While this sounds all nice and fanciful, it doesn't sound to me like he's thought it all the way through. Perhaps some people who know more about database design can point out any simple mistakes ive made...
They may be useful, but they need to avoid C++ing themselves with hard to read shortcut characters.
All military hardware today is shielded against EMP. I would think sentient robots 200 years from now would also be aware of their own weaknesses and be shielded as well.
which is why in most recent films, you're unlikely to see any computer without one.
A pixel is 'turned off' (goes clear) and lets the white backlight through it to display white.
This is why LCD's are not as good at displaying black as CRT's. The backlight is always on, and the pixels can have varying degrees of opaqueness when turned on. This makes black more of an uneven very dark gray (well, uneven on my LCD anyway..newer ones might be better at it)
Also, they won't necessarily be black or white. I have some green dead pixels.
I got up this morning and SETI was reporting a fatal error i've never seen before - coincidence?
This is why most harddrives on their labels say "DO NOT COVER EXHAUST HOLES"
Im surprised it worked for 2 days. Maybe thats how long it took for the air already in the drive to get pushed out the complex baffle filters that are behind the exhaust holes.
Seems like a big 'duh' today, but at the time it was revolutionary. As was her assembler.
If you want popular or semi-popular things, kazaa works well. For rare things, you might, if lucky, find one person somewhere who has it and it almost always returns 'Needs more sources'.
Today for example had interviews with some engineers at USA regarding the Cult of Safety, and a bunch of other things.
They've got a whole ongoing section dedicated to the investigation and how its going.
Only poorly programmed general purpose computers crash. The brake-by-wire system not only uses redundant wires and chips, all software within is proven correct through mathematical analysis. And the system is task specific - it only brakes.
Further, your mechanical brakes are far more likely to fail catastrophically. What do you do when a hydraulic brake line gets cut? All fluid will seep out from the entire system, rendering all your brakes useless.
With brake by wire, not only is it easy and feasible to run 'back up' wires to each wheel, a cut will only affect that specific wheel rather than all of them.
Avionics code is bug free because its proven mathematically for correctness and undergoes tens of thousands of hours of rigorous lab testing. At least that's how its supposed to work.
For example, the code flying the Boeing 777 flew literally millions of flights in every possible variable and condition before a single prototype of the plane ever left the ground. This was to catch any possible faults that the rigorous mathematical analysis didn't involve.
How or why ariane failed to do apparently either of these is shameful and shocking. But perhaps it is incidents like the ariane that led honeywell - and all modern avonics designers - to go the lengths they do now to validate the code.
Agents
6 million.
While this might be annoying to some, in general, its a good thing on public computers. Besides undoing any software installs a user might've tried, it also removes old cookies and temp files that might contain someone's personal info.
Most public uni. computers ive seen all have zip drives. If you want to download and save something, I suggest putting this to use. It is, after all, a public computer.
they dont ignite exactly on contact, its a highly exothermic reaction, but the needed component is powdered aluminum. This is probably what the long white trail following the orbiter was, as its altitude was much too high for contrails.
All in all, I dont think it was THAT bad. I kinda liked the little bugger. Sure the red on black was odd, but the effect was neat and it worked. The sound was pretty good too, since your ears were right by the speakers and it used true stereo sound with a fairly high sampling rate for the time.
while hydrogen is fairly common throughout space (obviously), theres really not that much free hydrogen floating around. Most of the kind that would be accessible with such a method is found in gaseous clouds, of which are nowhere near earth.
A few months ago in that very same office one of the UPS's begin emitting a high pitched whistle from within (described as a 'tea kettle' sound), likely very hot gases escaping.