This could be interesting, and the total logistics impact is staggering.
When the post office first started using letter sorters, there were those times when the system glitched and a letter clogged the sorter path. After a few hundred letters piled onto the stack, the sorter stopped.
Now consider a conveyer moving at 1200 feet per minute. If one package redirector misses, the system will have to stop for manual intervention. So how fast can the conveyer be stopped to avoid slipping and pile-up of other packages. Then how long will it take to un-pile those items that were moving at 1200 fpm and get the conveyer started again. It's like taking a trip. If you have to stop for gas and rest-room, the average trip speed is greatly reduced.
How does Wal-mart plan to get packages onto the conveyer at that rate? It will take twice as many people and twice the number of conveyer entry points and exit points. Then they will have to double the number of people doing the final stocking to shelves or taking to trucks. Also the number of trucks entering and leaving the warehouse will have to double and the roads will have to handle the increased traffic.
RFID speeds things all along the route and will allow much faster distribution, especially perishables like fruit and vegetables, and that also translates to less refridgeration time and lower cost in keeping environments cool or hot or in special gasses to control ripening rate.
Now consider what happens at the high speed checkout when one of the items registers as alcohol and the buyer is less than proper age. The line manager will be over helping at the cashierless line since the stupid system stops because the weight is not what it expected, and if you think Wal-mart is going to add another line manager just so you can get through faster...!
One of my fears is that with the new handling speed the bananas will be too green to eat and I will have to buy them two days early.
This never ends. When I first started work in the early '70s, an editorial titled "Design Engineers wanted. 10 year's experience. Older men need not apply." caught my attention. Company's want experience, but they also don't want to pay for it.
At that time, Texas Instruments was still hiring mostly new graduates and working them overtime (unpaid) for four years until they left voluntarily and that allowed TI to hire fresh new talent with the latest education.
Companies need experience, but they also need the fresh new talent and work philosophy of new graduates so that someone will be around to "keep the mainframes" running.
"before we even get to comparing men vs. women, we need to define what "intelligence" is and how to measure it.
Untill those 'blah', 'foo' and 'x' are defined the statement will make no sense."
"Blah" is the sound one makes after putting bad tasting food into his mouth.
"Foo" is the word an older woman uses when her husband trys to tell her something that she thinks is not true, as "Oh, foo to you!"
"x",unfortunately, is an unknown, and can never be defined, meaning that that we will be forced into an eternity of not knowing what intelligence really is.
Re:...the same features we delivered seven years a
on
Windows 95 Turns 10
·
· Score: 1
Have you yet stopped to ask, "How can I prove that what I just wrote is really true? Do I have any testing or data to back me up?"
Like you, I have a lot of opinions, but if I don't have data to support what I am saying, then it is just an opinion. I might even be right, but just because I (or you in this case) like something a lot doesn't make it better or best.
SO!
Without supporting data other than our own opinions, we are making it up and dishing it out. What else can I say?
As the old saying goes, "All generalities, including this one, are untrue."
You have made a number of statements that sound good to most engineers,including me, but still has no supporting data, at least none I have seen. Adding fuses and protection devices cannot increase reliability. They only protect after a failure has occurred. Extra bypass caps are added to reduce stress ratio in each, but if the failure is a short condition, you have two devices that can fail instead of one. Redundant circuits can increase the up-time, since up-time is more important than the mil 217 calculations of reliability.
Fans don't fail all the time. Only poorly designed fans fail, and two of those do not increase reliability. One good fan will increase reliability.
I would like to rephrase your last lines. You can overpay for a good PSU but you will always overpay for a poorly designed PSU. Like every component in a low cost computer, you can expect to get what you pay for.
I like the idea of the four legged version. A three foot high dog; snarling, barking loudly enough for the neighbors to hear, and then firing off one or two rounds of ammo, just might make a burglar think twice. You could probably track the burglar by the odor trail running out of his pants.
If it is controlled by Windows, I would not enter my own house without making sure the "dog" had not turned into a wild trojan horse.
Re:...the same features we delivered seven years a
on
Windows 95 Turns 10
·
· Score: 1
"This is, in fact, far ahead of anything currently available on Unix or Windows.".......
"Ya, when you're making shit up you can pump it out like a champ."
It's finally happened! SCO is not even good enough to make it into the "bad guy" list. They did not even make it into the "linux good" MS bad comparison. I'll bet their stock jumps back up to more than zero.
Any good hardware systems designer knows that a system as you describe is not condsidered as having a redundant power supply.
And if the plan is to continue operation after one of three dies, then the design criteria is to specify a power supply size such that the other two are capable of continuous operation at full system load and at maximum air temperature. If it does not do that, then it is not a redundant system. It is just a three power supply system.
It must be true. You need two video cards to play 3-D video games.
Although you are somewhat right, your guesswork about power supply quality is mostly quessing. Higher power rating and higher cost do not contribute to quality. And the more iternal components you have, the lower your reliability. It's true that bigger heatsinks weigh more if both are from the same material, and larger transformers weigh more than smaller ones. However a good fan offsets the heatsink size, a larger heatsink allows a slower and quieter fan. A heavy steel case is no better than a lightweight aluminum case, but one weighs more. Your one statement is mostly true. A higher quality product generally costs more than a lower cost one, but with power supplies that is not necessarily so.
I have designed power supplies for PCs and other computer systems. Most of the world requires that power supplies of this size have PFC, (power factor correction). One of the things that require it is the need to limit conducted electrical interference on the power lines. Power supplies without PFC cannot meet the requirements in most cases. The low power (up to 350 or 400 watts) can meet requirements without PFC.
Power supplies use switching technology, so the actual power drawn from the wall in in proportion to the power needed in the system. Most switching power supplies have an efficiency of around 85 percent,and that will vary about 5 percent up or down based on load. PFC suplies are a little less efficient since there is some additional internal power needed to power the PFC circuit. The difference is about 5 percent.
One application for this supply may be to power the hard drives of a SCSI raid system which can have up to 32 drives in parallel. (Note that SCSI drives can be controlled to start up under processor control so all do not start at once.)
One problem with high power supplies is that the connector to the mother board can only carry a limited power before it overheats at the connector pins and frys the MB and cable connector.
I expect the need for 1000 watts is mostly to insure that there is power to spare for a reliable system. If you run a component at half or less of its rated power the reliability figures (can) go up dramatically, and many companies demand high reliability.
The cost of components for 1000 watts vs 500 watts is minimal since both must use the circuit topology is the same for either power, and the number of components is about the same either way. Main difference is the size of fans and heatsinks in the PS along with the magnetics.
I understand that Google uses a very efficient compression technology to compress documents before they are indexed, thereby making characters so small that they can only be read with a magnifying glass or microscope.
In contrast, Yahoo, unless I misunderstand, only compresses the file after it has been indexed. Since only the file is compressed and not the individual characters, they indeed have a larger index file as the study concluded.:)
You must be very very young or very uninformed. MS has always had competition. Apple was there all along, and so was IBM, Honeywell, NCR and other large companies.
However, MS started deceptive and illegal business practices from early on by even having the "install" program say that non-MS based DOS would cause the computer to crash. Later, MS avoided competition by not allowing vendors to install other competitor software, and either would not sell Windows to them or would charge much higher prices to vendors who also sold computers with non-MS software on them. MS would not allow companies to install competitor's software on PC's that had Windows installed. When you have managed to build a company to the size that they have a very high market share, and then still do not allow competition, your company becomes a monolopy.
Right now, MS is still watched very closely by the government to make sure that it does not return to those really "bad" proctices.
Unfortunately, under the Republican administration, MS has been able to continue many practices that were considered monopolistic.
Slashdot Mirror
I would welcome a wide berth for some of the things I have not yet done! Heck, I might even settle for narrow.!
What was?
He used 'ambiguity' and 'terminate'. With words like that, he can't be one of us Alabama folks.
I still think Beowulf was a writer. How did things get so out-of-hand. (no out-of-hand comments allowed.)
When the post office first started using letter sorters, there were those times when the system glitched and a letter clogged the sorter path. After a few hundred letters piled onto the stack, the sorter stopped.
Now consider a conveyer moving at 1200 feet per minute. If one package redirector misses, the system will have to stop for manual intervention. So how fast can the conveyer be stopped to avoid slipping and pile-up of other packages. Then how long will it take to un-pile those items that were moving at 1200 fpm and get the conveyer started again. It's like taking a trip. If you have to stop for gas and rest-room, the average trip speed is greatly reduced.
How does Wal-mart plan to get packages onto the conveyer at that rate? It will take twice as many people and twice the number of conveyer entry points and exit points. Then they will have to double the number of people doing the final stocking to shelves or taking to trucks. Also the number of trucks entering and leaving the warehouse will have to double and the roads will have to handle the increased traffic.
RFID speeds things all along the route and will allow much faster distribution, especially perishables like fruit and vegetables, and that also translates to less refridgeration time and lower cost in keeping environments cool or hot or in special gasses to control ripening rate.
Now consider what happens at the high speed checkout when one of the items registers as alcohol and the buyer is less than proper age. The line manager will be over helping at the cashierless line since the stupid system stops because the weight is not what it expected, and if you think Wal-mart is going to add another line manager just so you can get through faster...!
One of my fears is that with the new handling speed the bananas will be too green to eat and I will have to buy them two days early.
I did not know that!
Those are Eskimos from France, aren't they?
Companies need experience, but they also need the fresh new talent and work philosophy of new graduates so that someone will be around to "keep the mainframes" running.
Untill those 'blah', 'foo' and 'x' are defined the statement will make no sense."
"Blah" is the sound one makes after putting bad tasting food into his mouth.
"Foo" is the word an older woman uses when her husband trys to tell her something that she thinks is not true, as "Oh, foo to you!"
"x",unfortunately, is an unknown, and can never be defined, meaning that that we will be forced into an eternity of not knowing what intelligence really is.
Build your own
It don't work.
You say "Damn Recorder!"
Well, somebody is fast!
Set the clock on your computer. You are a few minutes fast. http://www.karenware.com/powertools/ptsync.asp
Like you, I have a lot of opinions, but if I don't have data to support what I am saying, then it is just an opinion. I might even be right, but just because I (or you in this case) like something a lot doesn't make it better or best.
SO!
Without supporting data other than our own opinions, we are making it up and dishing it out. What else can I say?
You have made a number of statements that sound good to most engineers,including me, but still has no supporting data, at least none I have seen. Adding fuses and protection devices cannot increase reliability. They only protect after a failure has occurred. Extra bypass caps are added to reduce stress ratio in each, but if the failure is a short condition, you have two devices that can fail instead of one. Redundant circuits can increase the up-time, since up-time is more important than the mil 217 calculations of reliability.
Fans don't fail all the time. Only poorly designed fans fail, and two of those do not increase reliability. One good fan will increase reliability.
I would like to rephrase your last lines. You can overpay for a good PSU but you will always overpay for a poorly designed PSU. Like every component in a low cost computer, you can expect to get what you pay for.
If it is controlled by Windows, I would not enter my own house without making sure the "dog" had not turned into a wild trojan horse.
"Ya, when you're making shit up you can pump it out like a champ."
Are any other comments needed?
Wait......maybe that should be three-letter-word profanity.
A little dose of CX717 and you would have been alert enough to avoid making that rude remark in the first place.
It's finally happened! SCO is not even good enough to make it into the "bad guy" list. They did not even make it into the "linux good" MS bad comparison. I'll bet their stock jumps back up to more than zero.
And if the plan is to continue operation after one of three dies, then the design criteria is to specify a power supply size such that the other two are capable of continuous operation at full system load and at maximum air temperature. If it does not do that, then it is not a redundant system. It is just a three power supply system.
Although you are somewhat right, your guesswork about power supply quality is mostly quessing. Higher power rating and higher cost do not contribute to quality. And the more iternal components you have, the lower your reliability. It's true that bigger heatsinks weigh more if both are from the same material, and larger transformers weigh more than smaller ones. However a good fan offsets the heatsink size, a larger heatsink allows a slower and quieter fan. A heavy steel case is no better than a lightweight aluminum case, but one weighs more. Your one statement is mostly true. A higher quality product generally costs more than a lower cost one, but with power supplies that is not necessarily so.
Power supplies use switching technology, so the actual power drawn from the wall in in proportion to the power needed in the system. Most switching power supplies have an efficiency of around 85 percent,and that will vary about 5 percent up or down based on load. PFC suplies are a little less efficient since there is some additional internal power needed to power the PFC circuit. The difference is about 5 percent.
One application for this supply may be to power the hard drives of a SCSI raid system which can have up to 32 drives in parallel. (Note that SCSI drives can be controlled to start up under processor control so all do not start at once.)
One problem with high power supplies is that the connector to the mother board can only carry a limited power before it overheats at the connector pins and frys the MB and cable connector.
I expect the need for 1000 watts is mostly to insure that there is power to spare for a reliable system. If you run a component at half or less of its rated power the reliability figures (can) go up dramatically, and many companies demand high reliability.
The cost of components for 1000 watts vs 500 watts is minimal since both must use the circuit topology is the same for either power, and the number of components is about the same either way. Main difference is the size of fans and heatsinks in the PS along with the magnetics.
In contrast, Yahoo, unless I misunderstand, only compresses the file after it has been indexed. Since only the file is compressed and not the individual characters, they indeed have a larger index file as the study concluded. :)
Yes, and I suppose you think the spare tire in your car is free just because you did not have to order it seperately.
However, MS started deceptive and illegal business practices from early on by even having the "install" program say that non-MS based DOS would cause the computer to crash. Later, MS avoided competition by not allowing vendors to install other competitor software, and either would not sell Windows to them or would charge much higher prices to vendors who also sold computers with non-MS software on them. MS would not allow companies to install competitor's software on PC's that had Windows installed. When you have managed to build a company to the size that they have a very high market share, and then still do not allow competition, your company becomes a monolopy.
Right now, MS is still watched very closely by the government to make sure that it does not return to those really "bad" proctices.
Unfortunately, under the Republican administration, MS has been able to continue many practices that were considered monopolistic.