This sounds sort of like the metal storm guns I saw on the Discovery Channel a few weeks ago (anybody else see this on "Future Weapons"?). No firing pin needed -- bullets are triggered via RF and lauch from the gun like mini-rockets. Pretty much zero recoil and rediculous firing rates (since you don't need to cycle the firing pin). All transmissions are inside the gun -- no exterior RF triggers. It seems it would be pretty easy to make the bullets coded at that point. Oh, and it's super reliable because there are no moving parts (except the bullet). I know some of you hate the RF technology, but pure electrics are far more reliable than a mechanical firing pin. Also having bullets that won't fire unless I've "unlocked" my gun with the appropriate password doesn't sound like too bad an idea for consumer use. The article is a little sketchy on details or even if these systems are related, but it sounds very similar -- like an extension of the technology.
Your experience seems to indicate there may be a market for native HD material that isn't massively compressed -- like HD_DVDs and Blu-Ray discs. I've come to the conclusion through many posts that comcast sucks (everybody say it together!) for HD content and that the guys at ciruit city and best buy couldn't properly set up an HD system to save their jobs (let's daisy chain 20 HD sets together and see what the picture looks like!). It's sad that our retailers are doing such a disservice to themselves. Maybe HD discs will change that by making good HD content easier to see and set up correctly.
You are exactly correct -- and this is how the inforamtion gathered is used the most effectively. Professional athletes develop muscle memory patters that can be quite predictable. Witness pitch counting in baseball (the best and simplest example I can think of). By counting what pitches have been thrown by a particular pitcher, and where, it can offer a significant advantage to a team versus having no data on a pitcher. That doesn't mean you'll know what he'll throw every time -- it means you have a better idea, down to a couple pitches (which is still a lot). The same can be done with any professional athlete -- particularly in "flowing" sports like soccer where decisions must be made in split seconds (versus paced out like baseball), it's just harder to collect and analyze the data. I think that soccer(football) is a more flowing game;) than American football but not that it makes it less relevant to statistical analysis (which I think is your point).
Don't give up on statistics yet. I have to agree, however, with a large part of what you're saying. All this statisitc mongering is really no better (or different) than sports betting in any casino taken to the nth degree. The cold hard fact is that most of the time -- they WILL be right, but good gamblers know that anything can happen -- it's much more than just the odds. Outliers are in every sport (even US football with instant replay - ugh!)-- that's what makes them sport! Statistics can be used in various degrees of effectiveness, but it's more in the intelligence of how the information is used. I honestly don't think it's a matter of the type of statistics -- they all can be effective or equally useless. The Las Vegas guys (and Monte Carlo and everywhere else) get it right an amazing percentage of the time based on statistics -- it doesn't matter if it's soccer or table tennis. Makes you wonder who they get their numbers from.
Not so. This software, and many more like it used by professional sports teams around the worls -- particularly in the US NFL football clubs -- analyzes patterns in video footage. It calculates STATISTICS based on how often certain formations are used, what attacks, defenses are constructed, and analyzes weakneses based on previous performances. It is not a real-time process. Even if it was a real-time process, it only allows you to see what the current pattern is -- it's up to the coach to decide how to exploit it. Think of it more as an analysis program that looks at past events. There is no guarantee such events will transpire the same in the future. If you are familiar with baseball they have used a similar analysis system on pitchers (and batters now) for YEARS. Just counting pitches, what's been thrown and where you can get a VERY good idea of what a particular pitcher will throw at any time based on previous patterns. It works VERY well, and EVERYBODY uses it. The trick is to not operate in predicatable patterns -- a trick that can be very difficult for highly trained athletes with well formed muscle memory to achieve. Australia's loss to Brasil today proves nothing. Only the coach and players (of the Aussie team) may realize the full effectiveness of the information. Seems to me they did OK (although a loss is a loss).
"Seriously, why do you think China would benefit greatly from electric cars? The population centers are very dense - so cars aren't too useful there, outside of the cities, the roads are not like America's and the majority of people who live outside the cities generally could not afford a card of any sort anyway."
Actually, dense population centers are the perfect location for electric cars because of their short range and (typically) very small size. It's the largest reason the EV1 failed here in the US -- short range, long recharge time. Even the "rapid" charges took over an hour to charge up, and there were serious safety issues with those types of systems. Most setups were an 8 hour charge with inductive paddles. In the mid-nineties I worked on an "intelligent sharing" electric vehicle concept program that is still in place (as far as I know) at University of California Riverside. A small fleet of electric vehicles that could be picked up by registered users and driven from one charge station to another (where someone elese might use the car). It was called "ICVS" or some such acronym (Intelligent COmmunity Vehicle System). It was funded by Honda with EV1, smartcard access and GPS units in each car. It was being researched for major poulation centers like in Japan (in particular) but with possible applications in large cities in the US. THe big problem: we like to drive our cars FAR (and fast). Pure-electric vehicles only made sense in small areas for short trips. Hybrids offer a MUCH better solution for the typical US driver. I'd agree China is probably still too poor for such a system, but it's not inconceivable to start seeing similar systems in Japan (already have), Europe, and even the US -- anywhere where the small size of electric vehicles (combined with community-use programs) would benefit city trips.
I noticed that too. I cruised by Target just for kicks yesterday to see if they had any -- they had a whole locked up shelf full, with no sign, fanfare or anything. Wouldn't have even known they had the Lite version unless you looked closely, but it was ALL they had. Lot's of 'em. I was somewhat surprised. I'll pick one up in a few weeks maybe.
It also has to do with ease and accuracy of scaling. Scaling 24 bit requires more bits than that to accomplish without errors. If you go to higher source bit amounts any degredation will still be beyond what is visible, hence no visible scaling artifacts in the color area.
" Hall effect sensors can easily measure current with no resistance added or interfering with the circuit.
No resistance? Even a hundred micro-ohms would be significant, and it takes a pretty impressive chunk of copper to get that low. Using a Hall effect sensor in the midst of a 100 amp circuit is tricky, too."
C'mon. Now we're simply talking board layout issues on one of those $250k ATEs (or at least the test socket for that part). A hall effect sensor is almost trivial in such a customized setup (not that it's not without effort, but for those used to dealing with such issues - designing test sockets - it's not a big deal). Test sockets and ATEs are already a necessity in producing the part. You speak as one who has experience in the semiconductor industry, so you know this isn't an extra cost. I've never claimed in my posts that it's not "difficult to gauge in the slighteset", but I think it's the benchmark that's the real issue. "No issues" is pretty subjective, but I assure you no one has a significant problem testing current or performance on 100W+ amplifiers and other high power ICs -- micro processors are nothing special in this regard. Maybe the particular IC manufacturers in question WANT it to be difficult. End conspiracy theory (and goodnight);)
Actually, I don't think it's that bad an example. The SAT and most standardized tests (like the LSAT) are time limited for a reason. With the LSAT, the test is designed specifically so you CAN'T finish all the questions unless you can read (and remember) with near photographic recognition. These tests are not just testing if you can get the right answer. It's wether you can budget your time, get the right answers, and complete the test. These are all parts of the result. Achieving results in a set amount of time is a critical metric in many fields. I agree with the GP that maybe this should be noted on the tests. I agree with all your statements regarding ADD and ritalin though (non-ADD users will react quite differently), but this issue goes beyond ritalin. Of course in reality, these test scores don't have a shit to do with your skills or how successful you will be. Just how well you took that particular test that day.
Hall effect sensors can easily measure current with no resistance added or interfering with the circuit. Many IC manufacturers test VERY complex integrated systems every day with NO issues -- and IC test setups use nowhere near 1ohm for current tests. Maybe 0.1 ohm. I think the real problem here is the lack of a standardized setup and method for measuring. Having different supply points or loads on a single chip is pretty trivial. The biggest difference is the performance -- what the chip is doing -- when it is measured. I'd almost think you'd want a ramp up of processor performance and measure current use while this happens. There is no guarantee with todays processors that performace is linear with power consumption. Either that or measure at standard "benchmark" levels.
I see you have released the magic smoke before! Tantalums and electrolytics always seem to fail in entertaining ways (often exploding). I've heard of manufacturers avoiding using these types of caps for this very reason. I thought the number of cycles reference was funny -- it's such a huge number when you think of it that way -- but others might be confused;)
I think cycle useage for a capacitor is somewhat useless. Age is a better value -- as from when the cap was made. AC/DC rectifier circuits don't allow full discharge of caps -- ripple smoothing is just that: smoothing. If you want to talk full charge/discharge that's another thing, but that's not how caps are (typically) used (for RC timer circuits). Generally they are blocking DC values or filtering high frequency values -- none of which require full charge/discharge. Different types of caps will obviously have different shelf lifes. What you'll typically see is a drift from the prescribed value. Outright failure is typically from an over-voltage condition.
"You mean the post that told you exactly how to find UWB info in your precious Wikipedia (when you were obviously stumped)? The post that you completely failed to comprehend even though it directly answered your question? Wow."
Original question(s):
"What's the security on UWB? How do you keep these things on seperate networks? What's the max number of connections?"
Really, I just cared about the security -- that was my point -- but regardless, here is your answer:
"A quick Google search for "UWB" && "ultra wide band" reveals a number of good sources, including Intel and the UWB Forum. But hey, since those apparently aren't good enough, Wikipedia is the second link on the first page of results. Go nuts."
Nowhere did you answer my question. You simply referred me to other sources of information (one of which I'd already referred) that anyone would use. Not the valuable sources I hope to discourse with here on/. Who's failing reading comprehension?!? You must make lots of friends this way. Thanks again. Karma, sir. Karma.
While I'd agree that systems in certain markets are integrating VERY fast (cell phones, PDAs), that's certainly not true everywhere -- and I can't see system integration keeping up it's breakneck pace -- particulary moore's law. The reason being: there are only so many applications for semi-custom integrated solutions. The cell phone industry really drives some of this, because of its rediculously huge market, but everyone want to be different. It's tough selling an integrated package when someone just wants to use a few functions of it. Large volume applications in particular. Thus, when we design an integrated system we have at least one major player lined up, and hopefully at 2. And then there's still no guarantee they will buy when it's done so we can recoup our million dollar investment in the design. We also look at more general market applications.
Nah... The devices this applies to are of the sort that if something busts, you replace the whole damn thing anyway. Integration on ICs is a whole 'nother can of worms than a TV with a DVD player in. The chips DON'T break. The mechanical connections break (traces, solder joints). Reliability actually can significantly improve.... Think of it more as the nodules you describe. The audio section craps out -- you replace the audio section chip. The baseband/uController craps out, you replace it. That doesn't happen, but it's like what you describe. Discretes in these devices are tested discretely, and then a few tests are run on a final product to ensure it works. Integrated systems are tested as a functioning whole, leading to much greater reliability, as well as fewer process variations (also improving reliability). This isn't about making things more complicated (although integration often results in "extra" features to justify the cost) but rather putting similar or close working components in the same package.
Even more confusing: there is a standard SO package for ICs already (small outline package or SOP). I'm surprised IEEE used this nomenclature. This type of integrated system setup is more typically called system on a chip, since it's typically several systems mapped onto one piece of silicon.
This is true. Integrating discrete components generally greatly improves reliability. Since what was seperate components now is made in one fab at one time and is tested and designed to work in such an integrated fashion -- typically it's FAR more reliable that discretes. In fact, that integrated IC has been through more testing than discretes ever would. It's justifying the cost of the integration that's the biggest issue. Sometimes it's tough to justify putting all those parts in one package without several prospective customers.
Actually smaller is worse. It's cheaper, yes, becasue it's less silicon area, but the smaller you get, the more issues you have in creating the chip, and coupling issues with high-speed lines in closer proximity, the less reliable it tends to be. The semiconductor industry is also continuously updating silicon to justify keeping prices the same level or higher. Since silicon cost goes down once its been created, that's the only way to increase ASP on silicon -- add more features in the same space. Thus we see cell phones that continue to get smaller and smaller, with more features, yet less reliability. Hopefully the continued integration process we see in cell phone ICs will provide more reliability by having many systems (now integrated) be made in one fab at one time, thus decreasing fab process differences.
"My thing is if they got camera/multimedia phones right like: -High quality camera with at leat 2MP or 1600x1200 -removable flash card that stores the pictures -external screen that acts like a viewfinder, great for taking pictures of yourself with someone -simple usb connectivity -an mp3 player that also works off the flash card that would work on a bluetooth stereo headset that auto pauses the music on a call. I could ride my bike with my phone in my bag listening to some tunes and take a call easily without fumbling with the phone or wires all over. -maybe a video service BUT it has to be based on open standars and can to connect to any video server even one I might host. Also simple video recording to the flash card that can record longer then 15 seconds. -good 3 + day standby time."
Every one of those features is on current smart phones by DoCoMo in asia. I've had a smart phone in my possesion for almost a year now that features a rotating flip screen that acts as a viewfinder when in cideo mode. The video camera is in the hinge of the phone. It's got a couple gig hard drive on board, and had an SD card slot. Takes several megapixel pictures, better than TV video. USB connection, mp3 player. We're about 2 years behind on phone tech in the US. Europe is somewhat better.
Mostly because it IS a reliable source of information. I could easily comb through the 500+ pages of UWB security standards, but I ask here on/. because I like the push-pull of human interaction and info and every once in a while you find someone who actually knows what they're talking about. And then soemtimes I don't. Like your useless post. You are the chief.
Everything is started to be integrated. Working for an IC manufacturer, I've seen our progress into integrated subsystems happen with great speed. It's essentially taking individual components and packaging them together in one IC. This works great for the right customer, but makes it hard for the general market sell. That's why you see more of this integration with cell phone manufacturers, since their (HUGE) markets can still drive innovations like this. I work in audio, and we've seen the audio systems rapidly integrate. DACs are being packaged with mixers and headphone/loudspeaker amps. Supply regulation may be onboard. I2C or SPI control may be available for your analog systems. As more features are being added to phones, more is being integrated into single ICs. This saves both space and money -- critical areas for cell phones. It's also a self-sustaining circle: making new chips means more features/integration which leads to further integration/features requiring more chips. You've always got to have something new and better in your phone to justify the increased cost vs basic no-nonsense silicon (which nobody wants for marketing reasons). You don't see "simpler more reliable" phones. They're always better (subjective), smaller, and more feature laden. Oh, and they're built to last a max of two years. Planned obsolesence at its finest.
on UWB? How do you keep these things on seperate networks? What's the max number of connections? Anybody know this yet? I can't find it on wikipedia...
" Again, your posting (modded +3 -- nice job slashdotters) shows little experience with this in reality.
I hope you're not talking about THIS comment - it hasn't been moderated. If you're talking about my grandparent comment, well, it's a completely subjective comparison. Either you think that HD is worth it, or you don't. I don't, and the group (FWIW) doesn't think so either, obviously, if my net moderation is positive. Don't get all butt-hurt just because you're wrong. It's tedious. "
Well, I'll submit that I must be in the minority here on this -- but more as a factor of having seen true HD, and I'll still maintain most people here haven't. My experiences (and others I've observed) say otherwise. My comment, however, comes from the fact that every one of your reasons (cost, no discernable difference) is untrue, yet you still get modded up (all your posts). I'll go cry in my beer somewhere else! To each their own! Cheers!
Whoops -- I should have said mastered. You are certainly correct that current film video formats are huge. Unfortuantely, not as much is being shot directly in high-def digital formats (yet) but it is all still transferred to digital for editing/mastering. I'm sure some industry person will correct me and detail. And to only see that in 480 lines? It's garbage compared to the real thing.
Slashdot Mirror
This sounds sort of like the metal storm guns I saw on the Discovery Channel a few weeks ago (anybody else see this on "Future Weapons"?). No firing pin needed -- bullets are triggered via RF and lauch from the gun like mini-rockets. Pretty much zero recoil and rediculous firing rates (since you don't need to cycle the firing pin). All transmissions are inside the gun -- no exterior RF triggers. It seems it would be pretty easy to make the bullets coded at that point.
Oh, and it's super reliable because there are no moving parts (except the bullet). I know some of you hate the RF technology, but pure electrics are far more reliable than a mechanical firing pin. Also having bullets that won't fire unless I've "unlocked" my gun with the appropriate password doesn't sound like too bad an idea for consumer use.
The article is a little sketchy on details or even if these systems are related, but it sounds very similar -- like an extension of the technology.
Your experience seems to indicate there may be a market for native HD material that isn't massively compressed -- like HD_DVDs and Blu-Ray discs. I've come to the conclusion through many posts that comcast sucks (everybody say it together!) for HD content and that the guys at ciruit city and best buy couldn't properly set up an HD system to save their jobs (let's daisy chain 20 HD sets together and see what the picture looks like!). It's sad that our retailers are doing such a disservice to themselves. Maybe HD discs will change that by making good HD content easier to see and set up correctly.
You are exactly correct -- and this is how the inforamtion gathered is used the most effectively. Professional athletes develop muscle memory patters that can be quite predictable. Witness pitch counting in baseball (the best and simplest example I can think of). By counting what pitches have been thrown by a particular pitcher, and where, it can offer a significant advantage to a team versus having no data on a pitcher. That doesn't mean you'll know what he'll throw every time -- it means you have a better idea, down to a couple pitches (which is still a lot). The same can be done with any professional athlete -- particularly in "flowing" sports like soccer where decisions must be made in split seconds (versus paced out like baseball), it's just harder to collect and analyze the data. ;) than American football but not that it makes it less relevant to statistical analysis (which I think is your point).
I think that soccer(football) is a more flowing game
Don't give up on statistics yet. I have to agree, however, with a large part of what you're saying. All this statisitc mongering is really no better (or different) than sports betting in any casino taken to the nth degree. The cold hard fact is that most of the time -- they WILL be right, but good gamblers know that anything can happen -- it's much more than just the odds.
Outliers are in every sport (even US football with instant replay - ugh!)-- that's what makes them sport! Statistics can be used in various degrees of effectiveness, but it's more in the intelligence of how the information is used. I honestly don't think it's a matter of the type of statistics -- they all can be effective or equally useless.
The Las Vegas guys (and Monte Carlo and everywhere else) get it right an amazing percentage of the time based on statistics -- it doesn't matter if it's soccer or table tennis. Makes you wonder who they get their numbers from.
Not so. This software, and many more like it used by professional sports teams around the worls -- particularly in the US NFL football clubs -- analyzes patterns in video footage. It calculates STATISTICS based on how often certain formations are used, what attacks, defenses are constructed, and analyzes weakneses based on previous performances. It is not a real-time process. Even if it was a real-time process, it only allows you to see what the current pattern is -- it's up to the coach to decide how to exploit it.
Think of it more as an analysis program that looks at past events. There is no guarantee such events will transpire the same in the future.
If you are familiar with baseball they have used a similar analysis system on pitchers (and batters now) for YEARS. Just counting pitches, what's been thrown and where you can get a VERY good idea of what a particular pitcher will throw at any time based on previous patterns. It works VERY well, and EVERYBODY uses it. The trick is to not operate in predicatable patterns -- a trick that can be very difficult for highly trained athletes with well formed muscle memory to achieve.
Australia's loss to Brasil today proves nothing. Only the coach and players (of the Aussie team) may realize the full effectiveness of the information. Seems to me they did OK (although a loss is a loss).
"Seriously, why do you think China would benefit greatly from electric cars? The population centers are very dense - so cars aren't too useful there, outside of the cities, the roads are not like America's and the majority of people who live outside the cities generally could not afford a card of any sort anyway."
Actually, dense population centers are the perfect location for electric cars because of their short range and (typically) very small size. It's the largest reason the EV1 failed here in the US -- short range, long recharge time. Even the "rapid" charges took over an hour to charge up, and there were serious safety issues with those types of systems. Most setups were an 8 hour charge with inductive paddles.
In the mid-nineties I worked on an "intelligent sharing" electric vehicle concept program that is still in place (as far as I know) at University of California Riverside. A small fleet of electric vehicles that could be picked up by registered users and driven from one charge station to another (where someone elese might use the car). It was called "ICVS" or some such acronym (Intelligent COmmunity Vehicle System). It was funded by Honda with EV1, smartcard access and GPS units in each car. It was being researched for major poulation centers like in Japan (in particular) but with possible applications in large cities in the US. THe big problem: we like to drive our cars FAR (and fast). Pure-electric vehicles only made sense in small areas for short trips. Hybrids offer a MUCH better solution for the typical US driver.
I'd agree China is probably still too poor for such a system, but it's not inconceivable to start seeing similar systems in Japan (already have), Europe, and even the US -- anywhere where the small size of electric vehicles (combined with community-use programs) would benefit city trips.
I noticed that too. I cruised by Target just for kicks yesterday to see if they had any -- they had a whole locked up shelf full, with no sign, fanfare or anything. Wouldn't have even known they had the Lite version unless you looked closely, but it was ALL they had. Lot's of 'em. I was somewhat surprised. I'll pick one up in a few weeks maybe.
It also has to do with ease and accuracy of scaling. Scaling 24 bit requires more bits than that to accomplish without errors. If you go to higher source bit amounts any degredation will still be beyond what is visible, hence no visible scaling artifacts in the color area.
" Hall effect sensors can easily measure current with no resistance added or interfering with the circuit.
No resistance? Even a hundred micro-ohms would be significant, and it takes a pretty impressive chunk of copper to get that low. Using a Hall effect sensor in the midst of a 100 amp circuit is tricky, too."
C'mon. Now we're simply talking board layout issues on one of those $250k ATEs (or at least the test socket for that part). A hall effect sensor is almost trivial in such a customized setup (not that it's not without effort, but for those used to dealing with such issues - designing test sockets - it's not a big deal).
Test sockets and ATEs are already a necessity in producing the part. You speak as one who has experience in the semiconductor industry, so you know this isn't an extra cost.
I've never claimed in my posts that it's not "difficult to gauge in the slighteset", but I think it's the benchmark that's the real issue. "No issues" is pretty subjective, but I assure you no one has a significant problem testing current or performance on 100W+ amplifiers and other high power ICs -- micro processors are nothing special in this regard. Maybe the particular IC manufacturers in question WANT it to be difficult.
End conspiracy theory (and goodnight);)
Actually, I don't think it's that bad an example. The SAT and most standardized tests (like the LSAT) are time limited for a reason. With the LSAT, the test is designed specifically so you CAN'T finish all the questions unless you can read (and remember) with near photographic recognition. These tests are not just testing if you can get the right answer. It's wether you can budget your time, get the right answers, and complete the test. These are all parts of the result. Achieving results in a set amount of time is a critical metric in many fields. I agree with the GP that maybe this should be noted on the tests. I agree with all your statements regarding ADD and ritalin though (non-ADD users will react quite differently), but this issue goes beyond ritalin.
Of course in reality, these test scores don't have a shit to do with your skills or how successful you will be. Just how well you took that particular test that day.
Hall effect sensors can easily measure current with no resistance added or interfering with the circuit. Many IC manufacturers test VERY complex integrated systems every day with NO issues -- and IC test setups use nowhere near 1ohm for current tests. Maybe 0.1 ohm. I think the real problem here is the lack of a standardized setup and method for measuring. Having different supply points or loads on a single chip is pretty trivial. The biggest difference is the performance -- what the chip is doing -- when it is measured. I'd almost think you'd want a ramp up of processor performance and measure current use while this happens. There is no guarantee with todays processors that performace is linear with power consumption. Either that or measure at standard "benchmark" levels.
I see you have released the magic smoke before! Tantalums and electrolytics always seem to fail in entertaining ways (often exploding). I've heard of manufacturers avoiding using these types of caps for this very reason. ;)
I thought the number of cycles reference was funny -- it's such a huge number when you think of it that way -- but others might be confused
I think cycle useage for a capacitor is somewhat useless. Age is a better value -- as from when the cap was made. AC/DC rectifier circuits don't allow full discharge of caps -- ripple smoothing is just that: smoothing. If you want to talk full charge/discharge that's another thing, but that's not how caps are (typically) used (for RC timer circuits). Generally they are blocking DC values or filtering high frequency values -- none of which require full charge/discharge.
Different types of caps will obviously have different shelf lifes. What you'll typically see is a drift from the prescribed value. Outright failure is typically from an over-voltage condition.
"You mean the post that told you exactly how to find UWB info in your precious Wikipedia (when you were obviously stumped)? The post that you completely failed to comprehend even though it directly answered your question? Wow."
/. Who's failing reading comprehension?!? You must make lots of friends this way. Thanks again.
Original question(s):
"What's the security on UWB? How do you keep these things on seperate networks? What's the max number of connections?"
Really, I just cared about the security -- that was my point -- but regardless, here is your answer:
"A quick Google search for "UWB" && "ultra wide band" reveals a number of good sources, including Intel and the UWB Forum. But hey, since those apparently aren't good enough, Wikipedia is the second link on the first page of results. Go nuts."
Nowhere did you answer my question. You simply referred me to other sources of information (one of which I'd already referred) that anyone would use. Not the valuable sources I hope to discourse with here on
Karma, sir. Karma.
While I'd agree that systems in certain markets are integrating VERY fast (cell phones, PDAs), that's certainly not true everywhere -- and I can't see system integration keeping up it's breakneck pace -- particulary moore's law. The reason being: there are only so many applications for semi-custom integrated solutions. The cell phone industry really drives some of this, because of its rediculously huge market, but everyone want to be different. It's tough selling an integrated package when someone just wants to use a few functions of it. Large volume applications in particular. Thus, when we design an integrated system we have at least one major player lined up, and hopefully at 2. And then there's still no guarantee they will buy when it's done so we can recoup our million dollar investment in the design. We also look at more general market applications.
"Integration stinks"
Nah...
The devices this applies to are of the sort that if something busts, you replace the whole damn thing anyway. Integration on ICs is a whole 'nother can of worms than a TV with a DVD player in. The chips DON'T break. The mechanical connections break (traces, solder joints). Reliability actually can significantly improve....
Think of it more as the nodules you describe. The audio section craps out -- you replace the audio section chip. The baseband/uController craps out, you replace it. That doesn't happen, but it's like what you describe. Discretes in these devices are tested discretely, and then a few tests are run on a final product to ensure it works. Integrated systems are tested as a functioning whole, leading to much greater reliability, as well as fewer process variations (also improving reliability). This isn't about making things more complicated (although integration often results in "extra" features to justify the cost) but rather putting similar or close working components in the same package.
Even more confusing: there is a standard SO package for ICs already (small outline package or SOP). I'm surprised IEEE used this nomenclature. This type of integrated system setup is more typically called system on a chip, since it's typically several systems mapped onto one piece of silicon.
This is true. Integrating discrete components generally greatly improves reliability. Since what was seperate components now is made in one fab at one time and is tested and designed to work in such an integrated fashion -- typically it's FAR more reliable that discretes. In fact, that integrated IC has been through more testing than discretes ever would. It's justifying the cost of the integration that's the biggest issue. Sometimes it's tough to justify putting all those parts in one package without several prospective customers.
Actually smaller is worse. It's cheaper, yes, becasue it's less silicon area, but the smaller you get, the more issues you have in creating the chip, and coupling issues with high-speed lines in closer proximity, the less reliable it tends to be.
The semiconductor industry is also continuously updating silicon to justify keeping prices the same level or higher. Since silicon cost goes down once its been created, that's the only way to increase ASP on silicon -- add more features in the same space.
Thus we see cell phones that continue to get smaller and smaller, with more features, yet less reliability.
Hopefully the continued integration process we see in cell phone ICs will provide more reliability by having many systems (now integrated) be made in one fab at one time, thus decreasing fab process differences.
"My thing is if they got camera/multimedia phones right like:
-High quality camera with at leat 2MP or 1600x1200
-removable flash card that stores the pictures
-external screen that acts like a viewfinder, great for taking pictures of yourself with someone
-simple usb connectivity
-an mp3 player that also works off the flash card that would work on a bluetooth stereo headset that auto pauses the music on a call. I could ride my bike with my phone in my bag listening to some tunes and take a call easily without fumbling with the phone or wires all over.
-maybe a video service BUT it has to be based on open standars and can to connect to any video server even one I might host. Also simple video recording to the flash card that can record longer then 15 seconds.
-good 3 + day standby time."
Every one of those features is on current smart phones by DoCoMo in asia. I've had a smart phone in my possesion for almost a year now that features a rotating flip screen that acts as a viewfinder when in cideo mode. The video camera is in the hinge of the phone. It's got a couple gig hard drive on board, and had an SD card slot. Takes several megapixel pictures, better than TV video. USB connection, mp3 player.
We're about 2 years behind on phone tech in the US. Europe is somewhat better.
Mostly because it IS a reliable source of information. I could easily comb through the 500+ pages of UWB security standards, but I ask here on /. because I like the push-pull of human interaction and info and every once in a while you find someone who actually knows what they're talking about.
And then soemtimes I don't. Like your useless post. You are the chief.
Everything is started to be integrated. Working for an IC manufacturer, I've seen our progress into integrated subsystems happen with great speed. It's essentially taking individual components and packaging them together in one IC. This works great for the right customer, but makes it hard for the general market sell. That's why you see more of this integration with cell phone manufacturers, since their (HUGE) markets can still drive innovations like this.
I work in audio, and we've seen the audio systems rapidly integrate. DACs are being packaged with mixers and headphone/loudspeaker amps. Supply regulation may be onboard. I2C or SPI control may be available for your analog systems. As more features are being added to phones, more is being integrated into single ICs. This saves both space and money -- critical areas for cell phones. It's also a self-sustaining circle: making new chips means more features/integration which leads to further integration/features requiring more chips. You've always got to have something new and better in your phone to justify the increased cost vs basic no-nonsense silicon (which nobody wants for marketing reasons). You don't see "simpler more reliable" phones. They're always better (subjective), smaller, and more feature laden. Oh, and they're built to last a max of two years. Planned obsolesence at its finest.
on UWB? How do you keep these things on seperate networks? What's the max number of connections? Anybody know this yet? I can't find it on wikipedia...
" Again, your posting (modded +3 -- nice job slashdotters) shows little experience with this in reality.
I hope you're not talking about THIS comment - it hasn't been moderated. If you're talking about my grandparent comment, well, it's a completely subjective comparison. Either you think that HD is worth it, or you don't. I don't, and the group (FWIW) doesn't think so either, obviously, if my net moderation is positive. Don't get all butt-hurt just because you're wrong. It's tedious. "
Well, I'll submit that I must be in the minority here on this -- but more as a factor of having seen true HD, and I'll still maintain most people here haven't. My experiences (and others I've observed) say otherwise. My comment, however, comes from the fact that every one of your reasons (cost, no discernable difference) is untrue, yet you still get modded up (all your posts).
I'll go cry in my beer somewhere else! To each their own! Cheers!
Whoops -- I should have said mastered. You are certainly correct that current film video formats are huge. Unfortuantely, not as much is being shot directly in high-def digital formats (yet) but it is all still transferred to digital for editing/mastering. I'm sure some industry person will correct me and detail.
And to only see that in 480 lines? It's garbage compared to the real thing.