The RIAA/MPAA/etc. have been making fortunes off dead people's backs for decades, it would be a logical next step to eventually extend this to dead customers.
I think the same too... for me, speaking while driving is just as much of a distraction on the phone or in person. Actually, in-person might be the bigger distraction for me since conversations can start randomly and unnecessary warnings usually distract me quite a bit.
On the phone, nobody can talk to me until I take the call and I have the option of taking calls only while I am comfortable with current driving conditions.
Re:Tablespork, you must have been the only one
on
Apple Updates PowerBooks
·
· Score: 2, Insightful
OS/2 2.0 was a joint project between Microsoft and IBM. When the two "divorced", IBM continued to use the OS/2 brand and Microsoft forked its side of things into what would become NT's starting point.
Having common origins does not mean one has to stick to the original's APIs, file systems and other implementation details. The current Linux kernel probably looks and behaves like it has no relation with the 1.0 code as well.
How is dropping "all DOS support" going to make Longhorn any more native than any NT has ever been? WinNT has its roots in OS/2, not DOS.
The "DOS support" is only the DOS virtual machine. Virtual machines are just OS plug-ins to support alternative runtime environments (may include platform emulation) and executable formats, adding and removing VMs does not make the rest of the OS any more or less native. Dropping DOS support is simply a matter of dropping that VM, no big deal unless you use some peerless/irreplaceable DOS apps.
Nocona has a complete AMD64-compatible instruction set modulus two or three minor discrepancies which Intel claims were due to flaws in AMD's specifications.
Nocona can run the latest WinXP-64 betas and most AMD64 software, not limited to simple address space expansion even though this is what "EMT64" might cause one to believe.
The reason why Intel is only promoting the extended addressing range is because it wants uninformed customers to believe Itanium is the only true 64bits ticket from Intel.
I have been fighting off a pretty bad cold for nearly a week and this is apparently giving me some mild hallucinations... I apparently hallucinated the period before "Stealing" and the capital letter one word to the left - at the time of writing, I really thought I read "Physically stealing physical property(...)", doh!
AES lends itself fairly well to both ASIC/hardware and software implementations. Because we are talking about cryptographic messages most likely in the sub-kilobit size range, the amount of processing in question is fairly limited.
I remember about at least one company advertising RFID tag microcontrollers. The rest is a simple matter of balancing power and time... and since the RFID microcontroller can start processing before the key is in the ignition switch, a processing delay up to a few seconds should be acceptable, allowing the microcontroller to run its core at most likely less than 100kHz or even less than 10kHz if the chip contains dedicated AES logic - we are taking 8bits microcontrollers here.
The only reason why RFID tags are the only thing we commonly see is because demand for tags far exceeds demand for everything else that could possibly be handled by RFID techniques. If demand for AES-enabled RFID microcontrollers becomes large enough, microcontroller companies will make them.
BTW, the RFID microcontroller summary did mention that an external capacitor was necessary to smooth the power but I do not remember the rest.
As far as size is concerned, keep in mind that typical microcontrollers contain well under a milion transistors so a microcontroller suitable for secure authentication for an ignition system should be well under 10 square milimeters on 180nm process.
In the RIAA/MPAA's dream world, they would manage to stop unlicensed reproduction with DRM. In the real world, DRM only managed to slow down casual copying - at least up to now.
Personally, I do not buy into new technology until I have an interoperability guarantee... like DeCSS and derrivative works before buying my first DVD.
But really, the "proper" 686 is the Pentium Pro and it is the Pentium Pro's feature set that was used as the foundation for the P2 (basically a P-Pro+MMX) and P3 series.
What the PPro brought to the table compared to the plain old Pentium was an independent cache bus, an extra execution pipeline, out-of-order execution and a bunch of other tweaks common in other architectures.
eXeem simply makes everyone a tracker for what files people are downloading. On the legal side of things, it seems eXeem is a guarantee that the tracker is also a downloader or seeder and therefore more legally problematic for the hosts than plain trackers.
After all, even with eXeem, some initial tracker must be contacted before clients can start tracking on each other because they initially only know about the tracker(s) from the.torrent file.
Actually, Microsoft Office's XML exporter has two or three different behaviors depending on which version is used. IIRC, only the Professional and Enterprise editions export clean XML while the Personal editions only save in some proprietary description.
Artificial market segmentation appears to have become a primary hobby at Microsoft. First there was XP Home and Pro but now there is also Starter, MediaCenter and TabletPC editions. It really bugs me how MS labels the standard edition "Pro" and how it artificially cripples all other editions only for product "diversification" and the ability to ask $200 more for the standard edition.
Remember how the P3-Coppermine ran into a brick wall that prompted a recall at 1.13GHz?
The reason why the P4s have such huge pipelines is to meet timing margins required for high clock speeds. Pipelining helps by breaking down combinational logic blobs into sub-clock-tick-sized (settling time) logic bites and this costs glue-logic/clocking transistors, power and time efficiency.
The Pentium-M is designed to do the most work with the shortest pipeline on each clock tick using the least power. Because of its short pipeline, each stage (combinational logic blob) can have longer settling times. Since the pipeline cannot run faster than its slowest stage, the Pentium-M cannot scale much in frequency at stock voltage.
You will never be able to clock a PM at the same rates as the P4s - no matter how much extra power you throw at the PM, timing margins dictated by internal geometry will make it impossible just as it did for the 1.13GHz Coppermine P3.
At least this is one step in the generally right direction... but might mean I should have said "stay away" from Samsung instead if all their good pannels go to Korea first.
I wonder what the defect stats look like now. A few years ago, four defects per pannel were standard fare with many manufacturers not offering replacements until a dozen defects - how (un)usual is it to get a "perfect" LCD today? They used to be fairly rare but appear to have become far more common over the last 2-3 years.
About dead pixels: Samsung has announced a strict no-dead-pixel policy late last year. For people worried about this, the solution is simple: buy a Samsung-branded LCD manufactured in 2005.
My main gripe with LCDs is that affordable pannels lack high enough resolution at a reasonable price to make them useful for me so I will have to pass until affordable UXGA/WUXGA pannels become available. (My preferred resolution on my 19" CRT is 1600x1200x85Hz because pixels start blurring beyond this. For games, I use this fact as free anti-aliasing and run them at 2048x1536.)
First they export production jobs to cheap labor countries to increase profits by reducing production costs...... then they also introduce artificial pricing barriers to prevent people from capitalizing on pricing discrepancies.
If this trend is allowed to continue, employment will suffer, sales will suffer and CEOs will continue doing dumb things until they completely ruined their businesses, completely oblivious to the real-world impacts of their short-sighted greedy decisions.
If large corporations capitalize on globalization to reduce production costs, consumers should be allowed to do the exact same thing to cut their own expenses - corporations should not be allowed to monopolize globalization's "benefits". They should either globalize fairly/equitably (both ways) or not at all.
Well, as long as HP does not add counters on their cartridges' ID chip, one can always melt a hole through one side and refill the same cartridge until the drum (or print-head for inkjets) is ruined.
Electricity might not be free but where I live, it is -25C outside and with electric heating, PCs or heaters makes no difference on the power bills. Equatorial residents might be worried about global warming but as a selfish northern resident, it sounds great.
As far as CPUs (and electronics in general) are concerned though, I prefer sticking with cooler and (hopefully) more reliable chips since I rarely need extreme processing power and I definitely do not want whatever power I have at my disposal sending out smoke signals when I start using it.
If you pick up EMI, you pick up the combined EMI from all three color channels across the same spectrum with no way of splitting them, not really useful if the exercise is to capture "secure" video signals to circumvent copy protection.
As far as Tempest is concerned, the proof-of-concept was only good up to little more than VGA resulution, not really suitable for high-definition contents. There was a story about Tempest here a year or two ago and one of the suggested work-arounds was to use higher resolutions and refresh rates - this covers a wider spectral range which makes it much harder to recover a consistent signal since propagation parameters can vary wildly across the spectrum. I would be far more concerned about my WAP while I am writing this off my laptop than I would be about someone tuning on to my desktop's 1600x1200x85Hz signals.
Next step, hook up probes to your CRT's RGB electron gun gates and record from there.
If this becomes common practice, I wonder how long it will be until they find some sort of reason to outlaw CRTs or accelerate their discontinuation.
As long as the RIAA/MPAA's precious signals go through some sort of analog conversion, FPGA-DSP wizards can readily capture and convert streams, the two main challenges being not frying chips or themselves with the HV when capturing RGB signals off the CRT.
The only definite cost of unlicensed copying is bandwidth, there is no way of knowing how many of these copies actually are lost sales. Some will download it simply because it is available, I have a few friends who almost systematically download everything they see as long as they have spare HDD space or blank DVDs. So, the only undisputable copying costs are bandwidth and storage.
If a fast-food place was assaulted by 13000 clients, they would run out of stock and this stock is a direct recurrent cost, unlike bits which cost nearly nothing and are paid for by the ISP's customers at no direct cost for the RIAA/MPAA's partners.
If unlicenced copying ("piracy") was hurting the RIAA/MPAA members this bad, one would expect their revenue growth to be below-average or negative but in reality, they were one of the few to remain relatively steady through the recesion and still post above-average growth. (During the recession, average growth was -25% while the RIAA/MPAA members stood around -10%, I am somewhat surprised the milions of job-less people did not cut entertainment expenses first. Still, the RIAA/MPAA members claimed most of that -10% was due to "piracy")
Yes, people want recording contracts... and some time after they get them, may realize this may have been the single biggest mistake of their life because many record labels practically own people they pick for sign-up - some of these folks are desperate while others are so happy they blindly do whatever they are told and later realize they are on a trip to hell, though they might not realize that until they try to leave, sort of like the mafia again:)
IIRC, in the 2003 CRIA document, they wanted a $20/GB levy on all consumer-class digital storage and adjust existing media levies to match. Thankfully, this did not happen, most hikes were either denied or capped and HDD levies only apply if the HDD is integrated in special function hardware (player/recorder) such as the iPod - again, thankfully not at the requested rate but a more sane ~$1.50/GB.
There is a little table at the site below... http://www.ccfda.ca/subsections/eng_faqs .html (Probably not up to date since there is no mention of DVD +/- R/RW/RAM)
$2 under 1GB $15 under 10GB $25 beyond.
I wonder what customs would do if MP3 players started shipping without internal memory - barebone players, add your own HDD or CF/other card.
My guess is CRIA&friend will come back with the fixed $/GB on everything plan if the barebone player trend comes up.
A couple of times each year, there is a story about the RIAA (mistakenly) suing folks who offer their own (unlicensed as far as RIAA members are concerned) creations on P2P networks or direct downloads. Thankfully, it does not appear the RIAA insists too much once the folks in question tell the RIAA to buzz off and have their heads examined. (At least two such stories made it to/. in 2004)
This simply shows the RIAA sees piracy and unauthorized distribution everywhere there are MP3s or other such format... simple cases of shoot first, ask questions later. The RIAA/MPAA's automated tools are well known to make mistakes. (At least three stories about this on slashdot last year)
As for the levies' redistribution, do you seriously think air time reflects download and subsequent listening or copying patterns? No-name independents have a pretty hard time getting any air time, often having to buy it to get any, sometimes bankrupting themselves in the process.
The RIAA is somewhat like the air time mafia... if you are not family, good luck getting any - the big labels RIAA represents have more than enough cash to finance air time. No air time = nobody gets to know about your stuff on air = nobody will ask to hear your stuff = no levy money if it is the only thing that counts. It becomes a chicken&egg kind of problem.
The lithography process has a fair amount of randomness to it. Chips across the wafer will have different characteristics. In fact, characteristics can even vary significantly within the same die. Lucky chips will have evenly distributed dopants/insulation/conductor layers but this is rarely if ever the case.
The first step completed cores go through after manufacturing is functional testing. Functional testing seeks to eliminate chips which do not meet the minimum manufacturing expectations. Such expectations could be Vcore = max allowed, Icore = 100A max, f=2.4GHz, Tcore=50C, etc. Functional testing would reject CPUs (say Prescotts) which are obviously going to be out of specs. It would also identify bad cache banks so they can be disabled and the chip binned as a Celeron.
Next step would be speed-binning, probably in combination with electrical profiling to decide which chips qualify for low-power applications.
Functional testing (and possibly binning as well) is often done before the dies even leave the wafer so there is no labeling or packaging associated with it. The fully assembled parts are usually tested once again to verify that they successfully survived the packaging procedures and still meet the target parameters.
At this point, the package is probably still unmarked, the fuse bits unset and the parts ready to be re-qualified as something else if/as necessary. I would not be too surprised if some of AMD and Intel's (and others) long-term stock was stored as such.
Thankfully, the federal court ruled that levies were redundant and should be terminated.
While we might not know where all the levy money went, we at least now know the levies will be gone soon (in Canada), assuming they have not already been abolished thanks to the court's decision.
I wonder what kind of share independents manage to get from royalty claims. Since the RIAA regularly forgets that a free/independent market exists and sues intependents who distribute their own stuff, it must be quite a hassle for independents to fight off the RIAA if it accidentally (but happily) files levy claims for unlicensed productions.
People should realize that governments are becoming a system of "by rich people for rich people" instead of the "by the people for the people" they used to be and should be - governments are another area where the barriers to entry are rising every round, effectively keeping most people out.
Slashdot Mirror
The RIAA/MPAA/etc. have been making fortunes off dead people's backs for decades, it would be a logical next step to eventually extend this to dead customers.
I think the same too... for me, speaking while driving is just as much of a distraction on the phone or in person. Actually, in-person might be the bigger distraction for me since conversations can start randomly and unnecessary warnings usually distract me quite a bit.
On the phone, nobody can talk to me until I take the call and I have the option of taking calls only while I am comfortable with current driving conditions.
OS/2 2.0 was a joint project between Microsoft and IBM. When the two "divorced", IBM continued to use the OS/2 brand and Microsoft forked its side of things into what would become NT's starting point.
Having common origins does not mean one has to stick to the original's APIs, file systems and other implementation details. The current Linux kernel probably looks and behaves like it has no relation with the 1.0 code as well.
How is dropping "all DOS support" going to make Longhorn any more native than any NT has ever been? WinNT has its roots in OS/2, not DOS.
The "DOS support" is only the DOS virtual machine. Virtual machines are just OS plug-ins to support alternative runtime environments (may include platform emulation) and executable formats, adding and removing VMs does not make the rest of the OS any more or less native. Dropping DOS support is simply a matter of dropping that VM, no big deal unless you use some peerless/irreplaceable DOS apps.
Nocona has a complete AMD64-compatible instruction set modulus two or three minor discrepancies which Intel claims were due to flaws in AMD's specifications.
Nocona can run the latest WinXP-64 betas and most AMD64 software, not limited to simple address space expansion even though this is what "EMT64" might cause one to believe.
The reason why Intel is only promoting the extended addressing range is because it wants uninformed customers to believe Itanium is the only true 64bits ticket from Intel.
I have been fighting off a pretty bad cold for nearly a week and this is apparently giving me some mild hallucinations... I apparently hallucinated the period before "Stealing" and the capital letter one word to the left - at the time of writing, I really thought I read "Physically stealing physical property(...)", doh!
AES does not require beefy hardware to implement.
AES lends itself fairly well to both ASIC/hardware and software implementations. Because we are talking about cryptographic messages most likely in the sub-kilobit size range, the amount of processing in question is fairly limited.
I remember about at least one company advertising RFID tag microcontrollers. The rest is a simple matter of balancing power and time... and since the RFID microcontroller can start processing before the key is in the ignition switch, a processing delay up to a few seconds should be acceptable, allowing the microcontroller to run its core at most likely less than 100kHz or even less than 10kHz if the chip contains dedicated AES logic - we are taking 8bits microcontrollers here.
The only reason why RFID tags are the only thing we commonly see is because demand for tags far exceeds demand for everything else that could possibly be handled by RFID techniques. If demand for AES-enabled RFID microcontrollers becomes large enough, microcontroller companies will make them.
BTW, the RFID microcontroller summary did mention that an external capacitor was necessary to smooth the power but I do not remember the rest.
As far as size is concerned, keep in mind that typical microcontrollers contain well under a milion transistors so a microcontroller suitable for secure authentication for an ignition system should be well under 10 square milimeters on 180nm process.
In the RIAA/MPAA's dream world, they would manage to stop unlicensed reproduction with DRM. In the real world, DRM only managed to slow down casual copying - at least up to now.
Personally, I do not buy into new technology until I have an interoperability guarantee... like DeCSS and derrivative works before buying my first DVD.
> Stealing physical property deprives the owner of its use.
Does this imply that one can intelectually steal physical property or physically steal intellectual property as well?
But really, the "proper" 686 is the Pentium Pro and it is the Pentium Pro's feature set that was used as the foundation for the P2 (basically a P-Pro+MMX) and P3 series.
What the PPro brought to the table compared to the plain old Pentium was an independent cache bus, an extra execution pipeline, out-of-order execution and a bunch of other tweaks common in other architectures.
eXeem simply makes everyone a tracker for what files people are downloading. On the legal side of things, it seems eXeem is a guarantee that the tracker is also a downloader or seeder and therefore more legally problematic for the hosts than plain trackers.
.torrent file.
After all, even with eXeem, some initial tracker must be contacted before clients can start tracking on each other because they initially only know about the tracker(s) from the
Actually, Microsoft Office's XML exporter has two or three different behaviors depending on which version is used. IIRC, only the Professional and Enterprise editions export clean XML while the Personal editions only save in some proprietary description.
Artificial market segmentation appears to have become a primary hobby at Microsoft. First there was XP Home and Pro but now there is also Starter, MediaCenter and TabletPC editions. It really bugs me how MS labels the standard edition "Pro" and how it artificially cripples all other editions only for product "diversification" and the ability to ask $200 more for the standard edition.
Remember how the P3-Coppermine ran into a brick wall that prompted a recall at 1.13GHz?
The reason why the P4s have such huge pipelines is to meet timing margins required for high clock speeds. Pipelining helps by breaking down combinational logic blobs into sub-clock-tick-sized (settling time) logic bites and this costs glue-logic/clocking transistors, power and time efficiency.
The Pentium-M is designed to do the most work with the shortest pipeline on each clock tick using the least power. Because of its short pipeline, each stage (combinational logic blob) can have longer settling times. Since the pipeline cannot run faster than its slowest stage, the Pentium-M cannot scale much in frequency at stock voltage.
You will never be able to clock a PM at the same rates as the P4s - no matter how much extra power you throw at the PM, timing margins dictated by internal geometry will make it impossible just as it did for the 1.13GHz Coppermine P3.
At least this is one step in the generally right direction... but might mean I should have said "stay away" from Samsung instead if all their good pannels go to Korea first.
I wonder what the defect stats look like now. A few years ago, four defects per pannel were standard fare with many manufacturers not offering replacements until a dozen defects - how (un)usual is it to get a "perfect" LCD today? They used to be fairly rare but appear to have become far more common over the last 2-3 years.
About dead pixels: Samsung has announced a strict no-dead-pixel policy late last year. For people worried about this, the solution is simple: buy a Samsung-branded LCD manufactured in 2005.
My main gripe with LCDs is that affordable pannels lack high enough resolution at a reasonable price to make them useful for me so I will have to pass until affordable UXGA/WUXGA pannels become available. (My preferred resolution on my 19" CRT is 1600x1200x85Hz because pixels start blurring beyond this. For games, I use this fact as free anti-aliasing and run them at 2048x1536.)
First they export production jobs to cheap labor countries to increase profits by reducing production costs... ... then they also introduce artificial pricing barriers to prevent people from capitalizing on pricing discrepancies.
If this trend is allowed to continue, employment will suffer, sales will suffer and CEOs will continue doing dumb things until they completely ruined their businesses, completely oblivious to the real-world impacts of their short-sighted greedy decisions.
If large corporations capitalize on globalization to reduce production costs, consumers should be allowed to do the exact same thing to cut their own expenses - corporations should not be allowed to monopolize globalization's "benefits". They should either globalize fairly/equitably (both ways) or not at all.
Well, as long as HP does not add counters on their cartridges' ID chip, one can always melt a hole through one side and refill the same cartridge until the drum (or print-head for inkjets) is ruined.
Electricity might not be free but where I live, it is -25C outside and with electric heating, PCs or heaters makes no difference on the power bills. Equatorial residents might be worried about global warming but as a selfish northern resident, it sounds great.
As far as CPUs (and electronics in general) are concerned though, I prefer sticking with cooler and (hopefully) more reliable chips since I rarely need extreme processing power and I definitely do not want whatever power I have at my disposal sending out smoke signals when I start using it.
If you pick up EMI, you pick up the combined EMI from all three color channels across the same spectrum with no way of splitting them, not really useful if the exercise is to capture "secure" video signals to circumvent copy protection.
As far as Tempest is concerned, the proof-of-concept was only good up to little more than VGA resulution, not really suitable for high-definition contents. There was a story about Tempest here a year or two ago and one of the suggested work-arounds was to use higher resolutions and refresh rates - this covers a wider spectral range which makes it much harder to recover a consistent signal since propagation parameters can vary wildly across the spectrum. I would be far more concerned about my WAP while I am writing this off my laptop than I would be about someone tuning on to my desktop's 1600x1200x85Hz signals.
Next step, hook up probes to your CRT's RGB electron gun gates and record from there.
If this becomes common practice, I wonder how long it will be until they find some sort of reason to outlaw CRTs or accelerate their discontinuation.
As long as the RIAA/MPAA's precious signals go through some sort of analog conversion, FPGA-DSP wizards can readily capture and convert streams, the two main challenges being not frying chips or themselves with the HV when capturing RGB signals off the CRT.
The only definite cost of unlicensed copying is bandwidth, there is no way of knowing how many of these copies actually are lost sales. Some will download it simply because it is available, I have a few friends who almost systematically download everything they see as long as they have spare HDD space or blank DVDs. So, the only undisputable copying costs are bandwidth and storage.
If a fast-food place was assaulted by 13000 clients, they would run out of stock and this stock is a direct recurrent cost, unlike bits which cost nearly nothing and are paid for by the ISP's customers at no direct cost for the RIAA/MPAA's partners.
If unlicenced copying ("piracy") was hurting the RIAA/MPAA members this bad, one would expect their revenue growth to be below-average or negative but in reality, they were one of the few to remain relatively steady through the recesion and still post above-average growth. (During the recession, average growth was -25% while the RIAA/MPAA members stood around -10%, I am somewhat surprised the milions of job-less people did not cut entertainment expenses first. Still, the RIAA/MPAA members claimed most of that -10% was due to "piracy")
Yes, people want recording contracts... and some time after they get them, may realize this may have been the single biggest mistake of their life because many record labels practically own people they pick for sign-up - some of these folks are desperate while others are so happy they blindly do whatever they are told and later realize they are on a trip to hell, though they might not realize that until they try to leave, sort of like the mafia again :)
IIRC, in the 2003 CRIA document, they wanted a $20/GB levy on all consumer-class digital storage and adjust existing media levies to match. Thankfully, this did not happen, most hikes were either denied or capped and HDD levies only apply if the HDD is integrated in special function hardware (player/recorder) such as the iPod - again, thankfully not at the requested rate but a more sane ~$1.50/GB.
s .html
There is a little table at the site below...
http://www.ccfda.ca/subsections/eng_faq
(Probably not up to date since there is no mention of DVD +/- R/RW/RAM)
$2 under 1GB
$15 under 10GB
$25 beyond.
I wonder what customs would do if MP3 players started shipping without internal memory - barebone players, add your own HDD or CF/other card.
My guess is CRIA&friend will come back with the fixed $/GB on everything plan if the barebone player trend comes up.
A couple of times each year, there is a story about the RIAA (mistakenly) suing folks who offer their own (unlicensed as far as RIAA members are concerned) creations on P2P networks or direct downloads. Thankfully, it does not appear the RIAA insists too much once the folks in question tell the RIAA to buzz off and have their heads examined. (At least two such stories made it to /. in 2004)
This simply shows the RIAA sees piracy and unauthorized distribution everywhere there are MP3s or other such format... simple cases of shoot first, ask questions later. The RIAA/MPAA's automated tools are well known to make mistakes. (At least three stories about this on slashdot last year)
As for the levies' redistribution, do you seriously think air time reflects download and subsequent listening or copying patterns? No-name independents have a pretty hard time getting any air time, often having to buy it to get any, sometimes bankrupting themselves in the process.
The RIAA is somewhat like the air time mafia... if you are not family, good luck getting any - the big labels RIAA represents have more than enough cash to finance air time. No air time = nobody gets to know about your stuff on air = nobody will ask to hear your stuff = no levy money if it is the only thing that counts. It becomes a chicken&egg kind of problem.
The lithography process has a fair amount of randomness to it. Chips across the wafer will have different characteristics. In fact, characteristics can even vary significantly within the same die. Lucky chips will have evenly distributed dopants/insulation/conductor layers but this is rarely if ever the case.
The first step completed cores go through after manufacturing is functional testing. Functional testing seeks to eliminate chips which do not meet the minimum manufacturing expectations. Such expectations could be Vcore = max allowed, Icore = 100A max, f=2.4GHz, Tcore=50C, etc. Functional testing would reject CPUs (say Prescotts) which are obviously going to be out of specs. It would also identify bad cache banks so they can be disabled and the chip binned as a Celeron.
Next step would be speed-binning, probably in combination with electrical profiling to decide which chips qualify for low-power applications.
Functional testing (and possibly binning as well) is often done before the dies even leave the wafer so there is no labeling or packaging associated with it. The fully assembled parts are usually tested once again to verify that they successfully survived the packaging procedures and still meet the target parameters.
At this point, the package is probably still unmarked, the fuse bits unset and the parts ready to be re-qualified as something else if/as necessary. I would not be too surprised if some of AMD and Intel's (and others) long-term stock was stored as such.
Thankfully, the federal court ruled that levies were redundant and should be terminated.
While we might not know where all the levy money went, we at least now know the levies will be gone soon (in Canada), assuming they have not already been abolished thanks to the court's decision.
I wonder what kind of share independents manage to get from royalty claims. Since the RIAA regularly forgets that a free/independent market exists and sues intependents who distribute their own stuff, it must be quite a hassle for independents to fight off the RIAA if it accidentally (but happily) files levy claims for unlicensed productions.
People should realize that governments are becoming a system of "by rich people for rich people" instead of the "by the people for the people" they used to be and should be - governments are another area where the barriers to entry are rising every round, effectively keeping most people out.