Back in the early 2000's AMD had a clear and convincing lead in both absolute performance and price-performance for 2.5-3 years. Intel successfully kept them out of mass-market OEM products and cash-starved AMD was not able to keep up with Intel's research budget-- eventually paying a $1.25B settlement to AMD but this was not sufficient to make AMD whole.
Following that, we've had an extended period of stagnation on Intel's side until this point where AMD is again neck and neck with them.
The processor market is a whole lot better for everyone when this competition exists.
> All else being equal, if you use N times as many satellites to compute a position/velocity/time solution, your expected accuracy improves by a factor of sqrt(N)
To the extent all the errors are uncorrelated. If there's something that's equally wrong with all observations, e.g. the ionosphere delay model is bad for a direction, adding more observations doesn't eliminate that error.
Not to mention that more overdetermination is good-- lets you average out more non-systemic error.
There's also the whole issue of commercial providers offering high quality differential corrections that get centimeter or better accuracy out of theoretically less-capable signals.
It's normal for extradition treaties. It makes sense, because you need to present lots of evidence in a radically different jurisdiction. Exact same timeframe applies to Canada's requests of the US, etc.
> NO formal charge against her so far, only a provisional charge?
?? There's an arrest warrant. If you find out someone is going to Canada that you have a warrant for, you send a provisional request to Canada so they can act immediately. Then within 45 days you follow up with a formal extradition request. That's what the treaty says.
RCMP didn't just arrest her on their own-- they were irequested by US DOJ to detain her. Now they are awaiting a formal request for extradition; if none comes, they'll let her go in 60 days.
I suspect there is a lot more wiggle room in the USA on acting on arrest warrants than there is in Canada.
US arrest warrants have no standing in Canada. Instead, there was a request issued to Canada per Article 11 of the treaty:
> (1) In case of urgency a Contracting Party may apply for the provisional arrest of the person sought pending the presentation of the request for extradition through the diplomatic channel. Such application shall contain a description of the person sought, an indication of intention to request the extradition of the person sought and a statement of the existence of a warrant or arrest or a judgment of conviction against that person, and such further information, if any, as would be necessary to justify the issue of a warrant of arrest had the offense been committed, or the person sought been convicted, in the territory of the requested State.
I must have missed the part where the US is building up military bases on those islands, adding artificial military platforms, and is using them to try and control trade and freedom of navigation of our neighbors (the 9 dash line, as rejected by a UN arbitral tribunal).
Quote it in full, pls, because it is not wonderful but not nearly as damning in context.
What was your role in the original Haynes nomination and decision to renominate him? And at the time of the nomination, what did you know about Mr. Haynes’s role in crafting the administration’s detention and interrogation policies?
Mr. KAVANAUGH. Senator, I did not—I was not involved and am not involved in the questions about the rules governing detention of combatants or—and so I do not have the involvement with that. And with respect to Mr. Haynes’s nomination, I’ve—I know Jim Haynes, but it was not one of the nominations that I handled. I handled a number of nominations in the Counsel’s Office. That was not one of the ones that I handled.
You guys pay 3 times more than the US. If the surcharge for renewables is 23% of the price, and the ecological tax is a further 7%, it's a big fraction of the total price of electricity in the US-- 90%-- right there-- which is almost half of the 3x difference. Not to mention that the grid fees are higher because your operations are considerably more complicated due to the large fraction of renewables on the grid.
> Or, put it another way, no given xray (or cigarette) is likely to give you cancer. But getting 100 xrays a day (or smoking 5 packs a day) is likely to cause you to get cancer. That's why it's illegal to split a project into smaller pieces.
At the same time it's kinda ridiculous here. Yes, you may have grand plans for what you'd eventually like to do-- tunnels everywhere. But you should still be able to build logical, standalone pieces and not somehow try to get through EIR/permitting for everything you've ever contemplated.
More like 12. (But can we please use MJ/kg instead?) But you also need to take into account that you only convert about 25% of the energy in gasoline to useful work, and that you need to carry all the mass of an internal combustion engine to extract the energy.
The most real problem is that this is a way for motherboard and CPU vendors to segment the market, and prevent commodity PC hardware from being used for critical things. Home users "don't need" ECC, so it can be left off the cheap stuff.
It's just that the stuff will have sat for an indeterminate, long time while the clock is stopped-- providing an unusually long window for a bit to flip-- and resuming even from S1 is a relatively costly operation.
I think overall it is silly, but if you have ECC RAM and non-ECC cache, and spend most of the time in S1, it's not completely crazy.
Yes, you can always reinvent the wheel and oversee more of the process yourself/produce something directly, as opposed to buying an existing commercial product. As a rule, you spend a lot more resources doing this UNLESS there is some kind of unique synergy from vertical integration or you need something that is so unlike everything else out there that it justifies this otherwise inefficient approach.
NASA hasn't had a launch vehicle that's been competitive in $/kg or reliability with private offerings in 30 years. SLS is not forecast by NASA's own optimistic projections to be better on the former and only forecasts parity on the latter. It will be "to market" later. Why would you ever build SLS instead of buying cheaper commodity launch?
I have some degree of sympathy--- the whole concept of being subject to secret criminal prosecution is scary and unfortunate. (A problem I don't know how to handle: when people break the rules about classified things, classified information is likely needed to prove and contextualize it. At the same time, denying people a portion of the normal due process of law and throwing them in jail is a disturbing result).
I don't think he's wrong-- I think you two are talking past each other.
The '970 family was a purpose-built processor for Apple, and the first and only time of IBM selling processors in quantity to other vendors. IBM basically didn't use it themselves (yes, it showed up in a couple blade server platforms early on).
Yes, it was an offshoot of IBM's successful POWER processor lineup.
The funny thing is, you don't need anywhere near 1% daytime solar radiation. The eye has crazy dynamic range.
The sun is ~100000 Lux. People read indoors at ~150 Lux. Moderate local streetlighting is ~5 Lux. A full moon is ~0.1 Lux.
If you do 1/100,000th of the brightness of the sun, that's more light than a full moon provides. This makes the mirror size more like ~30 meters, which can be provided by a big e.g. mylar structure.
Yah-- that's a separate issue: statistically, women tend weight quality of life/happiness/work-life balance higher than men do, and as a result tend to shy away from male professions.
Of course, there's exceptions-- my wife is a mech eng.
Yah-- everyone needs to have the opportunity. But it may not be "fair" in numbers afterwards.
Testosterone seems to cause *increased variability* in outcomes. Women appear to be slightly smarter on average than men (depending on the metric you choose), but men have a greater variability in intelligence and performance. That is, men are over-represented at the very dumb and brilliant ends of the spectrum.
Equal opportunity may still result in an excess of men at the very top of many professions...
(And again, these are just broad statistical trends-- any individual should have the full chance to show what he or she can do, because broad statistics do not really hold at a sample size of 1).
> Some will have better immune systems. Some will have a more natural propensity to succumb, randomly.
Actually, one of the ironies of the 1918 flu is that it disproportionately killed the young and healthy-- it is thought this is because it provoked an excessive immune response ("cytokine storm") in people with the best immune systems. While the research is still young, we're coming up with a number of drug regimens to effectively treat this.
Slashdot Mirror
Back in the early 2000's AMD had a clear and convincing lead in both absolute performance and price-performance for 2.5-3 years. Intel successfully kept them out of mass-market OEM products and cash-starved AMD was not able to keep up with Intel's research budget-- eventually paying a $1.25B settlement to AMD but this was not sufficient to make AMD whole.
Following that, we've had an extended period of stagnation on Intel's side until this point where AMD is again neck and neck with them.
The processor market is a whole lot better for everyone when this competition exists.
> All else being equal, if you use N times as many satellites to compute a position/velocity/time solution, your expected accuracy improves by a factor of sqrt(N)
To the extent all the errors are uncorrelated. If there's something that's equally wrong with all observations, e.g. the ionosphere delay model is bad for a direction, adding more observations doesn't eliminate that error.
Not to mention that more overdetermination is good-- lets you average out more non-systemic error.
There's also the whole issue of commercial providers offering high quality differential corrections that get centimeter or better accuracy out of theoretically less-capable signals.
It's normal for extradition treaties. It makes sense, because you need to present lots of evidence in a radically different jurisdiction. Exact same timeframe applies to Canada's requests of the US, etc.
> NO formal charge against her so far, only a provisional charge?
?? There's an arrest warrant. If you find out someone is going to Canada that you have a warrant for, you send a provisional request to Canada so they can act immediately. Then within 45 days you follow up with a formal extradition request. That's what the treaty says.
It really sucks, but we have a history of letting people go when it's politically expedient -- e.g. swaps with the russkies.
RCMP didn't just arrest her on their own-- they were irequested by US DOJ to detain her. Now they are awaiting a formal request for extradition; if none comes, they'll let her go in 60 days.
I suspect there is a lot more wiggle room in the USA on acting on arrest warrants than there is in Canada.
US arrest warrants have no standing in Canada. Instead, there was a request issued to Canada per Article 11 of the treaty:
> (1) In case of urgency a Contracting Party may apply for the provisional arrest of the person sought pending the presentation of the request for extradition through the diplomatic channel. Such application shall contain a description of the person sought, an indication of intention to request the extradition of the person sought and a statement of the existence of a warrant or arrest or a judgment of conviction against that person, and such further information, if any, as would be necessary to justify the issue of a warrant of arrest had the offense been committed, or the person sought been convicted, in the territory of the requested State.
You're right. It's ~472,000,000 cubic feet.
Cubic units are always misleading for how "small" they are.
> They have disputed islands (just like the US)..
I must have missed the part where the US is building up military bases on those islands, adding artificial military platforms, and is using them to try and control trade and freedom of navigation of our neighbors (the 9 dash line, as rejected by a UN arbitral tribunal).
Quote it in full, pls, because it is not wonderful but not nearly as damning in context.
What was your role in the original Haynes nomination and decision
to renominate him? And at the time of the nomination, what
did you know about Mr. Haynes’s role in crafting the administration’s
detention and interrogation policies?
Mr. KAVANAUGH. Senator, I did not—I was not involved and am
not involved in the questions about the rules governing detention
of combatants or—and so I do not have the involvement with that.
And with respect to Mr. Haynes’s nomination, I’ve—I know Jim
Haynes, but it was not one of the nominations that I handled. I
handled a number of nominations in the Counsel’s Office. That was
not one of the ones that I handled.
It's poorly written in the summary. There was a deadline of Nov 16 for him to agree; he didn't; so Intel sued.
Sensitive to about a centimeter actually, with a sustained observation.
You guys pay 3 times more than the US. If the surcharge for renewables is 23% of the price, and the ecological tax is a further 7%, it's a big fraction of the total price of electricity in the US-- 90%-- right there-- which is almost half of the 3x difference. Not to mention that the grid fees are higher because your operations are considerably more complicated due to the large fraction of renewables on the grid.
> Or, put it another way, no given xray (or cigarette) is likely to give you cancer. But getting 100 xrays a day (or smoking 5 packs a day) is likely to cause you to get cancer. That's why it's illegal to split a project into smaller pieces.
At the same time it's kinda ridiculous here. Yes, you may have grand plans for what you'd eventually like to do-- tunnels everywhere. But you should still be able to build logical, standalone pieces and not somehow try to get through EIR/permitting for everything you've ever contemplated.
More like 12. (But can we please use MJ/kg instead?) But you also need to take into account that you only convert about 25% of the energy in gasoline to useful work, and that you need to carry all the mass of an internal combustion engine to extract the energy.
The most real problem is that this is a way for motherboard and CPU vendors to segment the market, and prevent commodity PC hardware from being used for critical things. Home users "don't need" ECC, so it can be left off the cheap stuff.
It's just that the stuff will have sat for an indeterminate, long time while the clock is stopped-- providing an unusually long window for a bit to flip-- and resuming even from S1 is a relatively costly operation.
I think overall it is silly, but if you have ECC RAM and non-ECC cache, and spend most of the time in S1, it's not completely crazy.
Kilowatts per kg? Really?
Your argument is ridiculous.
Yes, you can always reinvent the wheel and oversee more of the process yourself/produce something directly, as opposed to buying an existing commercial product. As a rule, you spend a lot more resources doing this UNLESS there is some kind of unique synergy from vertical integration or you need something that is so unlike everything else out there that it justifies this otherwise inefficient approach.
NASA hasn't had a launch vehicle that's been competitive in $/kg or reliability with private offerings in 30 years. SLS is not forecast by NASA's own optimistic projections to be better on the former and only forecasts parity on the latter. It will be "to market" later. Why would you ever build SLS instead of buying cheaper commodity launch?
I have some degree of sympathy--- the whole concept of being subject to secret criminal prosecution is scary and unfortunate. (A problem I don't know how to handle: when people break the rules about classified things, classified information is likely needed to prove and contextualize it. At the same time, denying people a portion of the normal due process of law and throwing them in jail is a disturbing result).
I don't think he's wrong-- I think you two are talking past each other.
The '970 family was a purpose-built processor for Apple, and the first and only time of IBM selling processors in quantity to other vendors. IBM basically didn't use it themselves (yes, it showed up in a couple blade server platforms early on).
Yes, it was an offshoot of IBM's successful POWER processor lineup.
The funny thing is, you don't need anywhere near 1% daytime solar radiation. The eye has crazy dynamic range.
The sun is ~100000 Lux. People read indoors at ~150 Lux. Moderate local streetlighting is ~5 Lux. A full moon is ~0.1 Lux.
If you do 1/100,000th of the brightness of the sun, that's more light than a full moon provides. This makes the mirror size more like ~30 meters, which can be provided by a big e.g. mylar structure.
Yah-- that's a separate issue: statistically, women tend weight quality of life/happiness/work-life balance higher than men do, and as a result tend to shy away from male professions.
Of course, there's exceptions-- my wife is a mech eng.
Yah-- everyone needs to have the opportunity. But it may not be "fair" in numbers afterwards.
Testosterone seems to cause *increased variability* in outcomes. Women appear to be slightly smarter on average than men (depending on the metric you choose), but men have a greater variability in intelligence and performance. That is, men are over-represented at the very dumb and brilliant ends of the spectrum.
Equal opportunity may still result in an excess of men at the very top of many professions...
(And again, these are just broad statistical trends-- any individual should have the full chance to show what he or she can do, because broad statistics do not really hold at a sample size of 1).
> Some will have better immune systems. Some will have a more natural propensity to succumb, randomly.
Actually, one of the ironies of the 1918 flu is that it disproportionately killed the young and healthy-- it is thought this is because it provoked an excessive immune response ("cytokine storm") in people with the best immune systems. While the research is still young, we're coming up with a number of drug regimens to effectively treat this.