We are still a long way from being able to simulate the climate with a true first principles calculation. While many basic aspects of physics can be included (conservation of mass, energy etc.), many need to be approximated for reasons of efficiency or resolutions (i.e. the equations of motion need estimates of sub-gridscale turbulent effects, radiative transfer codes approximate the line-by-line calculations using band averaging), and still others are only known empirically (the formula for how fast clouds turn to rain for instance). With these approximations and empirical formulae, there is often a tunable parameter or two that can be varied in order to improve the match to whatever observations exist. Adjusting these values is described as tuning and falls into two categories. First, there is the tuning in a single formula in order for that formula to best match the observed values of that specific relationship. This happens most frequently when new parameterisations are being developed.
Secondly, there are tuning parameters that control aspects of the emergent system. Gravity wave drag parameters are not very constrained by data, and so are often tuned to improve the climatology of stratospheric zonal winds. The threshold relative humidity for making clouds is tuned often to get the most realistic cloud cover and global albedo. Surprisingly, there are very few of these (maybe a half dozen) that are used in adjusting the models to match the data. It is important to note that these exercises are done with the mean climate (including the seasonal cycle and some internal variability) â" and once set they are kept fixed for any perturbation experiment.
So, even with the half-dozen parameters that are left to tune the model to the observed climate, perturbations are not tuned, so they can still be falsified.
When the model is mostly based on physics, there's very little to twiddle or tune in the first place. Falsification is also easy. When the observed temperature falls outside the error bars, the model is most likely incorrect.
Unless you state ahead of time what observations will falsify a given model calculation, you're not really doing science, even if you are doing something interesting.
This must apply doubly to denialists, who don't even have a working alternative model.
CO2 only helps plant life if CO2 happened to be the most limiting factor. Experiments in artificial greenhouses, where both water and nutrients are abundant, CO2 may well be the limiting factor to further growth.
In nature, water is often a bigger limiting factor than CO2, so additional CO2 may not make much of a difference.
In particular, one where CO2 leads, rather than lags temperature change
You first need to understand that CO2 both leads and lags temperature change, with different delays. Higher temperatures can lead to higher CO2, and higher CO2 will lead to higher temperatures. Right now, it is clear that CO2 is leading.
What observations will show us that 3C is too much?
If you can properly explain the glacial cycles without assuming 3C effect on doubling CO2,, for instance. Go ahead, you'd be famous.
and then fudge factor everything else until it comes to some close match
That's not how GCMs work. Almost everything in the model has a physical basis that cannot be tuned. Of course, the physical models depend on our understanding of the physics, which may change as a result of experiments and observations, but the scientists do not have the opportunity to simple twiddle some parameters to make the model fit, as you seem to think.
If we can also determine how big the effects are of other climate drivers, like albedo, solar radiation, earth orbit changes, and aerosols (manmade and volcanic) we can add everything together in a global climate model, and see how well it matches the measured temperature. This can be done for current century, but we can also run the models on historic data, and see if they accurately predict glacial cycles, for instance.
But wait, we haven't seen a doubling of CO2 in the past 100 years...or even the past 1000 years, and we certainly haven't seen 4-5C of warming that would go with it.
We have seen many CO2 doublings in the climate history, and temperature changes that go with it.
Also, the 4-5C degrees was a bit too much. Best current estimate is around 3C temperature rise after a CO2 doubling, and that would be the new equilibrium, which takes several decades to establish. Even if we could prevent any further CO2 increases, the global temperature would go up for a few more decades before settling on a new equilibrium until CO2 goes back down.
Information relevant to the market comes in at all times of the day. For instance, in the middle of a trading day, news could come out that a company has a major setback, like an explosion on a big plant. If a trader was just in the middle of a stock transaction, having bought some cheap shares in Asia, and trying to sell these in London for a penny more, and this news hits the market, all of a sudden the buy offer may be canceled, and the trader is stuck with overpriced stock.
For Joe the investor, it is pretty unlikely that he's going to have a buy order outstanding right at the moment that the bad news hits the market, and somebody dumps his shares on him for his overpriced offer. For a high frequency trader, who's constantly doing trades in thousands of different companies, he has a much bigger exposure to these upsetting events, especially if he's required to hold them for a minute. The fact that bad news for a particular company is a low frequency event doesn't matter.
So if Joe Schmo buys 100 shares of Shell, paying $6132 he would have saved $1. I fail to see the great benefit for Joe...
In this example, the benefit is small. Of course, if Joe is an active day trader, and trades his 100 shares 20 times a day, he could have saved $20 in total.
Also, the arbitrage trader only took $1 off the table for his services, so while the benefit is small, the fee is similarly small too.
Doing it in milliseconds rather than a minute reduces the risk for the trader.
Buying some stock in Asia, and selling it a minute later in New York carries a much bigger risk than selling it 0.1 seconds later. A minute may seem short to you, but when you spend all day doing those trades, a window of a minute means 600 times the risk of a 100 millisecond window that the price of the stock will change.
To offset this 600x bigger risk, the trader needs to have bigger profit margins on the trade as an insurance, which means a bigger price difference between global markets, which means that the average investor will get a worse price on their stock transaction.
Faster trading therefore benefits the real world and real economy.
Of course, that's a form of front running, which is illegal. In itself it has nothing to do with high speed trading, just like your other examples of fraud and corruption. No doubt that these things are happening in the stock market, just like they are happening anywhere else. That's why these things are illegal.
That would be tricky. A lot of things tend to reflect the 2.4 GHz signals, and the reflections interfere with the signal. A fun experiment is to attach a 2.4 GHz antenna to a spectrum analyzer, and move it around. Just moving it a few inches back and forth has enormous effect on the signal strength (that's why a lot of gear has two antennas).
There is no point. Your solution would cost more money than the margin currently collected by arbitrage specialists. Since the latter are competing amongst each other, the margins are always under pressure.
Joe Schmo just wants to buy a share of Shell, and would rather pay $61.33 on a US market, through his own broker, than to open up an account in London, quickly exchange some dollars for pounds, and buy it there for $61.31 before the price goes up. It's just too much hassle and risk, and extra fees.
Now comes in the arbitrage specialist. buys the share in London for $61.31 for the instant it is available, and re-sells that same share in New York for $61.32 milliseconds later. Joe Schmo can then take advantage of that. Instead of $61.33, he's now paying $61.32, but without all the hassle and risk of doing the trade himself.
Now, if you're going to enforce limits on what the arbitrage trader can do, Joe Schmo will have no choice, and buy the share locally for $61.33 instead.
No, it's not a coincidence. A lot of people that are producing things that are beneficial to others are contributing to their own bottom line when doing so. In fact, most people do exactly that.
Making the situation even worse is the fact that there is a complete lack of standardization on the ARM platform, especially for all the peripherals. But even for the core itself there are many different variants. This can be an advantage for embedded developers, because it gives you lots of choices.
For binary software vendors, it's a nightmare, because they would have to support all these different versions.
The government could also invest in different ares of science and technology, such as renewable energy. This would bring similar benefits in jobs and technological advances, while at the same time producing something useful.
Human space exploration is in its core a useless stunt. Exploration is much better done by unmanned craft.
The only advantage I see is that an 128GB SSD is a lot smaller than 128GB ram
Also cheaper, lower power. and non-volatile. A battery wouldn't work, because DRAM needs to be refreshed, and with the size of the RAM, the current drain on the battery would be substantial.
Making them cost effective also means making fuel cells without platinum. There just isn't enough platinum in the world to make a billion fuel cells.
Re:And Intel isn't happy about it.
on
Arduino Goes ARM
·
· Score: 1
I think Intel's job is easier. The changes required for high performance require very complex design to be added. Intel already has that design. ARM is still working on it.
To get lower power just means a combination of better manufacturing (equally hard for both), and removing transistors, which is much easier than adding them in.
You can't be talking about normal statistical variation, since the temperature over the past century is in a statistically significant upward trend. Since there is an effect, there must be a cause.
So, what is the cause of this "natural climate change" ?
Slashdot Mirror
source: http://www.realclimate.org/index.php/archives/2008/11/faq-on-climate-models/
So, even with the half-dozen parameters that are left to tune the model to the observed climate, perturbations are not tuned, so they can still be falsified.
When the model is mostly based on physics, there's very little to twiddle or tune in the first place. Falsification is also easy. When the observed temperature falls outside the error bars, the model is most likely incorrect.
This must apply doubly to denialists, who don't even have a working alternative model.
CO2 only helps plant life if CO2 happened to be the most limiting factor. Experiments in artificial greenhouses, where both water and nutrients are abundant, CO2 may well be the limiting factor to further growth.
In nature, water is often a bigger limiting factor than CO2, so additional CO2 may not make much of a difference.
You first need to understand that CO2 both leads and lags temperature change, with different delays. Higher temperatures can lead to higher CO2, and higher CO2 will lead to higher temperatures. Right now, it is clear that CO2 is leading.
If you can properly explain the glacial cycles without assuming 3C effect on doubling CO2,, for instance. Go ahead, you'd be famous.
That's not how GCMs work. Almost everything in the model has a physical basis that cannot be tuned. Of course, the physical models depend on our understanding of the physics, which may change as a result of experiments and observations, but the scientists do not have the opportunity to simple twiddle some parameters to make the model fit, as you seem to think.
If we can also determine how big the effects are of other climate drivers, like albedo, solar radiation, earth orbit changes, and aerosols (manmade and volcanic) we can add everything together in a global climate model, and see how well it matches the measured temperature. This can be done for current century, but we can also run the models on historic data, and see if they accurately predict glacial cycles, for instance.
We have seen many CO2 doublings in the climate history, and temperature changes that go with it.
Also, the 4-5C degrees was a bit too much. Best current estimate is around 3C temperature rise after a CO2 doubling, and that would be the new equilibrium, which takes several decades to establish. Even if we could prevent any further CO2 increases, the global temperature would go up for a few more decades before settling on a new equilibrium until CO2 goes back down.
Information relevant to the market comes in at all times of the day. For instance, in the middle of a trading day, news could come out that a company has a major setback, like an explosion on a big plant. If a trader was just in the middle of a stock transaction, having bought some cheap shares in Asia, and trying to sell these in London for a penny more, and this news hits the market, all of a sudden the buy offer may be canceled, and the trader is stuck with overpriced stock.
For Joe the investor, it is pretty unlikely that he's going to have a buy order outstanding right at the moment that the bad news hits the market, and somebody dumps his shares on him for his overpriced offer. For a high frequency trader, who's constantly doing trades in thousands of different companies, he has a much bigger exposure to these upsetting events, especially if he's required to hold them for a minute. The fact that bad news for a particular company is a low frequency event doesn't matter.
In this example, the benefit is small. Of course, if Joe is an active day trader, and trades his 100 shares 20 times a day, he could have saved $20 in total.
Also, the arbitrage trader only took $1 off the table for his services, so while the benefit is small, the fee is similarly small too.
Doing it in milliseconds rather than a minute reduces the risk for the trader.
Buying some stock in Asia, and selling it a minute later in New York carries a much bigger risk than selling it 0.1 seconds later. A minute may seem short to you, but when you spend all day doing those trades, a window of a minute means 600 times the risk of a 100 millisecond window that the price of the stock will change.
To offset this 600x bigger risk, the trader needs to have bigger profit margins on the trade as an insurance, which means a bigger price difference between global markets, which means that the average investor will get a worse price on their stock transaction.
Faster trading therefore benefits the real world and real economy.
Of course, that's a form of front running, which is illegal. In itself it has nothing to do with high speed trading, just like your other examples of fraud and corruption. No doubt that these things are happening in the stock market, just like they are happening anywhere else. That's why these things are illegal.
That would be tricky. A lot of things tend to reflect the 2.4 GHz signals, and the reflections interfere with the signal. A fun experiment is to attach a 2.4 GHz antenna to a spectrum analyzer, and move it around. Just moving it a few inches back and forth has enormous effect on the signal strength (that's why a lot of gear has two antennas).
There is no point. Your solution would cost more money than the margin currently collected by arbitrage specialists. Since the latter are competing amongst each other, the margins are always under pressure.
Joe Schmo just wants to buy a share of Shell, and would rather pay $61.33 on a US market, through his own broker, than to open up an account in London, quickly exchange some dollars for pounds, and buy it there for $61.31 before the price goes up. It's just too much hassle and risk, and extra fees.
Now comes in the arbitrage specialist. buys the share in London for $61.31 for the instant it is available, and re-sells that same share in New York for $61.32 milliseconds later. Joe Schmo can then take advantage of that. Instead of $61.33, he's now paying $61.32, but without all the hassle and risk of doing the trade himself.
Now, if you're going to enforce limits on what the arbitrage trader can do, Joe Schmo will have no choice, and buy the share locally for $61.33 instead.
No, it's not a coincidence. A lot of people that are producing things that are beneficial to others are contributing to their own bottom line when doing so. In fact, most people do exactly that.
And if so, does it really matter if you hit the wrong number ?
Making the situation even worse is the fact that there is a complete lack of standardization on the ARM platform, especially for all the peripherals. But even for the core itself there are many different variants. This can be an advantage for embedded developers, because it gives you lots of choices.
For binary software vendors, it's a nightmare, because they would have to support all these different versions.
If you want to do hard things, why not pick something that's both hard and useful instead ?
The government could also invest in different ares of science and technology, such as renewable energy. This would bring similar benefits in jobs and technological advances, while at the same time producing something useful.
Human space exploration is in its core a useless stunt. Exploration is much better done by unmanned craft.
Most software is distributed in binary form, which is hard enough to reverse engineer that patent protection doesn't really add that much benefit.
Oh, you had cards with a "1" on them? Lucky bastard.
Also cheaper, lower power. and non-volatile. A battery wouldn't work, because DRAM needs to be refreshed, and with the size of the RAM, the current drain on the battery would be substantial.
Making them cost effective also means making fuel cells without platinum. There just isn't enough platinum in the world to make a billion fuel cells.
I think Intel's job is easier. The changes required for high performance require very complex design to be added. Intel already has that design. ARM is still working on it.
To get lower power just means a combination of better manufacturing (equally hard for both), and removing transistors, which is much easier than adding them in.
Is that a concern for single core designs such as in TFA ?
You can't be talking about normal statistical variation, since the temperature over the past century is in a statistically significant upward trend. Since there is an effect, there must be a cause.
So, what is the cause of this "natural climate change" ?