Simulations

I like simulations: I’ve written lots of them. I’ve used them to understand questions better – questions like the electronic structure of semiconductors, the evolution of Ashkenazi intelligence, and methods of combining adaptive optics and imaging processing.

There are systems that are hard to simulate: stuff that’s highly nonlinear, or where the some of the key factors are poorly understood and/or poorly measured, when the required calculations grow exponentially, etc.

There are also systems that can be simulated essentially perfectly. The relevant physical laws are known exactly, as are the system parameters – while the required computations scale nicely.

This is the case for optics: there are codes that can accurately predict the performance of conventional optical systems. I used to work in an optical design shop [EDSG, part of Hughes Aircraft] where they did this every day.

Once they had designed and built a prototype of a periscope-type system (for an AFV, if memory serves). A general arrived for the big demo and looked at a target through the prototype.
“It’s upside down.”, he said. and so it was. That’s why they pay these guys the big bucks.

The prototype had only been tested on a crosshair reticle: looked the same upside down as right side up. Fixed with a prism.

To err is human, true in writing sims as well as everywhere else in life: but tests can expose those errors.

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

47 Responses to Simulations

  1. Is that general still available for hire? Would be interested to hear what he said about simulations that show that slightly rising temperatures are inevitably bad for living things.

  2. magusjanus says:

    I don’t think we’ve ever had a solid Westhunt post dedicated to global warming. Would be good to get his take on it, how much/how soon, important feedbacks (clouds?), likely econ effects, best approaches to it (and worst).

    I suspect I have a decent Guess what most would be but would be great to see nonetheless.

    • Slimboy Fat says:

      There was some talk on one of the James Miller interviews on it, i forget which one.

    • Anonymous says:

      I think it is an interesting subject area but It draws emotional people who have no clue what they are talking about. Talk about politics or talk about global warming and the comment section quickly plunges into ideological rants. Cochran once talked a little bit about global warming and said (if I recall right) feedback loops make simulation predictions really tricky and difficult to be accurate with. I am very leery of simulations that plug in factors that are a lot more unknown than the simulation designer admits to. They can be a great tool in the right hands and a perfect example of GIGO (garbage in-garbage out) in the wrong ones.

      • magusjanus says:

        I get what you’re saying but let’s be real here, he talks bout black-white race differentials and viruses causing male obligate homosexuality. I think the “don’t talk bout controversial topics that might incite clueless emotional people” ship sailed a while ago.

        And regardless, f those people. I want to read his thoughts on the matter (and those of intelligent commenters).

  3. AppSocRes says:

    This is a recent simulation involving both optics and general relativity: Kip Thorne and a team of special effects animators who specialize in optical effects collaborated to generate an accurate portrayal of what human observers would see if they could view a black hole much more closely than is now possible. They wound up with a picture completely different than what one sees on all those TV science documentaries about black holes: https://www.wired.com/2014/10/astrophysics-interstellar-black-hole/

  4. pyrrhus says:

    I wrote a completely unsophisticated simulation program for the world economy back in the mid seventies based solely on projected resource depletion. It predicted major problems beginning in this decade, about 2013 as I recall….Sometimes focusing on just one variable produces very interesting results.

    • Florida resident says:

      I have a rather universal answer to
      -How are you ?
      -Worse than it has been, but better than it will be.
      To pyrrhus:
      what was the criterion, that satisfied conclusion “major problems” ?

  5. RCB says:

    What languages do you code in?

  6. dearieme says:

    “There are systems that are hard to simulate”: that’s how I made a living at one point. Goodness, it was fun. It did, however, mean that when I first read some of the papers on Global Warming I reacted with a mixture of amazement and coarse cackling. Dear God those chaps were incompetent. It was only later (I suspect) that they decided to try to hide their incompetence by simply lying.

    • whyteablog says:

      Where did they go wrong? I’ve been hankering to see competent people speak out on this.

      • dearieme says:

        It would take a book.

        Before noting how they compared output temperatures from their models with measured temperatures, I contemplated how they fabricated “measured temperatures”. Perhaps the most outrageous technique was how they coped with what they thought was a defective measurement. Instead of discarding it, they would take an intricate average of the bad reading and of other readings “nearby”, where nearby might be hundreds of kilometres away. The bad measurement would then be used repeatedly in other averaging calculations, thus spreading badness around the planet. And, of course, “nearby” might be on, oh, the other side of a mountain range, perhaps.

        The funniest individual set of readings was from Antarctica, where it had clearly not occurred to the clowns that their thermometer was buried deep under snow for part of the year. Oh yes, and they did, from time to time, muddle up their latitudes and longitudes, or longitudes East or West: not careful workers, you might say. Having thus assembled their “measurements”, they wanted to compare them with predicted temperatures from the models. The grid points on the models didn’t coincide with the locations of the thermometers, so some sort of interpolation would be necessary. Since the models used finite element methods there were already functions that would provide a natural way to do the interpolation, ‘natural’ in the sense of being consistent with the numerical technique used effect the simulations. They, presumably unfamiliar with the internal workings of the codes they used, just slapped on primitive linear interpolations covering absurdly large distances.

        But that wasn’t the half of it. The models themselves were conspicuously unlikely to be anywhere near good enough to predict temperatures in a complicated physico-chemico-biological system like Earth’s. Some bits looked reasonable; some bits were sweeping empirical simplifications, the parameters in which were little better than guesses; and no doubt lots of necessary bits were simply missing – omitted for lack of knowledge, or lack of the imagination to see they were needed. Nor was there any sign that the “climate scientists” had given much thought to the possibility that their differential equations were exposed to the vagaries of what’s sometimes called Chaos Theory. The minimum requirement to enter this mathematical world is three coupled, nonlinear ordinary differential equations: their large sets of coupled, nonlinear partial differential equations were therefore exposed to the risk of producing predictions that were infected with meaningless variation. Was the output so affected? I don’t know, but more important there was no evidence that they knew either.

        It was all grisly: most physical scientists are quite clever people, and especially they are pretty clever at their speciality. By contrast hese people were mostly duds. Much of the work was done to a standard below that which you might hope to see in a final year undergraduate research project.

        If you want to get an idea of how slapdash they were, and how the transition from poor standards to downright dishonesty was occurring, you might like to read some of the emails that were leaked – or hacked – from the Climate Research Unit at the University of East Anglia. Google “climategate”.

        For a series of investigations of a high intellectual standard, your best bet on the web is Steve McIntyre’s Climate Audit blog. if I had my way he’d now be Sir Steve McIntyre.

        • whyteablog says:

          Interesting; thanks for the response. I must read further into this.

          • kot says:

            Nor was there any sign that the “climate scientists” had given much thought to the possibility that their differential equations were exposed to the vagaries of what’s sometimes called Chaos Theory. The minimum requirement to enter this mathematical world is three coupled, nonlinear ordinary differential equations: their large sets of coupled, nonlinear partial differential equations were therefore exposed to the risk of producing predictions that were infected with meaningless variation.

            Don’t bother, dearieme is a moron.

          • dearieme says:

            One of the more interesting things you’ll find (apart from commentary on the intellectual level exhibited here by kot) is an aspect of this chaos point. The difficulties in numerical solution of PDEs had been known about for ages; if my memory is right they were first remarked on by Richardson at about the time of WWI. They figure large in the problem of weather forecasting: that’s why forecasters didn’t make any claim, or at least any plausible claim, to be able to produce useful forecasts more than about 100 hours ahead. But, as if by magic, climate scientists claimed to be able to forecast 100 years ahead with useful precision. You’d think that this trick would have been carefully explained in numerous papers. If it was, these key papers weren’t cited often enough for me to stumble across them. In fact the papers I did see gave me the impression that people were running software the innards of which they knew little about.

            Another thing worth following is the revelation in the climategate emails of how careless the climate science community was in safeguarding facts. One of their tricks over the years was to “adjust” and readjust temperature measurements until they suited their case better. It turned out that the CRU at East Anglia had pulled this stunt so often that they had lost track of the original data altogether. In other words, when asked how their preferred set of numbers compared to the original set, they could give no honest answer. Nobody knew, so fundamental was their incompetence.

            You’ll see this pattern again and again; the field was full of duds – people whose intellects and conscientiousness were just not good enough for the tasks they’d set themselves. And like incompetent people in other walks of life – the building trades might be a good example – the incompetent turned to crookedness to try to preserve their ability to make a living. There resulted the jerry-built “science” of climate change.

            • kot says:

              One of the more interesting things you’ll find (apart from commentary on the intellectual level exhibited here by kot) is an aspect of this chaos point. The difficulties in numerical solution of PDEs had been known about for ages; if my memory is right they were first remarked on by Richardson at about the time of WWI. They figure large in the problem of weather forecasting: that’s why forecasters didn’t make any claim, or at least any plausible claim, to be able to produce useful forecasts more than about 100 hours ahead.

              Richardson was talking about NUMERICAL ANALYSIS, the problem of efficiently computing accurate solutions to differential equations. As in, given an ODE for f:R->R and an initial value f(0), get accurate estimates f(1), f(2)…. The question is how much work you have to do to get the desired number of accurate digits. This has nothing to do with the chaotic dynamics that are the real reason weather forecasts eventually fail. In fact the main use of “Chaos Theory” is identifying idiot laymen who think they can learn math from SciAm articles. With this statement alone you’ve already proven you know jack shit about math.

              But, as if by magic, climate scientists claimed to be able to forecast 100 years ahead with useful precision.

              Because the models are totally different on that scale. You don’t need to predict every eddy and current in a cup of coffee to figure out how fast it cools.

            • Ursiform says:

              There is a significant, fundamental difference between modeling trends and predicting instantaneous conditions. The limitations on modeling weather too far out has nothing to do with the ability to model climate trends. They are very different problems.

              Do you believe that computational fluid dynamics doesn’t work because the models can’t tell you where a particular molecule will be in 3.27 seconds?

      • pyrrhus says:

        Consult the sites Energy Matters or Wattsupwiththat. Many scientists have poked holes in the global warming scam….

  7. brendan r says:

    Introduction to Algorithms is the only textbook among the books you’ve recommended, I think. Should a non-coder read it? Do you think your algo experience helps you think about problems in general?

  8. Cpluskx says:

    How to find out if the universe is a simulation?

    • MawBTS says:

      Look for odd events that don’t conform to any known laws of physics. A perpetual motion machine would make me less certain of a natural universe.

      Also, it would take a massively powerful computer to model the interactions of every single atom and particle in the universe, so the dev might have implemented “hacks” to save computational resources. Maybe only the solar system is rendered in full detail, and objects outside our observation point use mipmapping. If so, we could build lots of Hubble telescopes and scatter them around the universe, so that the simulation is forced to render massive areas in full detail. The universe might then bog down, and the dev would step it. “Hey, guys, cut it out. My boss doesn’t know I’m doing this, and you’re making his PowerPoint run slowly.”

      • MawBTS says:

        The universe might then bog down, and the dev would step it.

        …Or he might turn down the graphical display settings.

        Imagining waking up one day, and the smallest object you can see is a molecule. No matter how powerful a microscope you use, you can no longer see atoms. Also, the electromagnetic spectrum has narrowed. Nuclear fission is now impossible and your microwave oven no longer works.

      • ziel says:

        Haha – that’s pretty good – but maybe in the Sim world they have better bosses?

        Anyway, things that violate the laws of physics would perhaps be too obvious. I look for things that on a case-by-case basis are perfectly ordinary, but in the aggregate, seem incredible. Things like the world supply chain. For example when drinking a bottle of wine, I think about how many bottles of wine there must be within a a couple square miles of my house, and how many more there are in the world, and how many grapes it must take to fill a single bottle, and could there possibly be enough grapes grown to fill all those bottles?

        Or oil? Can we really be pumping enough oil to be running all this electricity, all these cars, planes, trains, ships? Oil has the right defined properties so that it is plausible as a power source – who would ever believe solar panels and wind as our chief energy source? – at least as it is currently configured. But oil is perfect because it “comes out of the ground” – no one can see it, whose to say how much there is?

        When I see WWII naval documentaries, and all those ships, and the re-fueling ships – c’mon. They were really managing to fill these ships up with enough oil to be criss-crossing the oceans? – think about how much fuel a boat uses – we’re supposed to believe all these ships – and aircraft carriers – could be fueled up and travel thousands of miles across the Pacific?

        But then again, my Dad was on one of those ships, and he had some stories, so that’s an argument for “the real world” – but he didn’t have many stories. In fact, only a very few WWII vets ever talk about it – “too upsetting” everyone says. Well, ok, that’s convenient, but maybe there just isn’t much there, just a little backstory inserted by the sims, with only a select few to have enough details to fill out the narrative.

        But I think things are starting to spiral out of control in the sim – look for another 300-year reset in the ‘history’ books to push us forward a few centuries like in the first millenium.

        • gcochran9 says:

          +2, for a faked WWII and a ref to Fomenko’s New Chronology.

          You’ve earned a guest post, if you want.

          • anon says:

            I recently discovered Kasparov believes in new chronology.

            • gcochran9 says:

              You’re kidding.

            • pyrrhus says:

              I had never heard of Fomenko’s New Chronology until now, quite fascinating! It started when Fomenko, a mathematician working on astrophysics, checked out some NASA calculations based on historical reports of lunar eclipses. NASA’s mathematician, named Newton, concluded that the moon had been doing some very strange (and no doubt impossible) things in historical times. Fomenko concluded that the history was bunk, but subsequent calculations showed that the reports would work if they were transposed to the Middle Ages….

          • ziel says:

            Guest post – oh that’s tempting, but I think I’m out of ideas

            • gcochran9 says:

              I have plenty to spare. But if you write one of them up, it’s all your fault!.

              General policy: anyone that says something seriously interesting (or hilarious) is a candidate for a guest post. I get a veto on content though.

        • Cpluskx says:

          ”But I think things are starting to spiral out of control in the sim”
          Why?

          • dave chamberlin says:

            it’s all bullshit. The sim is a lovely concept that I like and appreciate but has a very small chance of being true. Wanna sell bullshit?

            !)Make it scary
            2)Use magic in a psuedo scientific way
            3)Tell them what they want to hear as if you aren’t a bullshitter

            Now I can come across a few folks that I can’t bullshit. They don’t want bullshit.They care about the complex truth.They even consider such ideas as possible but improbable. Those are my friends.

        • ChrisA says:

          If we are in a Sim and there are mistakes, won’t we simply be programmed to ignore the mistakes? Furthermore, you don’t have to simulate the entire universe, just the bits of it being observed by the “intelligent” observers in your sim. Which might be just me for all I know.

      • st says:

        “…odd events that don’t conform to any known laws of physics”. Really? That would be everything? This is easy. It seems that space background radiation does not conforms to Copernican principle – results from Planck space probe CMB mapping: https://www.youtube.com/watch?v=MjSoY3MnJBE
        https://www.researchgate.net/publication/1737826_Goodness_in_the_Axis_of_Evil etc.

        There could be more tricks however, instead of mipmapping the space, the hyphotetical programmer could have mipmap the time – rendering fully only the present, rendering partially not so distant past – say, last 50 000 years and only sketching anything older than one billion years – or so. That would have saved him a lot more computing power than mipmapping the space. But would get any scientific discipline dealing with the past in trouble.

  9. Peripatetic commenter says:

    Wait.

    “It’s upside down.”, he said. and so it was. That’s why they pay these guys the big bucks.

    Was that a reference to upside down proxies?

    https://climateaudit.org/2009/09/03/kaufmann-and-upside-down-mann/

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s