Yeah and they can fail because there's e.g. no noise to make an oscillator start which falls under "initial assumptions" I'd say.
Guess I've been fortunate to never use a model that didn't seem to represent reality in the cases I used it but I don't use Maxim regularly (at all.)
the audio industry tends to be a world of incremental improvements rather than revolutions. The average "workstation synthesizer" synthesis engine tends to have a lifespan of a decade. Many variations and updates and "Special Editions" and "Extreme Editions" are manufactured at various price points.
Products are released in 2019 that cost the better part of $4000 and have a 3" non-touch-sensitive display and 384 megabytes of editing memory and the reviewers will gush about how revolutionary it all is. The polyphony has been upgraded from 64 to 72. amazingggg
People try clean-sheet totally radical new designs from time to time and musicians mostly sulk and complain it doesn't sound like the good ol' days.
Space shuttle orbiter was dynamically unstable. Turn off the computers at 40k feet and it heads straight for the ground like a lawn dart. Can't just "trim" it and let it glide forget it.
It still handled and fell like a brick with wings at low altitude with feedback loops constantly working to stabilize it but enough not like a lawn dart to make it landable.
Vertical landing is a nice feature, but vertical takeoff with full bomb load and maximum fuel load is questionable. If it is possible, it as least consumes a lot of fuel, reducing the combat radius.
At least during the Falkland war, the Harriers were launched horizontally with a catapult and a ramp. After the mission with the ammunition and fuel used, the light plane landed vertically.
While the ships were much smaller than US style carrier vehicles, special ships with catapults were required, not just any ship would do.
There was no catapault; the Harrier engines provided the oomph.
The F35 is sometimes called STVL, slow takeoff and vertical land.
Aircraft mechanics, who used to learn on mock-ups of engines and systems, now learn on simulations. The Vaughn College of Aeronautics in New York discarded all their mock-ups. There could be interesting results.
Yeah, that's what they thought in Viet Nam when they sent Phantoms to do the job with electronics and missiles. Then they regretted it. Yes, electronics and missiles are much more reliable today. But don't let that fool you.
You could be right. Missiles on a drone are a formidable combination.
Both the military and the airlines are having trouble recruiting pilots.
Over here the military pilots are having difficult finding aircraft to fly.
Over here we have more rear-admirals than capital ships in the navy.
Pratt has a giant mechanics training facility in East Hartford, the size of a modest stadium. They use real engines.
No, based on observations (simulations are the all-terms-considered final step, NOT THE BASE). Several technologies, but NOT the world (and not 'the economy') will have to change.
The claim you just made, wrenches ALL those nouns out of their proper meanings.
Do you have mind picture, or a simulation in mind? Why not? How else can you support an answer to that question?
Imagination and simulations are ways of dealing with the future. If you don't use them, you don't plan, and folk who DO will outcompete you.
In a complex, nonlinear, chaotic simulation, you can tune its parameters to give any results you want. Climate sims are obviously tuned to hindcast accurately (they won't get published if they don't) but that doesn't make them predictive. Past sims have not been very predictive. The Statue of Liberty is still above water. The polar bears are doing fine.
But, of course, we have more powerful computers now.
I am not so sure about the hindcasts.
For example how does the simulation explain the Atlantic warm period (Holocene climate optimum) in which some tree spices grow at the Arctic Circle (based on tree trunks found in the bottom of lakes) while currently the most northern areas are the southern Scandinavia.
That's now how one handles a simulation. Is that how you use SPICE? I'm not trying to lie to myself when I fiddle with a model. Why would anyone?
It sounds like you're trying to sneak up on a conspiracy theory, but it would have to be of the sort that starts with underpants and ends with
Not if you want the simulation to predict anything useful.
Specific example? The first IPCC report came out in 1990, and what it said about what the world was gong to be like in 2019 wasn't too far off the mark.
The Statue of Liberty is still above water. The polar
And where are the published simulation results that suggested that they would be under water in 2019?
Not that John Larkin has a clue what they would be telling him if he ever bothered to find out about what the science is saying.
He vastly prefers denialist propaganda web-sites which presumably provide him with all the flattery he craves, even though the "informaton" they present is deliberately misleading.
For that they need information on deep ocean currents, which the Argo buoys are still collecting.
We know about the effect of the El Nino/La Nina alternation on local and global climate because they happen over a few year.
The Multidecadal Atlantic Oscillation
is a lot slower, and wasn't named until 1993.
It does seem to reflect ocean currents moving around. We can see the surface currents but the return currents flowing in the depths of the oceans are less easily observed (which is why the Argo Buoy observations were set under way).
That doesn't explain the warm period from 9000 to 5000 years ago
It is interesting to note that in the beginning of this period, the sea level rose by 60 m to current level apparently due to melting glaciers. Since a lot of energy was lost to melting ice, the air would have ben even warmer.
The hindcasts usually curve-fit the (poorly) measured temp data since the LIA. I don't know that anyone has good therories or models for the giant swings over millions of years. Short-term sims are what's needed to sell massive political intervention.
We'll all be dead in 12 years if we don't act NOW sort of thing.
I don't model nonlinear chaotic systems in Spice, I model electronic circuits that I know I can simulate usefully. I do of course tune the model to get the results that I can put into production; the trick there is to not fool myself. Extreme sensitivity to any parameter or value is one red flag that the circuit is not suitable for production. Except when that sensitivity is in fact a mathematical artifact of Spice, like the giant femtosecond spikes that someone here mentioned recently.
Mike gets a lot of feedback about how LT Spice actually models circuits. Climate modelers don't.
Al Gore did pretty well.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required