Science: Everything explained by PV=nRT, F=ma=Gm(1)•m(2)/r^2

[aparch]

This might have fit in the TV or Movies thread, but this is more "Science-y!" than either threads are....

How IMAX sized ultra-high resolution LED panels have revolutionized VFX and is looking to make chroma-key movie sets obsolete.

 

[dxmnkd316]

Amazing

 

[dxmnkd316]

https://www.quantamagazine.org/the-black-hole-information-paradox-comes-to-an-end-20201029/

I hope these papers pan out. The information paradox just seemed as insane as entanglement.

 

[Kepler]

https://www.quantamagazine.org/the-b...-end-20201029/

I hope these papers pan out. The information paradox just seemed as insane as entanglement.

The Ads/CFT video within that link is great.

 

[Kepler]

Meanwhile, something important.

 

[MinnesotaNorthStar]

Meanwhile, something important.

I absolutely love the fact that cake pans helped...

 

[Kepler]

OK my science and infrastructure people. This bullsh-t or what?

Which is it?

+ IT'S HAPPENING!!!11!
+ Solid idea, needs time to work out the kinks
+ Unclear, check back later
+ Complete and utter snake oil

 

[WisconsinWildcard]

OK my science and infrastructure people. This bullsh-t or what?

Which is it?

+ IT'S HAPPENING!!!11!
+ Solid idea, needs time to work out the kinks
+ Unclear, check back later
+ Complete and utter snake oil

Snake Oil. Believe it when it actually happens. Let me know when they figure out to have a 100+ mile tube with a vacuum and handle the temperature changes from day/night and the resulting expansion/contraction of the material. That and like 50 other "and then a miracle happens" problems.

 

[LynahFan]

OK my science and infrastructure people. This bullsh-t or what?

Which is it?

+ IT'S HAPPENING!!!11!
+ Solid idea, needs time to work out the kinks
+ Unclear, check back later
+ Complete and utter snake oil

The science, and even the technology, is fine. I just can't fathom the economics, so I guess that's still option D. I was going to try to WAG my own number, but I'll just go with the Wiki-referenced estimate: "Michael Anderson, a professor of agricultural and resource economics at UC Berkeley, predicted that costs [for a LA-SF route] would amount to around US$100 billion."

If the pods are going 700 mi/hr and you need to space them by say, 30 seconds, for safety and each pod carries 5 people, the throughput is only 5 people per every 30 seconds = 0.17 people/sec, or 600 people per hour. Figure there's only about 15 hrs per day when people would want to travel, and that's 9,000 people per day, or 3.285M/year. That sounds like a lot, but if you want to pay off your $100B investment in, say, 10 years, then you have to amortize your $100B investment over just 32.85M passengers, or $3044 per round trip. That doesn't even factor in the cost of money (investors will want to be paid for agreeing to a 10-year payoff!), operating and maintenance costs, etc. You would definitely have to charge at least $5K per trip to even come close to breaking even.

Saving a few hours of time is easily worth $5k per trip to a guy like Musk, but there are definitely not 9,000 people per day who would agree with him. The basic economics ( investment / (throughput * payoff period)) are probably two orders of magnitude off - the target price should probably be in the $50-$200 range to actually fill the pipe. The only way it works is if you can change 2 of those 3 factors by an order of magnitude - e.g. cut the investment by 90% AND increase the throughput by 10x. It's just insane to contemplate how you could ever actually do that.

Not too surprising, given that this scheme was cooked up by the guy who dreamed up Tesla's financial plan (current P/E ratio = 843, auto sector average = 25)....

 

[dxmnkd316]

98% nonsense. For all of the reasons stated above.

My first thought was exactly along the lines of WW. I just don’t see a way to hold vacuum in a system that large. Lose vacuum and it would take days to pull it back down. Maybe (probably) longer. It takes minutes to tens of minutes to pull down to a medium vacuum in a 6,000-gallon reactor without an enormously oversized vac pump. To maintain that would fucking insane.

 

[Kepler]

The science, and even the technology, is fine. I just can't fathom the economics, so I guess that's still option D. I was going to try to WAG my own number, but I'll just go with the Wiki-referenced estimate: "Michael Anderson, a professor of agricultural and resource economics at UC Berkeley, predicted that costs [for a LA-SF route] would amount to around US$100 billion."

If the pods are going 700 mi/hr and you need to space them by say, 30 seconds, for safety and each pod carries 5 people, the throughput is only 5 people per every 30 seconds = 0.17 people/sec, or 600 people per hour. Figure there's only about 15 hrs per day when people would want to travel, and that's 9,000 people per day, or 3.285M/year. That sounds like a lot, but if you want to pay off your $100B investment in, say, 10 years, then you have to amortize your $100B investment over just 32.85M passengers, or $3044 per round trip. That doesn't even factor in the cost of money (investors will want to be paid for agreeing to a 10-year payoff!), operating and maintenance costs, etc. You would definitely have to charge at least $5K per trip to even come close to breaking even.

Saving a few hours of time is easily worth $5k per trip to a guy like Musk, but there are definitely not 9,000 people per day who would agree with him. The basic economics ( investment / (throughput * payoff period)) are probably two orders of magnitude off - the target price should probably be in the $50-$200 range to actually fill the pipe. The only way it works is if you can change 2 of those 3 factors by an order of magnitude - e.g. cut the investment by 90% AND increase the throughput by 10x. It's just insane to contemplate how you could ever actually do that.

Not too surprising, given that this scheme was cooked up by the guy who dreamed up Tesla's financial plan (current P/E ratio = 843, auto sector average = 25)....

This is all assuming only one snake. I assumed they would scale:

If you're going to go to all that trouble to build one lane just build 100 or 1000 stacked in some pretty configuration:

 

[Kepler]

https://twitter.com/wikkit/status/1302628302742487040?s=21

Holy cow. If this is true, it’s insane.

I always meant to say, I don't understand something.

So a spike in the composition of solar radiation triggers an earthquake. And if we park a detector next to the sun we get early warning. Which is great.

But... how does the signal beat the radiation back to Earth? Aren't they both traveling the speed of light? And why is it a full day's warning, not 8 minutes 30 seconds?

 

[LynahFan]

This is all assuming only one snake. I assumed they would scale:

If you're going to go to all that trouble to build one lane just build 100 or 1000 stacked in some pretty configuration:

Sure - that's a way to scale your throughput, but it also adds cost. You need a 100x *relative* improvement in the cost/throughput ratio. You'll never achieve that - the 10th tunnel is marginally cheaper, but still not free.

 

[dxmnkd316]

I always meant to say, I don't understand something.

So a spike in the composition of solar radiation triggers an earthquake. And if we park a detector next to the sun we get early warning. Which is great.

But... how does the signal beat the radiation back to Earth? Aren't they both traveling the speed of light? And why is it a full day's warning, not 8 minutes 30 seconds?

I don’t remember if we’re talking rays hit and then immediately an earthquake hits. But any warning is a good thing.

that said, I’m pretty sure this type of solar radiation takes hours (or more) to hit.

edit: oh the tweet itself says a day of warning. Which makes sense. It’s why we get solar weather forecasts for auroras days in advance.

 

[dxmnkd316]

Sure - that's a way to scale your throughput, but it also adds cost. You need a 100x *relative* improvement in the cost/throughput ratio. You'll never achieve that - the 10th tunnel is marginally cheaper, but still not free.

Isn’t the rule of scale something like 2^1.67

Edit: ah this is what I was thinking of:
”Overall costs of capital projects are known to be subject to economies of scale. A crude estimate is that if the capital cost for a given sized piece of equipment is known, changing the size will change the capital cost by the 0.6 power of the capacity ratio (the point six to the power rule).[SUP][18][/SUP][SUP][19][/SUP]
”

 

[Kepler]

Sure - that's a way to scale your throughput, but it also adds cost. You need a 100x *relative* improvement in the cost/throughput ratio. You'll never achieve that - the 10th tunnel is marginally cheaper, but still not free.

I assumed scaling costs would get better and better as you add more lanes. So if 1 lane costs x 10 costs 2x and 100 costs 3x. And from $5k a ride now you're down to $150. Which everybody will do.

Build one cluster from DC to LA and another from Chicago to Mexico City with a big f-cking Central Hub in St. Louis. The Flyover Fulcrum!

They can do a loop-de-loop inside the Gateway Arch!

 

[LynahFan]

I don’t remember if we’re talking rays hit and then immediately an earthquake hits. But any warning is a good thing.

that said, I’m pretty sure this type of solar radiation takes hours (or more) to hit.

edit: oh the tweet itself says a day of warning. Which makes sense. It’s why we get solar weather forecasts for auroras days in advance.

It's not that it takes hours to "hit." Every photon leaving the sun gets here in its own 8.5 minutes. But one photon will not trigger an earthquake - it would be an accumulation of "excess" photons over time that would eventually trigger an earthquake. So if we detect the rate of photon arrival increasing now, then we can start thinking that there might be an earthquake several/many/24 hours from now.

 

[Kepler]

I don’t remember if we’re talking rays hit and then immediately an earthquake hits. But any warning is a good thing.

that said, I’m pretty sure this type of solar radiation takes hours (or more) to hit.

edit: oh the tweet itself says a day of warning. Which makes sense. It’s why we get solar weather forecasts for auroras days in advance.

Or maybe there's a time interval in which the spike increases to maximum, so if you trigger the warning at the start of the interval you get variable prep time (depending on how far into the spike the quake is released).

Or maybe if you know it's ground currents you can seed the area bordering the fault with "earthquake rods" to dissipate it. And, I dunno, power cities?

Or maybe you can actually continually trigger microquakes, like avalanches, to prevent the Big One from building up?

 

[Kepler]

It's not that it takes hours to "hit." Every photon leaving the sun gets here in its own 8.5 minutes. But one photon will not trigger an earthquake - it would be an accumulation of "excess" photons over time that would eventually trigger an earthquake. So if we detect the rate of photon arrival increasing now, then we can start thinking that there might be an earthquake several/many/24 hours from now.

In that case an early detector satellite is completely superfluous. You just have one of those cool detectors like the ones they have for neutrinos. 'Cept you don't even need a buried tank for these, they are basically photochromatic sunglasses.

 

[LynahFan]

I assumed scaling costs would get better and better as you add more lanes. So if 1 lane costs x 10 costs 2x and 100 costs 3x. And from $5k a ride now you're down to $150. Which everybody will do.

Build one cluster from DC to LA and another from Chicago to Mexico City with a big f-cking Central Hub in St. Louis. The Flyover Fulcrum!

They can do a loop-de-loop inside the Gateway Arch!

Sure, but eventually you hit the ceiling for demand. Maybe the financing would work out with 100 tunnels. But supply and demand still bites you in the butt, because to fill those 100 tunnels with passengers (so they're each paying a tiny fraction of the startup costs), now you need 900,000 passengers per day. Are there really 900,000 Angelinos who want to get to NoCal every single day (or vice versa)? Even if you made it *free*, you wouldn't get 900,000 takers per day - the value proposition is simply not there if it costs $100B up front.

Musk keeps stringing this along by pretending that his startup cost will be less than $10B, but there is no freaking way that's remotely accurate.