Comments by "Mikko Rantalainen" (@MikkoRantalainen) on "Sabine Hossenfelder" channel.

  1. One would think that building 2 GW solar panel setup with battery storage for 12 hours in Sahara or nearby plus high power DC transmission lines would be cheaper than building 2 GW (times inefficiency) panel setup in space.
    1
  2. Great summary of the current state and I have to say that there's more promise for the cold fusion in the future than I expected before this video.
    1
  3.  @Nill757  Yes, I'd be happy to pay even $5000 if I could get $100/kWh. That means that the $5000 unit should have at least 50 kWh battery. Typically you get something like 10–20 kWh battery for that price.
    1
  4.  @Nill757  B NEF pricing is just the full battery pack. I would say at least half the price of the Tesla Powerwall is the inverter. High quality inverters are still rare and the one in Tesla Powerwall is really high quality. Unfortunately, Tesla doesn't sell an unit with a single inverter and a lot of cells which would make more sense for the buyer. And Tesla still makes at least 50% profit per Powerwall because there's no competition. Especially compared to their cars where they sell the replacement battery with much less profit. Telsa Model S 100 kWh battery pack costs about 19000–22000 USD (full replacement price without any reductions) which would put battery cost (including the whole pack with integrated cooling but without inverter) around 190–220 USD/kWh. Considering that the minimum cost for a pack should be around 140 USD/kWh, I'd guess Tesla is actually selling those replacement batteries with nearly zero profit.
    1
  5.  @Nill757  I'm using the B NEF pricing as the minimum. Of course, I cannot know what Tesla internally needs to manufacture their cells but commerically available cells are still around 100 USD/kWh at best. Yes, Tesla is making about 50% profit per Powerwall because there's no competition for "high performance inverter that can be connected to grid combined with Li-ion battery storage suitable for single home power needs" product class. And they cleverly integrate the cells into the unit so that you cannot simply get their inverter and use e.g. recycled Tesla Model S battery. And since their inverter probably includes BMS, too, you couldn't use Model S battery in any case unless you modify the firmware of the inverter / BMS. The heavy competition is in battery cells, not in full products. And Tesla doesn't cell you battery cells. The closest you can get is the full battery pack for one of their cars and I'm not even sure if you can get the battery without paying for the service to install it into one of their cars, too.
    1
  6. @ I agree. The accurate term would be "renewable by Sun resources".
    1
  7. It appears that the true status is that we're running out of cheap Helium. This is no different than "running out of oil".
    1
  8.  @reclaimer2019  Of course, there are sports where the skill matters so much that the pure physical strength cannot give you huge edge. I think my point still holds, you should first decide what are you trying to measure? If you take something like figure skating, you need some minimum level of physical strength to be able to execute the jumps but then there's also choreography and execution. On the other hand, e.g. 100 m sprint is about pure force only. And there people are trying to use illegal drugs and the anti-doping organizations do their best to prevent that.
    1
  9. I think cats demonstrate why "Earth is on infinite flat sea" is the only possible flat earth theory even if you believed things are flat. It definitely cannot be finite flat disk.
    1
  10. Great video about an important issue in modern science research! I would say that game theory strongly suggests that when the rules of the game suck, the end result will suck, too.
    1
  11. 6:51 Loved these reviews!
    1
  12. Great video but I wasn't the fan of the thumbnail. It seemed so clickbaity that I almost skipped this video even though I have appreciated many videos on this channel. I would say that I was happy to watch this video despite the thumbnail raising heavy yellow journalism feels.
    1
  13. Spin-offs: I don't think you can turn off the spin according to laws of physics.
    1
  14. 5:58 Loved the "spare IKEA part" photo! I'm pretty sure that didn't come from IKEA so it fits perfectly here.
    1
  15. I'm worried that social unstability is going happen sooner than politicians figure out that we should switch to UBI (Universal Basic Income) before AI and robots take majority of the jobs. We still have no idea how the few skilled humans that can do things that AI cannot (yet) do should be rewarded to keep working. If majority of the people can use their time as they see fit thanks to UBI, what kind of resources the society can provide for the skilled individuals to keep them working? It cannot be simply a bit above average salary because then it wouldn't be worth their time. Right now, the competition is between minimum pay and well paid job. When AI has taken over majority of the works, it's UBI + do whatever you like vs do the hardest possible work and have very little free time. And if you have very little free time, you cannot use the money from your work that well either. It might well turn out that the highly skilled people will ask for higher salary than nowadays and still work less hours because otherwise 100% free time + doing whatever you like might feel like the better option. Of course, one way to combat this would be the set the monthly UBI payments so low that you can barely live on UBI, which would go against the idea of UBI. In long run, it should be assumed that majority of the humans cannot do any of the jobs majority of the people are currently doing. Only highly skilled individuals able to do things that AI cannot (yet) done or individuals that do the job for cheaper than an AI or robot, can keep their jobs.
    1
  16. @onics I think it was supposed to be a joke to refer toasters or CRT televisions (which typically do have an electromagnet).
    1
  17. 18:35 "The free version will suggest you marry the prince of Nigeria." Seems legit.
    1
  18. It appears that the highly corrosive effects will be hardest to deal with in near term future.
    1
  19. I wish this variant of post quantum gravity were true. Dark energy and dark matter seem so huge a hack that the true nature of the universe must be something else. And since we know for sure that the don't understand the world between quantum physics and general relativity, it would be safe to assume that the solution lies there instead of inventing new energy and matter variants like dark energy and dark matter fans are suggesting. That said, I understand that dark energy and dark matter do offer plausible predictions so the math must be close to reality even though the explanation seems to suck. And assuming gravity has similar non-deterministic behavior to quantum physics at small scales seems like a reasonable assumption. Now the question is how to fix the linearity issue?
    1
  20. As quantum computers are inherently inaccurate and can output only approximate answers, it could also turn out that for practical tasks, AI generated output that simulates the output of quantum computers may often be good enough solution and it can run with less resources than building a real big enough quantum computer. And AI systems are very good with linear problems.
    1
  21. AI is a good possibility to dismantle the whole copyright system and I'm looking forward for it. Artists should use the same format as everybody else: sign a contract about working and get salary. Don't work for free and ask for payment later, as is the supposed model with copyright law today.
    1
  22. I think the most interesting question is how strong magnetic fields would the real world currents cause if this was used for long distance power grid power transfer? As far as I know, all known superconducting materials stop being superconductive when placed in "too strong" magnetic field and the "too strong" is often surprinsingly little. The obvious problem being that when you try to transfer more energy over the wire (superconductive or not) you increase the strength of the magnetic field around the wire. And in case your wire goes from superconductive to normal wire when you exceed the magnetic field strength with too high current, it will instantly fry your wire because if it was too much for the superconductive state, it will be definitely too much for the non-superconductive state and the cable will melt.
    1
  23. How about Quantized Inertia (QI)?
    1
  24. I think it would be better not to take down any videos that are not against the law. Simply change the description to say that you've later understood that the content is incorrect one way or another and link to improved knowledge, such as this video. If you're not sure if the content is accurate but don't want to declare it as incorrect yet, just add something like "[currently disputed]" to the title until you know the truth.
    1
  25. I think if you're mainly worried about air pollution, you should use battery EV cars nowadays. If, however, you're interested in total pollution, the optimal choice is less clear. EV cars cause more particle matter from wheels and brakes than other case because more torque to the wheels and greater mass. In addition, EV cars are more expensive which is the result of more resources getting used to manufacture those. And all that extra money over cheaper cars goes to different workers around the world in the end. And what those workers do with that extra money should also be considered. Imagine this: if that EV car salesman makes good bonus and goes to around the world trip, he just caused more pollution thanks to you purchasing the EV car. I'm driving an old diesel car and I maintain its engine as well as possible and drive as little as possible. I believe that the total cost of the car ownership correlates how much money you distribute around and how much resource usage is possible with that money globally. Of course, you still want to keep your machinery in top condition, even if it were old.
    1
  26. Great video! The only thing which I don't understand is why Lorentzian Distance is the true distance? I understood that is has been empirically validated so it just works but how can it be explained otherwise?
    1
  27. I think that for accessibility reasons alone, nobody should ever try to transmit information with light orange or dark orange. How about adding a star, dagger or some other clearly visible symbols instead like in every other study that doesn't intentionally mislead you?
    1
  28. I agree that the technology needed to use dialup modems must be investigated with carbon dating.
    1
  29. I agree. I would like to see one video like this every week. Make it the paper that got most traction in news that reach the common people. If the review results in "this actually looks like a good paper" that would be a great video, too. Of course, reacting to crap papers will be more entertaining.
    1
  30. I think the VW scandal was actually caused by the fact that VW tried to avoid using EGR so much. Even their previous engines before the engines with the scandal software had sometimes problems with EGR causing the intake manifold to get blocked over time. I don't understand the exact cause for this but I guess it's related to sulfur in the fuel (this problem was way more common in the US than Europe around year 2010) and/or the engine oil that was used in the engine.
    1
  31.  @garypeatling7927  We're still running year 2000 TDI PD (engine code AJM in case you're interested) with a fully stock EGR setup and it has never glogged. The intake manifold has carbon inlining similar to thin paint, not a thick layer.
    1
  32. If you can reserve 87 km² of space for the receiver antenna, you can instead build 87 km² of solar panels. Combined with flow battery that should provide more energy than smaller space solar panel. Plus maintenance on Earth based system is much cheaper. Plus did they make any estimates how much it would cost to build 87 km² antenna? That kind of sounds "pretty big" to me.
    1
  33. I would argue that even humans are not self-monitoring. Sure, human can typically notice if they have a bleeding wound but they rarely can notice problems in their own brain.
    1
  34. I'm not sure if MOND is correct either but it seems much more plausible explanation than dark matter. If MOND turns out to be true, how would it affect the calculations that gave us dark energy?
    1
  35. I think ability to pass entrance exams of many universities for different areas of study should give pretty good estimate of intelligence of an AI. I think GPT-4 has already excelled in US Bar Exam, required to be passed before one can work as a lawyer. In fact, it was good enough to place 90th percentile compared to all human exam takers and GPT-4 had zero Bar Exam specific training material or special knowledge. Had it been given all the official material, it would have probably done much better.
    1
  36. I guess one could argue that if you decide which of the two videos recommended by YouTube you watch next by tossing a fair coin, you'll affect what kind of information you receive and it will affect the course of rest of your life. We are deterministic machines but we are continously taking external input and if you add true randomness to that input, you'll change the results – but you cannot know how it will change. All the input your brain gets will rewire the brain a little bit (this is called learning) and the interpretation of next input you receive will depend on the whole history of all input to the brain. As a result, even if you watch both of the recommended videos after the toss of coin and use the coin only to decide which one you watch first, your future will depend on that toss of coin. I definitely agree that we can only enjoy the ride.
    1
  37. 1:00 I would like to see a well made video about the details of how the normal matter (4%), dark matter (22%) and dark energy (74%) totals are computed. I think it might well turn out that our understanding of gravity is simply incorrect and there isn't in fact any dark matter or dark energy. Both of those seem to arise from the fact that we assume our current model of gravity is correct in large scales and those things are then needed to match the measurements to theoretical models. I would not be that surprised to find that we don't need dark matter nor dark energy with some improved theory for gravity. That said, we know that the current understanding of gravity is very close to any measurements we have done so any new theory would need to match current theory very closely in small scales (say one solar system or smaller).
    1
  38.  @Nihlux  Sure, if you want to use gravity to ignite the fusion, the minimum reactor size is somewhere between the mass of Jupiter and the mass of Sun. My point was that we know for sure that fusion reaction and happen and it emits energy. The question is if we can harness it effectively for electricity production. When it comes to quantum computers, we do not know if a computer that can hold the state long enough to be usable is possible even in theory. Right now, it appears plausible that there's always some minimum amount of noise in such quantum system which prevents using it for practical implementation of complex computations, ever.
    1
  39.  @marco21274  Yes, "bigger reactors have a higher efficiency" in theory but in practice these bigger reactors have been so fragile that the reactor is in service mode more often than in production mode. Doesn't improve efficiency in any practical terms.
    1
  40.  @marco21274  At least here in Finland, you get power from solar panels with about 4 month offset from the actual peak power needs from the grid. As long as we lack a method to efficiently store electricity for 4–6 months, solar panels make no sense here for large scale energy production. With wind turbines, you could get with "only" a week or two worth of energy storage but event that's not practical with currently existing technology. In countries where you don't have max energy consumption in sync with the darkness of Winter while it doesn't wind, renewable energy sources are much easier to use with very little energy storage.
    1
  41.  @wrobi73  Who is paying for the hardware implementing the static prices? The reason the price fluctuates is that there isn't enough energy production for the peak hours and this is worked around with dynamic pricing which forces some electricity users to stop using electricity which allows available supply and demand to meet. You cannot simply decide that you have static prices and have enough electricity for everybody willing to pay the static price. You have to implement the peak power delivery for real.
    1
  42.  @marco21274  The climate that Finland has every year happens once in 50 or 100 years in Germany so you should design the system to cope with that.
    1
  43.  @alfvomall9509  We can definitely produce enough batteries but the problem is cost. Current battery tech is not cost effective for grid scale electricity storage but neither is hydrogen. And the improvements needed for either tech makes me think that flow batteries are the solution in long run and for small storage requirements right now something like Tesla Megapack is the least bad option. Australia has been pretty happy with their Tesla Megapack installations if I've understood correctly.
    1
  44. The whole point of AI systems is not to explicitly program them. I see AI development as ultimate expression of "no-code" movement.
    1
  45. 5:48 Am I the only one that reads the next line in parenthesis as "this is the range of values which would mean that the result is statistically non-significant"? And since value 1.37 belongs in range 0.48 and 3.6, it is clearly statistically non-significant and not even close to end of the range which would allow p-hacking to make it appear as significant.
    1
  46. As a Finn, I'm disappointed that two of the authors were from Finland. I would expect more from fellow Finns.
    1
  47.  @donaldhobson8873  If you can get old car batteries for cheap enough and can transport those for cheap enough, you could definitely build a battery out of them. However, because lead batteries are so heavy compared to energy capacity, moving a LOT of old batteries around country is probably worse option than just purchasing enough LFP cells for now. Any solution is better than having no grid scale storage at all.
    1
  48. Thanks for the tip about making the rolling R in the back of your throat. My native tongue is Finnish and Finnish has rolling R with the tip of the tongue, exactly opposite to German accent.
    1
  49. 7:00 Do you know about studies that would have measured the particles from gasoline engines? I've seem claims that say gasoline engines emit even more particles than diesel engines but the particles emitted by gasoline engines are really small and don't end up causing lots of mass, hence won't get caught in current legal limits.
    1
  50. The interesting question is will it ever be illegal to kill an AI? Does it change the ruling if you have intact backups? I would argue that AI in offline state (full model in offline storage) is not that different from a human being under general anesthesia. Would you be charged for murder if you lose the backup?
    1