Comments by "Tony Wilson" (@tonywilson4713) on "Thunderf00t"
channel.
-
280
-
62
-
48
-
38
-
32
-
31
-
31
-
30
-
29
-
25
-
22
-
21
-
18
-
18
-
17
-
15
-
14
-
14
-
12
-
11
-
11
-
10
-
10
-
9
-
9
-
9
-
9
-
9
-
8
-
8
-
8
-
@jameskelly3502 Great point, YOU ARE RIGHT and I checked it out and there's an explanation.
Here's the second paragraph of that press release.
"This is a firm fixed-price, indefinite-delivery/indefinite-quantity contract modification for the Crew-10, Crew-11, Crew-12, Crew-13, and Crew-14 flights. The value of this modification for all five missions and related mission services is $1,436,438,446. The amount includes ground, launch, in-orbit, and return and recovery operations, cargo transportation for each mission, and a lifeboat capability while docked to the International Space Station. The period of performance runs through 2030 and brings the total CCtCap contract value with SpaceX to $4,927,306,350."
SHOR EXPLANATION
When people are talking about the US$70Million that's the LAUNCH. What you are talking about is the ENTIRE PROJECT with all the other stuff added in. As you can see there's a difference and quite often it can be a massive difference.
Those 5 Crew Dragon missions average just under US$300million which means that on top of the $70M for the launch there's almost another $230M for each flight. This is not simply a NASA problem. Its actually a major problem with projects EVERYWHERE.
For Example: Right now in Australia (as I'll explain) we have the AUKUS submarine project. The current Block 5 Virginia subs have a cost AUD$5 Billion each. The project cost for the 8 subs is AU$33-46 Billion each. From what information that's available there's AU$20-32 Billion for each sub that is currently unaccounted for.
LONGER EXPLANATION and again I am Sorry to all if this if this is long.
With almost every wonderful announcement the devil is in the details and the magic words in that announcement are indefinite-delivery/indefinite-quantity and related mission services.
BACKGROUND
I'm Australian but did my degree in America. It was in the late 80s during Reagan's Star Wars Program. Most of the department was on DARPA funding as were most of the postgrads. We all sort of new it was BS but we liked the funding and a lot of people got their MS & PhDs.
I came back to Australia afterwards which wasn't good timing and I ended up in industrial control systems, automation and robotics. In 2002 I met Apollo 17's Harrison Schmitt and he mentioned Helium-3, which meant we might be going back to the moon for mining. So I thought I'd go off to the Australian Mining sector and get some experience building mines so I could then do the same on the Moon.
Yeah I know that didn't work out, but what I got was an education in large multi-billion dollar contracting jobs and how contractors milk them for all they are worth. I also learned how to build complex systems in remote places and I know NASAs plans for a lunar base are crap because of these very issues of "other stuff."
Its not only in engineering and in fact the worst cases In Australia are in government department consulting. Go and check out the PwC scandal.
ISSUES WITH ENGINEERING and COST PLUSS CONTRACTS
1 mine site I worked on was the BHP Ravensthorpe Nickel project. Now you'd think BHP the largest mining company in the world would know how to get a mine built - WRONG. That job was budgeted at AU$1.5 Billion and ended up costing over AU$3.5 Billion and then they found out that someone had skimped on the drilling program and the ore body was nowhere near what they had expected in either quality or quantity. WHERE it really blew out was the cost plus contracting.
We had an electrician just not show for work one day on that project. He turned up at dinner time in the mess wearing another companies shirt and proudly announced he was getting an extra AU$10 and hour. Working 60+ hours a week, which you do on site that adds up to a lot of money. Within days other electricians were being snapped up in similar ways and the pay rate went from about AU$35/hr to AU$65/hr in about a week as people bounced from company to company.
For those who have never worked on cost plus contracting it goes like this. You have expenses (labor & stuff) and you hand them in and if the contract is cost plus 20% (which is common) then for every $1 of expense you get $1.20 in cash back. The reason why cost plus happens is that for large projects that go for several years you just cant plan everything. They can be made to work but the managers running them have to know what they are doing. This is why we see so many government and private sector projects blow out on their costs.
So when those electricians went from AU$35/hr to AU$65/hr their actual employers went from charging about $60/hr to over $100/hr. Here's where that adds up. If you have a 100 people and they suddenly cost an extra $10 that gets passed onto the company and they return (at 20%) $12 for which means your profit margin just went up $200 per hour.
So with something like SpaceX every time NASA makes an adjustment to a mission it means extra profit to SpaceX and its in their interests as a commercial company to max out those expenses.
Here in Australia we have a litany of projects both in the government and in the private sector that have blown out with some projects going billions over budget. The worst private sector project I heard of was the Gorgon Gas project which blew out by $15 Billion. You'd think Chevron would know their job and know how to manage a project BUT THEY DIDN'T. The Australian Navy is not only buying submarines but new frigates and that project recently jumped from AUD$30 Billion to AUD$45 Billion. These things happen from contract variations and that word "variation" is the sound of cash being printed to a cost+ contractor.
MORE EXAMPLES
Back in the day before the ISS came into being there was the Space Station Freedom project. Me and Classmates all believed that was what we'd be building before heading back to the Moon.
The 1st budget was USD$20 Billon and VP George Bush told them that was too expensive and to redesign it.
The 2nd budget was USD$30 Billon and VP George Bush told them stop being ridiculous.
The 3rd budget was USD$40 Billion and VP George Bush scrapped it, but not before a lot of money got spent doing those design studies.
In the end the ISS cost America $120 Billion to build and I think the current estimate puts it over $220 Billion so far when you add in the operations AND NOBODY has ever explained where its all gone.
The F35 program cost over a $Trillion in development AND NOBODY has ever explained where its all gone.
Here in Oz other than submarines and frigates we also have a patrol boat project underway. The previous class cost under AUD$30 Million each and these new ones are AUD$300 Million each - more than 10x the cost to do the same job AND NOBODY can explain the costs.
As part of Australia's AUKUS submarine project there was a AU$4.3 Billion dollar upgrade to facilities at the base near Perth. Knowing what they are basically doing I checked with a couple of people I know and that project shouldn't cost more than AU$1 Billion. Just 2 days ago they announced new plans and its now budgeted at AU$8 Billion with NO EXPLANTION what this extra AU$3.7 Billion is for let alone what most of the AU$4.3 was for.
BACK TO FALCON 9/CREW DRAGON
We know the cost of each Falcon 9 Crew Dragon launch is $70 Million but what you have shown is a fundamental problem in all these sorts of contracts. There's a lot more than just the basic costs than can be itemised and a lot of it we know nothing about.
So you are right there's a lot more, but when we are comparing apples to apples we have to compare what we can. The reason we talk about the at $70 Million is because we know its real and we can compare it to other things. We know the Shuttle flights cost US$350-450 million each, but that doesn't include the development and operational costs. We know the Soyuz seats at the end cost US$80 million each but we don't know what other costs with training (including language training) were incurred.
Hope that all explains it.
8
-
8
-
7
-
7
-
@grantadamson3478 Its was actually the classmate who's at NASA you should thank. About 20 years ago I was the one being mouthy and very "Elon like," and she rammed some hard truths back down my throat.
What I am really frustrated with the se days is the very same problems THEN are still the same problems we have NOW. Half the problem is that we take so much of what nature does in terms of life support. No matter how you want to consider it - if we want manned spaceflight beyond LEO the life support is an absolute. Only 24 men have been beyond LEO and they did it in equipment that was incredibly limited. Its just kept them alive and just got there and only just got back. One reason why they stopped was they were afraid of losing a crew. By 1972 they really had pushed their luck with 1960s technology.
One analogy I like to use for the moon is a remote mine site here on Earth. I actually met Harrison Schmitt (Apollo 17) in 2002 and he told me we were headed back tot he moon to mine it for He3. So I went off into the Australian mining industry for some hard on the ground experience with remote mining. The actual number of similarities between the Australian Outback and the Moon are more than you'd think.
For starters the first thing is a long range survey (Satellite & aerial) kind of like they did with the Ranger & Mariner probes.
Then they do an on the ground survey with a couple of geologists and an SUV. They go out with all their food, water & supplies check the place out and come home with a few samples. The only go for a few days or weeks.
Then they send out drilling crews for deeper exploration. The big difference at that point is the amount of equipment and men and supplies and for the first time accommodation, water storage, toilets, showers, communications and power generation. They don't simply go for a few days - they go for weeks. Some stay out there for months.
If they find a suitable site with suitable resources then they go with huge amounts of gear. They send in a cast of 1000s to build a mine, the processing plant, the trucks, diggers, more accommodation, more water, more sewerage, more of everything. All that infrastructure takes time effort and construction people and construction equipment.
Apollo was just like those couple of geologists doing the on the ground survey who bring back samples from site. The Apollo LM was like a space SUV, but for the next phase we need trucks not SUVs. I can build a mine site here only using SUVs but it would take 1000s of trips. What we need next is the space equivalent of a Kenworth and I don't know if Elon's new rocket is a Kenworth or just a bigger SUV. 🤷♂️🤷♀️
6
-
6
-
6
-
6
-
6
-
6
-
5
-
5
-
5
-
5
-
5
-
5
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
4
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
3
-
2
-
2
-
@Thunderf00t I don't know if you saw my last comment on the other video, but I do control systems and one of my main projects between 2006 & 2011 was an industrial bio-conversion plant in Perth, Western Australia that processed Municipal Solid Waste (NSW). As a Brit I'll assume you know where Perth is.
Sorry if this is a lengthy reply but if you want any further material I'll see what I can get you.
Keep up the good work. I just hate technology scammers.
I don't know if in your work you've had to deal with a control system engineers but I have to deal with all types of engineers and technologists - mechanical, chemical and in that case microbiologists. Early on I asked what made their process special so I got an personalized education in the microbiology of composting. My job wasn't to make the bacteria do its funky stuff it was to build a control system that included the tools to let the microbiologists make the bacteria do its funky stuff.
So there are many details about the bacteria I never was privy to. It was just this stuff they called "water" except it was dark brown and smelt awful. I'd describe it as biologically active liquor.
Their process was pretty novel in how fast it worked (21 days). In the natural environment composting can take anywhere from months to years depending on all sorts of factors - the microorganism types (worms, slugs, etc.), bacteria types (single cell bugs), temperature, moisture, oxygen levels, existing soil type, light or lack of light,.... etc. The list is quite lengthy. That's not that great for industrial processes where they want consistency.
Most of the OTHER processes developed did things like heap up the compost into mounds out in paddocks and left to ferment in situ. That releases lots of stuff (CO2, CH4,.... etc.). Other places simply dumped it into an unused quarry or open cut mine and cap it with a giant rubber sheet and tap the gas as its produced. Other processes munch/mince it all up and put it into silos so they can tap off the combustibles like methane. If not watched those silos can get so hot they catch on fire.
In these processes you do get more than methane. I know I had to engineer the electrical system to handle hydrogen. There wasn't much of it but there was enough to influence the electrical design because hydrogen ignites so easily. And yes many of the clowns promoting hydrogen have no idea.
Mostly those systems that put compost into silos and leave it alone take 60-90 days (not hours) to decompose a batch into compost. The variation mainly comes from the time of year. Municipal waste has EVERYTHING we put in the wheely bins - food waste, cloth, plastics, metals, grass clippings, leaves,.... etc.
So just giving people the impression that Lomi can provide "composted" material in 4 hours is pretty much garbage.
The company I was working with had come up with a way to handle the seasonal changes and make the process extremely rapid by tailoring the bacterial decomp. They could do a batch of the organic fraction of municipal waste in 21 days complete cycle. Raw waste came in and was smashed in a giant tumbling garbage smasher. Almost a Lomi on Superman level steroids. After that the metals, sand, glass and plastics were removed and finally it was loaded into a vessel (about 4-5 stories high) where it was introduced to the bacteria. At that stage there was oxygen and the process was aerobic and produced CO2. After the vessel was full it was sealed and once the bacteria had finished consuming all the oxygen was when the real magic started.
Certain bacteria are not like mammals, birds, etc. When the oxygen runs out they don't die they switch their metabolism. In a sealed container that process is gradual and we could watch it happen on the gas analyzer data. The bacteria would start consuming organics and oxygen producing C02 but as the oxygen ran out it would switch and start consuming organics and CO2 producing methane and a few other hydrocarbons. Eventually the bacteria would consume all it could. We'd see that on the gas analyzer as methane had stopped being produced. After that we reintroduced air and with it oxygen. Since this step was abrupt rather than gradual the sudden introduction of oxygen would kill the bacteria. After a few days of aeration all that was left was sterile very high quality compost.
All Lomi is, is the first step in a process like that. Its just a smashing and grinding machine.
*Laughably you can do the same thing with an old blender or nutri-bullet for a lot less than $500. Just blend up your scraps with some water and put it all in a cloth bag. Squeeze out the water and/or hang it somewhere to drain like they do for cheese making. Put the drained off water on you plants because it will have all sorts of organics. Then put the mashed up stuff in a composter and let nature do its stuff.
That or spend $20-40 million and build a system like I worked on.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
@shiftyjesusfish There's a little known subject in that some call "planetary mechanics." Its sort of the first step in any ideas about terraforming. Its about raw numbers regarding how much stuff you need. Your not worried about the dynamics of the planet as it rotates about the sun (or a star) its just about how much stuff you need.
Mars has a surface area of 144,000,000 km If you consider it like very big ball with 144,000,000 cubes stuck on the outside year sure there's gaps but then some of those cubes will have mountains inside. In quite simple terms you can approximate that to 144,000,000 cubic kilometers of AIR if you want to make the surface of Mars habitable. That's just the most basic thing getting enough air.
Atmospheric air at sea level is about 1.2kg per meter cubed at around 20C. So that 1 kilometer or air around Mars would weigh something like:
1.2 kg/m3 * 144,000,000 km3 * 1,000,000,000 m3/km3 = 172,800,000,000,000,000 kilograms of air.
Or dropping some zeros 172,800,000 Mega tons of air.
To put that in perspective the entire world production of iron ore is about 3,000 Megatons a year. So its an extraordinary amount of air and its not simply locked up in rocks or ice. Were talking about going comet hunting for suitable sources of Oxygen and Nitrogen.
If your wondering why this might seem crazy you'll now know why we don't discuss these things much. It just confuses people, but the most basic thing to realise is that planets are bloody big objects.
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
2
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
INDUSTRIAL CONTROL SYSTEMS engineer here:
This is WHY I KNOW FSD (full self driving) is a false and misleading concept at least for the moment.
AND APOLOGIES IF THIS IS LONGISH.
FYI - My degree was in aerospace but I have spent 30+ years in industrial control systems, automation and robotics. That has included working with many sensor systems including laser scanning systems. Although I don't work with vision systems I was introduced to the basics of vision systems in 1998 and am fully aware of many of the advancements in that area.
The actual problem with FSD is the amount of information that needs to be processed.
As human beings we just don't realise how much information our visual cortex processes every second and that's because most of it is processed by our peripheral system which is NOT part of our general conscious. Its all there in our periphery and we aren't focussing on it. Our peripheral system is extraordinary at clumping things together and dismissing irrelevant clumps while alerting our conscious system of potential threats or items of interest.
For example we don't see a 100,000 leaves attached to 1,000s branches attached to a trunk connected to a root system we see a tree. We don't see several million yellowish hairs covering 4 legs a body, a tail, a head, big teeth and an even bigger set of fangs we see a lion. Out on the African savannah people don't see millions of blades of grass, 1,000s and 1,000s of antelope, wildebeests, birds, insects and other wild life. OUR BRAIN via our peripheral system filters out the noise and will latch onto that 1 lion out of all those millions and millions of items in our visual range and SCREAM "that's a threat."
Similarly when driving a car down the average suburban street we see but don't focus on the millions of leaves - we see the trees and dismiss them as NOT a threat. We see don't see all the nuts, bolts, sheets of glass, sheet metal, paint and rubber - we see parked cars and dismiss them as NOT a threat. We see the bricks, boards, windows, window frames, paint - we see houses and dismiss them as NOT a threat. BUT WE DO SEE the bouncing ball coming down a driveway and our peripheral system SCREAMS that there's a dog or a child chasing after that ball OR we'll see a flash of something else and our peripheral system will alert our conscious brain to be aware of it. Like we'll suddenly notice one of the parked cars just moved.
This is what our peripheral system does with incredible speed. It processes a staggering mass of data every second and compares it to previous seconds and then filters out all the noise. This is why certain players in team sports seem so amazing in how they can suddenly pass to another player in a way that asks "How did they see them?" The answer is they are people whose peripheral system just operates better than average and in some rare cases a lot better.
NOW TRY AND CONSIDER HOW YOU MIGHT GET A COMPUTER TO DO THAT????
Remember no 2 trees are the same, and no 2 cars are ever parked the same, and no 2 houses are the same PLUS no 2 streets are the same anywhere on the planet. There's always something different. NOW CONSIDER that the perspective (as in the visual angles) on that scene is changing every second because your car is MOVING. You now have to process the next image and compare it to previous images to pick up that movement or notice that item that gets the wider scoping part of the system to flag an item of interest to the higher level decision making part of the system.
Suddenly you will realise that the scope of the technological task to get a computer to do what the human peripheral system does is monstrous.
Once you understand the scope of the task required to to do FSD you'll quickly realise that it MIGHT BE possible for some limited situations or MIGHT be possible once we get the visual scanning systems capable of sorting through all the noise to find those few items that need a higher level of evaluation we can't even begin the task BUT RIGHT NOW we don't have those systems because if they existed we hear all about it. We'd hear about the camera that's as good or better than a human eye and we'd hear about the processor that's as good as the human peripheral system AND NOBODY is even saying they have it under development or has made "the breakthrough".
Lets also NOT forget that a bunch of car manufacturers GAVE UP on FSD about 5 years ago. Uber sold off its FSD once they, (like the car manufacturers) realised just what it would take to do the job. This is also why, with the exception of a few tiny companies desperately trying for attention (and money) have stopped trying to build self FLYING air taxis.
Sorry if this was longish but I hope you get the gist of why it might be possible in future but NOT NOW.
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
@Fetidaf I all agree with that except for the general maintenance.
Outside of the engine & drivetrain there's not difference between electric, diesel, gas, hydrogen, LPG there's no difference between vehicle types.
That's one of the bullshit things regarding the push to electric vehicles. The cost of chassis, wheels, suspension, BRAKES, windscreens, door, lights,.. etc wont change. The Greenies also don't take into consideration the CO2 emissions regarding all those raw materials in a car or the production of components from those raw materials. And before you ask I used to build automated manufacturing cells for car parts in the supply chain.
If we really want to go electrical then the main push has to be in CONVERTING existing vehicles NOT replacing them.
We also need to be less concerned with trucks, power walls and mega batteries because we just don't have enough Lithium supply to do it all. We need to be looking at other batter technologies for stationary applications. Things like the Sadoway battery or the other batteries that don't use Lithium. In heavy vehicles like trucks and the giant dump trucks & diggers used in mining there's work underway in just changing the fuel over to hydrogen. Fortescue Metals in Australia is already testing that on dump trucks. Rolls Royce can supply stationary generators based based on existing diesel engines that use hydrogen as their fuel. It probably wont work for boats and ships because of the range they need. But I can see nuclear reactors taking over for the shipping industry.
1
-
@Fetidaf I don't why you think that the energy offset for a truck is a few 1000 miles.
Its little understood outside the manufacturing industries but in general your average family car consumes more energy and creates more pollution being made than it does in 20-25 years of normal driving.
I'll be honest the first time I heard that I called BS and the person telling me was actually a mechanic. He told me to FK-OFF and go look at how much goes into just making the raw steel, aluminum, plastic and glass.
Its why I say if the Greenies knew their stuff they campaign against new cars. People forget that the car industry is the biggest manufacturing industry in the world by raw materials and energy, because its doesn't just include the cars it includes all the stuff needed to make the cars. There's entire industries like industrial robots that primarily exist for car manufacturing. I know I used to program them.
Tesla's are full of metals that aren't found as much in other cars. There's a lot more copper and copper is incredibly energy intensive to refine it to where you can use it the way its used in a Tesla.
And before you ask, I have spent most of the last 20 years on mine sites. I have worked in the iron, ore, cocking coal, copper, aluminum, uranium and gold industries to name a few. One of the first mines I worked on was a copper mine that produced 99.999% pure copper. After getting it out of the ore with sulphuric acid they eventually got it into a near pure copper solution from which they electroplated it onto stainless steel sheets. That electroplating system used a lot of power.
Copper will be the next big issue in the energy transition. Like Lithium we don't produce enough. Its why people strip it out of old houses, factories and anywhere else they can.
1
-
1
-
1
-
1
-
1
-
1
-
1
-
INDUSTRIAL CONTROL SYSTEMS engineer here:
This is WHY I KNOW FSD (full self driving) is a false and misleading concept at least for the moment.
AND APOLOGIES IF THIS IS LONGISH.
FYI - My degree was in aerospace but I have spent 30+ years in industrial control systems, automation and robotics. That has included working with many sensor systems including laser scanning systems. Although I don't work with vision systems I was introduced to the basics of vision systems in 1998 and am fully aware of many of the advancements in that area.
The actual problem with FSD is the amount of information that needs to be processed.
As human beings we just don't realise how much information our visual cortex processes every second and that's because most of it is processed by our peripheral system which is NOT part of our general conscious. Its all there in our periphery and we aren't focussing on it. Our peripheral system is extraordinary at clumping things together and dismissing irrelevant clumps while alerting our conscious system of potential threats or items of interest.
For example we don't see a 100,000 leaves attached to 1,000s branches attached to a trunk connected to a root system we see a tree. We don't see several million yellowish hairs covering 4 legs a body, a tail, a head, big teeth and an even bigger set of fangs we see a lion. Out on the African savannah people don't see millions of blades of grass, 1,000s and 1,000s of antelope, wildebeests, birds, insects and other wild life. OUR BRAIN via our peripheral system filters out the noise and will latch onto that 1 lion out of all those millions and millions of items in our visual range and SCREAM "that's a threat."
Similarly when driving a car down the average suburban street we see but don't focus on the millions of leaves - we see the trees and dismiss them as NOT a threat. We see don't see all the nuts, bolts, sheets of glass, sheet metal, paint and rubber - we see parked cars and dismiss them as NOT a threat. We see the bricks, boards, windows, window frames, paint - we see houses and dismiss them as NOT a threat. BUT WE DO SEE the bouncing ball coming down a driveway and our peripheral system SCREAMS that there's a dog or a child chasing after that ball OR we'll see a flash of something else and our peripheral system will alert our conscious brain to be aware of it. Like we'll suddenly notice one of the parked cars just moved.
This is what our peripheral system does with incredible speed. It processes a staggering mass of data every second and compares it to previous seconds and then filters out all the noise. This is why certain players in team sports seem so amazing in how they can suddenly pass to another player in a way that asks "How did they see them?" The answer is they are people whose peripheral system just operates better than average and in some rare cases a lot better.
NOW TRY AND CONSIDER HOW YOU MIGHT GET A COMPUTER TO DO THAT????
Remember no 2 trees are the same, and no 2 cars are ever parked the same, and no 2 houses are the same PLUS no 2 streets are the same anywhere on the planet. There's always something different. NOW CONSIDER that the perspective (as in the visual angles) on that scene is changing every second because your car is MOVING. You now have to process the next image and compare it to previous images to pick up that movement or notice that item that gets the wider scoping part of the system to flag an item of interest to the higher level decision making part of the system.
Suddenly you will realise that the scope of the technological task to get a computer to do what the human peripheral system does is monstrous.
Once you understand the scope of the task required to to do FSD you'll quickly realise that it MIGHT BE possible for some limited situations or MIGHT be possible once we get the visual scanning systems capable of sorting through all the noise to find those few items that need a higher level of evaluation we can't even begin the task BUT RIGHT NOW we don't have those systems because if they existed we hear all about it. We'd hear about the camera that's as good or better than a human eye and we'd hear about the processor that's as good as the human peripheral system AND NOBODY is even saying they have it under development or has made "the breakthrough".
Lets also NOT forget that a bunch of car manufacturers GAVE UP on FSD about 5 years ago. Uber sold off its FSD once they, (like the car manufacturers) realised just what it would take to do the job. This is also why, with the exception of a few tiny companies desperately trying for attention (and money) have stopped trying to build self FLYING air taxis.
Sorry if this was longish but I hope you get the gist of why it might be possible in future but NOT NOW.
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
INDUSTRIAL CONTROL SYSTEMS engineer here:
This is WHY I KNOW FSD (full self driving) is a false and misleading concept at least for the moment.
AND APOLOGIES IF THIS IS LONGISH.
FYI - My degree was in aerospace but I have spent 30+ years in industrial control systems, automation and robotics. That has included working with many sensor systems including laser scanning systems. Although I don't work with vision systems I was introduced to the basics of vision systems in 1998 and am fully aware of many of the advancements in that area.
The actual problem with FSD is the amount of information that needs to be processed.
As human beings we just don't realise how much information our visual cortex processes every second and that's because most of it is processed by our peripheral system which is NOT part of our general conscious. Its all there in our periphery and we aren't focussing on it. Our peripheral system is extraordinary at clumping things together and dismissing irrelevant clumps while alerting our conscious system of potential threats or items of interest.
For example we don't see a 100,000 leaves attached to 1,000s branches attached to a trunk connected to a root system we see a tree. We don't see several million yellowish hairs covering 4 legs a body, a tail, a head, big teeth and an even bigger set of fangs we see a lion. Out on the African savannah people don't see millions of blades of grass, 1,000s and 1,000s of antelope, wildebeests, birds, insects and other wild life. OUR BRAIN via our peripheral system filters out the noise and will latch onto that 1 lion out of all those millions and millions of items in our visual range and SCREAM "that's a threat."
Similarly when driving a car down the average suburban street we see but don't focus on the millions of leaves - we see the trees and dismiss them as NOT a threat. We see don't see all the nuts, bolts, sheets of glass, sheet metal, paint and rubber - we see parked cars and dismiss them as NOT a threat. We see the bricks, boards, windows, window frames, paint - we see houses and dismiss them as NOT a threat. BUT WE DO SEE the bouncing ball coming down a driveway and our peripheral system SCREAMS that there's a dog or a child chasing after that ball OR we'll see a flash of something else and our peripheral system will alert our conscious brain to be aware of it. Like we'll suddenly notice one of the parked cars just moved.
This is what our peripheral system does with incredible speed. It processes a staggering mass of data every second and compares it to previous seconds and then filters out all the noise. This is why certain players in team sports seem so amazing in how they can suddenly pass to another player in a way that asks "How did they see them?" The answer is they are people whose peripheral system just operates better than average and in some rare cases a lot better.
NOW TRY AND CONSIDER HOW YOU MIGHT GET A COMPUTER TO DO THAT????
Remember no 2 trees are the same, and no 2 cars are ever parked the same, and no 2 houses are the same PLUS no 2 streets are the same anywhere on the planet. There's always something different. NOW CONSIDER that the perspective (as in the visual angles) on that scene is changing every second because your car is MOVING. You now have to process the next image and compare it to previous images to pick up that movement or notice that item that gets the wider scoping part of the system to flag an item of interest to the higher level decision making part of the system.
Suddenly you will realise that the scope of the technological task to get a computer to do what the human peripheral system does is monstrous.
Once you understand the scope of the task required to to do FSD you'll quickly realise that it MIGHT BE possible for some limited situations or MIGHT be possible once we get the visual scanning systems capable of sorting through all the noise to find those few items that need a higher level of evaluation we can't even begin the task BUT RIGHT NOW we don't have those systems because if they existed we hear all about it. We'd hear about the camera that's as good or better than a human eye and we'd hear about the processor that's as good as the human peripheral system AND NOBODY is even saying they have it under development or has made "the breakthrough".
Lets also NOT forget that a bunch of car manufacturers GAVE UP on FSD about 5 years ago. Uber sold off its FSD once they, (like the car manufacturers) realised just what it would take to do the job. This is also why, with the exception of a few tiny companies desperately trying for attention (and money) have stopped trying to build self FLYING air taxis.
Sorry if this was longish but I hope you get the gist of why it might be possible in future but NOT NOW.
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
@MrRaulstrnad What was your first thought?
For most people it's about 5 (usually 3-8) because they don't go beyond the switch or socket - they think of a desk lamp or a room light.
A few engineers (mostly with an electrical background) go onto the power station, but even fewer (without prompting go back to the basic metals needed.
Since I have worked in manufacturing and more recently mining I just go with the metals and my basic list is.
Tungsten (filament)
Tin (bulb cap)
Copper (low voltage wiring)
Aluminum (for the cores of the HV transmission lines)
Iron (transformers, transmission towers & lines)
Zinc (for galvanizing exposed iron components)
Nickel, Manganese & Chromium (for the stainless that wraps around HV transmission lines)
Coal (to process the iron ore)
Sulphur (to make sulphuric acid to process the other ores except aluminum)
Soda Ash (to make caustic soda to process the aluminum)
So that's at least 10 (or more) mines with all the factories that make mining equipment - diggers, trucks, crushers, conveyors, screens, tanks, pipes,..............etc
Then there's the trains, trucks, & ships that take the raw mining materials of to the processing plants.
Then there's the processing plants for taking the raw mining materials and producing raw stock - iron, nickel, aluminum smelters.....,etc.
Then there's the factories that make all the things that go into building a power station.
Then there's the factories that make all the things that go into building a power grid
Then there's the factories that make all the things that go into building a house with wiring.
Then there's the factories that make all the things that go into building a light.
And does not include any of the other infrastructure or many other raw materials required so that 1 person can have a home and turn on that light.
It's at least 20,000 and may be well over 50,000.
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1