As someone that switched from PCs to the Mac in the early 90s, then from the Mac to Windows 98, and the back to the Mac for 10.4, I think I can offer some insight.
First off, it's important to consider that the advantages and disadvantages vary over time. In most cases, historically, I switched back to the Mac just before 10.5 came out, and I can tell you that 10.5 made the contemporary Windows XP look like crap - and XP was great! 10.5 was just better. It was faster in general use, faster to boot, faster to switch users, had much better security (like it actually had some), Time Machine was nothing short of phenomenal, and Safari was easily the best browser available.
And around that time, the hardware was incomparable as well. TiBooks ate anything on the market in practically any measurement. The cheese grater Mac Pro is still so great people would rather have them than the new model.
Now? Not so much. macOS is fine, but the delta between it and Win10 is much thinner. The things I see now are much more limited. Chrome fixed the browser issue, but I still can't find a good mail client (the new version of Outlook, ugh, it never works right!). Performance is now on-par, and booting speed even better than the Mac. My Mac still gives me much less problems in terms of changing stuff, and I still don't have a single malware after 10 years hanging out on the 'net and downloading everything.
The current iMacs are still great, but its not alone, and on the laptop side the delta has closed a LOT. And Apple's insistence on thinner keyboards has made them craptastic. But many of those deltas that closed are only because Apple opened them - the MacBook Air remains a great machine, and it was years before anyone had something really comparable. But now they do.
Finally, "convergence". I have an iPhone and the way things magically move from the phone to the Mac to the web is great. I suspect the same is true on the PC/Android side, but I don't know because I don't have Android. Still, once you get used to it its great - like handing off a call from my phone to the computer when I'm at my desk at home.
>Are you saying that the heat generated by millions of degree temperature is the same than the one generated by about 100 degree?!
Sure, if there's less of it. Fusion reactors operate at conditions best described as a very good vacuum.
I'm sorry you find all of this so difficult to believe, but there's plenty of resources on the 'net you can read on the topic.
> It seems that the only one with problems to understand heat/temperature (and quite a few other basic concepts) is you.
I'm a physicist who worked in the energy industry for about a decade. I've written extensively on fusion technologies, and you've probably read some of those articles (assuming you've read any).
> Trying it for Minneapolis, I notice "system losses" figures in 0% for snow.
Actually it does figure it it, just not there. Snow often *increases* production, because it reflects more light onto clean panels. Snow coverage is modeled during the day-to-day simulation run.
> Something tells me that system isn't that accurate.
Is that "something" based on a single input and your decision to stop at that point complain about it rather than just put in a number? I'm sure you're capable of making this estimate. It's about 1/3rd of your roof's coverage because the cells heat up and cause the snow to slide off. Maybe 5 days a year.
The tool itself is considered to be accurate to within 10 to 12%, but there is widespread agreement it almost always underestimates actual production:
> Britain is a country with a huge number of poorly insulated homes.
I lived in Ireland for a year. One night I was getting cold on a windy night and noticed the drapes on the main window in the living room were blowing around. Ah ha, I just need to close the window!
I responded directly to your point. Don't blame me if you choose to redefine the terms "night" and "lots". And for that matter "cold", which no one in the UK should dare to define to someone who lives in Toronto!
In any event, its besides the issue anyway. As one can see on the National Grid's website:
There is plenty of demand during the day that PV can take. Every watt that comes from that is one that didn't come from something else, which is generally a good thing. Sure, if you keep moving the goalposts and coming up with new reasons why "it will never work" you could probably keep us going forever. But if you want to solve actual problems, PV is certainly part of that solution, as those very same CSV files demonstrate (they even have a separate column for it). Also surprising is the amount of pumped storage.
> They've been cooking soup in their Tokamaks since the Big Bang was invented in the 60's,
Actually, they started with stellarators in 1951.
They got a whole lot of money in the 70s and 80s to build the TFTR, which everyone was sure was going to reach break even.
It didn't.
So they got some more money to keep the ancient Alcator going, now in C-Mod form. It keeps running, zombie like, in spite of the fact that there's nothing left to learn from it. They also have some smaller-scale machines, but most of the interesting ones were cancelled.
>Dealing with temperatures You understand that temperature and heat are two different things, right?
Fusion takes place at high *temperatures*, but the *heat* loads are about the same as fission. That should not be surprising given that it outputs about the same amount of *power* (the rate of energy).
> Rooftop solar is about the least cost-effective way of running solar power
It is highly competitive above about 50 kW.
> but the cost of rooftop installation is high
For a large install on a flat roof it is close to even with ground mounts. The extra work of getting it all up there and installing around various HVAC and such is offset by the mechanically simpler and lighter install systems. It only really gets expensive in relative terms in small installs on tilted roofs.
> the cost of the inverter is relatively high
The inverters we sold scaled from about 50 cents/W for the smallest 250W models to about 22 cents (Canadian) for the 1 MW models. Unlike most forms of power, PV scales VERY linearly above about 50 kW.
> I don't want to be able to get one that's 50% more efficient a year later for the same price
You won't. I installed my system in 2010 using what were then one-down-from-the-best you could by panels, 230W a panel. That same panel today is 285W. That's a great improvement, but not 50% per year.
> I started looking at roof-top solar about 10 years ago. Back then, the cheap panels were 8% efficient
In 2007 most panels were around 160 W, but in the smaller form-factor using 5" cells. The jump that occurred between then and 2010 was due almost entirely to the move to 1 x 1.6m panels using 6" cells. Average efficiency during that period improved only slightly from about 10 to 12%, I'm not sure what type of panels you were looking at but apparently not A-quality examples?
We have largely reached the top of the S-curve, and price declines from here out will be smaller in conventional technologies. The epitaxial guys might have something to say, and the perkosites, but both are relatively low chances. $1/W is where it's going to be for a while.
No. Both only have an ROI of less than a decade IN CERTAIN SPECIFIC SITUATIONS.
Solar in SoCal, Nevada, New Mexico and Arizona generally have a fast payback. Because they see tons of son overall. Solar in Hawaii has a fast payback because the price of power in Hawaii is high due to geographic isolation.
Solar power in Chicago, or Minneapolis or Detroit or Seattle is a MUCH different story.
Really? Let's test that theory. Everyone reading this can make a single change and figure it out for their own location. Here we go...
Open http://pvwatts.nrel.gov/ Type in Chicago as the location and hit go Pick a weather station, likely the default, and hit go Change the system size to 1 (to unitize everything, you'll see that's the input to the steps below) and the module type to Premium (all panels made today are super-premium according to the definition in this program) Change the tilt to 30 degrees if you are between 40 and 50 north, and 35 if you are 50 to 60. I have not tried optimizing north of that Press go again. Write down the big blue number at the top. For Chicago I got 1312. I know this is lower than what people actually get, but PVWatts is a wonderful conservative estimate, so let's run with it.
Now that production number is useful,but for the next part what we really want to know is the Capacity Factor. That is easy to calculate, it's that number divided by the number of hours in a year, so in this case 1312 / (24 x 365) = 15% (which is less than my real-world array in Toronto, so this is why I say it's conservative).
Ok now we have the production side of things. That's half the problem. The other half is the money part. So here we go on that side...
Open the NREL LCoE calculator here: http://www.nrel.gov/analysis/tech_lcoe.html Change the period to 25, at a minimum. PV is generally thought to last about 40, but 25 is the warranty period so use that for now Change the Capital Cost to 1000, which is the current price for 1 kW (I told you that number would reappear) of commercial solar in the US Change the Capacity Factor to whatever you calculated above, in this case 15% Change the Fixed O&M to 17 (click the link to see why) Change the Variable O&M, Heat Rate and Fuel Cost to 0 (sunlight is free)
Look at the bottom line, the "Simple Levelized Cost of Renewable Energy". For the inputs above, that is 6.4 cents/kWh. That is competitive with wholesale averages (5.5), and *extremely* competitive for daytime peak.
Please, take 4 minutes of your life and do that calculation for your area. And if you want to know what it would be if you did it on your roof, change the 1000 to about 2500 to 3500, which accounts for the much higher costs of labor and construction on a small project.
So let's do one more? For the UK? Let's use York, which gets me 875 for the output of a 1k array at 30 degrees, which is 10% CF. That gets you a LCOE of 9.6 cents, which is highly competitive with wholesale rates in the UK.
Now your homework: repeat these calculations for your home using the closest weather station, with a Capital Cost of $3.50/W.
> just against stupid catch all remarks that say something is better than something else or has some specific ROI without taking into account any specifics.
Like your statement, which has no specifics?
Here, anyone can do this themselves. Go to pvwatts.nrel.gov and type in a nearby location. Change the system size to 1 kW and adjust the tilt angle - if you're above 40 degrees use 30 degrees instead of whatever they default to for California. You may also want to change the panel type to premium, because by the definitions of the program every panel made today is super-ultrapremium.
> Britain is a cold, dark country so we need lots of energy at night
No you don't. The night time load is about 1/2 the daytime, and that's why it's cheaper if you're on Economy 7 and 10.
> we had to import more than 10% of our energy requirements
Canada is awash with hydro and nuclear, and we import power all the time. It makes load balancing much easier if your grid spans as wide an area as possible.
Actually if you read it carefully, their claim of "first" refers only to the funding mechanism. And that's wrong too, there have been private reactors in the past.
"We are unveiling the first world-class controlled fusion device to have been designed, built, and operated by a private venture. "
What complete BS. Off the top of my head I can name the KMS ICF and the Riggatron as pure private-venture reactors that pre-date this one by*decades*. The later is named for the bank that funded it.
Pages and Numbers had many interesting features before they went cross platform, but both were incomplete. They needed another year of development before they would be really good.
That did not happen. They spent that time moving to iOS, eliminating any features that they couldn't implement on iOS. And that's still where we are today.
>Essentially, the EPA apparently considers any modification of the tractor
You can modify all sorts of crap on a JD tractor. Tires get changed all the time. You can change the entire cab if you want.
You just can't screw with the engine controls. Contrary to your line of argument, doing that has a very high probability of changing its emissions (like 100%).
> In other words, that EPA regulation should be considered unconstitutional
Then get a lawyer and sue them, and see if the court agrees with your asinine argument. And then we can put it on the list along with other nerd arguments like:
There's NO WAY Bell can stop our Blue Boxes! There's TOTALLY ILLEGAL for the government to spy on all our comms! There's NO WAY they can patent computer code! No one will ever get sued when using BitTorrent! etc.
I had an original Mac Pro that I got for a song so it became my primary machine. Over the years I added a big screen, drives, GPUs, memory, etc. It was not until 2014 that I considered an "upgraded" because I needed to move to 64bit in order to keep running Xcode. That left me with the decision of buying the top of the line iMac, or a Mac Pro. I went the later because I could keep my monitor, which I love. The price difference if you ignored the monitor was a couple of hundred bucks, so why not?
But quite frankly, the machine sucks. Oh, it's fast, and small, and very very quiet. And it looks good. But really, those are it's only good points. And there are lots of bad points...
1) You get two GPUs, one for rendering and one for calculations. However, I never (?) do GPU-hosted calculations, so that GPU is idle. I am certainly not alone in needing a single GPU. I would be happy if the second GPU could be used for rendering in a CrossFire-like way, but no one is bothering with that. So I have an expensive GPU doing nothing. Worse, it can't be used as a backup, as I understand it, so if the display GPU fails, my machine is dead.
2) There is a single "drive slot". It is non-standard (although such a standard did not really exist at the time). It also sprouts from one of the two GPUs, which is ridiculous. So Apple has to make two different GPU cards, one with and one without the SSD slot.
3) You may say it needs only one drive slot because you'll use external drives... right? Well here's the problem with that: most external drives are so much slower than the internal SSD that the machine is fully booted before the external is up and running (its FAST). Since you'll probably put your user account on that drive... odd things happen. Like your account is read-only. Or you get a sort of guest-like account. The only solution is to reboot.
4) It has FOUR USB ports. That isn't enough for anyone. Ever. All of them are on the back. So every time you want to plug in a USB key, you have to spin the machine. I gave up and left it back-to-front, so everyone gets to see my cable spaghetti.
5) It has SIX Thunderbolt. I have exactly one TB device, the screen.
6) All the ports are at the top of the machine, so the cables hang down and bend at the strain relief. If anything heavy ever falls on the cables, they're going to break. This is just bad design.
The good news is we can fix it all, easily:
1) put in at least two M2/U2 ports, preferably four. I shouldn't HAVE to use an external drive, and I shouldn't have to throw away the drive it came with if I want a larger option.
2) alternately, add a bay for a single (or two) conventional SATA laptop drives. You can get 1.5TBs for reasonable prices. It would make the case *slightly* larger, but who cares?
3) 10 USB-C, two of them on the front.
4) either move the ports down, or angle them downward to release the strain on the cables.
5) allow the system to run with a single GPU. And allow us to swap them! There's a number of small-form-factor GPU slots out there, and surely one of the companies you deal will with make one that can be mounted to the cooling block somehow.
"For his invention to work as described, they say, it would probably have to abandon the laws of thermodynamics, which say perpetual motion is not possible,"
I read this section of the article several times, and I cannot make heads nor tails.
The entire invention, assuming it is real, replaces the normally plastic-and-liquid electrolyte with a glass sheet. The major result of this change is that it prevents ion movement across layers, which suppresses dendrite growth. As a result, you can replace the electrodes with pure metal, which you can't do in a conventional design because this massively promotes dendrite growth. Using pure metal electrodes allows higher voltages.
That's it. It's a huge advance, if true, but there's certainly no new physics in here.
So when I real people not understanding the presence of pure electrodes, I wonder what they are thinking. There are lots of batteries with pure electrodes, not the least of which is the common dry cell, and on the other end things like ZEBRA which have pure sodium as one of the electrodes. The ZEBRA is a good example, because it too uses a solid electrolyte (beta something). I don't recall anyone saying it breaks the 2nd law.
Yet, reading the article, that appears to be the argument for this statement.
> when a modern ICE will cost you less than 4 grand in fuel for the entire life of the vehicle?:rolleyes:
The typical US car is driven 16,500 miles per year. It is used an average of 11 years. The average fuel economy is 24.5 miles/gallon. The average US gas price is around $2.
16000 x 11 / 24.5 = 7183 gallons x $2 ~= $14,500
Do math. It helps you avoid looking like an ultramaroon.
"Plus the speculation that these batteries will cost 10x as much when the inventor describes them as "cheap" is wild. If they cost 3x more to manufacture"
It's perfectly reasonable to expect such a battery to cost roughly the same as current technologies.
The details in all the articles I could find are very light, but it appears the basic construction of this battery replaces the plastic sheet separator with a glass one. The materials are otherwise the same or similar, with the exception of the electrodes which appear to be less complex and potentially less expensive.
There are no details on how the glass separator is constructed, but there are any number of ways this could be done in an automated fashion. The obvious one would be to make the plates and deposit the electrolyte on them then stack them into a cell. This would make cylindrical cells harder to make - they're just a long plastic sheet coated and then rolled up. However, prismatic cells require the sheet to be repeatedly folded, so stacking would be on the same order of complexity.
Overall, based on what little we know, I would expect these cells to cost about the same as current tech once they entered wide production.
> Look, I get it, but honestly, this is the same argument that was being advanced 100 years ago when electricity was automating things
Not really. If the device is *smarter* than you, then it really is something different. Automation to date has shifted employment from moving your arms to thinking about things. If the automation of the future outthinks you, then what do we do? Of course, if they are smarter than us, I suspect they will solve the problem for us anyway.
That said, I believe there is 0% chance of that happening. Modern AI is nothing more than applying a formula to a dataset, there's nothing "smart" at all, and it generally fails. In spite of BILLIONS spent on it, Google still gives me ads for products I bought months ago or never will buy. I really don't worry about the rise of the machines, at least not due to something we do.
"To wit: a common question we receive is, “Why did you send New Horizons to Pluto if it’s not a planet anymore?"
I call BS on this.
So there's all these people out there who are aware that Pluto exists, and that it was demoted to non-planet, and that we're sending a probe there, yet these same people cannot figure out why we would send a probe there? And these same people are also *unaware* that we send probes to things like the Moon and various asteroids, let alone deep space?
Slashdot Mirror
As someone that switched from PCs to the Mac in the early 90s, then from the Mac to Windows 98, and the back to the Mac for 10.4, I think I can offer some insight.
First off, it's important to consider that the advantages and disadvantages vary over time. In most cases, historically, I switched back to the Mac just before 10.5 came out, and I can tell you that 10.5 made the contemporary Windows XP look like crap - and XP was great! 10.5 was just better. It was faster in general use, faster to boot, faster to switch users, had much better security (like it actually had some), Time Machine was nothing short of phenomenal, and Safari was easily the best browser available.
And around that time, the hardware was incomparable as well. TiBooks ate anything on the market in practically any measurement. The cheese grater Mac Pro is still so great people would rather have them than the new model.
Now? Not so much. macOS is fine, but the delta between it and Win10 is much thinner. The things I see now are much more limited. Chrome fixed the browser issue, but I still can't find a good mail client (the new version of Outlook, ugh, it never works right!). Performance is now on-par, and booting speed even better than the Mac. My Mac still gives me much less problems in terms of changing stuff, and I still don't have a single malware after 10 years hanging out on the 'net and downloading everything.
The current iMacs are still great, but its not alone, and on the laptop side the delta has closed a LOT. And Apple's insistence on thinner keyboards has made them craptastic. But many of those deltas that closed are only because Apple opened them - the MacBook Air remains a great machine, and it was years before anyone had something really comparable. But now they do.
Finally, "convergence". I have an iPhone and the way things magically move from the phone to the Mac to the web is great. I suspect the same is true on the PC/Android side, but I don't know because I don't have Android. Still, once you get used to it its great - like handing off a call from my phone to the computer when I'm at my desk at home.
>Are you saying that the heat generated by millions of degree temperature is the same than the one generated by about 100 degree?!
Sure, if there's less of it. Fusion reactors operate at conditions best described as a very good vacuum.
I'm sorry you find all of this so difficult to believe, but there's plenty of resources on the 'net you can read on the topic.
> It seems that the only one with problems to understand heat/temperature (and quite a few other basic concepts) is you.
I'm a physicist who worked in the energy industry for about a decade. I've written extensively on fusion technologies, and you've probably read some of those articles (assuming you've read any).
So go ahead, tell me all about how wrong I am...
> Trying it for Minneapolis, I notice "system losses" figures in 0% for snow.
Actually it does figure it it, just not there. Snow often *increases* production, because it reflects more light onto clean panels. Snow coverage is modeled during the day-to-day simulation run.
> Something tells me that system isn't that accurate.
Is that "something" based on a single input and your decision to stop at that point complain about it rather than just put in a number? I'm sure you're capable of making this estimate. It's about 1/3rd of your roof's coverage because the cells heat up and cause the snow to slide off. Maybe 5 days a year.
The tool itself is considered to be accurate to within 10 to 12%, but there is widespread agreement it almost always underestimates actual production:
http://realmoney.thestreet.com/articles/06/18/2013/beware-misleading-solar-data
https://enphase.com/sites/default/files/Study-Performance-Versus-PVWatts_ENG_0.pdf
But of course, you're free to offer an alternative. You could try RETScreen.
> Britain is a country with a huge number of poorly insulated homes.
I lived in Ireland for a year. One night I was getting cold on a windy night and noticed the drapes on the main window in the living room were blowing around. Ah ha, I just need to close the window!
The window was closed.
> You're missing the point.
I responded directly to your point. Don't blame me if you choose to redefine the terms "night" and "lots". And for that matter "cold", which no one in the UK should dare to define to someone who lives in Toronto!
In any event, its besides the issue anyway. As one can see on the National Grid's website:
http://www2.nationalgrid.com/uk/Industry-information/Electricity-transmission-operational-data/Data-Explorer/
There is plenty of demand during the day that PV can take. Every watt that comes from that is one that didn't come from something else, which is generally a good thing. Sure, if you keep moving the goalposts and coming up with new reasons why "it will never work" you could probably keep us going forever. But if you want to solve actual problems, PV is certainly part of that solution, as those very same CSV files demonstrate (they even have a separate column for it). Also surprising is the amount of pumped storage.
> They've been cooking soup in their Tokamaks since the Big Bang was invented in the 60's,
Actually, they started with stellarators in 1951.
They got a whole lot of money in the 70s and 80s to build the TFTR, which everyone was sure was going to reach break even.
It didn't.
So they got some more money to keep the ancient Alcator going, now in C-Mod form. It keeps running, zombie like, in spite of the fact that there's nothing left to learn from it. They also have some smaller-scale machines, but most of the interesting ones were cancelled.
>Dealing with temperatures
You understand that temperature and heat are two different things, right?
Fusion takes place at high *temperatures*, but the *heat* loads are about the same as fission. That should not be surprising given that it outputs about the same amount of *power* (the rate of energy).
> Rooftop solar is about the least cost-effective way of running solar power
It is highly competitive above about 50 kW.
> but the cost of rooftop installation is high
For a large install on a flat roof it is close to even with ground mounts. The extra work of getting it all up there and installing around various HVAC and such is offset by the mechanically simpler and lighter install systems. It only really gets expensive in relative terms in small installs on tilted roofs.
> the cost of the inverter is relatively high
The inverters we sold scaled from about 50 cents/W for the smallest 250W models to about 22 cents (Canadian) for the 1 MW models. Unlike most forms of power, PV scales VERY linearly above about 50 kW.
> I don't want to be able to get one that's 50% more efficient a year later for the same price
You won't. I installed my system in 2010 using what were then one-down-from-the-best you could by panels, 230W a panel. That same panel today is 285W. That's a great improvement, but not 50% per year.
> I started looking at roof-top solar about 10 years ago. Back then, the cheap panels were 8% efficient
In 2007 most panels were around 160 W, but in the smaller form-factor using 5" cells. The jump that occurred between then and 2010 was due almost entirely to the move to 1 x 1.6m panels using 6" cells. Average efficiency during that period improved only slightly from about 10 to 12%, I'm not sure what type of panels you were looking at but apparently not A-quality examples?
We have largely reached the top of the S-curve, and price declines from here out will be smaller in conventional technologies. The epitaxial guys might have something to say, and the perkosites, but both are relatively low chances. $1/W is where it's going to be for a while.
No. Both only have an ROI of less than a decade IN CERTAIN SPECIFIC SITUATIONS.
Solar in SoCal, Nevada, New Mexico and Arizona generally have a fast payback. Because they see tons of son overall.
Solar in Hawaii has a fast payback because the price of power in Hawaii is high due to geographic isolation.
Solar power in Chicago, or Minneapolis or Detroit or Seattle is a MUCH different story.
Really? Let's test that theory. Everyone reading this can make a single change and figure it out for their own location. Here we go...
Open http://pvwatts.nrel.gov/
Type in Chicago as the location and hit go
Pick a weather station, likely the default, and hit go
Change the system size to 1 (to unitize everything, you'll see that's the input to the steps below) and the module type to Premium (all panels made today are super-premium according to the definition in this program)
Change the tilt to 30 degrees if you are between 40 and 50 north, and 35 if you are 50 to 60. I have not tried optimizing north of that
Press go again. Write down the big blue number at the top. For Chicago I got 1312. I know this is lower than what people actually get, but PVWatts is a wonderful conservative estimate, so let's run with it.
Now that production number is useful,but for the next part what we really want to know is the Capacity Factor. That is easy to calculate, it's that number divided by the number of hours in a year, so in this case 1312 / (24 x 365) = 15% (which is less than my real-world array in Toronto, so this is why I say it's conservative).
Ok now we have the production side of things. That's half the problem. The other half is the money part. So here we go on that side...
Open the NREL LCoE calculator here: http://www.nrel.gov/analysis/tech_lcoe.html
Change the period to 25, at a minimum. PV is generally thought to last about 40, but 25 is the warranty period so use that for now
Change the Capital Cost to 1000, which is the current price for 1 kW (I told you that number would reappear) of commercial solar in the US
Change the Capacity Factor to whatever you calculated above, in this case 15%
Change the Fixed O&M to 17 (click the link to see why)
Change the Variable O&M, Heat Rate and Fuel Cost to 0 (sunlight is free)
Look at the bottom line, the "Simple Levelized Cost of Renewable Energy". For the inputs above, that is 6.4 cents/kWh. That is competitive with wholesale averages (5.5), and *extremely* competitive for daytime peak.
Please, take 4 minutes of your life and do that calculation for your area. And if you want to know what it would be if you did it on your roof, change the 1000 to about 2500 to 3500, which accounts for the much higher costs of labor and construction on a small project.
So let's do one more? For the UK? Let's use York, which gets me 875 for the output of a 1k array at 30 degrees, which is 10% CF. That gets you a LCOE of 9.6 cents, which is highly competitive with wholesale rates in the UK.
Now your homework: repeat these calculations for your home using the closest weather station, with a Capital Cost of $3.50/W.
> just against stupid catch all remarks that say something is better than something else or has some specific ROI without taking into account any specifics.
Like your statement, which has no specifics?
Here, anyone can do this themselves. Go to pvwatts.nrel.gov and type in a nearby location. Change the system size to 1 kW and adjust the tilt angle - if you're above 40 degrees use 30 degrees instead of whatever they default to for California. You may also want to change the panel type to premium, because by the definitions of the program every panel made today is super-ultrapremium.
> Britain is a cold, dark country so we need lots of energy at night
No you don't. The night time load is about 1/2 the daytime, and that's why it's cheaper if you're on Economy 7 and 10.
> we had to import more than 10% of our energy requirements
Canada is awash with hydro and nuclear, and we import power all the time. It makes load balancing much easier if your grid spans as wide an area as possible.
Actually if you read it carefully, their claim of "first" refers only to the funding mechanism. And that's wrong too, there have been private reactors in the past.
"We are unveiling the first world-class controlled fusion device to have been designed, built, and operated by a private venture. "
What complete BS. Off the top of my head I can name the KMS ICF and the Riggatron as pure private-venture reactors that pre-date this one by*decades*. The later is named for the bank that funded it.
Keynote remains the jewel of the collection.
Pages and Numbers had many interesting features before they went cross platform, but both were incomplete. They needed another year of development before they would be really good.
That did not happen. They spent that time moving to iOS, eliminating any features that they couldn't implement on iOS. And that's still where we are today.
"federal Computer Fraud and Abuse Act broadly prohibits anyone from accessing a computer without authorization"
So someone makes a product that has no security, and we should sue the people who use that?
Yeah, that's par for the course of legal history I guess.
"extremely scientific and robust," relying on the employee's role, job level and location, as well as recent performance ratings"
So basically they are claiming that performance ratings are scientific, and that there's no possibility those are biased.
Right.
>Essentially, the EPA apparently considers any modification of the tractor
You can modify all sorts of crap on a JD tractor. Tires get changed all the time. You can change the entire cab if you want.
You just can't screw with the engine controls. Contrary to your line of argument, doing that has a very high probability of changing its emissions (like 100%).
> In other words, that EPA regulation should be considered unconstitutional
Then get a lawyer and sue them, and see if the court agrees with your asinine argument. And then we can put it on the list along with other nerd arguments like:
There's NO WAY Bell can stop our Blue Boxes!
There's TOTALLY ILLEGAL for the government to spy on all our comms!
There's NO WAY they can patent computer code!
No one will ever get sued when using BitTorrent!
etc.
I had an original Mac Pro that I got for a song so it became my primary machine. Over the years I added a big screen, drives, GPUs, memory, etc. It was not until 2014 that I considered an "upgraded" because I needed to move to 64bit in order to keep running Xcode. That left me with the decision of buying the top of the line iMac, or a Mac Pro. I went the later because I could keep my monitor, which I love. The price difference if you ignored the monitor was a couple of hundred bucks, so why not?
But quite frankly, the machine sucks. Oh, it's fast, and small, and very very quiet. And it looks good. But really, those are it's only good points. And there are lots of bad points...
1) You get two GPUs, one for rendering and one for calculations. However, I never (?) do GPU-hosted calculations, so that GPU is idle. I am certainly not alone in needing a single GPU. I would be happy if the second GPU could be used for rendering in a CrossFire-like way, but no one is bothering with that. So I have an expensive GPU doing nothing. Worse, it can't be used as a backup, as I understand it, so if the display GPU fails, my machine is dead.
2) There is a single "drive slot". It is non-standard (although such a standard did not really exist at the time). It also sprouts from one of the two GPUs, which is ridiculous. So Apple has to make two different GPU cards, one with and one without the SSD slot.
3) You may say it needs only one drive slot because you'll use external drives... right? Well here's the problem with that: most external drives are so much slower than the internal SSD that the machine is fully booted before the external is up and running (its FAST). Since you'll probably put your user account on that drive... odd things happen. Like your account is read-only. Or you get a sort of guest-like account. The only solution is to reboot.
4) It has FOUR USB ports. That isn't enough for anyone. Ever. All of them are on the back. So every time you want to plug in a USB key, you have to spin the machine. I gave up and left it back-to-front, so everyone gets to see my cable spaghetti.
5) It has SIX Thunderbolt. I have exactly one TB device, the screen.
6) All the ports are at the top of the machine, so the cables hang down and bend at the strain relief. If anything heavy ever falls on the cables, they're going to break. This is just bad design.
The good news is we can fix it all, easily:
1) put in at least two M2/U2 ports, preferably four. I shouldn't HAVE to use an external drive, and I shouldn't have to throw away the drive it came with if I want a larger option.
2) alternately, add a bay for a single (or two) conventional SATA laptop drives. You can get 1.5TBs for reasonable prices. It would make the case *slightly* larger, but who cares?
3) 10 USB-C, two of them on the front.
4) either move the ports down, or angle them downward to release the strain on the cables.
5) allow the system to run with a single GPU. And allow us to swap them! There's a number of small-form-factor GPU slots out there, and surely one of the companies you deal will with make one that can be mounted to the cooling block somehow.
Its shocking its taken this long.
> John Goodenough... ... did not invent this. It was largely developed by the first author on the paper prior to arriving in the US.
"For his invention to work as described, they say, it would probably have to abandon the laws of thermodynamics, which say perpetual motion is not possible,"
I read this section of the article several times, and I cannot make heads nor tails.
The entire invention, assuming it is real, replaces the normally plastic-and-liquid electrolyte with a glass sheet. The major result of this change is that it prevents ion movement across layers, which suppresses dendrite growth. As a result, you can replace the electrodes with pure metal, which you can't do in a conventional design because this massively promotes dendrite growth. Using pure metal electrodes allows higher voltages.
That's it. It's a huge advance, if true, but there's certainly no new physics in here.
So when I real people not understanding the presence of pure electrodes, I wonder what they are thinking. There are lots of batteries with pure electrodes, not the least of which is the common dry cell, and on the other end things like ZEBRA which have pure sodium as one of the electrodes. The ZEBRA is a good example, because it too uses a solid electrolyte (beta something). I don't recall anyone saying it breaks the 2nd law.
Yet, reading the article, that appears to be the argument for this statement.
> when a modern ICE will cost you less than 4 grand in fuel for the entire life of the vehicle? :rolleyes:
The typical US car is driven 16,500 miles per year.
It is used an average of 11 years.
The average fuel economy is 24.5 miles/gallon.
The average US gas price is around $2.
16000 x 11 / 24.5 = 7183 gallons x $2 ~= $14,500
Do math. It helps you avoid looking like an ultramaroon.
"Plus the speculation that these batteries will cost 10x as much when the inventor describes them as "cheap" is wild. If they cost 3x more to manufacture"
It's perfectly reasonable to expect such a battery to cost roughly the same as current technologies.
The details in all the articles I could find are very light, but it appears the basic construction of this battery replaces the plastic sheet separator with a glass one. The materials are otherwise the same or similar, with the exception of the electrodes which appear to be less complex and potentially less expensive.
There are no details on how the glass separator is constructed, but there are any number of ways this could be done in an automated fashion. The obvious one would be to make the plates and deposit the electrolyte on them then stack them into a cell. This would make cylindrical cells harder to make - they're just a long plastic sheet coated and then rolled up. However, prismatic cells require the sheet to be repeatedly folded, so stacking would be on the same order of complexity.
Overall, based on what little we know, I would expect these cells to cost about the same as current tech once they entered wide production.
> Look, I get it, but honestly, this is the same argument that was being advanced 100 years ago when electricity was automating things
Not really. If the device is *smarter* than you, then it really is something different. Automation to date has shifted employment from moving your arms to thinking about things. If the automation of the future outthinks you, then what do we do? Of course, if they are smarter than us, I suspect they will solve the problem for us anyway.
That said, I believe there is 0% chance of that happening. Modern AI is nothing more than applying a formula to a dataset, there's nothing "smart" at all, and it generally fails. In spite of BILLIONS spent on it, Google still gives me ads for products I bought months ago or never will buy. I really don't worry about the rise of the machines, at least not due to something we do.
"To wit: a common question we receive is, “Why did you send New Horizons to Pluto if it’s not a planet anymore?"
I call BS on this.
So there's all these people out there who are aware that Pluto exists, and that it was demoted to non-planet, and that we're sending a probe there, yet these same people cannot figure out why we would send a probe there? And these same people are also *unaware* that we send probes to things like the Moon and various asteroids, let alone deep space?
Suuuuuuureee.
I remember reading this *exact* story here on /.about five years ago.