You mean PCI-E 3.0 x4, SATA 3.0, and/or USB 3.0. Native, with pins dedicated to those purposes. (There's currently 5 different M.2 card keyings standardized.)
Also, from your own link:
I found stumbled across the Delock SATA to Thunderbolt adapter through a Google search, and I was hopeful that it might perform as well as it looks on paper. Unfortunately that's not the case.
The biggest issue is how difficult it is to get a drive working with it. After plugging it in and attaching a drive, it's hit or miss whether that drive will actually mount on my Macbook Pro Retina.
A secondary issue involves hot swapping disks. Once a drive is ejected from the Mac's interface and another is inserted, the Delock will not mount the second drive. I had to disconnect everything, re-attach, and go through the same hurdles I had to go through to get the drive working in the first place.
When it does finally mount the performance is far below the stated 6 gigabits per second potential, with benchmarks on a high-end SSD capping out at 3 gigabits per second even though the drive can go faster than that.
The bottom line? This product needs work. I can't recommend it.
The difficulty stems from the fact that Thunderbolt does not include SATA, requiring funky PCI-E to SATA chips that barely work, whereas M.2 supports SATA natively.
I can plug a SATA SSD into a USB adapter on my router and connect that with a serial cable to an ancient laptop with TCP/IP and NFS using SLIP or PPP or some other thing, and transfer fucking 1s and 0s. That doesn't mean that the serial port on the ancient laptop somehow groks SATA.
Thunderbolt has PCI Express and DisplayPort. It is used as an expansion bus for external peripherals.
M.2 has PCI Express and USB and SATA. It is used an an expansion bus for internal peripherals.
They're practically very dissimilar. Of the four electrical interfaces supported amongst them, they share just one in common. These aren't crazy words that only an engineer would understand.
I can't drive a DisplayPort monitor with M.2, and I can't connect a SATA drive to Thunderbolt.
SATA and eSATA are practically the same thing. M.2 and Thunderbolt are not.
In other words, Thunderbolt is NOT to M.2 as eSATA is to SATA.
In other words, both apples and oranges have a few things in common, but a lot more things that are not. At the end of the day, it's still apples and oranges.
Because a drone could never deploy a simple parachute, and/or have redundant propulsion (which can be done in software, today), and/or simply disassemble itself with a bang before falling out of the sky in small, low-mass chunks with terrible coefficient of drag and low terminal velocity.
I don't care if you block ads, or use CSS overlays to give your Youtube experience an OMG Kittens theme, or molest yourself with a bristle brush.
What would not be OK is if my ISP or some other third party blocked ads on my behalf, used CSS overlays to give my Youtube experience an OMG Kittens theme, or molest me with a bristle brush.
I use Adblock Edge on my PC, but also have a script that periodically downloads a magic list of hosts, tucks it into a format that dnsmasq likes, and runs on my Tomato-based router (there are a million variations on this).
The latter hosts-hacking always catches Youtube ads on the PS3 and Chromecast, and usually* gets rid of them on other devices on the network.
*Usually as in I see an ad so infrequently, and only on my Android phone, that I can't be bothered with doing anything more about it.
I used *I forget what* under MS-DOS to establish a PPP (SLIP? whatever) connection, ~1992, to a *nix host. It worked as well as MS-DOS could (and still does) allow.
Later, I used Telemate under MS-DOS to talk to the local Delphi dialup, to talk to Steve Jackson Games' Illuminati Online FreeBSD boxen.
Eventually, a local ISP showed up. I used Winsock on Windows, was disappointed: Things barely worked, which is saying a lot compared to all of the "barely worked" above.
I installed OS/2 on a 486SX with 4MB of RAM. The GUI loaded enough to see it, but then I discovered that OS/2 could run without a GUI: All command-line. It was fast. The TCP/IP stack robust enough to knock random other Internet users offline with a simple ping -f, all while my own connection was still useable: The pings would get longer and longer, and more and more infrequent, and then stop...even if I was on a different port of the exact same terminal server that they had been connected to, and even if asymmetric modem speeds said it shouldn't be that way.
Eventually, I got a Pentium 100 ("arguably overclocked" to a P120), and had 16MB of RAM on that board (16x1MB 30-pin SIMMS on carefully-stacked adapters). Worked a treat: I could finally use OS/2's GUI, and it was usable despite using 4x the RAM and about twice the CPU.
I used Linux after that, starting with Slackware 2.
I put on Windows 95 OSR2 after a then-employer handed me a copy of it and told me it was my job to do email support for his Windows-based software: I still did most of my work with a telnet/ssh session to SJ Games' io.com FreeBSD hosts.
As you can see, OS/2 was a blip on my own radar in those early days. But the Winsock days were really, really bad: Worse than the MS-DOS days.
And OS/2 was as solid as Linux, or the FreeBSD (then a mature thing) hosts that I paid by the month to use.
And OS/2's solid TCP/IP was included. With Windows, it was an extra, fickle (and not cheap, IIRC) third-party add-on.
95 OSR2 did OK, but meh. Nobody cared unless they were trying to get their new Packard Bell online, and then AOL by then the easiest answer. (They didn't get the money to buy Time Warner by accident.)
In ~2004, I bought a used, relatively clean and option-loaded 1995 BMW 325i with sport suspension for $6500.
I've spent about $4000 on repairs and general regular maintenance (from an auto-to-manual transmission swap, to the current valve cover/intake-parts rebuild, and including tires, wheels, and oil changes) in the past 11 years.
So I'm in for about $11,500 for over a decade of the very best car I've ever driven.
It sees some downtime, but I'm ~40k ahead of the game. And $40k buys at least a few $11,500 cars for fun spares.
All truly competitive free-market prices are (at least said to be) set at whatever the market will bear.
And contrary to what you say, an established and monopolized source of goods and/or services doesn't have that problem: They can charge whatever they feel like charging, and people will either pay it or be without that good or service. With a strong, non-competitive monopoly, it doesn't matter what the market will bear: You can provide minimal services at maximal pricing and reap maximum profits at the cost of those who can afford your good or service that you offer at a very self-serving price.
Hence, antitrust laws.
Hence, government.
Do you live under a rock, or do you just play someone who lives under a rock on TV?
What I think happened: COTS router was procured, cheap (Alibaba), and some kid was asked "Hey, kid: Do you think you can make this thing route everything over Tor?"
Kid agrees, and Kickstarter/Indigogo campaign happens.
Said kid then went through some Tomato source or forum posts, found the not-so-difficult bits that make Tor happen, implemented that (and only that) as requested, and said "I'll be taking that Porsche you offered me now, and it would be nice if you stuffed it with hookers and blow before delivery."
Product then ships with every gaping and elementary security flaw that the original Chinglish firmware had, PLUS automagic Tor....because "privacy."
The proceeds and remaining startup cash were then burned by the founders during a crazy weekend in Dubai. And that's that.
Many of us here can "create" such a device without these problems before we're even finished with our first pot coffee or get through a half bowl of cigarettes. But we wouldn't bother: We'd just post the sources and binaries on Github and make mention on the appropriate forums, and switch Mythbusters on in the TV in the basement and have a nice mid-morning nap.
2. PETE and polypropylene (the major components of single-use plastic bottles) have excellent compatibility with ethanol.
Indeed, the bottle of cheap 100 proof vodka that I have right over there says PETE on the bottom.
And separate seals aren't used anymore, AFAICT, on water bottles. The caps fit tightly enough into the neck of the bottle that additional gasketing is not needed.
So if the problem is transporting liquid alcohol in a more-convenient vessel, the solution is likely to already be in your recycling bin.
Or, you know: Stainless steel hip flask. Dissolve the soluables with a soak in strong isopropyl alcohol for a few days, wash with soap and water, install adult beverage, and insert into hip pocket.
Scroll up: I was talking-down the "and it's fast!" mentality of some OP, above.
But it's not fast, compared to any paid-for example of the very old things that I have in front of me, for the things I actually use computers for.
I mean, srsly, I don't care if it can render 1080p h.264 in perfect quality. I really don't: I've got a $23 Chromecast for that, plugged into the TV in my home theater The "difficult" tasks I have are all CPU-bound, and the CPU in question in TFS is anything but "fast." It may be low-power, and amazingly low-power at that, but it's not "fast" by any long stretch of any modern definition.
(In other news: 8088 CPUs also received low-power varients, some probably still in production. As did the 80386. None were commonly found in the field. My own Pentium-M undervolts to Low-Voltage Pentium-M specs, and then some, with perfect stability: If I decide to tweak it again (which I may not, since I've had the computer a very long time), its power consumption will also be very low for the work being accomplished.)
(and when I was undervolting my Pentium-M, it was because I was trying to minimize fan noise and radiant heat through the keyboard in the very quiet office environment I used it in at that time. I still don't care about TDP in portable computing: The first thing I do when things look like they may be lengthy is look for an available outlet, and I've (so far!) got enough extension cord in my bag to make it work.)
(Oh, and TDP is a lie these days, because CPUs tend to be both self- and dynamically-overclocking. If the chip gets hot, it'll just refuse to operate at the higher clock speeds that might meet demand, and will instead just slog along at a clock that keeps the temperature within reasonable ("TDP") spec. And in doing so, the chip vendor gets great numbers...which are based on lies.)
The airship cost $300,000 to buy. It doesn't matter if it cost someone else $90,000,000 to build it; the loss of $89,700,000 is the government's loss, not the current owners.
Car analogy: I once paid $6,500 for an excellent car that had a new sticker price of $53,000, and I've been driving it like a $6,500 car ever since.
105W is not astonishingly bad; it was simply the cost of performance at the time (before the i7 brand and DDR3 became a common thing). The power consumption game had barely started for desktop components.
I expect my portable computers to be just that: Portable computers. I do the same things with a portable computer as I do with a desktop computer.
For me, this lately means software decoding of many concurrent high-resolution video streams, and heavy single-threaded software.
I doubt this new Atom part is even as fast as my (even more ancient) 1.83GHz, 2MB cache, single-core Pentium-M laptop at these tasks.
It doesn't if the battery lasts twice as long, if it also takes twice as long to accomplish the work before me.
Perhaps I am a corner-case in that I actually want a CPU to be "fast" compared to products from a decade or so ago, especially if the device is bigger than a cell phone. I'm not buying anything slower than what I already have.
OS/2 Warp's killer feature was an excellent TCP/IP stack, enabling people to use the Internet without voluminous and hacked-together third-party software. It also included one of the better graphical web browsers of the era.
Because the only failure mode is the sudden and catastrophic sort of failure mode, and there is nothing that can be done to help promote safety after such a failure event occurs.
Also, as I'm sure you're aware: Airplanes don't glide, and helicopters don't auto-gyro.
Slashdot Mirror
You mean PCI-E 3.0 x4, SATA 3.0, and/or USB 3.0. Native, with pins dedicated to those purposes. (There's currently 5 different M.2 card keyings standardized.)
Also, from your own link:
The difficulty stems from the fact that Thunderbolt does not include SATA, requiring funky PCI-E to SATA chips that barely work, whereas M.2 supports SATA natively.
I can plug a SATA SSD into a USB adapter on my router and connect that with a serial cable to an ancient laptop with TCP/IP and NFS using SLIP or PPP or some other thing, and transfer fucking 1s and 0s. That doesn't mean that the serial port on the ancient laptop somehow groks SATA.
(Now are we done yet?)
No, not the same at all.
Thunderbolt has PCI Express and DisplayPort. It is used as an expansion bus for external peripherals.
M.2 has PCI Express and USB and SATA. It is used an an expansion bus for internal peripherals.
They're practically very dissimilar. Of the four electrical interfaces supported amongst them, they share just one in common. These aren't crazy words that only an engineer would understand.
I can't drive a DisplayPort monitor with M.2, and I can't connect a SATA drive to Thunderbolt.
SATA and eSATA are practically the same thing. M.2 and Thunderbolt are not.
In other words, Thunderbolt is NOT to M.2 as eSATA is to SATA.
In other words, both apples and oranges have a few things in common, but a lot more things that are not. At the end of the day, it's still apples and oranges.
(Are we done yet?)
Thunderbolt and M.2 are alike in that they both have an implementation of PCI Express. They're otherwise rather dissimilar.
So, they're essentially PCI Express.
Because a drone could never deploy a simple parachute, and/or have redundant propulsion (which can be done in software, today), and/or simply disassemble itself with a bang before falling out of the sky in small, low-mass chunks with terrible coefficient of drag and low terminal velocity.
Also: Delivery trucks are always perfectly safe.
Did I miss anything?
I don't care if you block ads, or use CSS overlays to give your Youtube experience an OMG Kittens theme, or molest yourself with a bristle brush.
What would not be OK is if my ISP or some other third party blocked ads on my behalf, used CSS overlays to give my Youtube experience an OMG Kittens theme, or molest me with a bristle brush.
Do you see the difference?
There is a third category in the form of an Xposed module that eradicates Youtube ads on Android.
I use Adblock Edge on my PC, but also have a script that periodically downloads a magic list of hosts, tucks it into a format that dnsmasq likes, and runs on my Tomato-based router (there are a million variations on this).
The latter hosts-hacking always catches Youtube ads on the PS3 and Chromecast, and usually* gets rid of them on other devices on the network.
*Usually as in I see an ad so infrequently, and only on my Android phone, that I can't be bothered with doing anything more about it.
Because AT&T didn't happen, etc.
Right, got it. Is there any more history that you'd like to re-state for the annals?
Typo: $5k in service and repairs, but whatever. If you want to pick on my old and awesome car, you'll do so no matter what, +/- $1k.
Here's my own progression:
I used *I forget what* under MS-DOS to establish a PPP (SLIP? whatever) connection, ~1992, to a *nix host. It worked as well as MS-DOS could (and still does) allow.
Later, I used Telemate under MS-DOS to talk to the local Delphi dialup, to talk to Steve Jackson Games' Illuminati Online FreeBSD boxen.
Eventually, a local ISP showed up. I used Winsock on Windows, was disappointed: Things barely worked, which is saying a lot compared to all of the "barely worked" above.
I installed OS/2 on a 486SX with 4MB of RAM. The GUI loaded enough to see it, but then I discovered that OS/2 could run without a GUI: All command-line. It was fast. The TCP/IP stack robust enough to knock random other Internet users offline with a simple ping -f, all while my own connection was still useable: The pings would get longer and longer, and more and more infrequent, and then stop...even if I was on a different port of the exact same terminal server that they had been connected to, and even if asymmetric modem speeds said it shouldn't be that way.
Eventually, I got a Pentium 100 ("arguably overclocked" to a P120), and had 16MB of RAM on that board (16x1MB 30-pin SIMMS on carefully-stacked adapters). Worked a treat: I could finally use OS/2's GUI, and it was usable despite using 4x the RAM and about twice the CPU.
I used Linux after that, starting with Slackware 2.
I put on Windows 95 OSR2 after a then-employer handed me a copy of it and told me it was my job to do email support for his Windows-based software: I still did most of my work with a telnet/ssh session to SJ Games' io.com FreeBSD hosts.
As you can see, OS/2 was a blip on my own radar in those early days. But the Winsock days were really, really bad: Worse than the MS-DOS days.
And OS/2 was as solid as Linux, or the FreeBSD (then a mature thing) hosts that I paid by the month to use.
And OS/2's solid TCP/IP was included. With Windows, it was an extra, fickle (and not cheap, IIRC) third-party add-on.
95 OSR2 did OK, but meh. Nobody cared unless they were trying to get their new Packard Bell online, and then AOL by then the easiest answer. (They didn't get the money to buy Time Warner by accident.)
In ~2004, I bought a used, relatively clean and option-loaded 1995 BMW 325i with sport suspension for $6500.
I've spent about $4000 on repairs and general regular maintenance (from an auto-to-manual transmission swap, to the current valve cover/intake-parts rebuild, and including tires, wheels, and oil changes) in the past 11 years.
So I'm in for about $11,500 for over a decade of the very best car I've ever driven.
It sees some downtime, but I'm ~40k ahead of the game. And $40k buys at least a few $11,500 cars for fun spares.
You've got it all backward.
All truly competitive free-market prices are (at least said to be) set at whatever the market will bear.
And contrary to what you say, an established and monopolized source of goods and/or services doesn't have that problem: They can charge whatever they feel like charging, and people will either pay it or be without that good or service. With a strong, non-competitive monopoly, it doesn't matter what the market will bear: You can provide minimal services at maximal pricing and reap maximum profits at the cost of those who can afford your good or service that you offer at a very self-serving price.
Hence, antitrust laws.
Hence, government.
Do you live under a rock, or do you just play someone who lives under a rock on TV?
In some areas, we already have multiple private hospitals and doctors and ambulance companies competing with eachother.
This doesn't seem to have any great impact on prices.
This level of security isn't hard. At all.
What I think happened: COTS router was procured, cheap (Alibaba), and some kid was asked "Hey, kid: Do you think you can make this thing route everything over Tor?"
Kid agrees, and Kickstarter/Indigogo campaign happens.
Said kid then went through some Tomato source or forum posts, found the not-so-difficult bits that make Tor happen, implemented that (and only that) as requested, and said "I'll be taking that Porsche you offered me now, and it would be nice if you stuffed it with hookers and blow before delivery."
Product then ships with every gaping and elementary security flaw that the original Chinglish firmware had, PLUS automagic Tor....because "privacy."
The proceeds and remaining startup cash were then burned by the founders during a crazy weekend in Dubai. And that's that.
Many of us here can "create" such a device without these problems before we're even finished with our first pot coffee or get through a half bowl of cigarettes. But we wouldn't bother: We'd just post the sources and binaries on Github and make mention on the appropriate forums, and switch Mythbusters on in the TV in the basement and have a nice mid-morning nap.
2. PETE and polypropylene (the major components of single-use plastic bottles) have excellent compatibility with ethanol.
Indeed, the bottle of cheap 100 proof vodka that I have right over there says PETE on the bottom.
And separate seals aren't used anymore, AFAICT, on water bottles. The caps fit tightly enough into the neck of the bottle that additional gasketing is not needed.
So if the problem is transporting liquid alcohol in a more-convenient vessel, the solution is likely to already be in your recycling bin.
Or, you know: Stainless steel hip flask. Dissolve the soluables with a soak in strong isopropyl alcohol for a few days, wash with soap and water, install adult beverage, and insert into hip pocket.
Either way, this is a solved problem.
Scroll up: I was talking-down the "and it's fast!" mentality of some OP, above.
But it's not fast, compared to any paid-for example of the very old things that I have in front of me, for the things I actually use computers for.
I mean, srsly, I don't care if it can render 1080p h.264 in perfect quality. I really don't: I've got a $23 Chromecast for that, plugged into the TV in my home theater The "difficult" tasks I have are all CPU-bound, and the CPU in question in TFS is anything but "fast." It may be low-power, and amazingly low-power at that, but it's not "fast" by any long stretch of any modern definition.
(In other news: 8088 CPUs also received low-power varients, some probably still in production. As did the 80386. None were commonly found in the field. My own Pentium-M undervolts to Low-Voltage Pentium-M specs, and then some, with perfect stability: If I decide to tweak it again (which I may not, since I've had the computer a very long time), its power consumption will also be very low for the work being accomplished.)
(and when I was undervolting my Pentium-M, it was because I was trying to minimize fan noise and radiant heat through the keyboard in the very quiet office environment I used it in at that time. I still don't care about TDP in portable computing: The first thing I do when things look like they may be lengthy is look for an available outlet, and I've (so far!) got enough extension cord in my bag to make it work.)
(Oh, and TDP is a lie these days, because CPUs tend to be both self- and dynamically-overclocking. If the chip gets hot, it'll just refuse to operate at the higher clock speeds that might meet demand, and will instead just slog along at a clock that keeps the temperature within reasonable ("TDP") spec. And in doing so, the chip vendor gets great numbers...which are based on lies.)
I engineer my systems and tools for me, not you.
I need ports, expandability, and the ability to plug random hardware in. I don't need light-weight, and I don't need to run all day on batteries.
I have all of that, along with what I believe to be comparable speed...instead of none of that, and $499 less in my pocket.
I've got better things to spend $499 on than a side-grade to a different form factor that doesn't fucking work for me. But thanks anyway, asshole!
If the helium can't get out, the mineral oil can't get in.
The airship cost $300,000 to buy. It doesn't matter if it cost someone else $90,000,000 to build it; the loss of $89,700,000 is the government's loss, not the current owners.
Car analogy: I once paid $6,500 for an excellent car that had a new sticker price of $53,000, and I've been driving it like a $6,500 car ever since.
105W is not astonishingly bad; it was simply the cost of performance at the time (before the i7 brand and DDR3 became a common thing). The power consumption game had barely started for desktop components.
I expect my portable computers to be just that: Portable computers. I do the same things with a portable computer as I do with a desktop computer.
For me, this lately means software decoding of many concurrent high-resolution video streams, and heavy single-threaded software.
I doubt this new Atom part is even as fast as my (even more ancient) 1.83GHz, 2MB cache, single-core Pentium-M laptop at these tasks.
It doesn't if the battery lasts twice as long, if it also takes twice as long to accomplish the work before me.
Perhaps I am a corner-case in that I actually want a CPU to be "fast" compared to products from a decade or so ago, especially if the device is bigger than a cell phone. I'm not buying anything slower than what I already have.
My car has a few bad failure modes, too, but I still load my kids into it and take them wherever.
So does my neighbor. And my neighbor's neighbor.
*shrug*
I've always done it that way: Buy a humble bundle, give every last dime to the EFF.
So it's half as fast as the positively ancient Q6600 I write this on.
Which is interesting, I guess, but not anything approaching "fast."
OS/2 Warp's killer feature was an excellent TCP/IP stack, enabling people to use the Internet without voluminous and hacked-together third-party software. It also included one of the better graphical web browsers of the era.
If anything, it was ahead of its time.
Because the only failure mode is the sudden and catastrophic sort of failure mode, and there is nothing that can be done to help promote safety after such a failure event occurs.
Also, as I'm sure you're aware: Airplanes don't glide, and helicopters don't auto-gyro.
[/sarcasm]