Yes, people now use FLASH memory but place it into read-only mode. It is cheaper when one requires relatively large amounts of memory - as would be required by a ROM. There is probably a way to program the memory if you interrupt the boot sequence before the OS is loaded. One would require a hardware connection - such as JTAG. But from the perspective of the OS, it behaves just like a ROM.
Or perhaps there is a jumper to enable read/write access. I believe the Asus Chrome Box units protected their boot ROM this way. Only instead of a jumper you had to remove a screw.
High performance ICs and minimal cooling result in high rates of change in temperature. This will put significant strain on the solder contacts. It is safer when parts heat up slowly so better cooling could prevent the damage from occurring.
Alternatively, Intel could modify their turbo mode so that it not only limits maximum temperature but also maximum rate of change in temperature. But before I place all the blame on Intel, do these things not contain NVidea GPUs? If so, that is the likely point of failure.
But that really has nothing to do a gas vs. electric motor,
It does, because the electric drive requires electric power which adds mass which reduces handling. Even Porsche will not be able to make an electric vehicle that matches the handling of their petrol-powered products - at least until energy storage becomes more efficient / lighter.
They stopped after TRIM support was reliably implemented on SSDs. If you recall, initial SSD implementations of TRIM had some reliability issues. The need for TRIM was not foreseen early enough for the initial implementations to be adequately tested. It is understandable that a manufacturer would want to limit new features on such a fundamentally important part of the system - or at least make them opt-in options. It is not like the SSD manufacturers were writing their firmware for, and testing compatibility with, MacOS.
There actually is a bit of truth here. People often refuse to admit mistakes they might have made. When confronted with facts that challenge their beliefs and validity of past actions, they dig in, deny, and challenge said facts. No one is really immune from these human tendencies. But scientists are more immune then most - evaluating data while remaining unbiased is a big part of the job.
The Xeons have ECC memory support - this is why Intel can charge so much more for a Xeon then an i7. It is basically a money grab by Intel. The Ryzen all have ECC memory support - assuming it is supported by the motherboard. So you can get your high reliability requirements from a standard Ryzen. Of course, Epyc has additional cores for more processing power and Epyc systems are designed expecting the use of ECC memory - so they are probably a safer bet if you have the spare cash. But they are a safer bet only because the non-CPU components in the system will be designed with matching reliability.
So it depends on your application but if you are really interested in data integrity, like most workstation applications, then the Ryzen with ECC memory should be fantastic.
Selling Intel stock might have been a bad long term prospect. For the first time in years we can expect the next generation of Intel CPUs to be noticeably faster than the previous generation. A year from now Intel is going to be selling a lot more CPUs then they are now.
You are correct, looks like the standard is to use the USB 2.0 data lines for L,R analog audio - just like MicroUSB. But they do fit a dedicated analog ground line which is good and support for a microphone which is unexpected - but great. In this configuration nothing is preventing a USB 3.0 data connection - but those ICs are more expensive to work with. As far as a FM antenna goes, the cable shielding could be used. The shield would typically be left floating from the headphone side and tied to ground (with RC) on the mobile. The mobile would just have to be smart enough to "un-tie" the shield when it intends to use it as an antenna. One could forget about digital data - but it would work for straight analog headphones.
Forums have the advantage of being easier to search for those just observing the conversation. Having a known viewing platform also has it's advantages - when using email one has to assume text only. The switch to a forum is a good idea so long as email alerts / updates to new content are provided. If people really do not want to abandon their email, perhaps an email bot could automate forum posts from an email reply. Perhaps they could write it in Swift?
USB-C do not require a DAC. There are additional lines designed for analog signals (amongst other things) - specifically, the SBU1 and SBU2 pins. Analog quality on a cable will be crap but if you are connecting directly to a dock it should be OK. This is actually an improvement to the Micro-USB connectors where the USB data lines are used for analog signals. With USB Type-C you get the analog lines while still preserving the USB data connection.
Of course a true digital connection with DAC is much better. There is no reason why mobile devices can not support both options at the same time - they probably already do. The cheap headphones can just take the hit on signal quality.
Distributed memory is just that - memory distributed along side the LUT/FF elements. Using this memory "consumes" some LUTs but it is far more efficient then 1 bit per LUT. See the following quote from the MachXO3 datasheet - note that a Slice has 2 LUTs and 2 FFs, 4 Slices per PFU block.
RAM Mode
In this mode, a 16x4-bit distributed single port RAM (SPR) can be constructed by using each LUT block in Slice 0 and Slice 1 as a 16x1-bit memory. Slice 2 is used to provide memory address and control signals.
All from 1 PFU block. And this is why a device with 640 LUTs can provide 5K of memory.
Because it is soft, it will conform to the required shape. Upon impact, it would harden and more evenly distribute the energy to whatever shape is required. It is therefor more suitable to protect shapes that are dynamic or otherwise unknown during manufacturing. At least, that is the theory. If the shape is known, you are correct is assuming that the benefits are minimal and a more traditional material will offer better performance.
The iCE40 is great - but not to synthesize a CPU. It lacks the dual-port memory required for registers. There is a reason whey Lattice offers their LatticeMicro processor for their MachXO series but not the iCE40.
The iCE40 is highly optimized and fantastic for most programmable logic. But if you compare the iCE40 to the MachXO3 you will see that the MachXO3 has "Distributed Memory" - an essential part for efficiently implementing a CPU.
Marvel is focused on cutting out the middle man - those that sell comics. They want to make digital sales direct to a customer tablet. This way they can offer attractive prices and leverage one comic to advertise another. For this they have attractive mobile applications that works quite well. While developing this comic distribution system it is inevitable that the physical comic sales will decline. It is not a sign that the comics are bad but a sign that they are moving to a different distribution channel. Just wait, next they will go after the theatres.
Oh, and as far as Wolverine goes, the character was overused as is and it was time for a break. He'll be back, badass as ever...
All Internet connection would not have to be severed - all it would take is the majority of underwater cables being severed at the same time. Such an event would cause sufficient gridlock to make the Internet useless. The "self healing" nature of the Internet can barely correct for accidents - let alone deliberate sabotage. That is why I described a scenario involving explosives triggered by a remote signal. Sending underwater transmissions is difficult. There are severe bandwidth limitations with huge antennas are required. But such signals will travel around the globe and it does not take much bandwidth to send an activation signal.
It is difficult to imagine a more efficient way of disabling global Internet communications. Admittedly, satellite communication would also need to be severed in order to be effective. But this would be one important step if Russia ever performs an aggressive move in Europe. Even if the military can use their dedicated radios, it prevents the vast majority of people from communicating.
The point would not be to kill the Internet so much as limit communications thereby blinding the enemy. It is a discussion involving tactics and theoretical scenarios. Only when it is reported that military subs are crawling the underwater cables does it become concerning. As you stated, all important information is encrypted. What other reason could Russia have for doing this?
You can bet that there will be explosives placed in various locations under the main internet communication links. Should it be required, a remote detonation signal could be sent and then say goodby to the internet. I suppose if they were going to go that far then they would also disable / destroy any satellites. Europe and the US would have almost all communications cut - good luck responding to a Russian threat in a timely manner.
This is all theoretical but if I were Russia, I would do it. Much easier then tapping into cables and there would be no way to find said explosives.
While they do have solar powered UAVs that can store sufficient energy to saw aloft 24/7, the energy requirements of the radio equipment would cause problems. If it was easy to do they would have done it already. First flights of a UAV flying 24/7 were done a while ago and since then there has been very little to report. But if someone knows of something I missed, please fill me in..
A bigger impact on company success is networking - you have to know the right people. Interacting with others, even when it involves purchasing overpriced coffee, is worth much more then the 10k you save by being frugal. Being frugal is good for a stable lifestyle but it is not conducive to becoming a millionaire. It is important to be effective before being efficient.
AC requires the whole grid to be fully synchronous.
Yes, and this is why your oven clock keeps accurate time for years. It is synchronized to the 60 (or 50) Hz input power. The grid then guarantees that they will generate, on average, exactly the specified 50/60 Hz frequency. The frequency changes slightly with load but it all averages out in the end.
Not long ago there was a/. article talking about how the US has announced an end to the 60Hz promise - in ~20 years. The introduction of numerous solar panel installations has made it more difficult to maintain a specific output frequency. In addition - GPS, cell, and accurate crystals have made the benefits of accurate mains timing far less important.
So the grid is already synchronous with but a few exceptions. The losses due to AC are minimal for shorter, lower voltage runs and will be less then losses resulting from converting from DC to AC. So for the most part, things run on AC. However, DC starts to look good when you are running at a very high voltage or if the surrounding environment is impacted by a changing magnetic field (think underwater). In such situations, DC is a good choice due to advancements in silicon switching technology.
The vast majority of people are not going to look for a USB certification logo on their USB cable. And even with the logo, there is a strong possibility they could be fake. Pushing 3A through such a cable is pushing the boundaries of what can be performed safely. No doubt the USB consortium knows this which is why anything above 3A requires an active cable.
My point is that 3A is not a low current draw. 3A is 1/5 the maximum current for most household outlets that use 14 gauge conductors. For comparison, a 24 gauge USB cable (typical) providing 3A is like a 15A circuit (12A continuous) providing 30A continuous. For the cheap USB cables that use 28 gauge conductors, it is > 60A. Then to make it even worse, USB cables are less efficient at dissipating energy due to the small size. Should someone have the cable bundled up or in some way insulated - a fire could easily result. Oh, the resistance also increases with temperature so to make a long story short - it is definitely dangerous.
I really like the USB-C power delivery design. It is great but not perfect. In my opinion the 3A limit is a bit much - 1.5A would have been more reasonable.
it's not possible to melt the USB-C cable for exceeding the power rating of a cable.
Only those cables rated for >3A are electronically identified. So cheap USB-C cables can still burn up if they can not handle 3A of current. With a 3A current, most standard cables will have to dissipate 2W per meter of cable. Not that bad but if people go cheap and use 28AWG power conductors, like with some current USB cables, the resulting 4W load could cause things to start smoking.
Power factor has nothing to do with it. Power factor defines how linear the load is - not the efficiency. Those old power bricks were horrible with regards to power factor. They just used diodes to turn the AC into DC resulting is surge loads on the main power line. Newer DCDC converters distribute the load making it act more like a resistor. No 60Hz transformer and less required capacitance are the real reasons why modern power supplies are so much smaller. Overall an efficiency of ~90% is relatively standard.
Any serious Russian effort in future will be run through American proxies
Good. If such an attempt is suspected, the authorities can raid the place, seize documentation, and shut them down. Much harder to do that when they are in Russia. It will probably be a lone individual but that is fine. What you are looking for is proof that they were working on behalf of Russia as this gives one leverage during future negotiations with Russia.
Slashdot Mirror
Yes, people now use FLASH memory but place it into read-only mode. It is cheaper when one requires relatively large amounts of memory - as would be required by a ROM. There is probably a way to program the memory if you interrupt the boot sequence before the OS is loaded. One would require a hardware connection - such as JTAG. But from the perspective of the OS, it behaves just like a ROM.
Or perhaps there is a jumper to enable read/write access. I believe the Asus Chrome Box units protected their boot ROM this way. Only instead of a jumper you had to remove a screw.
High performance ICs and minimal cooling result in high rates of change in temperature. This will put significant strain on the solder contacts. It is safer when parts heat up slowly so better cooling could prevent the damage from occurring.
Alternatively, Intel could modify their turbo mode so that it not only limits maximum temperature but also maximum rate of change in temperature. But before I place all the blame on Intel, do these things not contain NVidea GPUs? If so, that is the likely point of failure.
But that really has nothing to do a gas vs. electric motor,
It does, because the electric drive requires electric power which adds mass which reduces handling. Even Porsche will not be able to make an electric vehicle that matches the handling of their petrol-powered products - at least until energy storage becomes more efficient / lighter.
My third-party SSD was updated to APFS automatically. Mechanical hard drive was not.
They stopped after TRIM support was reliably implemented on SSDs. If you recall, initial SSD implementations of TRIM had some reliability issues. The need for TRIM was not foreseen early enough for the initial implementations to be adequately tested. It is understandable that a manufacturer would want to limit new features on such a fundamentally important part of the system - or at least make them opt-in options. It is not like the SSD manufacturers were writing their firmware for, and testing compatibility with, MacOS.
There actually is a bit of truth here. People often refuse to admit mistakes they might have made. When confronted with facts that challenge their beliefs and validity of past actions, they dig in, deny, and challenge said facts. No one is really immune from these human tendencies. But scientists are more immune then most - evaluating data while remaining unbiased is a big part of the job.
The Xeons have ECC memory support - this is why Intel can charge so much more for a Xeon then an i7. It is basically a money grab by Intel. The Ryzen all have ECC memory support - assuming it is supported by the motherboard. So you can get your high reliability requirements from a standard Ryzen. Of course, Epyc has additional cores for more processing power and Epyc systems are designed expecting the use of ECC memory - so they are probably a safer bet if you have the spare cash. But they are a safer bet only because the non-CPU components in the system will be designed with matching reliability.
So it depends on your application but if you are really interested in data integrity, like most workstation applications, then the Ryzen with ECC memory should be fantastic.
Selling Intel stock might have been a bad long term prospect. For the first time in years we can expect the next generation of Intel CPUs to be noticeably faster than the previous generation. A year from now Intel is going to be selling a lot more CPUs then they are now.
From the link: Payload of 225 kg at a cost of 7.16 mil.
You are correct, looks like the standard is to use the USB 2.0 data lines for L,R analog audio - just like MicroUSB. But they do fit a dedicated analog ground line which is good and support for a microphone which is unexpected - but great. In this configuration nothing is preventing a USB 3.0 data connection - but those ICs are more expensive to work with. As far as a FM antenna goes, the cable shielding could be used. The shield would typically be left floating from the headphone side and tied to ground (with RC) on the mobile. The mobile would just have to be smart enough to "un-tie" the shield when it intends to use it as an antenna. One could forget about digital data - but it would work for straight analog headphones.
Forums have the advantage of being easier to search for those just observing the conversation. Having a known viewing platform also has it's advantages - when using email one has to assume text only. The switch to a forum is a good idea so long as email alerts / updates to new content are provided. If people really do not want to abandon their email, perhaps an email bot could automate forum posts from an email reply. Perhaps they could write it in Swift?
USB-C do not require a DAC. There are additional lines designed for analog signals (amongst other things) - specifically, the SBU1 and SBU2 pins. Analog quality on a cable will be crap but if you are connecting directly to a dock it should be OK. This is actually an improvement to the Micro-USB connectors where the USB data lines are used for analog signals. With USB Type-C you get the analog lines while still preserving the USB data connection.
Of course a true digital connection with DAC is much better. There is no reason why mobile devices can not support both options at the same time - they probably already do. The cheap headphones can just take the hit on signal quality.
What is heck "Distributed Memory"?
Distributed memory is just that - memory distributed along side the LUT/FF elements. Using this memory "consumes" some LUTs but it is far more efficient then 1 bit per LUT. See the following quote from the MachXO3 datasheet - note that a Slice has 2 LUTs and 2 FFs, 4 Slices per PFU block.
RAM Mode
In this mode, a 16x4-bit distributed single port RAM (SPR) can be constructed by using each LUT block in Slice 0 and Slice 1 as a 16x1-bit memory. Slice 2 is used to provide memory address and control signals.
All from 1 PFU block. And this is why a device with 640 LUTs can provide 5K of memory.
MachXO3 Family Data Sheet
Because it is soft, it will conform to the required shape. Upon impact, it would harden and more evenly distribute the energy to whatever shape is required. It is therefor more suitable to protect shapes that are dynamic or otherwise unknown during manufacturing. At least, that is the theory. If the shape is known, you are correct is assuming that the benefits are minimal and a more traditional material will offer better performance.
these implemented on an iCE40
The iCE40 is great - but not to synthesize a CPU. It lacks the dual-port memory required for registers. There is a reason whey Lattice offers their LatticeMicro processor for their MachXO series but not the iCE40.
The iCE40 is highly optimized and fantastic for most programmable logic. But if you compare the iCE40 to the MachXO3 you will see that the MachXO3 has "Distributed Memory" - an essential part for efficiently implementing a CPU.
Marvel is focused on cutting out the middle man - those that sell comics. They want to make digital sales direct to a customer tablet. This way they can offer attractive prices and leverage one comic to advertise another. For this they have attractive mobile applications that works quite well. While developing this comic distribution system it is inevitable that the physical comic sales will decline. It is not a sign that the comics are bad but a sign that they are moving to a different distribution channel. Just wait, next they will go after the theatres.
Oh, and as far as Wolverine goes, the character was overused as is and it was time for a break. He'll be back, badass as ever...
All Internet connection would not have to be severed - all it would take is the majority of underwater cables being severed at the same time. Such an event would cause sufficient gridlock to make the Internet useless. The "self healing" nature of the Internet can barely correct for accidents - let alone deliberate sabotage. That is why I described a scenario involving explosives triggered by a remote signal. Sending underwater transmissions is difficult. There are severe bandwidth limitations with huge antennas are required. But such signals will travel around the globe and it does not take much bandwidth to send an activation signal.
It is difficult to imagine a more efficient way of disabling global Internet communications. Admittedly, satellite communication would also need to be severed in order to be effective. But this would be one important step if Russia ever performs an aggressive move in Europe. Even if the military can use their dedicated radios, it prevents the vast majority of people from communicating.
The point would not be to kill the Internet so much as limit communications thereby blinding the enemy. It is a discussion involving tactics and theoretical scenarios. Only when it is reported that military subs are crawling the underwater cables does it become concerning. As you stated, all important information is encrypted. What other reason could Russia have for doing this?
You can bet that there will be explosives placed in various locations under the main internet communication links. Should it be required, a remote detonation signal could be sent and then say goodby to the internet. I suppose if they were going to go that far then they would also disable / destroy any satellites. Europe and the US would have almost all communications cut - good luck responding to a Russian threat in a timely manner.
This is all theoretical but if I were Russia, I would do it. Much easier then tapping into cables and there would be no way to find said explosives.
While they do have solar powered UAVs that can store sufficient energy to saw aloft 24/7, the energy requirements of the radio equipment would cause problems. If it was easy to do they would have done it already. First flights of a UAV flying 24/7 were done a while ago and since then there has been very little to report. But if someone knows of something I missed, please fill me in..
A bigger impact on company success is networking - you have to know the right people. Interacting with others, even when it involves purchasing overpriced coffee, is worth much more then the 10k you save by being frugal. Being frugal is good for a stable lifestyle but it is not conducive to becoming a millionaire. It is important to be effective before being efficient.
AC requires the whole grid to be fully synchronous.
Yes, and this is why your oven clock keeps accurate time for years. It is synchronized to the 60 (or 50) Hz input power. The grid then guarantees that they will generate, on average, exactly the specified 50/60 Hz frequency. The frequency changes slightly with load but it all averages out in the end.
Not long ago there was a /. article talking about how the US has announced an end to the 60Hz promise - in ~20 years. The introduction of numerous solar panel installations has made it more difficult to maintain a specific output frequency. In addition - GPS, cell, and accurate crystals have made the benefits of accurate mains timing far less important.
So the grid is already synchronous with but a few exceptions. The losses due to AC are minimal for shorter, lower voltage runs and will be less then losses resulting from converting from DC to AC. So for the most part, things run on AC. However, DC starts to look good when you are running at a very high voltage or if the surrounding environment is impacted by a changing magnetic field (think underwater). In such situations, DC is a good choice due to advancements in silicon switching technology.
The vast majority of people are not going to look for a USB certification logo on their USB cable. And even with the logo, there is a strong possibility they could be fake. Pushing 3A through such a cable is pushing the boundaries of what can be performed safely. No doubt the USB consortium knows this which is why anything above 3A requires an active cable.
My point is that 3A is not a low current draw. 3A is 1/5 the maximum current for most household outlets that use 14 gauge conductors. For comparison, a 24 gauge USB cable (typical) providing 3A is like a 15A circuit (12A continuous) providing 30A continuous. For the cheap USB cables that use 28 gauge conductors, it is > 60A. Then to make it even worse, USB cables are less efficient at dissipating energy due to the small size. Should someone have the cable bundled up or in some way insulated - a fire could easily result. Oh, the resistance also increases with temperature so to make a long story short - it is definitely dangerous.
I really like the USB-C power delivery design. It is great but not perfect. In my opinion the 3A limit is a bit much - 1.5A would have been more reasonable.
it's not possible to melt the USB-C cable for exceeding the power rating of a cable.
Only those cables rated for >3A are electronically identified. So cheap USB-C cables can still burn up if they can not handle 3A of current. With a 3A current, most standard cables will have to dissipate 2W per meter of cable. Not that bad but if people go cheap and use 28AWG power conductors, like with some current USB cables, the resulting 4W load could cause things to start smoking.
Power factor has nothing to do with it. Power factor defines how linear the load is - not the efficiency. Those old power bricks were horrible with regards to power factor. They just used diodes to turn the AC into DC resulting is surge loads on the main power line. Newer DCDC converters distribute the load making it act more like a resistor. No 60Hz transformer and less required capacitance are the real reasons why modern power supplies are so much smaller. Overall an efficiency of ~90% is relatively standard.
Any serious Russian effort in future will be run through American proxies
Good. If such an attempt is suspected, the authorities can raid the place, seize documentation, and shut them down. Much harder to do that when they are in Russia. It will probably be a lone individual but that is fine. What you are looking for is proof that they were working on behalf of Russia as this gives one leverage during future negotiations with Russia.