Hell, Wi Fi can't saturate a 100mbit fast ethernet interface.
Not having a problem getting 250Mbps over WiFi to a single device here. And it's not even a good one. I could easily saturate a 1Gbps internet link over WiFi.
I didn't start using StackOverflow until later, but I'm still in the top 2% of answer givers there (so I've read and answered a good number of questions), and I think I did see one comment once that was racist/sexist, and it was quickly removed.
One other time, there was a user who had a huge chip on their shoulder who decided for no reason to tell everyone about how they were different, then got argumentative (about the technical subject) and rude, and then when he was told he was wrong, blamed it on the fact that he was different. I don't think that got closed, but it was obvious this person was just looking for attention and wanted to start an discussion about their differences.
Would you buy an autonomous car where heavy rain or a hint of snow would keep you from getting home while a human driver just gets into the car and drives home?
Nope, but I have PERSONALLY been affected by a lying police officer, so I know it happens. It wasn't a misunderstanding, he wasn't stating something he actually thought was correct, he just did it because he was in a position of "power", and there wasn't much I could do about it but take it to a judge who would immediately side with him anyways. And he told me as such. Sucks, and I'm out a couple hundred dollars, and a day going to court.
That said, I find most cops are nice and honest, and I still value the services that they provide, but assuming that just because a cop says something doesn't make it irrefutably true, unfortunately. Wish it was.
The same way the cop saying he saw something is considered "evidence". What guarantee can the cop give that his testimony isn't a complete lie for whatever purpose he might do so (prejudice, bored, envy, revenue, shorting stock)?
It's "presented as evidence" and the judge (or jury) decides based on the evidence what he believes really happened. The prosecution will also get a chance to argue that the data may have been altered, shot with some weird lenses that distort perspective, etc if they so choose, and if the company (or driver) is found to have falsified/perjured themselves then they look at real jail time and real fines rather than the $35 ticket they are charged with.
Well there is a simple answer to that... He didn't. The first version of linux was written for the 80386. A quote from Linus himself... His announcement of linux:
Hello everybody out there using minix -
I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu) for 386(486) AT clones. This has been brewing since april, and is starting to get ready. I'd like any feedback on things people like/dislike in minix, as my OS resembles it somewhat (same physical layout of the file-system (due to practical reasons) among other things).
I've currently ported bash(1.08) and gcc(1.40), and things seem to work. This implies that I'll get something practical within a few months, and I'd like to know what features most people would want. Any suggestions are welcome, but I won't promise I'll implement them:-)
Linus (torvalds@kruuna.helsinki.fi)
PS. Yes - it's free of any minix code, and it has a multi-threaded fs. It is NOT portable (uses 386 task switching etc), and it probably never will support anything other than AT-harddisks, as that's all I have:-(. —Linus Torvalds
If you scroll up, you will see that I said the 80286 had protected mode. How is it that you can't read? One of the things (of many) that made the 80386 different from the 80286 was that it could enter and exit protected mode without having to completely reset. Sure, there were other big architecture changes as well, like 32-bit registers, and a MMU that could do page level protections instead of segment level protections like the 80286. But the 80386 was the first intel chip that could (and still can really) run a modern operating system.
The 68000 didn't have a protected mode either. It kind of had a user/supervisor mode, but since the MMU wasn't integrated, it depended heavily on external chips to implement most of the functionality. That wasn't implemented until the 68010 or 68020 I believe (Maybe even 68030). And yes, I wrote a lot of 68xxx code as well as x86 code. I actually started on 6502 -> 68000 -> x86. Haven't done any ARM programming, but I try and keep up with the different architectures. I've run/used Windows on x86, ARM, Alpha, and MIPS. I haven't written an OS for anything other than the x86, but I did a lot of systems programming on the 6502 and 68000 (and x86), including multitasking when the OS wasn't multitasking itself. The concepts aren't all the much different.
Here's an online magazine from 1990 that highlights a shareware project I did on the 68000 in assembly (YMG), in the background. On an OS that didn't support multitasking, and using hardware devices (RS232 port): https://atarimax.com/freenet/f...
I'm too lazy to go googling, but I'm sure you can find references on the web to some of my past projects.. Super fast disk defragmenters, bulletin board systems, file transfer protocols, multi-user systems, online multiuser gaming systems, multitasking/multiprocessing systems for DOS, windows device drivers, proxy servers, conversations with the w3c groups on html and css specs, etc etc. I'm a pretty decent programmer, and I've been around enough to know a lot of things about a lot of esoteric things.
If you knew anything about OSes, you'd know exactly why I put up the link for the 386 processor. I even pointed you to the exact pages to read.
But since you obviously don't know, here it is. The 80386 processor was the first (and simplest) processor that allowed programs to be able to enter and exit into protected mode. Technically, the 80286 could run protected mode, but not without a complete reset. Without protected mode, the computer would act pretty much how you seem to think it does, with all programs having complete access to everything it wants. The 80386 added the descriptor tables which allowed for memory remapping (needed for virtualizing device memory and virtual memory). Maybe you never knew why, or maybe you were around to know, but there wasn't a UNIX or LINUX for 80286 processors, but there was for the 80386. That's why.
So if you want to learn the very basics of how modern operating systems actually work, the 80386 progammer's manual is one of the best places to start. But if you knew squat, you'd already know that.
There is no computer related topic that is to technically for me...
Then I suggest you actually read those links so you understand. From you:
Gaming normally wants to bypass a lot of the OS layer hence the popularity of DirectX
That is nonsense. There never was an "OS layer" to access the graphics card. What exactly should that be?
No, that is how the OS actually works. See when a program that is running in ring 3 (or an user-level application... Like a game for example) wants to access hardware, it calls a API call, or makes an I/O request, or read/writes to a specific set of memory. The later will cause an exception (Access exception) while the former will just cause a ring transition, but in either case it is then caught by the OS, when then translates the request into something acceptable to the hardware. That exception handling is very costly in terms of performance (partially because of the ring transition). In order to reduce the number of ring transitions, you can either try and allow more things to be directly accessible from user mode without OS oversight (which then can cause stability issues, especially when you have multiple applications who want access to the same hardware), or you start to allow more drivers to run in a ring above ring 3 (usually ring 0) -- That of course leads to entire system stability issues -- AKA BSOD, kernel panic, or whatever the OS has.
So yes, one of the many things that a "server" OS will do is push as much of the drivers into userland so that a poorly performing driver won't take down the entire kernel. But that almost always has a fairly large performance hit. "Desktop" OSes will push as much of the drivers into kernel space in order to gain performance, but at the expense of system stability. Depending on the OS and what the needs of the OS is, dictates where along line of performance vs stability it will use. Obviously the best performance being that absolutely everything runs in Ring 0, and the most stable is that only the core kernel runs in Ring 0 and everything else (drivers included) get pushed down to rings 1-3.
And yes, in many cases, some of the server vs desktop tuning can changed at run-time and use the exact same binary. And yes, in many cases you can take an OS that is designed to be server-class, and run it on the same exact desktop or vice versa. That doesn't mean there isn't a difference, you've just changed the difference from a compile-time difference to a run-time difference, or of course optionally, running the OS sub-optimally (which apparently you don't seem to care about at all).
You may program computers, but for a long number of years, I actually wrote OSes, so yeah, I kind of understand exactly how they work. Apparently you don't. I gave you links so that you could actually read them, and at least understand the basics of what you are talking about, but apparently you'd rather just continue stating what you ignorantly believe to be true rather than educate yourself.
Using a common slashdot metaphor, you are arguing that there is no difference between engines.. They are all engines, they all burn something and make something go, and you can throw a semi-truck desiel engine in a ferrari or a ferrari engine in a truck and they are exactly the same. Therefore there is no such difference between them, and they work exactly the same and you can't understand why a ferrari engine costs as much as it does when you can buy a honda engine for 1/100 the cost. And since you've driven a car before, well you know what you are talking about, and with $2000 you can win nascar, and put tesla out of business. Good luck to you.
Since I can't actually force you to learn or to stop saying ignorant things, and you apparently refuse to do either on your own, I think I'm done here.
Hmmm... Well, I'll just suggest that you brush up on your technical skills. Here is a list of suggested materials that cover the topic: https://css.csail.mit.edu/6.85...
Pay particular attention to pages 106-128 which discusses the protection mechanisms, 145-149 which discusses I/O privileges, and 152-172 which describes interrupts and exceptions.
If you don't have the time, or the topics are too technical for you, then you can just read this, which granted, has a few faults, but from a high level, it's describes things pretty well without getting overly technical: https://blog.codinghorror.com/...
As for the second part of my quote, you can read this: https://www.microsoftpressstor... please pay attention to the section labeled "Quantum" (about 1/3rd of the way through the document). That describes in further detail the differences between desktop and server optimizations. Again, old information, but it still applies today.
Sure there is/was. The OS layer abstracted the graphics hardware away from the user level programs. DirectX (where the "direct" nomenclature came from) was the ability for user-level applications to interface with the graphics hardware either directly or at a much lower level than before.
The difference between a "server OS" and a "desktop OS" still exists, but the differences a much smaller than they used to be. Server OS's will tend to be on the security side more, while desktop os's tend to favor performance more. Also server OS's will prefer throughput (more work done before a task switch increases efficiency), while desktop OS's will prefer low latency/responsiveness -- usually.
And bicycles are a good replacement for semi-trucks, because if you can just scale up the speed at which they go, and scale them up to 5 tons, they will carry more than our current semi-trucks and use no gasoline at all!
I think my first sentence bears repeating...
ARM has come a long way, but it has serious problems trying to scale up.
So completely ignoring that they have issues scaling up, if they COULD just scale up (which they can't) and do it at 100% efficiency (which they can't), it would be competitive.
ARM has come a long way, but it has serious problems trying to scale up. Those issues haven't been solved yet. Taking the current top of the line desktop processor vs the current top of the line A11 from apple, it has a LONG way to go: https://browser.geekbench.com/...
That's the iPhone X putting up just shy of 10k in geekbench, and the intel platform putting up just shy of 70k. That is a pretty big difference to make up, but I look forward to seeing what Apple does.
I would guess they figured with graphics cards having 3500+ cores and ample memory for massive lookup tables, suddenly it seems feasible. That or a patent just expired, or both.
Now you are just being a troll. The point was that he said that a human would learn to drive more defensively in this situation, where this situation is picture perfect for NOT expecting someone to jump out in the middle of the road.
Slashdot Mirror
Hell, Wi Fi can't saturate a 100mbit fast ethernet interface.
Not having a problem getting 250Mbps over WiFi to a single device here. And it's not even a good one. I could easily saturate a 1Gbps internet link over WiFi.
I didn't start using StackOverflow until later, but I'm still in the top 2% of answer givers there (so I've read and answered a good number of questions), and I think I did see one comment once that was racist/sexist, and it was quickly removed.
One other time, there was a user who had a huge chip on their shoulder who decided for no reason to tell everyone about how they were different, then got argumentative (about the technical subject) and rude, and then when he was told he was wrong, blamed it on the fact that he was different. I don't think that got closed, but it was obvious this person was just looking for attention and wanted to start an discussion about their differences.
That's so 1990's. Upgrade your compiler to the 2000's version. It'll give you a silver star for participation instead of those silly cryptic messages.
I paid $1750 for my 130" 4K/HDR/60Hz "TV", but it's a projection TV: https://www.amazon.com/Optoma-...
I think you are confused. UHD is 4k, and HD is (often) 1080p. I think you meant to say that you can tell the difference between SDR and HDR.
Would you buy an autonomous car where heavy rain or a hint of snow would keep you from getting home while a human driver just gets into the car and drives home?
Yes.
I wish my $80,000 car got free updates. Or any updates at all.
Nope, but I have PERSONALLY been affected by a lying police officer, so I know it happens. It wasn't a misunderstanding, he wasn't stating something he actually thought was correct, he just did it because he was in a position of "power", and there wasn't much I could do about it but take it to a judge who would immediately side with him anyways. And he told me as such. Sucks, and I'm out a couple hundred dollars, and a day going to court.
That said, I find most cops are nice and honest, and I still value the services that they provide, but assuming that just because a cop says something doesn't make it irrefutably true, unfortunately. Wish it was.
The same way the cop saying he saw something is considered "evidence". What guarantee can the cop give that his testimony isn't a complete lie for whatever purpose he might do so (prejudice, bored, envy, revenue, shorting stock)?
It's "presented as evidence" and the judge (or jury) decides based on the evidence what he believes really happened. The prosecution will also get a chance to argue that the data may have been altered, shot with some weird lenses that distort perspective, etc if they so choose, and if the company (or driver) is found to have falsified/perjured themselves then they look at real jail time and real fines rather than the $35 ticket they are charged with.
Why would Linus build a Linux/Unix for x286
Well there is a simple answer to that... He didn't. The first version of linux was written for the 80386. A quote from Linus himself... His announcement of linux:
Hello everybody out there using minix -
I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu) for 386(486) AT clones. This has been brewing since april, and is starting to get ready. I'd like any feedback on things people like/dislike in minix, as my OS resembles it somewhat (same physical layout of the file-system (due to practical reasons) among other things).
I've currently ported bash(1.08) and gcc(1.40), and things seem to work. This implies that I'll get something practical within a few months, and I'd like to know what features most people would want. Any suggestions are welcome, but I won't promise I'll implement them :-)
Linus (torvalds@kruuna.helsinki.fi)
PS. Yes - it's free of any minix code, and it has a multi-threaded fs. It is NOT portable (uses 386 task switching etc), and it probably never will support anything other than AT-harddisks, as that's all I have :-(.
—Linus Torvalds
If you scroll up, you will see that I said the 80286 had protected mode. How is it that you can't read? One of the things (of many) that made the 80386 different from the 80286 was that it could enter and exit protected mode without having to completely reset. Sure, there were other big architecture changes as well, like 32-bit registers, and a MMU that could do page level protections instead of segment level protections like the 80286. But the 80386 was the first intel chip that could (and still can really) run a modern operating system.
The 68000 didn't have a protected mode either. It kind of had a user/supervisor mode, but since the MMU wasn't integrated, it depended heavily on external chips to implement most of the functionality. That wasn't implemented until the 68010 or 68020 I believe (Maybe even 68030). And yes, I wrote a lot of 68xxx code as well as x86 code. I actually started on 6502 -> 68000 -> x86. Haven't done any ARM programming, but I try and keep up with the different architectures. I've run/used Windows on x86, ARM, Alpha, and MIPS. I haven't written an OS for anything other than the x86, but I did a lot of systems programming on the 6502 and 68000 (and x86), including multitasking when the OS wasn't multitasking itself. The concepts aren't all the much different.
Here's an online magazine from 1990 that highlights a shareware project I did on the 68000 in assembly (YMG), in the background. On an OS that didn't support multitasking, and using hardware devices (RS232 port): https://atarimax.com/freenet/f...
I'm too lazy to go googling, but I'm sure you can find references on the web to some of my past projects.. Super fast disk defragmenters, bulletin board systems, file transfer protocols, multi-user systems, online multiuser gaming systems, multitasking/multiprocessing systems for DOS, windows device drivers, proxy servers, conversations with the w3c groups on html and css specs, etc etc. I'm a pretty decent programmer, and I've been around enough to know a lot of things about a lot of esoteric things.
And you?
If you knew anything about OSes, you'd know exactly why I put up the link for the 386 processor. I even pointed you to the exact pages to read.
But since you obviously don't know, here it is. The 80386 processor was the first (and simplest) processor that allowed programs to be able to enter and exit into protected mode. Technically, the 80286 could run protected mode, but not without a complete reset. Without protected mode, the computer would act pretty much how you seem to think it does, with all programs having complete access to everything it wants. The 80386 added the descriptor tables which allowed for memory remapping (needed for virtualizing device memory and virtual memory). Maybe you never knew why, or maybe you were around to know, but there wasn't a UNIX or LINUX for 80286 processors, but there was for the 80386. That's why.
So if you want to learn the very basics of how modern operating systems actually work, the 80386 progammer's manual is one of the best places to start. But if you knew squat, you'd already know that.
There is no computer related topic that is to technically for me ...
Then I suggest you actually read those links so you understand. From you:
Gaming normally wants to bypass a lot of the OS layer hence the popularity of DirectX
That is nonsense. There never was an "OS layer" to access the graphics card. What exactly should that be?
No, that is how the OS actually works. See when a program that is running in ring 3 (or an user-level application... Like a game for example) wants to access hardware, it calls a API call, or makes an I/O request, or read/writes to a specific set of memory. The later will cause an exception (Access exception) while the former will just cause a ring transition, but in either case it is then caught by the OS, when then translates the request into something acceptable to the hardware. That exception handling is very costly in terms of performance (partially because of the ring transition). In order to reduce the number of ring transitions, you can either try and allow more things to be directly accessible from user mode without OS oversight (which then can cause stability issues, especially when you have multiple applications who want access to the same hardware), or you start to allow more drivers to run in a ring above ring 3 (usually ring 0) -- That of course leads to entire system stability issues -- AKA BSOD, kernel panic, or whatever the OS has.
So yes, one of the many things that a "server" OS will do is push as much of the drivers into userland so that a poorly performing driver won't take down the entire kernel. But that almost always has a fairly large performance hit. "Desktop" OSes will push as much of the drivers into kernel space in order to gain performance, but at the expense of system stability. Depending on the OS and what the needs of the OS is, dictates where along line of performance vs stability it will use. Obviously the best performance being that absolutely everything runs in Ring 0, and the most stable is that only the core kernel runs in Ring 0 and everything else (drivers included) get pushed down to rings 1-3.
And yes, in many cases, some of the server vs desktop tuning can changed at run-time and use the exact same binary. And yes, in many cases you can take an OS that is designed to be server-class, and run it on the same exact desktop or vice versa. That doesn't mean there isn't a difference, you've just changed the difference from a compile-time difference to a run-time difference, or of course optionally, running the OS sub-optimally (which apparently you don't seem to care about at all).
You may program computers, but for a long number of years, I actually wrote OSes, so yeah, I kind of understand exactly how they work. Apparently you don't. I gave you links so that you could actually read them, and at least understand the basics of what you are talking about, but apparently you'd rather just continue stating what you ignorantly believe to be true rather than educate yourself.
Using a common slashdot metaphor, you are arguing that there is no difference between engines.. They are all engines, they all burn something and make something go, and you can throw a semi-truck desiel engine in a ferrari or a ferrari engine in a truck and they are exactly the same. Therefore there is no such difference between them, and they work exactly the same and you can't understand why a ferrari engine costs as much as it does when you can buy a honda engine for 1/100 the cost. And since you've driven a car before, well you know what you are talking about, and with $2000 you can win nascar, and put tesla out of business. Good luck to you.
Since I can't actually force you to learn or to stop saying ignorant things, and you apparently refuse to do either on your own, I think I'm done here.
Hmmm... Well, I'll just suggest that you brush up on your technical skills. Here is a list of suggested materials that cover the topic:
https://css.csail.mit.edu/6.85...
Pay particular attention to pages 106-128 which discusses the protection mechanisms, 145-149 which discusses I/O privileges, and 152-172 which describes interrupts and exceptions.
If you don't have the time, or the topics are too technical for you, then you can just read this, which granted, has a few faults, but from a high level, it's describes things pretty well without getting overly technical:
https://blog.codinghorror.com/...
As for the second part of my quote, you can read this: https://www.microsoftpressstor... please pay attention to the section labeled "Quantum" (about 1/3rd of the way through the document). That describes in further detail the differences between desktop and server optimizations. Again, old information, but it still applies today.
Enjoy!
Sure there is/was. The OS layer abstracted the graphics hardware away from the user level programs. DirectX (where the "direct" nomenclature came from) was the ability for user-level applications to interface with the graphics hardware either directly or at a much lower level than before.
The difference between a "server OS" and a "desktop OS" still exists, but the differences a much smaller than they used to be. Server OS's will tend to be on the security side more, while desktop os's tend to favor performance more. Also server OS's will prefer throughput (more work done before a task switch increases efficiency), while desktop OS's will prefer low latency/responsiveness -- usually.
And bicycles are a good replacement for semi-trucks, because if you can just scale up the speed at which they go, and scale them up to 5 tons, they will carry more than our current semi-trucks and use no gasoline at all!
I think my first sentence bears repeating...
ARM has come a long way, but it has serious problems trying to scale up.
So completely ignoring that they have issues scaling up, if they COULD just scale up (which they can't) and do it at 100% efficiency (which they can't), it would be competitive.
ARM has come a long way, but it has serious problems trying to scale up. Those issues haven't been solved yet. Taking the current top of the line desktop processor vs the current top of the line A11 from apple, it has a LONG way to go: https://browser.geekbench.com/...
That's the iPhone X putting up just shy of 10k in geekbench, and the intel platform putting up just shy of 70k. That is a pretty big difference to make up, but I look forward to seeing what Apple does.
We can that completely separate CPU, video hardware, OS, and network stack... Your phone.
For the price of one, I can get a decent set of decent speakers.
Where can I find a decent set of speakers for $29? That's what I paid for my alexa dot.
Correction, you can't walk everywhere either: http://www.newsweek.com/apple-...
Until people learn how to walk, we should revoke their walking license.
I would guess they figured with graphics cards having 3500+ cores and ample memory for massive lookup tables, suddenly it seems feasible. That or a patent just expired, or both.
Am I the only one who noticed?
an FBI agent applied the bloodied body's index finger
Uh... Perhaps using the thumbprint instead would have been better since that is what the iPhone uses?
Q: Why didn't the chicken cross the road?
A: Because there might have been an uber car on it.
Well, uber is fixing that my lowering the once-in-every-1500-mile occurrences to once every 11 miles. That'll keep the drivers engaged. LOL.
Already did. Read below.
Now you are just being a troll. The point was that he said that a human would learn to drive more defensively in this situation, where this situation is picture perfect for NOT expecting someone to jump out in the middle of the road.