The funny thing is that it teeter-totters back and forth from one core to the other. I wish I knew what made it do that.
The OS runs the process a few milliseconds at a time, then kicks the process of the cpu for another process to run (if there is one, including OS tasks such as I/O routines). When the OS starts up the process again for a few more milliseconds, it may start it up on a different core. That is why both cores will show 50% average utilization.
Now if you set CPU affinity for that process to be on one core, then it will max that core out at 100% and the other core will be idle. This may result in better performance, because you get better cache utilization if the process stays on the same core.
On a related topic, this can also be the case if the app is multithreaded -- sometimes it is more efficient to run multiple threads on the same CPU instead of across CPUs, if each thread is accessing the same region of memory. Otherwise, if the threads are on different CPUs or cores, then the threads are constantly invalidating the cache on the other core, causing more (expensive) reads/writes to main memory.
Just wondering, how hot do your drives run? I haven't had hardly any significant drive failures in my home equipment in the last 15 - 20 years or so. But then again my drives run at close to room temprature (either because of the drive brand itself, or the case has good cooling). For people that I know that had premature drive failures, the drives ran excessively hot (you couldn't touch them for more than a second or so).
Also from TFA "The termination shock is the region of the heliosphere where the supersonic solar wind slows to subsonic speed"
Last I checked wasn't sonic speed something only relative to earth? Wouldn't that make this point completly arbitrary in a cosmic sense?
This was covered in the Slashdot post a while back about Voyager 2 crossing the termination shock. It boils down to the fact that the plasma from the solar wind does conduct waves, although due to the density of the particles and the nature of a plasma, the waves are much faster than the speed of sound through earth's atmosphere. So sonic speed does have a point (and related phenomena in this context. See this article, or google "super sonic speed heliopause".
My problem with most of the split-keyboards is they put the "6" key on the left side of the split, not the right side. The way I was taught typing is the left hand hits numbers 1-5, and the right does 6-10 -- I can't get used to hitting the "6" with my left hand.
I think the best option would be to pick up a "slug" (nslu2), which has ethernet (or 2 ethernet, I can't remember) + 2 usb. Then add a usb wifi dongle that can run in AP mode, along with a 4-gig usb thumbdrive, and you're golden. Another option is to use a ethernet -> wifi box (they are sold as wifi adapters for things like the sony ps2). But then you are back to having multiple power bricks.
Except, when I bought a new car, there was a small defect in the paint job -- a nearly unnoticeable paint bubble. I'm sure that every car that comes off the lot has a blemish somewhere. Doesn't cause the car to crash, and life goes on. Same thing ain't true with software -- that same "blemish" could easily be turned around and allow someone to break into the software.
The only way we could get comparable results with software v.s. physical objects is if computer systems develop the ability to withstand a certain percentage of defects without adverse affects. Kind of like how a few bolts on a bridge can be bad, because the way it is engineered is if a section calls for 5 bolts, they put 8 bolts in there just to be safe. Not a lot of practical ways to do that with software, at least not without a performance trade off.
The definition I follow is as follows: An operating system is composed of two major components -- a Kernel, and a collection of utility applications. The OS kernel is utilized by other software to provide coordinated access via a stable API to the underlying hardware resources. The bundled utility applications are provided to facilitate user interaction and manipulation of the underlying hardware and logical constructs that are contained within the hardware. So utilities such as ls, cp, mv,... are for the user to manipulate the logical data on a storage device. Text editors can fall in this category also, so they can be part of the OS.
The gray area is with utilities such as ftp, telnet, minicom, etc. They provide user access to logical data constructs that are accessed through the network hardware, so they can be reasonably considered OS components. But a web browser does essentially the same thing as ftp (allows access to the http protocol instead of ftp protocol), but many people consider a web browser to be an application and not an OS utility.
Now a level higher than OS is an Operating Environment. This is what I would classify most Linux distributions, along with Mac OSX -- they include the Kernel, Utilities, and an integrated set of applications that allow a user to do useful work (not just maintain the system's data structures).
Better yet, create the primary partition on a 16 gig card to be, say, 2 gig. Modify your encryption code to look at the raw device starting at over 2 gig. Peal off the label from the card, and replace it with one that says "2GB".
What I do is go into Edit mode, hit Z to create a cut list from the commercial flags, then validate that visually (check a few frames before/after each cut mark). Also, I check to make sure the cut out sections are all about the same size and positioned fairly evenly.
I noticed that the latest version tends to be a bit more accurate, requiring fewer manual fixes.
UMPCs are a bit smaller (similar size or slightly bigger than a Sharp Zaurus SL-C3100), but have a more powerful processor, more ram, and a much shorter battery life than these new devices. Oh, and they also run about $2000.00 or so. These netbooks are a hybrid of the laptop and UMPC concept -- keep shrinking the laptop until the point that the price would start going up, then cut back the processing power & memory until it is the minimum to run their target apps (web browser, email, productivity apps).
The idea of a stable distribution release isn't that all the shipped software is bug free on the day it is shipped; it is that all the components will have security & bug fixes applied to them throughout the distribution's life cycle (v.s. replacing a buggy software component with a newer version which may have new bugs). For example, if the distribution ships with OpenSSH 3.9 and a bug is found, a stable distribution will ship a patched version of OpenSSH 3.9 instead of shipping the latest OpenSSH release. This way the distro gets less buggy as time goes on instead of having old bugs replaced with new bugs over time.
Who in their right minds would turn down an opportunity to see new places
That's what I used to think. But in reality, when you travel for work, you get to see the inside of airplains, hotels, the client's office, and a taxi, and very little else. What you don't get to see a lot of is friends and family.
The "inputdev" tag comes up automatically when the spoon icon is attached to the story. My guess is that kdawson picked the spoon icon for this story because spoons go with icecream, not because it is a story about input devices.
Actually, that fob is only on the OEM keys, and is only needed for starting the car (assuming it has an anti-theft chip in it). A cheap copy made at the hardware store works fine for the door, and is flat. Also, they make plastic credit-card shaped spare keys where the key folds out from the plastic card. Made for the wallet.
I've read that, and the ideas that it links to. What you are proposing is that everything be converted to a form of multi-branch pipeline programming, correct? That is, think of a standard Unix pipe. Then imagine a process having multiple inputs / outputs, and each of those outputs can be connected to a input on a different process. So once the base modules are done, programming would be a matter of connecting various inputs and outputs, like designing an electronic circuit.
I can see how this would help with multiple cores, but can every construct in computing be represented by these "circuit" modules? Also, is this similar to Hartmann pipelines?
Actually, I was thinking it would be perfect for a coach to use on the Sunday Afternoon battlefield, known as the grid-iron. Note, only one player on the team is allowed to have a radio in his helmet (usually it's the quarterback) -- this would be a good way to subvert that rule.
I was more upset when I ordered a case of hard drives -- the shipping container said "Quantity: 1K", and I only got 1000 hard drives, not the 1024 I was expecting.
The only way I know of to get a phone to work with any carrier is to produce a phone without a radio, and use a radio plugin from the carrier. This can be accomplished by the phone having a CF slot, since nearly all the carriers have a CF data card which can be used for voice also.
Just producing a multi-band phone won't work, since many carriers (such a Verizon and Sprint) won't activate a phone that didn't come from them. But they can't controll where you use the CF card you get from them though.
A lot of people are satisfied with t-mobile. What you do is get one of the "free" phones from them, then go and get whatever unlocked GSM phone you want -- I picked up a moto a780 for less than 200, and it works great with my t-mobile sim.
So far they seem to have the most responsive customer service i've dealt with (was on nextel before). Only thing that bugs me is they don't have any way of blocking SMS spam (unless it is sent to your t-mobile email account, then you can put filters on that one). Of course, i've only gotten sms spam once every couple of months though.
The main problem is a combination of the short spindown time, and something wanting to write out to the drive every 30 seconds or so. The main culprit could be the fact that by default, a files last access time (atime) gets updated on every read, even if that read comes from cache. So when the drive is spun down, it gets spun up even on cached reads (to write out the atime). Add "-o noatime" to the filesystems in/etc/fstab, and that should clear up the issue.
They aren't doing that great of a job keeping the website up to date -- the people running this are new at this type of activity (FIC normally sells to OEMs an resellers, not directly to end users). From following the mail list, I'd say December (or possibly January). The main things that have been added: Faster CPU WIFI Different GPS chip They also had to take out one of the speakers to fit the WIFI chip (the GTA-01 had stereo speakers). My biggest issues with the Neo 1973 are that it has very few external buttons (I'd like at least a 4-way nav buttons plus a couple of select buttons on the front), and the fact that it doesn't have EDGE support (GPRS only). So I'm torn between getting a gta-02 model, or keeping my current cellphone (Moto A780) and pairing that up with a Nokia N800.
Actually it does exist. However it is currently only in prototype form -- a few have been made going to key developers who are putting it through its paces to see if any final hardware bugs are found, before the production run starts.
If no showstoppers are found in the current prototype run, then production should ramp up and they should be available by December. Of course that could slip again if major problems are found (which is a good thing -- I don't want broken hardware, I'd rather wait an extra month to get something that works).
Slashdot Mirror
The funny thing is that it teeter-totters back and forth from one core to the other. I wish I knew what made it do that.
The OS runs the process a few milliseconds at a time, then kicks the process of the cpu for another process to run (if there is one, including OS tasks such as I/O routines). When the OS starts up the process again for a few more milliseconds, it may start it up on a different core. That is why both cores will show 50% average utilization.
Now if you set CPU affinity for that process to be on one core, then it will max that core out at 100% and the other core will be idle. This may result in better performance, because you get better cache utilization if the process stays on the same core.
On a related topic, this can also be the case if the app is multithreaded -- sometimes it is more efficient to run multiple threads on the same CPU instead of across CPUs, if each thread is accessing the same region of memory. Otherwise, if the threads are on different CPUs or cores, then the threads are constantly invalidating the cache on the other core, causing more (expensive) reads/writes to main memory.
um, old days of cd cases? I remember the good old days of free floppies!.
Just wondering, how hot do your drives run? I haven't had hardly any significant drive failures in my home equipment in the last 15 - 20 years or so. But then again my drives run at close to room temprature (either because of the drive brand itself, or the case has good cooling). For people that I know that had premature drive failures, the drives ran excessively hot (you couldn't touch them for more than a second or so).
Also from TFA "The termination shock is the region of the heliosphere where the supersonic solar wind slows to subsonic speed"
Last I checked wasn't sonic speed something only relative to earth? Wouldn't that make this point completly arbitrary in a cosmic sense?
This was covered in the Slashdot post a while back about Voyager 2 crossing the termination shock. It boils down to the fact that the plasma from the solar wind does conduct waves, although due to the density of the particles and the nature of a plasma, the waves are much faster than the speed of sound through earth's atmosphere. So sonic speed does have a point (and related phenomena in this context. See this article, or google "super sonic speed heliopause".
See if you can sell the warped one to the guy above that prefers the Microsoft Natural keyboard.
My problem with most of the split-keyboards is they put the "6" key on the left side of the split, not the right side. The way I was taught typing is the left hand hits numbers 1-5, and the right does 6-10 -- I can't get used to hitting the "6" with my left hand.
I think the best option would be to pick up a "slug" (nslu2), which has ethernet (or 2 ethernet, I can't remember) + 2 usb. Then add a usb wifi dongle that can run in AP mode, along with a 4-gig usb thumbdrive, and you're golden.
Another option is to use a ethernet -> wifi box (they are sold as wifi adapters for things like the sony ps2). But then you are back to having multiple power bricks.
Except, when I bought a new car, there was a small defect in the paint job -- a nearly unnoticeable paint bubble. I'm sure that every car that comes off the lot has a blemish somewhere. Doesn't cause the car to crash, and life goes on. Same thing ain't true with software -- that same "blemish" could easily be turned around and allow someone to break into the software.
The only way we could get comparable results with software v.s. physical objects is if computer systems develop the ability to withstand a certain percentage of defects without adverse affects. Kind of like how a few bolts on a bridge can be bad, because the way it is engineered is if a section calls for 5 bolts, they put 8 bolts in there just to be safe. Not a lot of practical ways to do that with software, at least not without a performance trade off.
The definition I follow is as follows: ... are for the user to manipulate the logical data on a storage device. Text editors can fall in this category also, so they can be part of the OS.
An operating system is composed of two major components -- a Kernel, and a collection of utility applications. The OS kernel is utilized by other software to provide coordinated access via a stable API to the underlying hardware resources. The bundled utility applications are provided to facilitate user interaction and manipulation of the underlying hardware and logical constructs that are contained within the hardware. So utilities such as ls, cp, mv,
The gray area is with utilities such as ftp, telnet, minicom, etc. They provide user access to logical data constructs that are accessed through the network hardware, so they can be reasonably considered OS components. But a web browser does essentially the same thing as ftp (allows access to the http protocol instead of ftp protocol), but many people consider a web browser to be an application and not an OS utility.
Now a level higher than OS is an Operating Environment. This is what I would classify most Linux distributions, along with Mac OSX -- they include the Kernel, Utilities, and an integrated set of applications that allow a user to do useful work (not just maintain the system's data structures).
Better yet, create the primary partition on a 16 gig card to be, say, 2 gig. Modify your encryption code to look at the raw device starting at over 2 gig. Peal off the label from the card, and replace it with one that says "2GB".
What I do is go into Edit mode, hit Z to create a cut list from the commercial flags, then validate that visually (check a few frames before/after each cut mark). Also, I check to make sure the cut out sections are all about the same size and positioned fairly evenly.
I noticed that the latest version tends to be a bit more accurate, requiring fewer manual fixes.
UMPCs are a bit smaller (similar size or slightly bigger than a Sharp Zaurus SL-C3100), but have a more powerful processor, more ram, and a much shorter battery life than these new devices. Oh, and they also run about $2000.00 or so.
These netbooks are a hybrid of the laptop and UMPC concept -- keep shrinking the laptop until the point that the price would start going up, then cut back the processing power & memory until it is the minimum to run their target apps (web browser, email, productivity apps).
The idea of a stable distribution release isn't that all the shipped software is bug free on the day it is shipped; it is that all the components will have security & bug fixes applied to them throughout the distribution's life cycle (v.s. replacing a buggy software component with a newer version which may have new bugs). For example, if the distribution ships with OpenSSH 3.9 and a bug is found, a stable distribution will ship a patched version of OpenSSH 3.9 instead of shipping the latest OpenSSH release. This way the distro gets less buggy as time goes on instead of having old bugs replaced with new bugs over time.
I know what you mean about feeling dated. I watched an episode recently, and kept on thinking, Why doesn't he just grab his cell phone?
That's what I used to think. But in reality, when you travel for work, you get to see the inside of airplains, hotels, the client's office, and a taxi, and very little else. What you don't get to see a lot of is friends and family.
The "inputdev" tag comes up automatically when the spoon icon is attached to the story. My guess is that kdawson picked the spoon icon for this story because spoons go with icecream, not because it is a story about input devices.
Actually, that fob is only on the OEM keys, and is only needed for starting the car (assuming it has an anti-theft chip in it). A cheap copy made at the hardware store works fine for the door, and is flat. Also, they make plastic credit-card shaped spare keys where the key folds out from the plastic card. Made for the wallet.
I've read that, and the ideas that it links to. What you are proposing is that everything be converted to a form of multi-branch pipeline programming, correct? That is, think of a standard Unix pipe. Then imagine a process having multiple inputs / outputs, and each of those outputs can be connected to a input on a different process. So once the base modules are done, programming would be a matter of connecting various inputs and outputs, like designing an electronic circuit.
I can see how this would help with multiple cores, but can every construct in computing be represented by these "circuit" modules? Also, is this similar to Hartmann pipelines?
Actually, I was thinking it would be perfect for a coach to use on the Sunday Afternoon battlefield, known as the grid-iron.
Note, only one player on the team is allowed to have a radio in his helmet (usually it's the quarterback) -- this would be a good way to subvert that rule.
I was more upset when I ordered a case of hard drives -- the shipping container said "Quantity: 1K", and I only got 1000 hard drives, not the 1024 I was expecting.
The only way I know of to get a phone to work with any carrier is to produce a phone without a radio, and use a radio plugin from the carrier. This can be accomplished by the phone having a CF slot, since nearly all the carriers have a CF data card which can be used for voice also.
Just producing a multi-band phone won't work, since many carriers (such a Verizon and Sprint) won't activate a phone that didn't come from them. But they can't controll where you use the CF card you get from them though.
A lot of people are satisfied with t-mobile. What you do is get one of the "free" phones from them, then go and get whatever unlocked GSM phone you want -- I picked up a moto a780 for less than 200, and it works great with my t-mobile sim.
So far they seem to have the most responsive customer service i've dealt with (was on nextel before). Only thing that bugs me is they don't have any way of blocking SMS spam (unless it is sent to your t-mobile email account, then you can put filters on that one). Of course, i've only gotten sms spam once every couple of months though.
The main problem is a combination of the short spindown time, and something wanting to write out to the drive every 30 seconds or so. The main culprit could be the fact that by default, a files last access time (atime) gets updated on every read, even if that read comes from cache. So when the drive is spun down, it gets spun up even on cached reads (to write out the atime). /etc/fstab, and that should clear up the issue.
Add "-o noatime" to the filesystems in
They aren't doing that great of a job keeping the website up to date -- the people running this are new at this type of activity (FIC normally sells to OEMs an resellers, not directly to end users). From following the mail list, I'd say December (or possibly January).
The main things that have been added:
Faster CPU
WIFI
Different GPS chip
They also had to take out one of the speakers to fit the WIFI chip (the GTA-01 had stereo speakers).
My biggest issues with the Neo 1973 are that it has very few external buttons (I'd like at least a 4-way nav buttons plus a couple of select buttons on the front), and the fact that it doesn't have EDGE support (GPRS only).
So I'm torn between getting a gta-02 model, or keeping my current cellphone (Moto A780) and pairing that up with a Nokia N800.
Actually it does exist. However it is currently only in prototype form -- a few have been made going to key developers who are putting it through its paces to see if any final hardware bugs are found, before the production run starts.
If no showstoppers are found in the current prototype run, then production should ramp up and they should be available by December. Of course that could slip again if major problems are found (which is a good thing -- I don't want broken hardware, I'd rather wait an extra month to get something that works).