Because enforcing that every application use these would mean certain sorts of applications couldn't be written (or at least not as easily).
Unless setting "Turn on DEP for all programs and services except those I select" doesn't do what it says (i.e., a program can still "opt-out" in code), then there are very few apps that have a problem with DEP.
I have this set on dozens of machines (both server and desktop), and have had to make exceptions for less than 5 programs, with the only really annoying one being the driver installer for a TV tuner card (since I think that means that any program named "SETUP.EXE" would be exempted). After I ran the install, I revoked the exception just in case.
There are two numbers sensors should measure, to be able to determine usable signal quality... the strength of the signal, and the amount of noise (including the receiver's noise threshold and internal noise).
And, since this a digital signal with known characteristics (width of analog spectrum, encoding method, amount of error correction, etc.), the minimum S/N for no unrecoverable loss of data can be mathematically derived. For example, on US OTA digital TV, it's around 19dB (although it's been a while, so my memory might be a few dB off).
This level should be "zero bars", as the slightest intermittent drop would cause data loss, and the phone is using the most transmitter power to keep the connection. Five bars should be "the phone is using the lowest possible transmitter power". So, overall, the bars should be an inverse display of transmitter power being used.
So without claiming that this benchmark is the final word, I'd expect the AMD to be at least on a similar performance level to the Xeon.
Passmark is an excellent benchmark that thoroughly utilizes every core. So, it measures the absolute limit of the processor (in general...there could be slight variations based on the motherboard, RAM, etc.).
As such, it's a pretty good indicator of general server performance, and does give you a feel for "normal" deskstop performance (where at least some multitasking is involved), but it's not a good indicator of single-thread performance.
At least with verion 3.6.4 you can go to Edit -> Prefferences -> Advanced -> Network and specify the limit of what Firefox may cache.
That's the disk cache, not the memory cache.
Also, the numbers reported by Firefox as "used memory cache" are about 10% of the total memory used on my system. So, Firefox claims that only 80MB are used for memory cache, while Windows reports that 800MB is being used by the firefox.exe process.
Sure, these might not be the best motherboards for gaming (although they are pretty competitive compared to other socket 1156 motherboards), but for a workstation doing everything else, they're great.
And, this way you get a motherboard that is thoroughly tested with ECC RAM (as that's what is expected to be used), and likely far better BIOS control of the ECC.
I noticed that you have started to advertise through my ISPs Zip+4 locality based advertising system. Unfortunately, I believe that this system is intrusive and an infringement not only on my right to privacy and anonymity, but also encroaches on the data allowance I pay for which is already prohibitively capped by my ISP.
Thanks to the government allowing the ISP a monopoly of the "last mile" of the connection, I cannot switch my ISP to one not offering this intrusive advertising scheme. I can, however, refuse to offer my business to those who make use of it.
Sincerely,
A lost customer."
I suspect that the LWAA (Letter Writing Association of America) is going to be coming after both of us, though.
BTW, on my machine Opera behaves much better than Firefox with 20+ tabs open (I have 57 right now), it's still snappy and Firefox would be crawling and taking up loads of RAM. But of course YMMV.
I'm wondering how much other apps/OS settings have to do with this.
On my work machine, I see FF taking up to 1.3GB of RAM, and it becomes very unresponsive until I close and re-open the browser. On my home machine, it never takes more than about 300MB, despite essentially the same browsing pattern.
The work machine runs XP SP3 while at home I'm running XP 64-bit. I run the 32-bit version of Firefox on both.
Re:Does it have a monitor and full-size keyboard?
on
Flight of the Desktops
·
· Score: 1
As hard drives get larger, RAID becomes a lot more important. At the very least, you need a large backup drive, and USB 2.0 isn't enough to handle that much in a reasonable amount of time.
Likewise, as RAM increases, ECC becomes more important.
Currently, there are games that cannot run at even laptop resolutions at 30fps without multiple graphics cards. Microsoft Office 2010 uses 3D acceleration to draw the UI, so many laptops will become visibly slower. IE9 is supposed to use this same sort of 3D, so doing the "e-mail, web browsing, and word processing" that are all "most people" do with their computer will become a lot more tedious.
All of these can certainly be added to laptops, but battery technology isn't keeping up with the rest of the increases, so you have things like the current Core i7 only lasting an hour or so on battery. That's going to get worse before it gets better.
So, what's likely going to happen is that desktops will make a comeback as smaller laptops/netbooks and more powerful smart phones (and even the iPad) take over the portable computing market.
"And, to get that "live CD" behavior," just boot the CD image.
True, but you can customize an installed-to-HD distro to be exactly what you want, and keep it patched up (as long as you remember to temporarily disable the "revert to snapshot" while patching).
That's how I work from home instead of staying at the office all hours of the night.
If you can carry everything you need on your laptop, you're one of the lucky few. Every bit of data I need is available over a VPN. And, I don't need to lug any major hardware (just a RSA key) around to get to it.
For anybody who needs truly portable computing (e.g., at a client site, on the commute, etc.), some sort of laptop/netbook/whatever is probably the only choice (until smart phones really ramp up the computing power). For people who need access from a couple of fixed places, it's almost always easier (and sometimes cheaper) to have a desktop in each location.
So that I have choice! At work my laptop is plugged into external devices, including a 20" monitor. At the end of the work day, I just pick it up and take it home, and sit with it on the couch to surf the web.
First, I would never want to use a work laptop for my personal machine at home...any decent IT department will lock it down way too much.
Second, I would never want to work for a company that basically required me to bring my personal laptop to use to do work, nor would I trust a company that would just let somebody plug any computer into their network.
If you're not regularly using the laptop on the commute, at coffee shops, etc., you don't really need a laptop. I'd bet 99% of what most people want to move from home to work is data, so just carry a 500GB (or smaller, if you don't need as much) portable drive to and from work. Depending on the exact drive, it will weigh anywhere from less than an ounce (USB flash drive) to about 8 ounces, either of which is far less than a laptop.
Re:Does it have a monitor and full-size keyboard?
on
Flight of the Desktops
·
· Score: 2, Informative
It's also got 4TB of disk space, 6 powered USB ports (4 in back of which 3 are in use, 2 in front of which I use one), memory card reader, DVD burner, and a cable-TV video card so I can also use it as a DVR.
You do realize that laptops can have all those things, right?
Wake me up when you finish copying a few hundred gigabytes to a USB 2.0 hard drive. Also, once you plug 8 devices into that USB bus, it gets even worse. eSATA (which I have on my laptop) is a lifesaver, but you usually have only one port, while on a desktop, even two eSATA ports is a lot.
Seriously, if you want anything at all other than a basic computer (e.g., really fast processors, multiple processors, RAID, multiple video cards, serious amounts of RAM, etc.), a laptop won't cut it. Even if you can get those things on a laptop, they are a lot more expensive. If you use a laptop as a very rarely moved powerful desktop replacement, you're spending a lot more money than you need to...a good desktop and a cheap laptop will cost less and give you the same utility. Also, with current Intel and AMD desktop chips, overclocking is so easy and safe that the price/performance becomes even more in favor of the desktop.
I suspect you meant "etherrnet" instead of "Internet", since how you connect to the Internet doesn't control how you communicate between devices in your home.
With that assumption, if you are the type that reads/., there is no excuse for your home not to have wired ethernet if you want wired ethernet.
I know we had a new developer start who's surname started with O', and while there was no problem getting him set up in the email system, his [name].o'[surname]@[address].com email address wouldn't work with a particular instance of Bugzilla we had to use (it just silently threw away the emails to him).
You got lucky in being able to incorrectly set up the e-mail address in your system.
The single quote is a delimiter in e-mail addresses, and needs to be correctly paired or escaped. Otherwise, the address is not valid according to the RFC.
I built a Shuttle SFF PC once, and the barebones kit started around $200-$400, and that's before dropping in a CPU and RAM.
You can start at less than $180, with CPU at around $150 (for Core 2 Duo or Quad), and 4GB of RAM for less than $75. A video card 5-10 times as fast as the GeForce 320M is around $50.
So, for around $450 you can buy faster hardware in about the same form factor as the Mac Mini, but with far more future expandability.
It's impossible to determine if it's a human pressing on a remote control or a computer hitting an IR blaster changing the channel.
I suspect that cable boxes and their remotes will eventually have something like the "rolling codes" system that garage doors and cars use.
Then, only companies with lots of money to license the technology would be able to build universal remotes that work with the cable box. And, if you design a workaround, you get a DMCA lawsuit. All so that the cable company can charge you $10/month extra for their crappy DVR.
Most US warships have vending machines and stores that sell anything from snacks to clothing to DVD players (depending on the size of the ship and its store), so yes, money is still used.
Everything in those vending machines can be considered "luxuries", though. It's not like the sailors have to pay for their food in the mess hall, or pay rent for their rack.
You should easily be able to get 5% on a $5M investment. Assuming 50% taxes on the interest (way high for the US), and spending $300K/year, you have 22 years before the money runs out.
Setting the tax rate to a more realistic 30% and spending $250K/year gives you 33 years before the money runs out, while dropping to $200K/year spending is 55 years of money. For many, that's the rest of their lives.
The heart of the game is a big random number generator. There are many more possible outcomes than can be represented by the numbers/symbols on the reels. There is a mapping between the random numbers and what the payouts are, this sets the odds. After the next random number is determined, the machine decides what it wants to show on the reels. Usually it likes to show some kind of near miss to keep up player interest. There are other algorithm adjustments you can make to make a machine make frequent small payouts, or occasional big payouts.
Although this is the closest description in this thread of how slot machines work, it's still a bit off, but even Wikipedia gets it right.
To see all the gory details, and why you see the "near miss", check out The Wizard of Odds. The Q&A about slots (and other games) at that site is also a good resource.
Show me a documented example of a licensed casino owner keeping a machine on the floor after discovering that it was defective.
I suspect it happens all the time.
Suppose a machine was defective in such a way that it would never hit the biggest jackpot listed. Since that happens only once every few years, it may never be known. And, when such a machine is "inspected", it may not even be noticed if the error is subtle enough. Especially since it's in the casino's interest to not have the machine 100% functional. Then, any large jackpot payout can be avoided, even if the error isn't directly related to that jackpot. In my example, if the biggest jackpot is $X million progressive and can't be won due to an error in programming, but hit something like 5,000 times your bet, the casino could still void your win due to "malfunction".
An $11M jackpot is a prize where the winner moves to a new home and changes their name, giving up gambling forever - it provides little advertising benefit.
There are many people who win those kinds of large jackpots who later end up broke, with a good chunk of the losses coming from gambling.
The free market is as much an idealized unattainable as the Star Trek no money communism utopia.
Since pretty much every incarnation of Star Trek depicted the use of "money" (i.e., a token that has no intrinsic value of its own, but has been defined to have value by the government), I guess the free market is quite attainable.
What confuses people is that 95% of Star Trek episodes are set on the ship or in a diplomatic situation. On current US warships, sailors don't need money for very much other than gambling among themselves...meals are free, etc. Why should the starship Enterprise be any different?
In diplomatic situations, the host country (planet) doesn't make the guests pay for their food and entertainment at the state dinners, and often lets them use housing that is set aside for such occasions.
But, the crew paid for food and drinks in restaurants and bars, tribbles cost money to purchase, and bribes were paid for information.
I like the multiple tax bracket idea (with no taxes for the lowest bracket) and progressively stronger brackets (or possibly a smooth gradient?) better for that reason.
I suspect the only two real problems with the current system is that the brackets aren't adjusting with inflation, and there needs to be one or more brackets above the current "two-income professional family all the way up to Bill Gates" that is tops in the US.
Slashdot Mirror
Because enforcing that every application use these would mean certain sorts of applications couldn't be written (or at least not as easily).
Unless setting "Turn on DEP for all programs and services except those I select" doesn't do what it says (i.e., a program can still "opt-out" in code), then there are very few apps that have a problem with DEP.
I have this set on dozens of machines (both server and desktop), and have had to make exceptions for less than 5 programs, with the only really annoying one being the driver installer for a TV tuner card (since I think that means that any program named "SETUP.EXE" would be exempted). After I ran the install, I revoked the exception just in case.
The computation you should be looking at is Signal to noise ratio.
There are two numbers sensors should measure, to be able to determine usable signal quality... the strength of the signal, and the amount of noise (including the receiver's noise threshold and internal noise).
And, since this a digital signal with known characteristics (width of analog spectrum, encoding method, amount of error correction, etc.), the minimum S/N for no unrecoverable loss of data can be mathematically derived. For example, on US OTA digital TV, it's around 19dB (although it's been a while, so my memory might be a few dB off).
This level should be "zero bars", as the slightest intermittent drop would cause data loss, and the phone is using the most transmitter power to keep the connection. Five bars should be "the phone is using the lowest possible transmitter power". So, overall, the bars should be an inverse display of transmitter power being used.
So without claiming that this benchmark is the final word, I'd expect the AMD to be at least on a similar performance level to the Xeon.
Passmark is an excellent benchmark that thoroughly utilizes every core. So, it measures the absolute limit of the processor (in general...there could be slight variations based on the motherboard, RAM, etc.).
As such, it's a pretty good indicator of general server performance, and does give you a feel for "normal" deskstop performance (where at least some multitasking is involved), but it's not a good indicator of single-thread performance.
If that weren't enough, remember that they basically only run on EDGE, so you'll be pumping out packets at a blazing 4-10 KBytes/sec.
At even 1 KB/sec, if you have 150,000 phones in the botnet, you have the equivalent of an OC-24 for sending spam.
At least with verion 3.6.4 you can go to Edit -> Prefferences -> Advanced -> Network and specify the limit of what Firefox may cache.
That's the disk cache, not the memory cache.
Also, the numbers reported by Firefox as "used memory cache" are about 10% of the total memory used on my system. So, Firefox claims that only 80MB are used for memory cache, while Windows reports that 800MB is being used by the firefox.exe process.
Or, you know, go AMD. Because they don't limit ECC to only server parts.
Or, just buy any one of a half-dozen motherboards costing less than $200 and add a Xeon that is priced within 5% of the equivalent spec non-Xeon.
Sure, these might not be the best motherboards for gaming (although they are pretty competitive compared to other socket 1156 motherboards), but for a workstation doing everything else, they're great.
And, this way you get a motherboard that is thoroughly tested with ECC RAM (as that's what is expected to be used), and likely far better BIOS control of the ECC.
I will take a copy of your letter, however.
Here, I'll make that copy for you:
I suspect that the LWAA (Letter Writing Association of America) is going to be coming after both of us, though.
BTW, on my machine Opera behaves much better than Firefox with 20+ tabs open (I have 57 right now), it's still snappy and Firefox would be crawling and taking up loads of RAM. But of course YMMV.
I'm wondering how much other apps/OS settings have to do with this.
On my work machine, I see FF taking up to 1.3GB of RAM, and it becomes very unresponsive until I close and re-open the browser. On my home machine, it never takes more than about 300MB, despite essentially the same browsing pattern.
The work machine runs XP SP3 while at home I'm running XP 64-bit. I run the 32-bit version of Firefox on both.
As hard drives get larger, RAID becomes a lot more important. At the very least, you need a large backup drive, and USB 2.0 isn't enough to handle that much in a reasonable amount of time.
Likewise, as RAM increases, ECC becomes more important.
Currently, there are games that cannot run at even laptop resolutions at 30fps without multiple graphics cards. Microsoft Office 2010 uses 3D acceleration to draw the UI, so many laptops will become visibly slower. IE9 is supposed to use this same sort of 3D, so doing the "e-mail, web browsing, and word processing" that are all "most people" do with their computer will become a lot more tedious.
All of these can certainly be added to laptops, but battery technology isn't keeping up with the rest of the increases, so you have things like the current Core i7 only lasting an hour or so on battery. That's going to get worse before it gets better.
So, what's likely going to happen is that desktops will make a comeback as smaller laptops/netbooks and more powerful smart phones (and even the iPad) take over the portable computing market.
"And, to get that "live CD" behavior," just boot the CD image.
True, but you can customize an installed-to-HD distro to be exactly what you want, and keep it patched up (as long as you remember to temporarily disable the "revert to snapshot" while patching).
That's how I work from home instead of staying at the office all hours of the night.
If you can carry everything you need on your laptop, you're one of the lucky few. Every bit of data I need is available over a VPN. And, I don't need to lug any major hardware (just a RSA key) around to get to it.
For anybody who needs truly portable computing (e.g., at a client site, on the commute, etc.), some sort of laptop/netbook/whatever is probably the only choice (until smart phones really ramp up the computing power). For people who need access from a couple of fixed places, it's almost always easier (and sometimes cheaper) to have a desktop in each location.
So that I have choice! At work my laptop is plugged into external devices, including a 20" monitor. At the end of the work day, I just pick it up and take it home, and sit with it on the couch to surf the web.
First, I would never want to use a work laptop for my personal machine at home...any decent IT department will lock it down way too much.
Second, I would never want to work for a company that basically required me to bring my personal laptop to use to do work, nor would I trust a company that would just let somebody plug any computer into their network.
If you're not regularly using the laptop on the commute, at coffee shops, etc., you don't really need a laptop. I'd bet 99% of what most people want to move from home to work is data, so just carry a 500GB (or smaller, if you don't need as much) portable drive to and from work. Depending on the exact drive, it will weigh anywhere from less than an ounce (USB flash drive) to about 8 ounces, either of which is far less than a laptop.
It's also got 4TB of disk space, 6 powered USB ports (4 in back of which 3 are in use, 2 in front of which I use one), memory card reader, DVD burner, and a cable-TV video card so I can also use it as a DVR.
You do realize that laptops can have all those things, right?
Wake me up when you finish copying a few hundred gigabytes to a USB 2.0 hard drive. Also, once you plug 8 devices into that USB bus, it gets even worse. eSATA (which I have on my laptop) is a lifesaver, but you usually have only one port, while on a desktop, even two eSATA ports is a lot.
Seriously, if you want anything at all other than a basic computer (e.g., really fast processors, multiple processors, RAID, multiple video cards, serious amounts of RAM, etc.), a laptop won't cut it. Even if you can get those things on a laptop, they are a lot more expensive. If you use a laptop as a very rarely moved powerful desktop replacement, you're spending a lot more money than you need to...a good desktop and a cheap laptop will cost less and give you the same utility. Also, with current Intel and AMD desktop chips, overclocking is so easy and safe that the price/performance becomes even more in favor of the desktop.
Your home may not have wired Internet at all.
I suspect you meant "etherrnet" instead of "Internet", since how you connect to the Internet doesn't control how you communicate between devices in your home.
With that assumption, if you are the type that reads /., there is no excuse for your home not to have wired ethernet if you want wired ethernet.
I know we had a new developer start who's surname started with O', and while there was no problem getting him set up in the email system, his [name].o'[surname]@[address].com email address wouldn't work with a particular instance of Bugzilla we had to use (it just silently threw away the emails to him).
You got lucky in being able to incorrectly set up the e-mail address in your system.
The single quote is a delimiter in e-mail addresses, and needs to be correctly paired or escaped. Otherwise, the address is not valid according to the RFC.
I built a Shuttle SFF PC once, and the barebones kit started around $200-$400, and that's before dropping in a CPU and RAM.
You can start at less than $180, with CPU at around $150 (for Core 2 Duo or Quad), and 4GB of RAM for less than $75. A video card 5-10 times as fast as the GeForce 320M is around $50.
So, for around $450 you can buy faster hardware in about the same form factor as the Mac Mini, but with far more future expandability.
It's impossible to determine if it's a human pressing on a remote control or a computer hitting an IR blaster changing the channel.
I suspect that cable boxes and their remotes will eventually have something like the "rolling codes" system that garage doors and cars use.
Then, only companies with lots of money to license the technology would be able to build universal remotes that work with the cable box. And, if you design a workaround, you get a DMCA lawsuit. All so that the cable company can charge you $10/month extra for their crappy DVR.
So use a Linux VM instead. It's trivially easy nowadays.
And, to get that "live CD" behavior, configure the VM software to revert to the base snapshot each time the VM is rebooted.
Most US warships have vending machines and stores that sell anything from snacks to clothing to DVD players (depending on the size of the ship and its store), so yes, money is still used.
Everything in those vending machines can be considered "luxuries", though. It's not like the sailors have to pay for their food in the mess hall, or pay rent for their rack.
It doesn't matter.
You should easily be able to get 5% on a $5M investment. Assuming 50% taxes on the interest (way high for the US), and spending $300K/year, you have 22 years before the money runs out.
Setting the tax rate to a more realistic 30% and spending $250K/year gives you 33 years before the money runs out, while dropping to $200K/year spending is 55 years of money. For many, that's the rest of their lives.
The heart of the game is a big random number generator. There are many more possible outcomes than can be represented by the numbers/symbols on the reels. There is a mapping between the random numbers and what the payouts are, this sets the odds. After the next random number is determined, the machine decides what it wants to show on the reels. Usually it likes to show some kind of near miss to keep up player interest. There are other algorithm adjustments you can make to make a machine make frequent small payouts, or occasional big payouts.
Although this is the closest description in this thread of how slot machines work, it's still a bit off, but even Wikipedia gets it right.
To see all the gory details, and why you see the "near miss", check out The Wizard of Odds. The Q&A about slots (and other games) at that site is also a good resource.
Show me a documented example of a licensed casino owner keeping a machine on the floor after discovering that it was defective.
I suspect it happens all the time.
Suppose a machine was defective in such a way that it would never hit the biggest jackpot listed. Since that happens only once every few years, it may never be known. And, when such a machine is "inspected", it may not even be noticed if the error is subtle enough. Especially since it's in the casino's interest to not have the machine 100% functional. Then, any large jackpot payout can be avoided, even if the error isn't directly related to that jackpot. In my example, if the biggest jackpot is $X million progressive and can't be won due to an error in programming, but hit something like 5,000 times your bet, the casino could still void your win due to "malfunction".
An $11M jackpot is a prize where the winner moves to a new home and changes their name, giving up gambling forever - it provides little advertising benefit.
There are many people who win those kinds of large jackpots who later end up broke, with a good chunk of the losses coming from gambling.
The free market is as much an idealized unattainable as the Star Trek no money communism utopia.
Since pretty much every incarnation of Star Trek depicted the use of "money" (i.e., a token that has no intrinsic value of its own, but has been defined to have value by the government), I guess the free market is quite attainable.
What confuses people is that 95% of Star Trek episodes are set on the ship or in a diplomatic situation. On current US warships, sailors don't need money for very much other than gambling among themselves...meals are free, etc. Why should the starship Enterprise be any different?
In diplomatic situations, the host country (planet) doesn't make the guests pay for their food and entertainment at the state dinners, and often lets them use housing that is set aside for such occasions.
But, the crew paid for food and drinks in restaurants and bars, tribbles cost money to purchase, and bribes were paid for information.
I like the multiple tax bracket idea (with no taxes for the lowest bracket) and progressively stronger brackets (or possibly a smooth gradient?) better for that reason.
I suspect the only two real problems with the current system is that the brackets aren't adjusting with inflation, and there needs to be one or more brackets above the current "two-income professional family all the way up to Bill Gates" that is tops in the US.