With an "open" NAT (as required for xbox live), you can force UDP packets through. An open NAT doesn't check that UDP replies come from the same IP you sent the original outgoing packet to. So "A" sends an outgoing packet to a 3rd party, which tells "B" the port to send to to "reply". "A" can then reply directly to "B", and you have NAT-bypassing communication going on.
Is there a free/cheap virus scanner that won't refuse to work on Windows Server 2003? AVG and Avast antivirus both won't, without an expensive license. Comodo antivirus doesn't list Windows Server 2003 in it's supported OSs list.
I have the student (MSDNAA) edition of Windows Server 2003, and I would like to be able to put SOME virus protection on it.
It's been proven that people rely on technology to extend their memory (in fact use technology to do pretty much anything as a natural extension of their body). Instead of remembering everyones' phone numbers, you store them all in a device for the purpose and remember how to retrieve them. A persons' name gets cross-referenced in your mind with the information in the phonebook, and it all still works fine.
Most people don't even realise. To them it's perfectly natural that their phonebook is their memory of peoples' phone numbers. From their (subconcious) point of view, it improves their memory because they never forget the phone number.
On the other hand using their memory less means it gets weaker.
If they often have to actually dial the number themself (instead of just selecting "call" on a phone with a phonebook built in), then they will start to remember it (their memory will essentially cache the phone number). For example, I buy stuff online so often that I have my debit card number, expiry and code memorised.:(
Essentially in the modern age peoples' memories are relegated to remembering anything that they haven't yet written down, and acting as a cache for information they often look up.
If you had a device that identified everything for you, you'd start to rely on the device for the names of things, and people, even those that you'd normally remember. You'd remember the names of people close to you, but not the name of someone that you met for the first time yesterday, or even this morning. Not that everyone can manage this already.
So, basically, 70% of the heat from the chip heats up the air in the piston and increases the pressure, and 30% of it goes into the heatpipes, then the radiator to be dispersed into the air by the fan moved by the other 70% (of which most will probably just radiate away anyway, and only about 10% of the power will actually get to move the fan)?
What we really need is some information about how hot different power chips get when cooled by this thing. e.g. degrees C / Watt, and how this compares to a purely passive cooler.
That includes a game like Morrowind, which even ran poorly on a discrete Geforce 5600 (which could only manage to run it at 800x600 with FPS a lowly 15-30 range). To be fair, everything ran poorly on the nvidia 5000 series, due to it claiming shader support, despite being very bad at them. "Homeworld 2" particularly sticks in my mind for this, it actually ran worse (unplayable) on the 5600 than on the geforce 2 I was trying to replace.
If a game cannot achieve a minimum of 60fps at 1024x768 then it is thoroughly unplayable unless you like blocky slideshows. Tell that to console gamers, most console games are played on 640x480 60Hz interlaced, with the framerate capped at 30 fps. 45 fps is considered the minimum for twitch games, eg action shooters like UT. 30 fps is the desired framerate for most other games. 10 fps is acceptable in real-time strategy games.
wtf? I'm sure I closed that link after "Half-Empty". Like this:
even the stingiest of bosses won't sacrifice $100k+ in hardware to shave a few bucks off the electric bill. Really? "Half-Empty" Ok, so that was risking the expensive hardware by shaving money off of the maintenance bill, which admittedly is a bit larger than the electricity bill.
That's actually an interesting question -- how many of us here WOULD vote for Ron Paul?** I'm guessing no more than about half, since there are a fair number of left-wing liberals in the slashdot mix.
**I did, but I'm just one in, uh, a million and change... 17%
In others, they simply downsampled sounds and textures; the results were usually unnoticeable. All but a few games ended up online that way. Or downsampled video. I also heard that in one case the crackers patched support for a better video codec in to the game, allowing them to recode the video to a size that would fit on 700MB cds without compromising the quality (and also allowing it to play without skipping, as a CD has a slower read-rate than a GD in the Dreamcast).
After skill points, or in some cases before, it's all about your skills. Take someone who doesn't know how to put together a good ship, or how to fly one well, and then your 2m SP char is ruining the 10m SP char's day. Bought characters (or those with bought currency) are easy to spot because they're sitting in a nice shiny Navy-Issue Battleship, but easily lose their first fight, and so lose the battleship.
It's probably for the same reason we talk about thousands of kilograms instead of "just" saying "gigagrams". The term "MIPS" is not really an abbreviation anymore, it became a proper word describing a performance unit everyone in the industry is used to. Actually "thousands of kilograms" would be "megagrams", but we generally call them "tonnes".
five different models and feature 64-way chips Is that like 54 cores on a chip? Wowzers, we're gonna need a lot of salsa for this party. No, it means they have the data links for communicating with 63 other chips (i.e. part of a mainframe with 64 cpus)
... average uptime of a windows box is 11days... If that's true it's only because most Windows-running machines are desktops and so are shut down every night.
I have two Windows PCs, one runs Server 2003 Student Edition and its uptime is always the time since the last Microsoft patch I applied (generally gets to 1-3 months), and the other is running XP x64, is my desktop and so is shut down every night. The former has only ever had problems due to its dodgy pci bus (now worked around, only affected certain combinations of pci devices and pci sockets), and the only problems with the latter were some recent blue-screens due to bad ram. The ram has since degraded to the point that if I put it in my PC it doesn't get as far as turning the monitor on (let alone starting Windows) before it crashes.
I haven't had problems with Windows itself myself for 5 years or so.
I have had to perform repairs on other peoples' Windows PCs, and 90% of the time the only complaint is that it's slow e.g. 512MB of ram and 500MB of stuff launching at startup, so the machine is permanently paging, or a PC with a 1MB integrated graphics chip trying to play 3D games and only PCI sockets for upgrades (it's really hard to find a PCI gfx card nowadays). The other 10% is mostly a pc infested with viruses making the machine unstable (often with an expired Norton trial sitting helpless).
One person with the latter problem shared their PC with their younger brother, who had managed to contract so many viruses on it (despite an active and up-to-date virus scanner) that it had to be wiped. That wasn't a problem with Windows, IT didn't let the viruses in, HE did.
Games almost always use assembler for vector and matrix math functions, because speed is critical and explicitly using SSE is more reliable than hoping that your compiler can figure out how to turn a long list of multiplies and adds into the appropriate SSE instructions.
However the most common use of lots of assembler is compilers. Not just traditional source->executable compilers, but JIT recompilers are in every emulator that wants a sensible amount of speed.
I'm pretty sure everything you need to develop for Windows has been free for a LONG time (the SDK comes with a command-line compiler IIRC, MSDN is available online and there's windbg for debugging), so it's only the IDE they're giving free (and the express version of the IDE has been free since v2005).
Slashdot Mirror
You could have the dedicated servers provide 100kB/sec to each user, with the speed raised by people seeding back.
In the case of watching tv, you could have people seed while they're watching it (after downloading it).
Though the real solution to mass-distribution would use multicast to dramatically lower the amount of data transmitted.
Are you referring to "Think of a number, any number."?
For which "Er, five" is the wrong answer?
Except that Douglas Adams has said "I don't write jokes in base 13".
It's base 10, and intended to be wrong, to allow the punchline of "I always knew there was something fundamentally wrong about the universe".
Oh definitely "Aye"!
:P
I need some material for a d&d campaign, perhaps it'll give me inspiration
With an "open" NAT (as required for xbox live), you can force UDP packets through.
An open NAT doesn't check that UDP replies come from the same IP you sent the original outgoing packet to. So "A" sends an outgoing packet to a 3rd party, which tells "B" the port to send to to "reply". "A" can then reply directly to "B", and you have NAT-bypassing communication going on.
It's a horrible hack though.
Which seems to be how skype does it too. Unless they've changed it and I didn't notice.
Scratch that, found this forum post asking the same question: http://forums.microsoft.com/TechNet/ShowPost.aspx?PostID=938970&SiteID=17
Which says that ClamAV and BitDefender are both free for Server 2003.
Is there a free/cheap virus scanner that won't refuse to work on Windows Server 2003?
AVG and Avast antivirus both won't, without an expensive license. Comodo antivirus doesn't list Windows Server 2003 in it's supported OSs list.
I have the student (MSDNAA) edition of Windows Server 2003, and I would like to be able to put SOME virus protection on it.
It's been proven that people rely on technology to extend their memory (in fact use technology to do pretty much anything as a natural extension of their body). Instead of remembering everyones' phone numbers, you store them all in a device for the purpose and remember how to retrieve them. A persons' name gets cross-referenced in your mind with the information in the phonebook, and it all still works fine.
:(
Most people don't even realise. To them it's perfectly natural that their phonebook is their memory of peoples' phone numbers. From their (subconcious) point of view, it improves their memory because they never forget the phone number.
On the other hand using their memory less means it gets weaker.
If they often have to actually dial the number themself (instead of just selecting "call" on a phone with a phonebook built in), then they will start to remember it (their memory will essentially cache the phone number). For example, I buy stuff online so often that I have my debit card number, expiry and code memorised.
Essentially in the modern age peoples' memories are relegated to remembering anything that they haven't yet written down, and acting as a cache for information they often look up.
If you had a device that identified everything for you, you'd start to rely on the device for the names of things, and people, even those that you'd normally remember. You'd remember the names of people close to you, but not the name of someone that you met for the first time yesterday, or even this morning. Not that everyone can manage this already.
So, basically, 70% of the heat from the chip heats up the air in the piston and increases the pressure, and 30% of it goes into the heatpipes, then the radiator to be dispersed into the air by the fan moved by the other 70% (of which most will probably just radiate away anyway, and only about 10% of the power will actually get to move the fan)?
What we really need is some information about how hot different power chips get when cooled by this thing. e.g. degrees C / Watt, and how this compares to a purely passive cooler.
What does WoL matter when the machine is already being powered up?
Or you could set the temp shutoff.
Just saying.
"Homeworld 2" particularly sticks in my mind for this, it actually ran worse (unplayable) on the 5600 than on the geforce 2 I was trying to replace. If a game cannot achieve a minimum of 60fps at 1024x768 then it is thoroughly unplayable unless you like blocky slideshows. Tell that to console gamers, most console games are played on 640x480 60Hz interlaced, with the framerate capped at 30 fps.
45 fps is considered the minimum for twitch games, eg action shooters like UT.
30 fps is the desired framerate for most other games.
10 fps is acceptable in real-time strategy games.
"Stupider" is a perfectly valid word, referring to: Someone/something being "more stupid" than a given reference.
See Dictionary.
the electric bill. Really?
"Half-Empty"
Ok, so that was risking the expensive hardware by shaving money off of the maintenance bill, which admittedly is a bit larger than the electricity bill.
the electric bill. Really?
"Half-Empty" Ok, so that was risking the expensive hardware by shaving money off of the maintenance bill, which admittedly is a bit larger than the electricity bill.
**I did, but I'm just one in, uh, a million and change... 17%
I also heard that in one case the crackers patched support for a better video codec in to the game, allowing them to recode the video to a size that would fit on 700MB cds without compromising the quality (and also allowing it to play without skipping, as a CD has a slower read-rate than a GD in the Dreamcast).
Traditional security-hole exploiting viruses are rare now.
Instead a virus scanner protects users from their own stupidity.
You can't buy skill at playing EVE.
... average uptime of a windows box is 11daysI have two Windows PCs, one runs Server 2003 Student Edition and its uptime is always the time since the last Microsoft patch I applied (generally gets to 1-3 months), and the other is running XP x64, is my desktop and so is shut down every night. The former has only ever had problems due to its dodgy pci bus (now worked around, only affected certain combinations of pci devices and pci sockets), and the only problems with the latter were some recent blue-screens due to bad ram. The ram has since degraded to the point that if I put it in my PC it doesn't get as far as turning the monitor on (let alone starting Windows) before it crashes.
I haven't had problems with Windows itself myself for 5 years or so.
I have had to perform repairs on other peoples' Windows PCs, and 90% of the time the only complaint is that it's slow e.g. 512MB of ram and 500MB of stuff launching at startup, so the machine is permanently paging, or a PC with a 1MB integrated graphics chip trying to play 3D games and only PCI sockets for upgrades (it's really hard to find a PCI gfx card nowadays). The other 10% is mostly a pc infested with viruses making the machine unstable (often with an expired Norton trial sitting helpless).
One person with the latter problem shared their PC with their younger brother, who had managed to contract so many viruses on it (despite an active and up-to-date virus scanner) that it had to be wiped. That wasn't a problem with Windows, IT didn't let the viruses in, HE did.
Games almost always use assembler for vector and matrix math functions, because speed is critical and explicitly using SSE is more reliable than hoping that your compiler can figure out how to turn a long list of multiplies and adds into the appropriate SSE instructions.
However the most common use of lots of assembler is compilers. Not just traditional source->executable compilers, but JIT recompilers are in every emulator that wants a sensible amount of speed.
I'm pretty sure everything you need to develop for Windows has been free for a LONG time (the SDK comes with a command-line compiler IIRC, MSDN is available online and there's windbg for debugging), so it's only the IDE they're giving free (and the express version of the IDE has been free since v2005).
And the IDE is the best I've used TBH.