You completely misrepresent the facts. MNG support was TEMPORARILY removed from Mozilla because it had been without a maintainer for a long span of time, was terribly buggy, and extremely bloated (300KB just for MNG support). The code was no longer viable. The project now has a new maintainer, and will be remerged when repair work has been completed.
For those that really care, the old code is still available for use in the form of an extension.
Intel Prescott, the chip that's slated to set power dissipation records for mainstream CPUs, uses a new.09 micron process. Apparently Intel is seeing fewer benefits than they expected.
Overclockers.com maintains a nice database of the relative performance of various air and watercooling systems on a variety of platforms: The Heatsink and Watercooling Roundup.
The Koolance EXOS system does this already. The water flows to an external radiator that has built in fans and other such lovely things. This could also be done with heat pipes. The problem is that most users don't want a heavy, bulky block sitting next to their computer. Additionally, you have hoses to worry about.
The 103W figure for the Prescott 3.6Ghz is actually the Thermal Design Power. This is the amount of power the processor is expected to use during "normal" operation. A P4-C 3.0Ghz with HyperThreading has a TDP of about 80W, with an actual maximum power usage of 104W. Assuming a similar scale, a Prescott 3.6Ghz can be expected to dissipate around 130W. It's this maximum figure that really matters, since I don't think most people want their processor to throttle during gaming or whenever they are driving their CPU hard.
A strong, healthy emulator that allows Linux to run as many Windows apps as possible can only be in the best interest of the Linux community. If casual users know that the applications they use will run under emulation in Linux, they will be much more likely to make the switch. Thus, good, easily accessible emulators mean MORE USERS. This has to be the all-consuming goal of everyone who seeks to promote the Linux operating system, as well as the ideology that goes along with it. If we have the users, the applications will come. If we have the users, the manufacturer-supported drivers will come. If we have the users, MORE USERS will come.
When we have a large enough installed base of Linux users, three things will happen. Firstly, the market will be large enough that game companies can profitably support native development. Secondly, game companies will see that their games are going to be run under Linux ANYWAY, and it will be cheaper from a support aspect for them to just skip the emulation and do a native port. Lastly, we'll get widespread official driver support from manufacturers, which will make Linux much more robust on a wide variety of platforms.
What's the point of this long, rambling comment? Having a larger Linux community is a circle that starts with users. With users come official support, applications, and drivers, and easier implementation, and with these come more users. By discouraging quality emulation, you discourage users, thus compromising the future of Linux. Users are leverage, and trying to leverage game companies to produce native applications when we don't HAVE any leverage isn't going to work. If we can develop the market to the point where game companies want to develop for Linux, excellent. Trying to force them to develop for Linux is just stupid.
No, it's not the same type of propulsion. Furthermore, it doesn't use anything NEAR the same principles. That's like saying that walking and driving a car are the same types of propulsion, since they both produce energy from combustion reactions and use this energy to move you along the ground. Thanks for playing, though.
Also, I apologize for missing the memo that said that nuclear-bomb based propulsion actually existed. Oh wait, that's because THEY DON'T EXIST. Just because it's POSSIBLE does not make it CURRENTLY AVAILABLE. It will cease to be "scifi lore" when someone actually bothers to make one.
The story is sparse on details, but it seems that it works to help prevent the onset of AIDS by allowing an individual to become partially immune to the immune-system-damaging effects of the HIV virus. Partially, since they still die eventually, but it helps. If, instead of a drug that acts on the virus itself, it stimulates an immune response, it would be classified as a vaccine, methinks.
Quite frankly, the market for total newbie classes is rather frustrated. Most people who would want to learn the very basics have already done so. You have to offer unique choices if you want to appeal to a broad audience. How about "How not to get screwed online"? Teach people about spyware, trojans, the-nice-nigerian-man-really-doesn't-want-to-make- you-rich, etc. Things that have practical value to the average newbie who ALREADY has a working knowledge of AOL, windows, and office. The Hacker Ethics class and other ideas are good, and you should implement some to appeal to a wider audience beyond just newbies, which is going to be important.
Just remember, you have to get as many people interested and coming to class as possible. Trying to be a LAN party store is not going to work, as similar businesses in my home town discovered, for the simple reason that the kind of person that appeals to ALREADY has a capable computer and has no compunctions about bringing it to his friend's house and hooking it up to a network so he can play Counter-Strike. You're starting an uphill battle.
And nobody ever left this world regretting that they hadn't spent more time playing video games.
Are you sure about that? Just what DO people on their death beds wish they spent more time doing? Methinks it's what they enjoyed, be that helping people, raising their children, having sex, or playing videogames.
Any board that has to rely on having a tube amp onboard as a gimmick can be trusted to provide, at best, mediocre audio quality, as the article attests. To address your main point, it's just not worth it to integrate high quality sound chipsets into motherboards. The number of people that care enough to spend the extra $50 for a motherboard with great onboard audio as opposed to merely good audio is rather tiny.
Really, there isn't just one universal onboard audio chipset that you can compare. The nVidia Soundstorm APU found on some nForce2 motherboards provides EXCELLENT quality. The Realtek ALC650 chip used as an APU in others (most Soundstorm boards will use the ALC650 as a DAC) provides crap. The Via Envy24HT APU is so good that even the best discrete consumer audio solutions (the M-Audio Revolution 7.1, for example) use it. The older Via "AC'97"-labelled APU was utter crap. The C-Media controller used on many motherboards is decent. As the article describes, the solutions that are included with modern motherboards ARE beginning to rival the quality of discrete soundcards, even good ones.
While I will most certainly agree that tossing some random naked chicks into a movie can't make up for a bad movie overall, it's not like said random naked chick can actually turn a good movie into a bad movie (unless, I suppose, it was a kids movie). If they're not replacing good content with nudity, then who cares? When it's a question of breasts vs. another ten seconds of a car chase, I'll take the breasts, thank you very much.
A "melt-down" of an Athlon, in a worst case scenario, would result in a small amount of smoke. There would be no fire and no sparks. If your power-supply blows AND starts shooting flames far enough to hit your case side panel, I think that any flammable material inside the computer is the least of your concerns.
It's a computer. If any part of it gets hot enough that YOU wouldn't want to touch it, then you need to fix the cooling. It's not like your box is filled with wires and chips that are glowing red-hot.
Since quite often fans are pushed right up next to stamped metal fan guards, it can be QUITE significant. Fans also make very annoying noises when their airflow is restricted in either direction, especially on the intake side.
The main problem is poor design. Most cases today still use 80mm fans, which you need a lot of to get adequate cooling. This means lots of noise. A well designed case using 120mm fans can push a LOT of air at an extremely low noise level, and using large, well designed heatsinks with at least 80mm fans on processors can do the same thing. With intelligent fan placement and slot-exhaust videocard cooling (see the nVidia Geforce FX for a good idea poorly executed) you can keep the case temp down near room temp, dramatically reducing the need for noisy fans.
Freezers are made to keep cold things cold, or to bring things down from room temperature to freezing. They CANNOT deal with a continuous heat load in the 200W range, at least not for long (burned out compressor, anyone?).
That's for the Pentium Pro. I'm alsmost certain the same functionality existed all the way back to the 486, although I'll admin my recollection is a bit rusty, so I may be wrong.
The Pentium Pro was not a desktop processor. It was a server processor, analogous to Xeon. And, again, while the P2/P3 supported temperature monitoring, they DID NOT have any kind of internal throttling or shutdown. When they got too hot, they crashed because stuff stopped working at that temperature. Do it enough, or let them run really hot for awhile, and they died. Pentium and 486 had no thermal monitoring capabilities whatsoever.
I somehow suspect such a video would be displaying ideal conditions, both in terms of the proximity of the thermistor to the CPU and the reaction speed of the motherboard's monitoring software.
All AMD processors since the original Palomino core have had an onboard thermal diode. It's up to the motherboard to shut them down at a given temperature, but the temperature being worked off of IS the core temperature, on boards that support such overheat protection.
Again, I've *seen* this sort of spontaneous failure (well, more accurately, heard the snap from inside the case).
Such failures are rather unlikely. If you don't drop the system, the heatsink isn't going to come off. If it comes off in transit due to rough handling, you'd probably know it by the large object rattling around in the case. Regardless, overheat protection IS featured on current AMD processors and motherboards, so it really is a moot point.
Now, this contact is far from thermally ideal and probably wouldn't be good enough for extended use, but it is enough to transfer some heat away from the CPU. If there's some thermal compound there it's even better.
Spec mounting pressure for a CPU is usually around 30lbs of force. Gravity is exerting what, 0.5lbs on your average heatsink? There is most assuredly not enough contact to make a heatsink just sitting on top of a CPU noticeably better than no heatsink at all.
And there's plenty more where an XP 2800+ ties, beats, or dominates the P4 2.8Ghz. Athlon XPs generally come out ahead or equal in gaming, completely dominate in general usage and office applications, COMPLETELY own in hard integer and floating point math, and lose quite badly in hard 3D rendering (to distinguish from gaming) and media encoding. Basically, if it doesn't use SSE2, the Athlon XP wins.
All intel CPUs have had the ability to automatically shut themselves down in the case of excessive heat since at least the 486.
The first Intel desktop processor to feature an on-chip thermal diode was the Pentium II. The P4 was the first processor to throttle or shutdown itself based on CPU temperature. Earlier processors USUALLY didn't fry from overheating, simply because they didn't produce that much heat and it was thus difficult for them to get hot enough to die, but it DID happen. If you remove the heatsink from a Coppermine P3 while it's running at maximum load, it will FAIL, not shutdown. I think it's quite likely that it would be permanently damaged in the process, even if it doesn't burn up Tbird-style.
Most motherboard sensors can't react quickly enough to effectively throttle an Athlon.
They don't bother throttling, they just cut power if the CPU hits 70C (or another user configured temperature). They can, in fact, react quickly enough to prevent a CPU in which the heatsink is removed during maximum load operation from frying. There was a video released by AMD in answer to the THG video which showed this in action.
The HSF being knocked loose during shipping and the owner turning the machine on without realising.
In such a case, the motherboard is guaranteed to be destroyed already, and probably also the cards the heatsink hit on the way down. Realistically, the heatsink just doesn't come off without catastrophic impact or force to the system, something that just plain doesn't occur very often.
However, there's no reason a HSF wouldn't work to some extent without being clipped on.
A heatsink that is not affixed to the processor cannot cool the processor, because it is not in effective contact with it.
Slashdot Mirror
You completely misrepresent the facts. MNG support was TEMPORARILY removed from Mozilla because it had been without a maintainer for a long span of time, was terribly buggy, and extremely bloated (300KB just for MNG support). The code was no longer viable. The project now has a new maintainer, and will be remerged when repair work has been completed.
For those that really care, the old code is still available for use in the form of an extension.
Intel Prescott, the chip that's slated to set power dissipation records for mainstream CPUs, uses a new .09 micron process. Apparently Intel is seeing fewer benefits than they expected.
Overclockers.com maintains a nice database of the relative performance of various air and watercooling systems on a variety of platforms: The Heatsink and Watercooling Roundup.
The Koolance EXOS system does this already. The water flows to an external radiator that has built in fans and other such lovely things. This could also be done with heat pipes. The problem is that most users don't want a heavy, bulky block sitting next to their computer. Additionally, you have hoses to worry about.
The 103W figure for the Prescott 3.6Ghz is actually the Thermal Design Power. This is the amount of power the processor is expected to use during "normal" operation. A P4-C 3.0Ghz with HyperThreading has a TDP of about 80W, with an actual maximum power usage of 104W. Assuming a similar scale, a Prescott 3.6Ghz can be expected to dissipate around 130W. It's this maximum figure that really matters, since I don't think most people want their processor to throttle during gaming or whenever they are driving their CPU hard.
A strong, healthy emulator that allows Linux to run as many Windows apps as possible can only be in the best interest of the Linux community. If casual users know that the applications they use will run under emulation in Linux, they will be much more likely to make the switch. Thus, good, easily accessible emulators mean MORE USERS. This has to be the all-consuming goal of everyone who seeks to promote the Linux operating system, as well as the ideology that goes along with it. If we have the users, the applications will come. If we have the users, the manufacturer-supported drivers will come. If we have the users, MORE USERS will come.
When we have a large enough installed base of Linux users, three things will happen. Firstly, the market will be large enough that game companies can profitably support native development. Secondly, game companies will see that their games are going to be run under Linux ANYWAY, and it will be cheaper from a support aspect for them to just skip the emulation and do a native port. Lastly, we'll get widespread official driver support from manufacturers, which will make Linux much more robust on a wide variety of platforms.
What's the point of this long, rambling comment? Having a larger Linux community is a circle that starts with users. With users come official support, applications, and drivers, and easier implementation, and with these come more users. By discouraging quality emulation, you discourage users, thus compromising the future of Linux. Users are leverage, and trying to leverage game companies to produce native applications when we don't HAVE any leverage isn't going to work. If we can develop the market to the point where game companies want to develop for Linux, excellent. Trying to force them to develop for Linux is just stupid.
No, it's not the same type of propulsion. Furthermore, it doesn't use anything NEAR the same principles. That's like saying that walking and driving a car are the same types of propulsion, since they both produce energy from combustion reactions and use this energy to move you along the ground. Thanks for playing, though.
Also, I apologize for missing the memo that said that nuclear-bomb based propulsion actually existed. Oh wait, that's because THEY DON'T EXIST. Just because it's POSSIBLE does not make it CURRENTLY AVAILABLE. It will cease to be "scifi lore" when someone actually bothers to make one.
The article discusses using thrust pulses from combustible propellant, not the nuclear explosions of scifi lore.
The story is sparse on details, but it seems that it works to help prevent the onset of AIDS by allowing an individual to become partially immune to the immune-system-damaging effects of the HIV virus. Partially, since they still die eventually, but it helps. If, instead of a drug that acts on the virus itself, it stimulates an immune response, it would be classified as a vaccine, methinks.
Actually, I thought Warthog's article was some of the best writing I've seen in a long time.
Quite frankly, the market for total newbie classes is rather frustrated. Most people who would want to learn the very basics have already done so. You have to offer unique choices if you want to appeal to a broad audience. How about "How not to get screwed online"? Teach people about spyware, trojans, the-nice-nigerian-man-really-doesn't-want-to-make- you-rich, etc. Things that have practical value to the average newbie who ALREADY has a working knowledge of AOL, windows, and office. The Hacker Ethics class and other ideas are good, and you should implement some to appeal to a wider audience beyond just newbies, which is going to be important.
Just remember, you have to get as many people interested and coming to class as possible. Trying to be a LAN party store is not going to work, as similar businesses in my home town discovered, for the simple reason that the kind of person that appeals to ALREADY has a capable computer and has no compunctions about bringing it to his friend's house and hooking it up to a network so he can play Counter-Strike. You're starting an uphill battle.
And nobody ever left this world regretting that they hadn't spent more time playing video games.
Are you sure about that? Just what DO people on their death beds wish they spent more time doing? Methinks it's what they enjoyed, be that helping people, raising their children, having sex, or playing videogames.
Any board that has to rely on having a tube amp onboard as a gimmick can be trusted to provide, at best, mediocre audio quality, as the article attests. To address your main point, it's just not worth it to integrate high quality sound chipsets into motherboards. The number of people that care enough to spend the extra $50 for a motherboard with great onboard audio as opposed to merely good audio is rather tiny.
Really, there isn't just one universal onboard audio chipset that you can compare. The nVidia Soundstorm APU found on some nForce2 motherboards provides EXCELLENT quality. The Realtek ALC650 chip used as an APU in others (most Soundstorm boards will use the ALC650 as a DAC) provides crap. The Via Envy24HT APU is so good that even the best discrete consumer audio solutions (the M-Audio Revolution 7.1, for example) use it. The older Via "AC'97"-labelled APU was utter crap. The C-Media controller used on many motherboards is decent. As the article describes, the solutions that are included with modern motherboards ARE beginning to rival the quality of discrete soundcards, even good ones.
That said, it was incredibly funny. Good show.
While I will most certainly agree that tossing some random naked chicks into a movie can't make up for a bad movie overall, it's not like said random naked chick can actually turn a good movie into a bad movie (unless, I suppose, it was a kids movie). If they're not replacing good content with nudity, then who cares? When it's a question of breasts vs. another ten seconds of a car chase, I'll take the breasts, thank you very much.
Now let's play "guess my age and gender!"
A "melt-down" of an Athlon, in a worst case scenario, would result in a small amount of smoke. There would be no fire and no sparks. If your power-supply blows AND starts shooting flames far enough to hit your case side panel, I think that any flammable material inside the computer is the least of your concerns.
It's a computer. If any part of it gets hot enough that YOU wouldn't want to touch it, then you need to fix the cooling. It's not like your box is filled with wires and chips that are glowing red-hot.
Since quite often fans are pushed right up next to stamped metal fan guards, it can be QUITE significant. Fans also make very annoying noises when their airflow is restricted in either direction, especially on the intake side.
The main problem is poor design. Most cases today still use 80mm fans, which you need a lot of to get adequate cooling. This means lots of noise. A well designed case using 120mm fans can push a LOT of air at an extremely low noise level, and using large, well designed heatsinks with at least 80mm fans on processors can do the same thing. With intelligent fan placement and slot-exhaust videocard cooling (see the nVidia Geforce FX for a good idea poorly executed) you can keep the case temp down near room temp, dramatically reducing the need for noisy fans.
Freezers are made to keep cold things cold, or to bring things down from room temperature to freezing. They CANNOT deal with a continuous heat load in the 200W range, at least not for long (burned out compressor, anyone?).
Nice trolling.
Athlon XP 2400+ beats P4-C 3.0Ghz at General Usage test
Athlon XP beats P4-C at same speed grade at Unreal Tournament 2003
Athlon XP 1700+ beats P4 3.0Ghz at RC5-72 Encryption
In the interest of fairness, Athlon XP gets beatdown by P4 at Lightwave 3D Rendering
As I said, Athlon XPs are generally better at gaming, general usage, and hard math. P4s are generally better at hard 3D rendering and media encoding.
That's for the Pentium Pro. I'm alsmost certain the same functionality existed all the way back to the 486, although I'll admin my recollection is a bit rusty, so I may be wrong.
The Pentium Pro was not a desktop processor. It was a server processor, analogous to Xeon. And, again, while the P2/P3 supported temperature monitoring, they DID NOT have any kind of internal throttling or shutdown. When they got too hot, they crashed because stuff stopped working at that temperature. Do it enough, or let them run really hot for awhile, and they died. Pentium and 486 had no thermal monitoring capabilities whatsoever.
I somehow suspect such a video would be displaying ideal conditions, both in terms of the proximity of the thermistor to the CPU and the reaction speed of the motherboard's monitoring software.
All AMD processors since the original Palomino core have had an onboard thermal diode. It's up to the motherboard to shut them down at a given temperature, but the temperature being worked off of IS the core temperature, on boards that support such overheat protection.
Again, I've *seen* this sort of spontaneous failure (well, more accurately, heard the snap from inside the case).
Such failures are rather unlikely. If you don't drop the system, the heatsink isn't going to come off. If it comes off in transit due to rough handling, you'd probably know it by the large object rattling around in the case. Regardless, overheat protection IS featured on current AMD processors and motherboards, so it really is a moot point.
Now, this contact is far from thermally ideal and probably wouldn't be good enough for extended use, but it is enough to transfer some heat away from the CPU. If there's some thermal compound there it's even better.
Spec mounting pressure for a CPU is usually around 30lbs of force. Gravity is exerting what, 0.5lbs on your average heatsink? There is most assuredly not enough contact to make a heatsink just sitting on top of a CPU noticeably better than no heatsink at all.
And there's plenty more where an XP 2800+ ties, beats, or dominates the P4 2.8Ghz. Athlon XPs generally come out ahead or equal in gaming, completely dominate in general usage and office applications, COMPLETELY own in hard integer and floating point math, and lose quite badly in hard 3D rendering (to distinguish from gaming) and media encoding. Basically, if it doesn't use SSE2, the Athlon XP wins.
All intel CPUs have had the ability to automatically shut themselves down in the case of excessive heat since at least the 486.
The first Intel desktop processor to feature an on-chip thermal diode was the Pentium II. The P4 was the first processor to throttle or shutdown itself based on CPU temperature. Earlier processors USUALLY didn't fry from overheating, simply because they didn't produce that much heat and it was thus difficult for them to get hot enough to die, but it DID happen. If you remove the heatsink from a Coppermine P3 while it's running at maximum load, it will FAIL, not shutdown. I think it's quite likely that it would be permanently damaged in the process, even if it doesn't burn up Tbird-style.
Most motherboard sensors can't react quickly enough to effectively throttle an Athlon.
They don't bother throttling, they just cut power if the CPU hits 70C (or another user configured temperature). They can, in fact, react quickly enough to prevent a CPU in which the heatsink is removed during maximum load operation from frying. There was a video released by AMD in answer to the THG video which showed this in action.
The HSF being knocked loose during shipping and the owner turning the machine on without realising.
In such a case, the motherboard is guaranteed to be destroyed already, and probably also the cards the heatsink hit on the way down. Realistically, the heatsink just doesn't come off without catastrophic impact or force to the system, something that just plain doesn't occur very often.
However, there's no reason a HSF wouldn't work to some extent without being clipped on.
A heatsink that is not affixed to the processor cannot cool the processor, because it is not in effective contact with it.