"If you can't deal with this, don't shop on ebay."
Well put. This is exactly why I would never buy anything off ebay. My experience is "if it's too good to be true, it probably is."
At least when I buy something in a store and it doesn't go my way I can confront the store owner directly [usually get exchange/refund at that point;-)].
Not only is there no legal basis for this argument as another poster pointed out. But....
This argument is stupid. Microsoft sells Windows it doesn't give it away. The fact that you run a pirated copy is actually contrary to what Microsoft wanted you to do in the first place.
This argument could apply to Ford for making cars that can be stolen and used to run over school children [cuz lord knows the other type of children are not worth protecting].
Anyways, lay off the crack for a week and go get a job.
cooperation from vendors doesn't mean what you think it means. Incomplete [late] databooks are not that rare for most firms.
Just because MSFT is huge doesn't mean they can produce perfect work. Several things contribute to the ultimate non-perfectness
1. Moving target. Even while they are writing a version of Windows new hardware is being developed. They have to be able to accomodate late submissions.
2. Not all MSFT employees are developers and not all MSFT developers work on windows.
3. Diminishing returns. Adding more people produces smaller returns on investment.
These characteristics are not unique of MSFT though. The same could be said of say KDE. Not all KDE members are developers, piling on 1000s of developers won't make it 1000x better and KDE doesn't target features from 5 years ago.
lo-and-behold KDE has bugs in it. Shocking!!! The horror!!! OMG!!!
This is why the original posters type of tripe pisses me off. It's so fucking narrow minded.
That's right up their with pointing out the series of bugs in A-patchy webserver, or the various permission escalations in the lenux Kurnul.
Look, I'm l33t, I point out the flaws of one company to make myself look l33t. When those flaws are actually shared by basically all other software firms out there.
Now I'm not an MS fanboy [run Gentoo] but that doesn't mean I can stand idiots like you. Let's see you try and write an OS that can even *half* compete with Windows and not have any bugs. Then you can sit here and be all mighty about what a company should or should not do.
"It happens with ALL programs, including Internet Explorer. It's not a problem with one program locking up the GUI, it's a problem with the base system locking up ALL applications momentarily."
No, actually it isn't. How do you think things like dnetc can run 100% cpu yet still display nice graphs and such?
The OS is capable of handling threads just fine. The problem is [including IE] that control logic is placed in the GUI thread.
Basically the way GUIs work in Win32 is you have a callback that handles all messages [e.g. resize, repaint, controls clicked/moved, etc...].
Normally if responding to a message can block you spawn a thread to handle it and return to the callback asap.
Try this demo out [you'll have to write it...]
Make a simple dialog with one button on it "Do It" and two text labels.
Now write two applications both of which have a 100ms timer displayed in one of the text labels.
1st program. When the button is clicked call a function and multiply some variables a million times. Time the process and display the time on the dialog.
2nd program. When the button is clicked spawn a lower priority thread todo the same million multiplications. Time the process and display the time on the dialog.
I'm positive that if you don't have anything else running the 2nd program will display the 100ms timer and still compute the multiplications in roughly the same time as the first program. The first program will lock up the 100ms timer and finish the multiplications faster.
The fact that people [including MSFT] write bad code isn't surprising. Things like the control panel and IE basically do everything in the same thread. That's why loading device manager or any laggy FTP site causes the program to effectively halt.
But if you really don't believe me just write the two applications and test the theory.
"GCC is an inferior compiler for the x86, whether you like it or not. Intel's optimizing C/C++ compiler is much faster according to numerous benchmarks"
This simply is not true. Like all compiler [or benchmarks] good at one thing is not good at all things.
I found that even with profiling the Intel compiler [v8] is better at things like MP math [libtommath] but the same and worse at other crunching [ciphers/hashes in libtomcrypt].
In fact the Intel C compiler doesn't make much use of SSE2 properly at all [hint: SSE2 is like MMX but on the 128-bit XMM registers] which can be used to speed up 32-bit ALU work [more registers to play with]. The most I've seen is it uses SSE to clear temps in memory.
Not saying that the compiler isn't good. It's fast [to compile] and all. Just saying that GCC isn't that "lost" compared to it.
Part of GUI lockups is because people put work in the same thread as their GUI callbacks. That's why you get applications where if you do s load/save and the gui locks up, etc.. This isn't Microsofts fault it's the fault of the lousy programmer who doesn't know how to write a responsive application.
The second problem is kernel time. E.g. a huge write to disk from a buffer. Linux is somewhat better [2.4.26 running here] but you can still get the occasional hiccup from long I/O syncs.
Tom
Re:Smart cards: Terrible Idea in the US
on
RFID MasterCard
·
· Score: 1
Um, it's still a hassle to call them up, get the charges taken off, etc...
It's also bad for the credit card company which then has to clean up the trouble.
My definition of perfect means I can plug in my new USB scanner and have it work immediately. I can goto a website and see those new SVG images, I can....
All this "functionality" that is second thought to whiny little bitches like you *IS* what gives operating systems their complexity.
Not to mention keeping compability with so many years of prior operating systems. I'm sure if WinXP wouldn't run Win95/98/ME [heck even Win3.11] applications there would be a huge outcry of how evil MSFT is for limiting compability.
So basically people want a bloated featureful operating system and then expect it to be perfect.
Admitedly a lot of the bloat is self-induced by Microsoft subscribing to any new system [.NET, C#, ASP, etc...] just to sell more product. Why people go beyond the win32api in plain C is a mystery to me. GUI coding really ought to account for a minimum of the time not maximum...
Though why we are still using magcards is beyond me.
BTW your signature is full of it. "Freedom of Speech" is from the government not private citizens. I can tell you to shut up as much as I like and there ain't nothing in your charter of rights [Canadians don't have admendments] that says otherwise.
Europeans are smart and use "smart-cards" already. Why are Americans still playing around with new-fangled passive devices which are just not secure?
The reality of the situation is you can't trust the reader. Ever. This is why it's easy to scam debit [get their card no and pin], why it's easy to charge credit cards, etc...
Sure it might cost more per card but the cards would be subject to *less* abuse and you'd have to pay out *less* ultimately in fraud.
I'm not saying the PM is slow I'm just saying compared to an Athlon it's not faster. Well at least at ALU work. I'll bet the PM has a stronger SIMD unit which is where the AMD K8 comes in;-)
Point is a desktop cpu won't be a verbatim port of the PM for long...;-)
From what I've read the Pentium M is nothing more than a souped up P6 core [slightly longer pipeline, 1MB power optimized cache and SSE/SSE2 iirc].
So really it's the PIII v2.0.
Last I checked though the Athlon XP [and Barton] still whoop the PIII's ass and probably the Pentium M at equal clockrates [let's not forget that 2.2Ghz AMD Bartons exist...].
So my guess is if they do ditch the P4 it won't be directly to the Pentium M but instead a design based off it [but with a higher IPC than the P-M] cuz otherwise they'll get screwed in the processor speed wars and that would look awful stupid.
Socket/Super 7 boards.... I recall having an MI, MII and a K6-2 350 in the same board.
Of course that would require both Intel and AMD to sit down and design some Socket1000 board or something. But that gets trickier cuz many of the pin [in Socket478 for instance] are grounds and power. IIRC there are 166 pins dedicated to power management. So the layout of the actual processor would be dictated somewhat by the location of power.
But it would be nice to be able to take out an Intel core and slap in an AMD core in the same motherboard...
"running" you mean full load or idle? 45C is higher than the 43C I got with my "pos" cooler.
Also my 78CFM fan is at the lowest speed [not highest] and it only gets ~25CFM [iirc] at the current speed [of around 2300-2500RPMs].
And why I think newer is better? Because my P4 *actually* runs cooler and it *actually* doesn't shutdown because of heat during full load and it *actually* performs nicely for games, development and building.
I've been a diehard AMD fan for the longest while. But whatever you may *think* about the P4 my P4 box *actually* performs and is cool at very modest cooling conditions.
I'll take by your inability to spell Athlon you're just trolling but...
The P4 may be rated as higher wattage but it certainly doesn't run "hotter". I've had quite a few Athlons [tbird 1200, 1400, 1800+, 2400+, 3000+ and a duron 1.3Ghz for kicks] and all of them have idled hotter than my first P4 [2.80C Northwood].
I think those people who bitch about the P4 heat are
a) running first generation P4's or b) overclocking the device or c) using defective cooling.
Granted I have a later generation P4 [Stepping=9 iirc] but with a 50$ Thermaltake Polo cooler the cpu idles at only a few degrees over ambient (well around 10C or so) and under load gets luke warm to the touch (my guess 25C over ambient).
My Athlon 3000+ with an 80$ radiator Zen cooler idled at 23C over ambient. Certainly I was cooling it correctly, heck I had the sides off my case and a 12" desk fan blowing on the thing...
Honestly I'm either super lucky or just have my box setup correctly because I can't fathom overheating my P4 processor.
1. Intel cpus have better power management. HLT will actually power down a P4 whereas it won't an Athlon [at least not fully and even then you need to disconnect a bus to get it to work].
2. The P4 has a heat spreader. That does help.
3. The P4 is probably made such that the components which generate the most heat are not in the same spot. [speculation...].
4. The P4 has fewer execution units. The Athlon has three very complete ALUs and a powerful FPU. The P4 has two fast ALUs which are incomplete, a slow ALU and a highly pipelined FPU. It's meant for SIMD and trivial ALU ops [compares, add, subtract, move, etc...]
If you own a decent speed Athlon [Barton 3000+ for instance] just put your hand near the cpu while it's churning something like md5crk [or a gentoo build]. You can feel the heat.
Do the same thing with a similarly scaled P4 [for all intents an purposes during my Gentoo build my 2.8Ghz P4 took roughly the same time [if not faster] than my Barton system]].
Sure if you overclock the thing by 800Mhz or something you will notice heat but the chip wasn't designed for that. Heck, try to bump an Athlon up 200Mhz and watch it roast...
I dunno why people bitch about the P4 though, even clocked at 3.2 [from 2.8] my P4 2.80C idles at a whole 10C over ambient. All with a relatively cheap Thermaltake Polo cooler.
Fuck, even with the best heat sink I could find at the time for my Athlon [a Zen Cooler] it idled at 43C [23C over ambient].
Under load the Athlon put out heat you could feel literally pouring out of the cpu. my P4 underload gets slightly warm to touch [the heat sink] but you can't feel the same heat pouring out of it.
By load I mean running the native md5crk core in linux [Gentoo linux]. I ran the md5crk program for a week straight with no heat issues with the cpu fan set to 2000RPM [the lowest it goes].
Maybe the prescott is hot but the Northwood [the 2.80C at least] is very mild.
Considering I routinely shutdown my Barton 3000+ by running various tests [such as the tests I perform on my LibTomMath library] I find the Pentium [while a tad slower at ALU math] a godsend. It's reliable and speedy enough for my liking.
I don't see AMD [or Intel] as having to have higher sales to "achieve" something. The fact that AMD mobo/cpus are prevalent at all [you can get them pretty much anywhere you can get an Intel setup, excluding say Dell;-)] means it's a success.
I mean how many retail stores can you walk into and order a C3 or Efficeon setup? [non-laptop anyways...].
That being said AMD Athlons are still being held back by a core showing some age. A 64-bit cache bus [makes SSE painful], low number of read/write buffers and huge heat factor make the Athlon rather annoying.
The K8 is somewhat improved though but still has the heat factor. [For those curious the K8 has 2 more steps in the instruction scheduler which gives a slightly higher IPC, more directpath that is fast ALU/FPU instructions and a 128-bit cache data path].
Fact of the matter though is that for most desktop uses [editing text, browsing the web, playing mp3s, playing games, watching dvds] both the Athlon and P4 series are equally capable of handling the task and then some. So it's really a matter of price, availability and fit.
In my particular case the P4's very low amount of heat [combined with a Thermaltake Polo cpu cooler] and comparable cost [Barton 3000+ costs 235$, P4 2.80C costs 255$ where I live] made the P4 the "winning" choice for me.;-)
How about they stop cutting horrible corners and follow the specs?
The reason why laptops often don't work 100% in linux is because developers don't follow specs then write windows drivers to cover up missing hardware functionality.
Slashdot Mirror
"If you can't deal with this, don't shop on ebay."
;-)].
Well put. This is exactly why I would never buy anything off ebay. My experience is "if it's too good to be true, it probably is."
At least when I buy something in a store and it doesn't go my way I can confront the store owner directly [usually get exchange/refund at that point
Tom
Not only is there no legal basis for this argument as another poster pointed out. But....
This argument is stupid. Microsoft sells Windows it doesn't give it away. The fact that you run a pirated copy is actually contrary to what Microsoft wanted you to do in the first place.
This argument could apply to Ford for making cars that can be stolen and used to run over school children [cuz lord knows the other type of children are not worth protecting].
Anyways, lay off the crack for a week and go get a job.
Tom
cooperation from vendors doesn't mean what you think it means. Incomplete [late] databooks are not that rare for most firms.
Just because MSFT is huge doesn't mean they can produce perfect work. Several things contribute to the ultimate non-perfectness
1. Moving target. Even while they are writing a version of Windows new hardware is being developed. They have to be able to accomodate late submissions.
2. Not all MSFT employees are developers and not all MSFT developers work on windows.
3. Diminishing returns. Adding more people produces smaller returns on investment.
These characteristics are not unique of MSFT though. The same could be said of say KDE. Not all KDE members are developers, piling on 1000s of developers won't make it 1000x better and KDE doesn't target features from 5 years ago.
lo-and-behold KDE has bugs in it. Shocking!!! The horror!!! OMG!!!
This is why the original posters type of tripe pisses me off. It's so fucking narrow minded.
Tom
It's not good enough to support 2004 hardware though. I want PCI-Express, AMD64 support, latest USB gadgets, etc...
Windows hasn't been feature frozen for 20 years so really your post is misleading.
Tom
How does this get an interesting mod?
That's right up their with pointing out the series of bugs in A-patchy webserver, or the various permission escalations in the lenux Kurnul.
Look, I'm l33t, I point out the flaws of one company to make myself look l33t. When those flaws are actually shared by basically all other software firms out there.
Now I'm not an MS fanboy [run Gentoo] but that doesn't mean I can stand idiots like you. Let's see you try and write an OS that can even *half* compete with Windows and not have any bugs. Then you can sit here and be all mighty about what a company should or should not do.
Tom
"It happens with ALL programs, including Internet Explorer. It's not a problem with one program locking up the GUI, it's a problem with the base system locking up ALL applications momentarily."
No, actually it isn't. How do you think things like dnetc can run 100% cpu yet still display nice graphs and such?
The OS is capable of handling threads just fine. The problem is [including IE] that control logic is placed in the GUI thread.
Basically the way GUIs work in Win32 is you have a callback that handles all messages [e.g. resize, repaint, controls clicked/moved, etc...].
Normally if responding to a message can block you spawn a thread to handle it and return to the callback asap.
Try this demo out [you'll have to write it...]
Make a simple dialog with one button on it "Do It" and two text labels.
Now write two applications both of which have a 100ms timer displayed in one of the text labels.
1st program. When the button is clicked call a function and multiply some variables a million times. Time the process and display the time on the dialog.
2nd program. When the button is clicked spawn a lower priority thread todo the same million multiplications. Time the process and display the time on the dialog.
I'm positive that if you don't have anything else running the 2nd program will display the 100ms timer and still compute the multiplications in roughly the same time as the first program. The first program will lock up the 100ms timer and finish the multiplications faster.
The fact that people [including MSFT] write bad code isn't surprising. Things like the control panel and IE basically do everything in the same thread. That's why loading device manager or any laggy FTP site causes the program to effectively halt.
But if you really don't believe me just write the two applications and test the theory.
Tom
"GCC is an inferior compiler for the x86, whether you like it or not. Intel's optimizing C/C++ compiler is much faster according to numerous benchmarks"
This simply is not true. Like all compiler [or benchmarks] good at one thing is not good at all things.
I found that even with profiling the Intel compiler [v8] is better at things like MP math [libtommath] but the same and worse at other crunching [ciphers/hashes in libtomcrypt].
In fact the Intel C compiler doesn't make much use of SSE2 properly at all [hint: SSE2 is like MMX but on the 128-bit XMM registers] which can be used to speed up 32-bit ALU work [more registers to play with]. The most I've seen is it uses SSE to clear temps in memory.
Not saying that the compiler isn't good. It's fast [to compile] and all. Just saying that GCC isn't that "lost" compared to it.
Tom
Um two fold problem.
Part of GUI lockups is because people put work in the same thread as their GUI callbacks. That's why you get applications where if you do s load/save and the gui locks up, etc.. This isn't Microsofts fault it's the fault of the lousy programmer who doesn't know how to write a responsive application.
The second problem is kernel time. E.g. a huge write to disk from a buffer. Linux is somewhat better [2.4.26 running here] but you can still get the occasional hiccup from long I/O syncs.
Tom
Um, it's still a hassle to call them up, get the charges taken off, etc...
It's also bad for the credit card company which then has to clean up the trouble.
Tom
A real "adult" [female or male] wouldn't do something so idiotically childish.
Kids do stupid things without thinking of the consequences. This person is nothing more than a mere child.
Tom
Independently many functions are bug free. It's their *interaction* that can cause bugs.
Beside if you think MSFT is so bad look up "kernel vulnerabilities" in google. MSFT is not the only company/group that produces code with bugs in it.
Tom
That's your definition of Perfect.
....
My definition of perfect means I can plug in my new USB scanner and have it work immediately. I can goto a website and see those new SVG images, I can
All this "functionality" that is second thought to whiny little bitches like you *IS* what gives operating systems their complexity.
Not to mention keeping compability with so many years of prior operating systems. I'm sure if WinXP wouldn't run Win95/98/ME [heck even Win3.11] applications there would be a huge outcry of how evil MSFT is for limiting compability.
So basically people want a bloated featureful operating system and then expect it to be perfect.
Admitedly a lot of the bloat is self-induced by Microsoft subscribing to any new system [.NET, C#, ASP, etc...] just to sell more product. Why people go beyond the win32api in plain C is a mystery to me. GUI coding really ought to account for a minimum of the time not maximum...
Tom
What services? My parents run XP and haven't had a virus yet [aside from the occasional spyware thingy they install].
I think if you wanted todo your friends a favour you'd tell them about new-fangled NAT boxes that cost all of 100$ and let them network their home.
Tom
Yeah, dude deserves any respect... he's some punk kid that thinks he owns the internet. No better than punks who spew in usenet or spam email.
Tom
Never heard of it.
Though why we are still using magcards is beyond me.
BTW your signature is full of it. "Freedom of Speech" is from the government not private citizens. I can tell you to shut up as much as I like and there ain't nothing in your charter of rights [Canadians don't have admendments] that says otherwise.
Tom
Europeans are smart and use "smart-cards" already. Why are Americans still playing around with new-fangled passive devices which are just not secure?
The reality of the situation is you can't trust the reader. Ever. This is why it's easy to scam debit [get their card no and pin], why it's easy to charge credit cards, etc...
Sure it might cost more per card but the cards would be subject to *less* abuse and you'd have to pay out *less* ultimately in fraud.
Tom
I'm not saying the PM is slow I'm just saying compared to an Athlon it's not faster. Well at least at ALU work. I'll bet the PM has a stronger SIMD unit which is where the AMD K8 comes in ;-)
;-)
Point is a desktop cpu won't be a verbatim port of the PM for long...
Tom
From what I've read the Pentium M is nothing more than a souped up P6 core [slightly longer pipeline, 1MB power optimized cache and SSE/SSE2 iirc].
So really it's the PIII v2.0.
Last I checked though the Athlon XP [and Barton] still whoop the PIII's ass and probably the Pentium M at equal clockrates [let's not forget that 2.2Ghz AMD Bartons exist...].
So my guess is if they do ditch the P4 it won't be directly to the Pentium M but instead a design based off it [but with a higher IPC than the P-M] cuz otherwise they'll get screwed in the processor speed wars and that would look awful stupid.
Tom
Socket/Super 7 boards.... I recall having an MI, MII and a K6-2 350 in the same board.
Of course that would require both Intel and AMD to sit down and design some Socket1000 board or something. But that gets trickier cuz many of the pin [in Socket478 for instance] are grounds and power. IIRC there are 166 pins dedicated to power management. So the layout of the actual processor would be dictated somewhat by the location of power.
But it would be nice to be able to take out an Intel core and slap in an AMD core in the same motherboard...
Tom
"running" you mean full load or idle? 45C is higher than the 43C I got with my "pos" cooler.
Also my 78CFM fan is at the lowest speed [not highest] and it only gets ~25CFM [iirc] at the current speed [of around 2300-2500RPMs].
And why I think newer is better? Because my P4 *actually* runs cooler and it *actually* doesn't shutdown because of heat during full load and it *actually* performs nicely for games, development and building.
I've been a diehard AMD fan for the longest while. But whatever you may *think* about the P4 my P4 box *actually* performs and is cool at very modest cooling conditions.
Tom
Um? How about not.
I'll take by your inability to spell Athlon you're just trolling but...
The P4 may be rated as higher wattage but it certainly doesn't run "hotter". I've had quite a few Athlons [tbird 1200, 1400, 1800+, 2400+, 3000+ and a duron 1.3Ghz for kicks] and all of them have idled hotter than my first P4 [2.80C Northwood].
I think those people who bitch about the P4 heat are
a) running first generation P4's
or
b) overclocking the device
or
c) using defective cooling.
Granted I have a later generation P4 [Stepping=9 iirc] but with a 50$ Thermaltake Polo cooler the cpu idles at only a few degrees over ambient (well around 10C or so) and under load gets luke warm to the touch (my guess 25C over ambient).
My Athlon 3000+ with an 80$ radiator Zen cooler idled at 23C over ambient. Certainly I was cooling it correctly, heck I had the sides off my case and a 12" desk fan blowing on the thing...
Honestly I'm either super lucky or just have my box setup correctly because I can't fathom overheating my P4 processor.
Tom
Well I think it's really for several reasons
1. Intel cpus have better power management. HLT will actually power down a P4 whereas it won't an Athlon [at least not fully and even then you need to disconnect a bus to get it to work].
2. The P4 has a heat spreader. That does help.
3. The P4 is probably made such that the components which generate the most heat are not in the same spot. [speculation...].
4. The P4 has fewer execution units. The Athlon has three very complete ALUs and a powerful FPU. The P4 has two fast ALUs which are incomplete, a slow ALU and a highly pipelined FPU. It's meant for SIMD and trivial ALU ops [compares, add, subtract, move, etc...]
If you own a decent speed Athlon [Barton 3000+ for instance] just put your hand near the cpu while it's churning something like md5crk [or a gentoo build]. You can feel the heat.
Do the same thing with a similarly scaled P4 [for all intents an purposes during my Gentoo build my 2.8Ghz P4 took roughly the same time [if not faster] than my Barton system]].
Sure if you overclock the thing by 800Mhz or something you will notice heat but the chip wasn't designed for that. Heck, try to bump an Athlon up 200Mhz and watch it roast...
Tom
I guess the heatspreader helps.
I dunno why people bitch about the P4 though, even clocked at 3.2 [from 2.8] my P4 2.80C idles at a whole 10C over ambient. All with a relatively cheap Thermaltake Polo cooler.
Fuck, even with the best heat sink I could find at the time for my Athlon [a Zen Cooler] it idled at 43C [23C over ambient].
Under load the Athlon put out heat you could feel literally pouring out of the cpu. my P4 underload gets slightly warm to touch [the heat sink] but you can't feel the same heat pouring out of it.
By load I mean running the native md5crk core in linux [Gentoo linux]. I ran the md5crk program for a week straight with no heat issues with the cpu fan set to 2000RPM [the lowest it goes].
Maybe the prescott is hot but the Northwood [the 2.80C at least] is very mild.
Considering I routinely shutdown my Barton 3000+ by running various tests [such as the tests I perform on my LibTomMath library] I find the Pentium [while a tad slower at ALU math] a godsend. It's reliable and speedy enough for my liking.
Tom
I don't see AMD [or Intel] as having to have higher sales to "achieve" something. The fact that AMD mobo/cpus are prevalent at all [you can get them pretty much anywhere you can get an Intel setup, excluding say Dell ;-)] means it's a success.
;-)
I mean how many retail stores can you walk into and order a C3 or Efficeon setup? [non-laptop anyways...].
That being said AMD Athlons are still being held back by a core showing some age. A 64-bit cache bus [makes SSE painful], low number of read/write buffers and huge heat factor make the Athlon rather annoying.
The K8 is somewhat improved though but still has the heat factor. [For those curious the K8 has 2 more steps in the instruction scheduler which gives a slightly higher IPC, more directpath that is fast ALU/FPU instructions and a 128-bit cache data path].
Fact of the matter though is that for most desktop uses [editing text, browsing the web, playing mp3s, playing games, watching dvds] both the Athlon and P4 series are equally capable of handling the task and then some. So it's really a matter of price, availability and fit.
In my particular case the P4's very low amount of heat [combined with a Thermaltake Polo cpu cooler] and comparable cost [Barton 3000+ costs 235$, P4 2.80C costs 255$ where I live] made the P4 the "winning" choice for me.
Tom
How about they stop cutting horrible corners and follow the specs?
The reason why laptops often don't work 100% in linux is because developers don't follow specs then write windows drivers to cover up missing hardware functionality.
Tom