First off, I've seen a lot of Intel-apologist posts, along the lines of "at least they acknowledged the problem quickly" or "these mistakes happen with bleeding-edge hardware". I find it quite interesting that Intel has stumbled (again) where AMD has had (nearly) flawless product launches for several product cycles. The G5 was also a painless launch.
Another issue with Intel is the "paper" nature of these launches. Try to find 3.6 GHz. parts - they aren't even listed on pricewatch.com. There is only one page for 3.4 GHz. Prescotts or 3.4 GHz. Extremely Expensive Edition. Is Intel having yield problems?
Let's look at some of the last few "issues" with Intel:
Switch from "RDRAM is great" to "no more RDRAM for us"
Switch from "GHz. matters" to "rating scheme"
Switch from "no need for 64 bit x86" to "look at our cool new 64 bit Xeons"
Newest/fastest processors dissipate ridiculous amounts of power
To me, it really looks like AMD has grabbed the "innovation hat" from Intel and is running with it. I hope fewer folks fall for Intel's marketing (the company's main strong point) and start using the better desktop/server technology - AMD64.
The one area that Intel has technology edge is in mobile processors, AMD has some work to do there - although the mobile Athlon64s are great desktop replacement chips.
The problem with having a trusted site list is that sometimes even the trusted sites have problems. In articles talking about this exploit, a few mention how large companies have exploits in their severs that allow hackers to blindly redirect users to Russian websites where various trojans and viruses are installed. Who is to say that my favorite website has not been hijacked without me or the company knowing it?
True, but the idea is to provide more options that make things safer - after all there is no absolute security. Plus, the idea is that each 'trusted' site would get it's own set of privileges, so if Javascript is turned off for a site because it doesn't normally need it a rogue Javascript attack would fail. That at least improves the situation somewhat...
OK, I've read plenty of "just use Mozilla" posts and backpatting here, but IMO we should be thinking about Mozilla/Firefox security as well.
True this particular exploit didn't affect Mozilla/Firefox, but it is certainly possible that something similar might in the future.
So, with that in mind, what new security features would help make Mozilla/Firefox even safer and better?
These come to my mind:
A trusted site list to which I can easily add the current site, and indicate whether it can load images, run scripts and/or download applets.
An option that will pop up a dialog asking for permission if an untrusted site tries to do any of the above.
Some type of "zone" concept similar to IEs so that internal (company) sites can have more privileges than external sites.
Capability of central administration and control (in a business setting) so that users can easily be protected from themselves in a business or large network environment.
Thoughts? Can some or all of this be easily implemented as Firefox extensions?
If Mozilla/Firefox is clearly a better, more secure solution, it will gain marketshare rapidly.
I work at a bank. A lot of the applications used internally are web apps that require IE... Mozilla/Opera aren't an option because those apps require MSJVM (Microsoft Virtual Machine - no joke), Active X or other proprietary MS technology.
Sounds like your IT director has done a horrible job and should be fired.
You would have been much better off implementing that stuff in a browser agnostic, standards compliant way, using Java for any heavy lifting required.
I don't think this is correct. If you turn off "Active Scripting" for the "Internet Zone" you should be invulnerable, AFAIK. Specifically, it is a Javascript exploit.
Due to the design of Dynamic RAM chips, memory bit flip errors are not influenced by how long the memory sits "idle". I emphise idle here because Dynamic ram is never really idle. Each cell in a DRAM chip contains a capacitor and a transistor. If a DRAM cell is left to its own devices, the capacitor soon discarges and the cell looses its state. To stop this from happening, in the background, the RAM controller on the chip is constantly recharging the capacitors. Each cell is read and rewritten about every few milliseconds.
Yes, and if the state is flipped by an outside influence before the read, the new "read and rewritten state" will be incorrect.
This is known to happen randomly from natural radiation (mainly cosmic rays). That is the main reason ECC memory exists (as the grandparent pointed out). You should do a little research before loudly proclaiming your incorrect thesis next time.
But only if you really need that performance. When a 3d game plays at 100 fps in a 2.4GHz CPU, this only means that game players are wasting their money if they buy anything better than an 800MHz CPU, which presumably will get the 30 fps they need to play a perfect game. But for a number-crunching program that needs to do teraflops, there's no substitute for a fast CPU in a well-matched system. Apple and AMD marketroids are doing users a disservice when they claim to have a magic substitute for CPU speed.
You are clearly a troll, but I'll try one last time.
[As an aside, 30 FPS is by no means optimal for gameplay - military flight sims are required to achieve 60 FPS, locked frame rate.]
The grandparent poster was quite correct in pointing to the SPEC results, which are a fair comparison of CPU architectures. They allow the CPU vendor to perform whatever compiler optimizations it desires on the system being tested. The SPEC subtests are all real-world codes that perform useful work.
What your Intel worship ignores is that Intel made very serious tradeoffs in order to achieve those high clockspeeds, primarily increasing the number of pipeline stages to almost 3x the number in the Opteron or G5.
That means on many real-world codes, the P4 underperforms considerably. That is reflected in real world performance, as well as the SPEC scores.
1) The fastest possible CPU, in *true* GHz, not in AMD's inflated "+" bogoghz.
I have mod points, but rather than simply mod you down for this silly statement, I'll reply instead.
AMD's 'inflated "+" bogoghz' are often conservative with regards to P4 performance. They have been a necessary marketing evil, due to lots of people like you who've been brainwashed by Intel's "GHz. matters" campaign. It is extremely telling that Intel is now adopting a "model number" approach, since the clockspeeds of CPUs within Intel's own line clearly demonstrate that clockspeed is a meaningless way to compare chips using different architectures. For instance, a 1.5 GHz. Pentium M might be faster than a 2.6 GHz. P4.
In the same vein, An Athlon 64 3800+ might well be faster (in terms of throughput) than a P4 3.6 GHz. Comparitive performance does vary based on the task performed, so the best way to compare is using tasks similar to your most performance-critical task. If you're talking about gaming, though, you'll find AMD wins most of the benchmarks, at much lower clockspeeds.
Macromedia Flash applications backed by SOAP look very interesting for apps requiring more GUI-like, real-time interaction. This is basically what Java applets were intended for.
Flash is OK for simple apps...for more complex things Java is still a better alternative. Java Web Start and other similar things make web deployment of full-fledged desktop apps easy.
Java is getting to be very good...sadly it's taken about five years longer than originally advertised. However, better late than never!:-)
But then float the idea of taxing corporations so that they'll keep jobs in america, and a lot of those same slashdotters will say it's insightfull; that it's got merit and should be considered. Here's an idea - why don't we just raise taxes on companies that fire anyone, companies that don't make cheap products, companies that smell funny... etc etc. It dosen't seem to bother the slashdotters because it isn't their freedom we're talking about; it's somebody else's.
I'm not in favor of "taxing corporations so that they'll keep jobs in america", but I am concerned that America is losing important capabilities through offshoring.
Do we really want a nation of burger-flippers making fast food for the CEOs who can afford it, while all the interesting technical work is done in Bangladesh and elsewhere? It's my gut feeling that transferring our technical prowess to other countries will, in the end, be very damaging to the U.S. There has to be a better way.
But the idea that nuclear fission was going to provide cheap, clean, limitless power was also stupid, but at one point widely believed.
Stupid? How so? There is a large school of thought that it's still the best option for us right now. It is, very arguably, far safer than any of the sufficient alternatives, which are most certainly killing thousands of people every year (not to mention contributing greenhouse gasses and depleting precious resources).
Nuclear power is also quite cheap, and the mid-ocean subduction zones potentially provide a viable, safe and long-term disposal method.
Modern reactor designs can't melt down or pull a Chernobyl.
Until something better comes along, I'd say nuclear power is the way to go...augmented by clean, renewable sources like solar, wind, biomass and so on.
This is obviously not the right way if you are worried about passwords being found years later on hard disks, as was mentioned in previous slashdot article.
Zeroing and/or encrypting the password buffer is the right solution there, as the article pointed out.
Shouldn't that read "attempted by a large ISP"? Could this result in mass-migration to other services, or are no others viable? As an aside, are cable modems available in Britain?
I do think this is a slippery slope, especially since "pornography" is always hard to define... Are "innocent" shots of (semi)naked teens on Scandinavian beaches "porn", for instance? Who decides?
But having the audio chip separate was a major perk.
Why? I'm (fairly) sure the audio quality is good on the T3, and battery life on my T665C is no great shakes. Of course, I was spoiled by a Visor to start with, with multiple weeks on two AAAs.;-)
The Clie's have great battery life as is. But with the audio chip, it's much more efficient for MP3 playback, then running it off the processor. (always have your screen disabled via the little button on the side).
In the tests I've seen, the current CLIEs don't do significantly better on battery life than the T3, and it is only running a 106 MHz. processor. (Caveat: battery life was no better when the PDA was being used intensively - I think the CLIE did do better when the load was light. I'm not sure about MP3 playback specifically.)
Battery life IMHO is a big part of Clie Life.
None of the current crop of PDAs really shines on battery life - better battery, electronic and display technology would really help!
I chose my T665C over Palm branded hardware for a reason: So much more for the cash.
I also have a T665C, which I like. I got it a while back (over a year) for $279, which was a good deal at the time. It only has a 66 MHz. Dragonball processor, 16 MB RAM, PalmOS 4.1 and a 320x320 (nice as you said) display.
Fast forward to now - I can get a new Tungsten T3, 400 MHz. ARM processor, 64 MB RAM, PalmOS 5, and a 320x480 (very nice!) display, new for $330 or so. I intend to as soon as circumstances permit. (Its also an MP3 player BTW.)
Sony's recent CLIEs have been sub-par in my opinion, and Palm's offerings have improved substantially... I'm not sad to see Sony exit the market.
Up to an estimated 100 times the rated output, in about a second. It takes 30 seconds on that reactor type to do a scram (emergency insertion of control rods). The power spike seems to have been a "prompt criticality" event, driven by the immediate neutrons from fission. Normally reactors keep their chain reactions going only by delayed neutrons that sputter out of fission products seconds to hours after the fission. That's why power reactors are controllable. Prompt criticality is how bombs work.
Yeah, that was quite a power surge - those nuclear reactions are quick compared to chemical...
It should be pointed out that this type of problem cannot physically happen with well-designed reactors. New designs will self-regulate due to negative void coefficient. Here's a relevant page.
chances are that a single catastrophic failure in a nuclear power plant _will_ affect me.
Really? More than the 500+ above-ground nuclear detonations in the 20th century?
Those detonations weren't a good thing for the environment, but they clearly show that some release of radioactive material is quite survivable. On the other hand, you're breathing coal and oil residues right now...nasty stuff really. It's even radioactive...
Personally, I vote for modern nuclear technology as one of the cornerstones of our energy policy going forward, along with solar, wind, hydro and renewable hydrocarbons.
Good comments, for the most part. You might be interested to look at the Scala language, it seems to have some nice higher-level (and functional) features. At least it shows that very interesting things can be built on top of the Java foundation.
My main problem with Java is verbosity, which is simply a function of the veryExplicitVariableNamesAreGood meme. Foolish consistency, perhaps?
For me, personally, I think programming languages should evolve more in the direction of mathematical notation, and extensibility is important. Mathematics is the tool (evolved over centuries) we've created to deal with the most complex concepts ever invented. Two observations about math notation - first it is minimalist, so complex concepts may be grasped with a minimum amount of notation. Second, it is extensible, so that new concepts may be expressed concisely. Interesting how this contrasts with modern programming language design... My feeling is that more power is better, damn the torpedoes...and all that power should ultimately translate to very tight, efficient, realtime-capable code. Interestingly, my current language for most development is Java...sigh.
To address one of your points:
Look at the size of the Java APIs. How many packages are there? How many classes? How many methods? This is making our lives, as programmers, easier... how?
By allowing us to not have to implement the underlying code ourselves, of course.:-)
That said, some of the library (*cough* Java I/O system) is pretty brain-dead. In that sense, though, Java is extensible enough to fix things.
Why do all the OCer's and Gamers always drool over Dual proc boards? There are *very few* SMP capable games at all.
[sarcasm]You fool! You should be ripping MP3s while you're playing like Intel says![/sarcasm].
Seriously, there is no reason that games couldn't take advantage of dual (or more) threads. As another poster pointed out, Doom3 will have the option of sound processing on another thread. AI seems another natural target for multithreading. As the mainstream processors start going dual core (both Intel and AMD have accounced upcoming multicore CPUs) developers will have no choice. Ditto for the new multiprocessor consoles (PS3 etc.).
Right now, though, game developers are obviously the ones who need these boxes most...;-)
AS for cache, I don't know about the 8Mb for the Opteron, but the G5 has 512k of L2 and 64Kb of L1. I guess that with 8Mb it might improve a lot also.
The Opterons have 1 MB (8 Mb) L2 cache where the G5 has.5 MB (4 Mb) L2.
At similar clockspeeds I think the performance is fairly similar, though the Opterons may do better in a dual-CPU configuration since they have on-chip memory controllers and thus more total memory bandwidth.
I'd like to see a head-to-head shootout using top compilers (an often overlooked issue) for both.
Gamma photons are not heat.
The heat radiation is in the infrared. VERY different part of the spectrum there. (X-ray patients don't warm up when they're being examined. They might get cancer though, through the ionization effects though.)
When atomic weapons explode, most of the energy is released in the soft X-ray region. This is simply a consequence of the black body curve and the (extremely high) temperature of a nuke blast.
In the atmosphere (note emphasis) these soft X-rays are quickly absorbed (average free path is something like 9 in. if I remember correctly) and then re-emitted as thermal radiation. That is why there's a "fireball" from a nuclear bomb.
If the bomb were to explode in space, there would be much less thermal effect.
Why do you think nuclear reactors work through heating water instead of X-ray absorption?
Hint: It's not just because it's easier.
No, it's because the reactor materials become hot. Note that in a fission reactor the temperature never reaches several million degrees, where the X-rays would be produced, as they are in a nuclear explosion.
I hope this cleared things up for you. If not, read more at TutorGig.
Read the part about 2/3 of the way down under "Effects of a Nuclear Explosion".
The competitor that comes to my mind is Sharp and their Transmeta Efficeon processor. That notebook is quoted as being 2 pounds and 1GHZ/512MB/20GB/10.4" for $1499.
I don't think there's any comparison on performance...the 2800+ AMD part should completely smoke the Transmeta. Battery life...well I'm not sure but how long are you really planning on running unplugged at a time? I'd guess the AMD parts are intended to run for ~5 hours with moderate use.
AMD outsold Intel in RETAIL desktop sales. Dell is obviously not retail. Here's a better read.
Yes, and lots of corporations and government entities buy Dell as a matter of course. Nonetheless, the fact is that AMD now has better technology in many ways (especially multiway Opteron boxes, which aren't retail either;). AMD is gathering momentum, and Dell would do well to not ignore it...or it will finally start to lose some marketshare over time.
Remember, Dell wasn't always #1...and another entity will be sometime down the road.
It's also quite telling that Intel was forced to adopt the AMD64 instruction set (even if it's calling it something else).;-)
I don't mean to troll, but every time there's a post about some new bleeding edge video card, there's always someone getting modded up to +5, insightful for saying he'd buy it if it weren't due to lack of driver support, and I'm left wondering what the hell for?
Seeing as how none of the other replies mentioned it, one reason is to do cutting-edge OpenGL development under Linux. There is significant interest in doing Linux game development using cross-platform toolkits of various types. One example is Garage Game's Torque engine. Write to that, and get Windows, Mac and Linux support with very little (if any) tweaking. IMO, Linux is the best and most cost-effective platform for game development.
This is why, once again, my next video card purchase will most likely be from NVIDIA. I'll get ATI if I manage a G5...;-) (I wonder how soon the G5s will get these cards?)
Slashdot Mirror
- Switch from "RDRAM is great" to "no more RDRAM for us"
- Switch from "GHz. matters" to "rating scheme"
- Switch from "no need for 64 bit x86" to "look at our cool new 64 bit Xeons"
- Newest/fastest processors dissipate ridiculous amounts of power
To me, it really looks like AMD has grabbed the "innovation hat" from Intel and is running with it. I hope fewer folks fall for Intel's marketing (the company's main strong point) and start using the better desktop/server technology - AMD64.The one area that Intel has technology edge is in mobile processors, AMD has some work to do there - although the mobile Athlon64s are great desktop replacement chips.
True, but the idea is to provide more options that make things safer - after all there is no absolute security. Plus, the idea is that each 'trusted' site would get it's own set of privileges, so if Javascript is turned off for a site because it doesn't normally need it a rogue Javascript attack would fail. That at least improves the situation somewhat...
I do hope that Mozilla never has an image-based vulnerability though... ;-)
True this particular exploit didn't affect Mozilla/Firefox, but it is certainly possible that something similar might in the future.
So, with that in mind, what new security features would help make Mozilla/Firefox even safer and better?
These come to my mind:
- A trusted site list to which I can easily add the current site, and indicate whether it can load images, run scripts and/or download applets.
- An option that will pop up a dialog asking for permission if an untrusted site tries to do any of the above.
- Some type of "zone" concept similar to IEs so that internal (company) sites can have more privileges than external sites.
- Capability of central administration and control (in a business setting) so that users can easily be protected from themselves in a business or large network environment.
Thoughts? Can some or all of this be easily implemented as Firefox extensions?If Mozilla/Firefox is clearly a better, more secure solution, it will gain marketshare rapidly.
Sounds like your IT director has done a horrible job and should be fired.
You would have been much better off implementing that stuff in a browser agnostic, standards compliant way, using Java for any heavy lifting required.
I don't think this is correct. If you turn off "Active Scripting" for the "Internet Zone" you should be invulnerable, AFAIK. Specifically, it is a Javascript exploit.
Check out the CERT advisory.
Yes, and if the state is flipped by an outside influence before the read, the new "read and rewritten state" will be incorrect.
This is known to happen randomly from natural radiation (mainly cosmic rays). That is the main reason ECC memory exists (as the grandparent pointed out). You should do a little research before loudly proclaiming your incorrect thesis next time.
You are clearly a troll, but I'll try one last time.
[As an aside, 30 FPS is by no means optimal for gameplay - military flight sims are required to achieve 60 FPS, locked frame rate.]
The grandparent poster was quite correct in pointing to the SPEC results, which are a fair comparison of CPU architectures. They allow the CPU vendor to perform whatever compiler optimizations it desires on the system being tested. The SPEC subtests are all real-world codes that perform useful work.
What your Intel worship ignores is that Intel made very serious tradeoffs in order to achieve those high clockspeeds, primarily increasing the number of pipeline stages to almost 3x the number in the Opteron or G5.
That means on many real-world codes, the P4 underperforms considerably. That is reflected in real world performance, as well as the SPEC scores.
I hope this cleared things up for you.
I have mod points, but rather than simply mod you down for this silly statement, I'll reply instead.
AMD's 'inflated "+" bogoghz' are often conservative with regards to P4 performance. They have been a necessary marketing evil, due to lots of people like you who've been brainwashed by Intel's "GHz. matters" campaign. It is extremely telling that Intel is now adopting a "model number" approach, since the clockspeeds of CPUs within Intel's own line clearly demonstrate that clockspeed is a meaningless way to compare chips using different architectures. For instance, a 1.5 GHz. Pentium M might be faster than a 2.6 GHz. P4.
In the same vein, An Athlon 64 3800+ might well be faster (in terms of throughput) than a P4 3.6 GHz. Comparitive performance does vary based on the task performed, so the best way to compare is using tasks similar to your most performance-critical task. If you're talking about gaming, though, you'll find AMD wins most of the benchmarks, at much lower clockspeeds.
I hope that cleared things up a bit!
Flash is OK for simple apps...for more complex things Java is still a better alternative. Java Web Start and other similar things make web deployment of full-fledged desktop apps easy.
Java is getting to be very good...sadly it's taken about five years longer than originally advertised. However, better late than never! :-)
I'm not in favor of "taxing corporations so that they'll keep jobs in america", but I am concerned that America is losing important capabilities through offshoring.
Do we really want a nation of burger-flippers making fast food for the CEOs who can afford it, while all the interesting technical work is done in Bangladesh and elsewhere? It's my gut feeling that transferring our technical prowess to other countries will, in the end, be very damaging to the U.S. There has to be a better way.
Stupid? How so? There is a large school of thought that it's still the best option for us right now. It is, very arguably, far safer than any of the sufficient alternatives, which are most certainly killing thousands of people every year (not to mention contributing greenhouse gasses and depleting precious resources).
Nuclear power is also quite cheap, and the mid-ocean subduction zones potentially provide a viable, safe and long-term disposal method.
Modern reactor designs can't melt down or pull a Chernobyl.
Until something better comes along, I'd say nuclear power is the way to go...augmented by clean, renewable sources like solar, wind, biomass and so on.
Zeroing and/or encrypting the password buffer is the right solution there, as the article pointed out.
Shouldn't that read "attempted by a large ISP"? Could this result in mass-migration to other services, or are no others viable? As an aside, are cable modems available in Britain?
I do think this is a slippery slope, especially since "pornography" is always hard to define... Are "innocent" shots of (semi)naked teens on Scandinavian beaches "porn", for instance? Who decides?
Why? I'm (fairly) sure the audio quality is good on the T3, and battery life on my T665C is no great shakes. Of course, I was spoiled by a Visor to start with, with multiple weeks on two AAAs. ;-)
The Clie's have great battery life as is. But with the audio chip, it's much more efficient for MP3 playback, then running it off the processor. (always have your screen disabled via the little button on the side).
In the tests I've seen, the current CLIEs don't do significantly better on battery life than the T3, and it is only running a 106 MHz. processor. (Caveat: battery life was no better when the PDA was being used intensively - I think the CLIE did do better when the load was light. I'm not sure about MP3 playback specifically.)
Battery life IMHO is a big part of Clie Life.
None of the current crop of PDAs really shines on battery life - better battery, electronic and display technology would really help!
Sorry 'bout the late reply, hope you catch it!
I also have a T665C, which I like. I got it a while back (over a year) for $279, which was a good deal at the time. It only has a 66 MHz. Dragonball processor, 16 MB RAM, PalmOS 4.1 and a 320x320 (nice as you said) display.
Fast forward to now - I can get a new Tungsten T3, 400 MHz. ARM processor, 64 MB RAM, PalmOS 5, and a 320x480 (very nice!) display, new for $330 or so. I intend to as soon as circumstances permit. (Its also an MP3 player BTW.)
Sony's recent CLIEs have been sub-par in my opinion, and Palm's offerings have improved substantially... I'm not sad to see Sony exit the market.
Yeah, that was quite a power surge - those nuclear reactions are quick compared to chemical...
It should be pointed out that this type of problem cannot physically happen with well-designed reactors. New designs will self-regulate due to negative void coefficient. Here's a relevant page.
Really? More than the 500+ above-ground nuclear detonations in the 20th century?
Those detonations weren't a good thing for the environment, but they clearly show that some release of radioactive material is quite survivable. On the other hand, you're breathing coal and oil residues right now...nasty stuff really. It's even radioactive...
Personally, I vote for modern nuclear technology as one of the cornerstones of our energy policy going forward, along with solar, wind, hydro and renewable hydrocarbons.
My main problem with Java is verbosity, which is simply a function of the veryExplicitVariableNamesAreGood meme. Foolish consistency, perhaps?
For me, personally, I think programming languages should evolve more in the direction of mathematical notation, and extensibility is important. Mathematics is the tool (evolved over centuries) we've created to deal with the most complex concepts ever invented. Two observations about math notation - first it is minimalist, so complex concepts may be grasped with a minimum amount of notation. Second, it is extensible, so that new concepts may be expressed concisely. Interesting how this contrasts with modern programming language design... My feeling is that more power is better, damn the torpedoes...and all that power should ultimately translate to very tight, efficient, realtime-capable code. Interestingly, my current language for most development is Java...sigh.
To address one of your points:
Look at the size of the Java APIs. How many packages are there? How many classes? How many methods? This is making our lives, as programmers, easier... how?
By allowing us to not have to implement the underlying code ourselves, of course. :-)
That said, some of the library (*cough* Java I/O system) is pretty brain-dead. In that sense, though, Java is extensible enough to fix things.
[sarcasm]You fool! You should be ripping MP3s while you're playing like Intel says![/sarcasm].
Seriously, there is no reason that games couldn't take advantage of dual (or more) threads. As another poster pointed out, Doom3 will have the option of sound processing on another thread. AI seems another natural target for multithreading. As the mainstream processors start going dual core (both Intel and AMD have accounced upcoming multicore CPUs) developers will have no choice. Ditto for the new multiprocessor consoles (PS3 etc.).
Right now, though, game developers are obviously the ones who need these boxes most... ;-)
The Opterons have 1 MB (8 Mb) L2 cache where the G5 has .5 MB (4 Mb) L2.
At similar clockspeeds I think the performance is fairly similar, though the Opterons may do better in a dual-CPU configuration since they have on-chip memory controllers and thus more total memory bandwidth.
I'd like to see a head-to-head shootout using top compilers (an often overlooked issue) for both.
When atomic weapons explode, most of the energy is released in the soft X-ray region. This is simply a consequence of the black body curve and the (extremely high) temperature of a nuke blast.
In the atmosphere (note emphasis) these soft X-rays are quickly absorbed (average free path is something like 9 in. if I remember correctly) and then re-emitted as thermal radiation. That is why there's a "fireball" from a nuclear bomb.
If the bomb were to explode in space, there would be much less thermal effect.
Why do you think nuclear reactors work through heating water instead of X-ray absorption? Hint: It's not just because it's easier.
No, it's because the reactor materials become hot. Note that in a fission reactor the temperature never reaches several million degrees, where the X-rays would be produced, as they are in a nuclear explosion.
I hope this cleared things up for you. If not, read more at TutorGig. Read the part about 2/3 of the way down under "Effects of a Nuclear Explosion".
Hrm, if I was going to spend $4000(!) on a laptop, I think I'd get a 17" Powerbook (Radeon 9700, yum!) along with some accessories and RAM expansion.
YMMV. :-)
I don't think there's any comparison on performance...the 2800+ AMD part should completely smoke the Transmeta. Battery life...well I'm not sure but how long are you really planning on running unplugged at a time? I'd guess the AMD parts are intended to run for ~5 hours with moderate use.
Anyone have any benchmarks on the Efficeon?
Yes, and lots of corporations and government entities buy Dell as a matter of course. Nonetheless, the fact is that AMD now has better technology in many ways (especially multiway Opteron boxes, which aren't retail either;). AMD is gathering momentum, and Dell would do well to not ignore it...or it will finally start to lose some marketshare over time.
Remember, Dell wasn't always #1...and another entity will be sometime down the road.
It's also quite telling that Intel was forced to adopt the AMD64 instruction set (even if it's calling it something else). ;-)
Seeing as how none of the other replies mentioned it, one reason is to do cutting-edge OpenGL development under Linux. There is significant interest in doing Linux game development using cross-platform toolkits of various types. One example is Garage Game's Torque engine. Write to that, and get Windows, Mac and Linux support with very little (if any) tweaking. IMO, Linux is the best and most cost-effective platform for game development.
This is why, once again, my next video card purchase will most likely be from NVIDIA. I'll get ATI if I manage a G5... ;-) (I wonder how soon the G5s will get these cards?)