No, it would not explode. What is so difficult about fusion is to get the process run at all. If you interrupt the process with some external accident it will die immediately. The reactor only contains fuel for a few seconds (and is constantantly refueled). That also means that little harmful material will be released, even in the worst possible case.
If you build it smaller you can increase the magnetic density thus increasing the plasma pressure, which is exactly what is needed. The drawback is that the reactor can only run in shorter "shots".
I listen to music, and normally I buy about one full price CD a month. I never download MP3s, and I have just a few CD copies.
If I can not be sure that a CD I buy works as it is supposed to (that is, being playable at every CD player I will ever own) I will not buy it!
The reason I use CDs is that I find it the most convenient way to listen to music, and to store music, right now. I know that when I no longer wants to listen to my CDs, I can convert them to any other format I like.
If I no longer will be able to convert the CDs, I probably wont buy them in the first place!
If CDs wont be playable at all in an ordinary CD player, ripping and copying music will suddenly be worth the effort even more than it is today.
The music industry just tells everybody that it is doomed by proposing these rotten changes to CDs.
And by the way, I have bought quite few CDs this year - not because I have copied them, but because little good music has been produced this year.
If I have understood things right, this Xeon@2.2 is essentially a Pentium 4. And it is supposed to be used in servers (as in fileserver, sqlserver, mailserver, webserver).
The P4 has proven to be really good executing SSE2 float instructions - is a server likely to ever do a single SSE2 float instruction? Any floating point operations at all? The P4 has also proven to have a really weak integer part.
What server applications use floating point operations at all? Mail, WWW, sql, php, file, asp, cvs, print, etc etc, I believe these applications rarely use any floating point operations. Also, I believe they these server applications primarily use integer operations and logical operations (read lots of branches and a 20-stage pipeline).
Are Intel targeting number crunching servers for scientific use? Then the primary issues are bus bandwidth and cache-size. 2.2 Ghz wont impress that much if it does not work with an SGI compiler.
Are Intel really going to put P4s in conventional servers? Will they succeed with this fantastic bluff?
Please, inform me what essential thing I have missed!
Seems like when (if) Berlin is finally out there Aqua and Cocoa will look like some simplified Berlin-light. Berlin obviously will look even cooler than Aqua but support more programming languages, network support etc.
But Aqua/Cocoa may well be successful because it is quite simple and uniform, always primarily focusing on usability, functionality and user experience. To be really successful Berlin must both be simple to use and configure, yet support powerful functions.
Did anyone find what hardware is required to test it on an i386-debain machine?
You are of course right - there will be a day when all computers are at least 64 bits. What I wanted to point out is that 64 bits are not just simply twice as good as 32 bits, just because it is twice as many bits. And especially, for "ordinary" PC-work (multimedia, games, image handling etc) there is no immediate gain. Performance in FLOPS or FPS are not automatically boasted because you have a 64 bit cpu.
Governments with lack of money have tried to print more money. This works pretty bad and have in famous cases led to hyper inflation (you all know about 1000 000 000 DM for a piece of bread).
Will this happen here? Will Intel and Motorola reply? Can we look forward to the Athlon 90 000, PIV 7.7T etc?
Leaving SMB would mean W95/98/ME and even NT4.0/5.0 would not be able to share files with the newer MS OSes. As long as these old Microsoft OS can communicate with the next MS OS, so will Samba. I doubt MS would break file sharing between NT4.0/5.0 and future OSes.
They threw out NTLM, in came Kerberos
They threw out WINS, in came DDNS
The question whether the decimals of pi are random has more to do with our definition of random than pi itself. Of course we all agree that pi remains the same forever. On the other hand, if you want series of random integers, why not start at some decimal in pi?
So, are the decimals random? The answer is not more interesting than that to the question whether a number can be "probable prime" (either it is prime, or it is not, so how could it probably be prime, the probaility is definately 0 or 1).
I had debian/testing running on my G4 a few weeks ago (but now I run OS X). I had ethernet, 1600x1200@85Hz, sound, 3-button scrollmouse and a fully working Swedish keyboard up and running... and I am no debian Guru.
Do you have any hardware problems, or have you not tried? I had problems with the modem in Yellowdog, and it seems like the ethernet driver dont like 10Mbit...
Is there any sane reason why I should put a CPU optimized for multimedia and with a lousy integer part in a file/mail/print/backup-server or a W2k Domain Controller? I guess not. How soon will Dell put 2 of these CPUs in there general purpose servers? I guess quite soon...
It would be more fair to compare UltraSparcs, Alphas and Itaniums when they start rolling out.
The Itanium will be at least a generation ahead of Alphas and UltraSparcs when they are released so I dont really agree it would be more fair, but it would look better for Intel. And it will be very interesting to see what the Itanium can perform.
Alright, it is of course no problem to come up with thousands of applications where unlimited CPU would be nice. But still:
Graphics can easily be handled in the Graphics card. The framerate in Q3 or whatever is not really there thanks to the CPU, its there thanks to the graphics card (I dont play games).
...and when it comes to most real applications (scientific work) memory is usually the showstopper. If the program does not fit in the cache it often wouldn't help even if you could do floating point operations in zero clockcycles - the cpu would just be waiting for memory all the time. And RAMBUS/Intel, memory latency is the main problem, not memory bandwidth.
If Intel put 32Mb full-speed L2 cache on their Xeon, then its high clockfrequency would be _really_ useful.
...and of course you can find lots of more or less useful applications for fast cpus with just 128kb of cache...
It is of course impressive that Intel made it run at 1.7GHz, but given this, the benchmarks are quite what could be expected. The Xeon is intended for High-End workstations and servers. I dont think many people will actually choose a system, just becaues it is Xeon-based (the OEM has made the choice).
When buying servers the CPUs are not really the most important thing (at least now where I work - we focus on disk/RAID/memory and takes whatever CPU comes with it, and maybe double it). If CPU is really important (massive database, scientific applications etc) 64-bit CPUs or vector machines might be more adequate even though the operate at lower clock freq.
Most Xeons will be in Windows (and of course, Linux/BSD) servers. On these servers PIII/Thunderbird is not really an option anyway.
So, what I have wanted to say all the time is that I'd rather see a benchmark/comparision between UltraSparcs, Alphas and Xeons.
Hmm, I dont really follow you. You say the major work was made by two persons at Stanford University, who did not patent RSA themselves. Then I think it sounds like we have Diffie/Hellman to thank for RSA rather than IP or Shannon & CO.
You do not claim that RSA was not invented thanks to intellectual property. Was it invented because it could be patented? If it was then MS can say that one of the most significant discoveries for cryptography was made thanks to IP, and without IP maybe we would not have RSA today - and that would be worse than living with the patent for years (I think).
Of course, I agree with everything our wrote, I just wanted to point this out. I GUESS intellectual property had nothing to do with the invention of RSA. Am I wrong?
Wouldnt it be better if Intel focused on improving performance without adding more instructions all the time. If they hadnt added more instructions to the P4 (for example) it would not have performed so badly in benchmarks (that are based on old code).
In the end, most applications and OSes will still be built on 386 or Pentium compatible code (with the possible exception of Multimedia applications - which by the way are the only ones that requires performance;)
NT/W2k are hardly secure without the service packs applied. On the other hand, they are quite secure "out of the box" with the service packs applied. The problem is that if people dont apply the service packs when released and production systems are not patched regularely then the system wont be secure anyway. This is of course true for most OSs, both UNIX dialects and NT/w2k.
Its no really big deal to secure most relevant OSs, if security is the only thing we are looking for. Unfortunately it is quite common that service packs are not applied to production MS systems because the SP may break something, because it requires a reboot and must be scheduled etc.
How many servers are actually reinstalled regularly to obtain "out of the box" security?
The important thing is not really how secure a system is out of the box. The important thing is how secure can the system really be expected to be after several upgrades and reconfigurations.
So, Microsofts actions against Netscape might be justified by the fact that Netscape actually threatened MS. Microsoft actions are not justified by the fact that Netscape did not give appropriate credit to the people behind the technologies you mention.
If Netscape did something bad to Spyglass, that does not automatically justify MS to do what they want to Netscape. If I hit you, anyone may not hit me, especially not if it is not in your defence. IANAL
What about dual Duron? Do Durons support this, and does the MoBo support it? I remember the rush for dual Celerons and Intels lies about the possibility of using dual Celeron.
Slashdot Mirror
No, it would not explode. What is so difficult about fusion is to get the process run at all. If you interrupt the process with some external accident it will die immediately. The reactor only contains fuel for a few seconds (and is constantantly refueled). That also means that little harmful material will be released, even in the worst possible case.
If you build it smaller you can increase the magnetic density thus increasing the plasma pressure, which is exactly what is needed. The drawback is that the reactor can only run in shorter "shots".
I listen to music, and normally I buy about one full price CD a month. I never download MP3s, and I have just a few CD copies.
If I can not be sure that a CD I buy works as it is supposed to (that is, being playable at every CD player I will ever own) I will not buy it!
The reason I use CDs is that I find it the most convenient way to listen to music, and to store music, right now. I know that when I no longer wants to listen to my CDs, I can convert them to any other format I like.
If I no longer will be able to convert the CDs, I probably wont buy them in the first place!
If CDs wont be playable at all in an ordinary CD player, ripping and copying music will suddenly be worth the effort even more than it is today.
The music industry just tells everybody that it is doomed by proposing these rotten changes to CDs.
And by the way, I have bought quite few CDs this year - not because I have copied them, but because little good music has been produced this year.
If I have understood things right, this Xeon@2.2 is essentially a Pentium 4. And it is supposed to be used in servers (as in fileserver, sqlserver, mailserver, webserver).
The P4 has proven to be really good executing SSE2 float instructions - is a server likely to ever do a single SSE2 float instruction? Any floating point operations at all? The P4 has also proven to have a really weak integer part.
What server applications use floating point operations at all? Mail, WWW, sql, php, file, asp, cvs, print, etc etc, I believe these applications rarely use any floating point operations. Also, I believe they these server applications primarily use integer operations and logical operations (read lots of branches and a 20-stage pipeline).
Are Intel targeting number crunching servers for scientific use? Then the primary issues are bus bandwidth and cache-size. 2.2 Ghz wont impress that much if it does not work with an SGI compiler.
Are Intel really going to put P4s in conventional servers? Will they succeed with this fantastic bluff?
Please, inform me what essential thing I have missed!
Seems like when (if) Berlin is finally out there Aqua and Cocoa will look like some simplified Berlin-light. Berlin obviously will look even cooler than Aqua but support more programming languages, network support etc.
But Aqua/Cocoa may well be successful because it is quite simple and uniform, always primarily focusing on usability, functionality and user experience. To be really successful Berlin must both be simple to use and configure, yet support powerful functions.
Did anyone find what hardware is required to test it on an i386-debain machine?
You are of course right - there will be a day when all computers are at least 64 bits. What I wanted to point out is that 64 bits are not just simply twice as good as 32 bits, just because it is twice as many bits. And especially, for "ordinary" PC-work (multimedia, games, image handling etc) there is no immediate gain. Performance in FLOPS or FPS are not automatically boasted because you have a 64 bit cpu.
So, what is the advantages of a 64-bit architecture?
It can address much more than 4Gb of memory (most people dont need this)
It can use larger integers (who the hell factorizes primes anyway)
But on the other hand, all pointers take up double space...
When will we start hearing about the 64-bit-myth?
I fail to see how designing an entire CPU with the primary objective that it should run at a high clockspeed helps anybody at all.
Governments with lack of money have tried to print more money. This works pretty bad and have in famous cases led to hyper inflation (you all know about 1000 000 000 DM for a piece of bread).
Will this happen here? Will Intel and Motorola reply? Can we look forward to the Athlon 90 000, PIV 7.7T etc?
Leaving SMB would mean W95/98/ME and even NT4.0/5.0 would not be able to share files with the newer MS OSes. As long as these old Microsoft OS can communicate with the next MS OS, so will Samba. I doubt MS would break file sharing between NT4.0/5.0 and future OSes.
They threw out NTLM, in came Kerberos
They threw out WINS, in came DDNS
I wouldn't worry too much yet...
Mac OS X 10.1 is supposed to share files on MS networks (using SMB).
The question whether the decimals of pi are random has more to do with our definition of random than pi itself. Of course we all agree that pi remains the same forever. On the other hand, if you want series of random integers, why not start at some decimal in pi?
So, are the decimals random? The answer is not more interesting than that to the question whether a number can be "probable prime" (either it is prime, or it is not, so how could it probably be prime, the probaility is definately 0 or 1).
I had debian/testing running on my G4 a few weeks ago (but now I run OS X). I had ethernet, 1600x1200@85Hz, sound, 3-button scrollmouse and a fully working Swedish keyboard up and running... and I am no debian Guru.
Do you have any hardware problems, or have you not tried? I had problems with the modem in Yellowdog, and it seems like the ethernet driver dont like 10Mbit...
Is there any sane reason why I should put a CPU optimized for multimedia and with a lousy integer part in a file/mail/print/backup-server or a W2k Domain Controller? I guess not. How soon will Dell put 2 of these CPUs in there general purpose servers? I guess quite soon...
The Itanium will be at least a generation ahead of Alphas and UltraSparcs when they are released so I dont really agree it would be more fair, but it would look better for Intel. And it will be very interesting to see what the Itanium can perform.
Alright, it is of course no problem to come up with thousands of applications where unlimited CPU would be nice. But still:
Graphics can easily be handled in the Graphics card. The framerate in Q3 or whatever is not really there thanks to the CPU, its there thanks to the graphics card (I dont play games).
...and when it comes to most real applications (scientific work) memory is usually the showstopper. If the program does not fit in the cache it often wouldn't help even if you could do floating point operations in zero clockcycles - the cpu would just be waiting for memory all the time. And RAMBUS/Intel, memory latency is the main problem, not memory bandwidth.
If Intel put 32Mb full-speed L2 cache on their Xeon, then its high clockfrequency would be _really_ useful.
...and of course you can find lots of more or less useful applications for fast cpus with just 128kb of cache...
It is of course impressive that Intel made it run at 1.7GHz, but given this, the benchmarks are quite what could be expected. The Xeon is intended for High-End workstations and servers. I dont think many people will actually choose a system, just becaues it is Xeon-based (the OEM has made the choice).
When buying servers the CPUs are not really the most important thing (at least now where I work - we focus on disk/RAID/memory and takes whatever CPU comes with it, and maybe double it). If CPU is really important (massive database, scientific applications etc) 64-bit CPUs or vector machines might be more adequate even though the operate at lower clock freq.
Most Xeons will be in Windows (and of course, Linux/BSD) servers. On these servers PIII/Thunderbird is not really an option anyway.
So, what I have wanted to say all the time is that I'd rather see a benchmark/comparision between UltraSparcs, Alphas and Xeons.
The advantage with LCD is that it takes up 2D space on your desktop. 3D-LCD must be a step in the wrong direction.
Hmm, I dont really follow you. You say the major work was made by two persons at Stanford University, who did not patent RSA themselves. Then I think it sounds like we have Diffie/Hellman to thank for RSA rather than IP or Shannon & CO.
You do not claim that RSA was not invented thanks to intellectual property. Was it invented because it could be patented? If it was then MS can say that one of the most significant discoveries for cryptography was made thanks to IP, and without IP maybe we would not have RSA today - and that would be worse than living with the patent for years (I think).
Of course, I agree with everything our wrote, I just wanted to point this out. I GUESS intellectual property had nothing to do with the invention of RSA. Am I wrong?
Wouldnt it be better if Intel focused on improving performance without adding more instructions all the time. If they hadnt added more instructions to the P4 (for example) it would not have performed so badly in benchmarks (that are based on old code).
;)
In the end, most applications and OSes will still be built on 386 or Pentium compatible code (with the possible exception of Multimedia applications - which by the way are the only ones that requires performance
NT/W2k are hardly secure without the service packs applied. On the other hand, they are quite secure "out of the box" with the service packs applied. The problem is that if people dont apply the service packs when released and production systems are not patched regularely then the system wont be secure anyway. This is of course true for most OSs, both UNIX dialects and NT/w2k.
Its no really big deal to secure most relevant OSs, if security is the only thing we are looking for. Unfortunately it is quite common that service packs are not applied to production MS systems because the SP may break something, because it requires a reboot and must be scheduled etc.
How many servers are actually reinstalled regularly to obtain "out of the box" security?
The important thing is not really how secure a system is out of the box. The important thing is how secure can the system really be expected to be after several upgrades and reconfigurations.
So, Microsofts actions against Netscape might be justified by the fact that Netscape actually threatened MS. Microsoft actions are not justified by the fact that Netscape did not give appropriate credit to the people behind the technologies you mention.
Good post! Its a pity you forgot your Nick...
If Netscape did something bad to Spyglass, that does not automatically justify MS to do what they want to Netscape. If I hit you, anyone may not hit me, especially not if it is not in your defence. IANAL
What about dual Duron? Do Durons support this, and does the MoBo support it? I remember the rush for dual Celerons and Intels lies about the possibility of using dual Celeron.